The Flat Earth Society
Other Discussion Boards => Technology & Information => Topic started by: Dr David Thork on April 25, 2018, 03:51:21 PM
-
So wanted a new computer. Actually I wanted a 4k monitor, and that was £400-£600 so thought to hell with it, I'll get an AIO.
Opted for a Dell Inspiron 27 7775.
Ryzen 1400
RX 580
8 GB DDR4 2400 Ram
128 Gb SSD
1TB HDD
Wireless card
Wondows 10
27" UHD (3840x2160) display
Shitty mouse and keyboard that I won't use.
(https://images.anandtech.com/doci/11455/dell_inspiron_7000_1.jpg)
(Click for big)
Total cost: £714.
The screen and GPU are worth about that on their own so pretty happy with the deal. I'll use it mostly for light work (Excel, web design, browsing, watching youtube), I might buy some games but 1080p is enough for me and I prefer less demanding games like e-sports titles and Tekken ... that kind of thing. This is more than enough machine for me ... it needs a name ... I'm not sure about that yet.
-
I should probably add, this is the first PC I haven't built myself for about 20 years.
This is mostly down to economics. GPU prices, RAM prices etc ... they are stupid right now and so it isn't possible to build a PC for anywhere near the price that the large retailers can. My top tip if you want to build a PC is become your own breaker.
Go to an outlet store and get a refurbished or scratch and dent type PC. Pick one with all the specs you want. It'll cost about half. Then use a bit of the cash you have left over to buy a new case, some RGB, cooling, whatever and then transplant the gubbings across. You'll save hundreds of dollars and you know all the parts will work first time as it has already been tested.
http://outlet.us.dell.com/ArbOnlineSales/Online/InventorySearch.aspx?c=us&cs=22&l=en&s=dfh&brandid=2202&sign=PXhcOSHtr1T4IOw%2fPR7UdfVNBAjJ6V7ceOFKXGOeFVYVYXOpRltf2KfjtVs4o4kpzLB4ISiTmrVBJpYNN9HbX%2fUkwG6tmx7LbVACedRxOcQBfXyL5PhNKBQRGJQqSn3S4AXazoZddFjRdxqTI%2bjXd87XIQeEqCzpQqNuocuowKVA6wJVYxFCoPgzY5ImBPc6mpWi6bzRYKDEmyRgwtcUcR9vzvTDnzQ1S%2b9i6u4yp9pAUje2eYJS4w%3d%3d
Refurbs with latest gen parts tend to be pretty much as new (guaranteed no scratches in the screen, all parts work, is clean, etc) and most likely just something someone opened and sent straight back if it has new components. Which is what I opted for here. Comes with a warranty from Dell as well. The PC I bought is £1,399 as new, so yeah, around half the cost.
-
Is that one of those "All in One" things?
-
Yeah, its an iMac for people who hate Apple.
-
Thork, I hate to tell you this but Wondows 10 is not a real operating system. You've been scammed.
-
I'd rather have Wondows than Lonix.
-
tfw it's 2018 so you buy a computer with a 4K monitor and 8 GiB of RAM
What's next on the list, an ethanol-powered Ferrari?
-
I don't really need more than 8Gb. The only time I would might be a game ... but the graphics card is 8gb VRAM. If I had a GTX 1060 3GB, the RAM may be an issue, but not with that card.
Also, this AIO is a little unusual in that it is easy to upgrade in the future. I can add RAM (its not soldered), change the hard drives, change out the CPU for any 65W AM4 thing in the future ... the only thing I can't upgrade is the graphics chip ... and this one will be fine for me for years. I don't game much.
https://youtu.be/PmYmneeET9M?t=9s
-
I don't really need more than 8Gb. The only time I would might be a game ... but the graphics card is 8gb VRAM. If I had a GTX 1060 3GB, the RAM may be an issue, but not with that card.
How are you always so terrible at understanding everything
-
Ssooo...
Super thin
High powered.
Lots of ram.
Not enough vents.
Something tells me this is gonna either be super under powered due to heat issues or it'll have issues sooner than others due to constant high heat.
But yes, Parsifal is right. You're getting a 4K tv to go with your 2005 DVD player.
-
Oh for fuck's sake, not another Thork thread where he says stupid things about technology and tricks everyone into responding for weeks.
Let the man spend his money however he wants. He's clearly not interested in actual technical expertise and just wants to feel good about the product he's bought. Let him have it. I'm sure it will work well enough for his needs - so what if he could have got something better at no extra cost?
Thork, I suggest you name the machine Denny, in the honour of my favourite character from The Room.
-
it needs a name ... I'm not sure about that yet.
My computers are called "You F******g C***s" because the only time I talk to them is when they are not working properly
-
Let the man spend his money however he wants. He's clearly not interested in actual technical expertise and just wants to feel good about the product he's bought. Let him have it. I'm sure it will work well enough for his needs - so what if he could have got something better at no extra cost?
Like what?
Spent £714. What could I have got that was better for no extra cost bearing in mind I wanted a new 4k screen into the bargain.
Ssooo...
Super thin
High powered.
Lots of ram.
Not enough vents.
Something tells me this is gonna either be super under powered due to heat issues or it'll have issues sooner than others due to constant high heat.
Well it isn't underpowered. It'll beat out any iMac with the exception of the iMac Pro that starts at £5000 and has a disease for an operating system.
But yes, Parsifal is right. You're getting a 4K tv to go with your 2005 DVD player.
I don't understand this analogy at all? This is a mid range PC to run a 4k screen. I don't see the problem here. I can watch 4 vids on it and edit 4k images which I need to do for work ... and it is a competent computer as well, to do all the stuff I need.
-
Like what?
Spent £714. What could I have got that was better for no extra cost bearing in mind I wanted a new 4k screen into the bargain.
No, I'm not gonna fall for that again. You've already made your purchase, there's no point in arguing, and I'm sure you'll enjoy it greatly.
-
???
I don't want to argue. I'm just interested.
-
Ssooo...
Super thin
High powered.
Lots of ram.
Not enough vents.
Something tells me this is gonna either be super under powered due to heat issues or it'll have issues sooner than others due to constant high heat.
Well it isn't underpowered. It'll beat out any iMac with the exception of the iMac Pro that starts at £5000 and has a disease for an operating system.
So the specs would have you believe. But I'm having difficulty wrapping my head around the heat issue. I mean, how DO they solve the tight space and large heat generation that comes with more power? Very powerful fan? Super efficient processors? I don't know but I remember you loving very quiet computers and I'm curious if you'll find this to be as quiet as you need.
But yes, Parsifal is right. You're getting a 4K tv to go with your 2005 DVD player.
I don't understand this analogy at all? This is a mid range PC to run a 4k screen. I don't see the problem here. I can watch 4 vids on it and edit 4k images which I need to do for work ... and it is a competent computer as well, to do all the stuff I need.
Ok, if you need to edit 4k images for work then that's another story. But from what you posted earlier, it sounded like you bought a VR ready 4k computer but had no intention of using anything that would run at 4k. ie. a 4k TV to run a DVD whose resolution is 640x480 or even require that much raw power.
-
Oh, yeah no 4k was a must as I need to test websites in 4k these days and have retina ready images. I can always downscale to 1080p to see what that looks like too, but I can't upscale my current monitor. 4k is the reason I had to purchase. I could have just got a monitor, but it wasn't a huge amount more for an entire PC and my little NUC is getting long in the tooth with its dual core CPU.
Heat.
Yeah, I do like my computer to stfu whilst I'm working. This thing has heat pipes. Whilst working, it should be relatively quiet because I won't be asking much of the graphics card. When I play a game, the fan will ramp up, but I won't be able to hear that over the sound of all that victory.
The processor is 65W which should be fine. I won't overclock it and have little need to push it to 100% during the day. I use a 1.8 boosting to 2.9 dual core i5 right now and its fine. The graphics card is the laptop version ... ie it runs 110W instead of 180W TDP. This will probably amount to about a 25% hit on graphics, but I don't have extreme requirements. I'm moving up from intel 540 HD graphics after all. It should be a huge improvement. It'll still be better than a 1050ti would perform and that would have been my choice for a desktop build based on my requirements.
Small is always a premium. But I built a shuttle when they first came out, I have a NUC now, I had 2 other small form factor PCs as well. I haven't had a tower in over 10 years. I just don't need workstation horsepower and put my focus onto things that don't take up a lot of real estate. This same machine can be specced with a 1700 8 core Ryzen as well (twice as many cores and higher base clock), so it wont be the hottest version of this machine they sell.
-
Eh, heat pipes usually means it's got one or two central fans that blast air out over the fins. Alot of laptops use this since they want heat to run out of the side and not the bottom. (since bottom means more heat buildup or a hot lap)
Still, hope you're satisfied with it once you get it.
-
Is basically a big laptop without a keyboard. I'm sure it'll be fine.
-
I think reading this conversation between Dave and Thork actually killed some portion of my brain.
-
I think reading this conversation between Dave and Thork actually killed some portion of my brain.
Mission accomplished.
-
I think reading this conversation between Dave and Thork actually killed some portion of my brain.
You too?
-
I don't see what the problem is. TFES has a very long tradition of someone getting a computer, they post their specs and everyone else tells them why they bought the wrong thing.
Here's my last computer for example.
https://forum.tfes.org/index.php?topic=4766.msg91913#msg91913
-
I think reading this conversation between Dave and Thork actually killed some portion of my brain.
Why?
Please, do point out where my concerns are invalid because i'm an idiot.
Or is it that I'm being nice to thork that makes your brain rot and die?
-
Its mostly jelly. I'm getting a new computer ... so I'm excited about that. You've been polite and asked me some questions. Rushy is looking at his potato PC and wants the world to burn, so he types mean stuff.
-
Rushy is looking at his potato PC
Okay I lost here. You really should come on IRC from time to time, Thork.
-
Why? You all hate me. That wouldn't be fun for any of us.
-
Why?
So that you know how ridiculously over-spec'd Rushy's PC is before you go calling it a potato.
-
Maybe he's just jelly that I didn't waste as much money as him?
Maybe he tells lies about the computer he owns and just talks about his fantasy PC?
Maybe he is really a she and shouldn't have left the kitchen again.
But I don't care about any of these very probable outcomes enough, to find out on IRC.
-
I think reading this conversation between Dave and Thork actually killed some portion of my brain.
Why?
Please, do point out where my concerns are invalid because i'm an idiot.
Or is it that I'm being nice to thork that makes your brain rot and die?
It's your extremely... interesting opinions on heat transfer. Most systems can passively radiate more heat than you'd expect, in addition to handling higher temperatures than you'd expect, and while there are certainly a lot of things to critique about Thork's new Wondows 10 machine, not being able to dissipate heat isn't one of them. A laptop or all-in-one can dissipate 200W pretty easily and you'd only notice them getting a little lukewarm to the touch, and that's assuming the device is sitting at 100% for extended periods of time. I doubt Thork does anything like that.
However, instead of just saying that, Thork responds with "yeah well it has heat pipes" which is also hilarious.
Maybe he's just jelly that I didn't waste as much money as him?
Maybe he tells lies about the computer he owns and just talks about his fantasy PC?
Maybe he is really a she and shouldn't have left the kitchen again.
But I don't care about any of these very probable outcomes enough, to find out on IRC.
Watching Thork quickly 180 from "Rushy has a potato" to "Rushy is just jealous he didn't buy a cheap all-in-one Dell" makes this thread worth the read.
-
I think reading this conversation between Dave and Thork actually killed some portion of my brain.
Why?
Please, do point out where my concerns are invalid because i'm an idiot.
Or is it that I'm being nice to thork that makes your brain rot and die?
It's your extremely... interesting opinions on heat transfer. Most systems can passively radiate more heat than you'd expect, in addition to handling higher temperatures than you'd expect, and while there are certainly a lot of things to critique about Thork's new Wondows 10 machine, not being able to dissipate heat isn't one of them. A laptop or all-in-one can dissipate 200W pretty easily and you'd only notice them getting a little lukewarm to the touch, and that's assuming the device is sitting at 100% for extended periods of time. I doubt Thork does anything like that.
However, instead of just saying that, Thork responds with "yeah well it has heat pipes" which is also hilarious.
Really?
I've got an old CPU(95W) and mid to high range powered GPU and it's constantly blasting out air even though it's got a lot of space(I mean, it's a desktop so the fan will run all the time). I assumed that passive heat dissipation wasn't easy given how much power is going through it especially since you've got the GPU and CPU squeezed into such a tight space.
How does it do it? Cause I remember the old laptop days of very hot, high fan speed gaming laptops. (still have one actually).
-
I've got an old CPU(95W) and mid to high range powered GPU and it's constantly blasting out air even though it's got a lot of space(I mean, it's a desktop so the fan will run all the time). I assumed that passive heat dissipation wasn't easy given how much power is going through it especially since you've got the GPU and CPU squeezed into such a tight space.
How does it do it?
Heat pipes.
...
However, instead of just saying that, Thork responds with "yeah well it has heat pipes" which is also hilarious.
Uh-huh.
Maybe he's just jelly that I didn't waste as much money as him?
Maybe he tells lies about the computer he owns and just talks about his fantasy PC?
Maybe he is really a she and shouldn't have left the kitchen again.
But I don't care about any of these very probable outcomes enough, to find out on IRC.
Watching Thork quickly 180 from "Rushy has a potato" to "Rushy is just jealous he didn't buy a cheap all-in-one Dell" makes this thread worth the read.
Get back to the kitchen. >:(
-
Really?
I've got an old CPU(95W) and mid to high range powered GPU and it's constantly blasting out air even though it's got a lot of space(I mean, it's a desktop so the fan will run all the time). I assumed that passive heat dissipation wasn't easy given how much power is going through it especially since you've got the GPU and CPU squeezed into such a tight space.
How does it do it? Cause I remember the old laptop days of very hot, high fan speed gaming laptops. (still have one actually).
Back in the olden days laptops used a lot more energy, were larger, and had bigger and worse batteries. This is also why, as you've probably noticed, the power bricks for mobile device power cords are much smaller. It's not that our transformer technology is better (it isn't) it's just that the devices we use now simply don't use as much energy as they used to.
This also happens to be why battery life is so much better in newer devices. Our batteries are pretty much the same, but the circuits utilizing them aren't nearly as power hungry. As a good comparison, each core on a Raspberry Pi 3 (a ~$35 computer) is about as powerful as a Pentium 2, 300 MHz CPU (a $2000 CPU at its release!) The TDP of the Pentium II was about 45W, but the TDP of the RPI is only 5W with all four cores running! With more modern laptops, you're looking at even more processor savings in computation versus electric usage. This is why, even when you see a tiny laptop with a ~60W+ TDP, you'll probably never see it actually use that amount of energy, because the amount of computational power in even something like a phone vastly and hilariously outweighs just about any task a consumer will put the computer through, which leads to the popular increase of passive cooling and savings to the consumer.
Also, overheating is never a problem in a modern computer. There are so many safety circuits in place that throttle power when overheating, you'd have to do some serious BIOS hacking just to get a computer to ignore those safety mechanisms and overheat itself. It's practically impossible to cook any computer components these days without the computer already being broken in some other way. Modern CPUs can even detect when they don't have a heatsink sitting on their die, and will instantly shut down instead of trying to boot.
All in all, heat dissipation in a modern, mainstream consumer all-in-one device is a complete nonissue.
-
That's a very eloquent answer to something, but it isn't the answer to Lord Dave's question.
He wants to know ... How is it my desktop is loud and noisy, and yet Thork is buying a PC of similar power in a smaller form factor and it won't be as noisy.
The answer to that is heat pipes.
Heat pipes haven't been in computers all that long. It is a relatively new technology. It gave rise to the 'gaming laptop' and is the reason in the past you couldn't have a gaming laptop. Of course now, people manage to stuff gtx 1080's into laptops. Heat pipes are a very efficient way of getting the heat away from the chip without any energy cost. It is a closed loop system and the vapour chamber gets heat away quickly to a place where it can be exhausted by a fan. So that leads the question ... why hasn't Lord Dave's desktop got heat pipes? And the answer to that is cost. They are expensive, which is one of the reasons sff computers have always costed more. This is evidenced by AMD's latest CPU coolers.
(http://www.relaxedtech.com/reviews/amd/wraith-max-and-wraith-spire-cooler/wraith-max-spire.jpg)
Notice the cheaper cooler on the right for lower end 65W CPUs. And as soon as they need to cool 95W monsters they use the one on the left, complete with heat pipes. But that is a more expensive heatsink.
Then consider the PC I'm getting.
(https://i.gadgets360cdn.com/large/Dell_Inspiron_27_7775_open_ndtv_1513326496647.jpg)
That is going to pull the heat right to the edge of the case, where it will be blown straight out by the fan. To engineer that into a case, you have to make the heat pipes fit your case and your exact motherboard + your GPU is going to need to be in a fixed place (hard with discrete cards). That isn't great for a desktop, because its going to push the price of your case up. So when you bought your PC or case, they didn't add any. But if you bought a sff case like a Shuttle or a Zotac thing (where heat will be an issue), the case wasn't $60. It is more likely $250 and comes with the motherboard, PSU and heat pipes all built in as barebones. You rarely see sff cases that aren't barebones builds, because they need a bespoke heat solution. This is why at the same time as gaming laptops arrived on the market ... bare bones and small form factor PCs became a thing.
Of course you could go one better than heat pipes in a desktop and add water cooling ... but that is expensive too.
So heat pipes. Make AIOs possible, but also push the cost of the device up compared with a desktop box.
Lord Dave doesn't care that processors are more efficient in this context. They aren't in this context. I have a desktop grade CPU in a sff ... how is that possible these days? Heat pipes.
-
Heat pipes haven't been in computers all that long. It is a relatively new technology.
Ah, fuck it, it's not like my non-involvement is making this any less painful.
Thork, I know you're old, but marvelling over 1940s breakthroughs is pushing it. Heat pipes were in use in modern computing since the early '90s. Fuck, it's only been a few months since I serviced an early-2000s laptop which used them extensively. This is not a new development, nor is it in any way revolutionary (anymore). Processors run much cooler than they did in the past. Compare something like a Pentium IV to a modern CPU and this becomes abundantly obvious.
-
Head pipes in computers have only been around 15 years or so. Don't be giving me the "they were invented in the 1940's" nonsense.
For examples of application - aerospace and astronautics in the 1970's; energy and conservation in the 1980's; industrial and natural energy utilization in the 1990's; computers and electronics cooling in the 2000's; and possibly global warming and environment protection in the 2010's.
And that technology took a while to proliferate. It wasn't like they just appeared over night or that the technology hasn't improved over time to allow better cooling solutions that weren't possible 15 years ago.
And no, processors don't run any cooler. Where do you get this shit from? You can go back to the early 2000's and the processors for desktops were still around 65W-95W TDP!
https://en.wikipedia.org/wiki/List_of_CPU_power_dissipation_figures#Pentium_4
Sure, today's do more and a 15W version would outperform those old P4s, but for desktop they kept the heat/power the same and improved performance to give you a tangible increase in performance generation over generation. They didn't make them run cooler as you just made up.
-
And no, processors don't run any cooler. Where do you get this shit from?
Honestly, it's knowledge that's so extremely common that I actually struggle to come up with a justification. High-end Pentium IVs could easily run at 70°C-80°C, while a modern i7 will usually oscillate somewhere around 55°C-65°C. The cooling technology has not changed, despite your assertion to the contrary (heat pipes have been commonplace since the 1990s).
And no, processors don't run any cooler. Where do you get this shit from? You can go back to the early 2000's and the processors for desktops were still around 65W-95W TDP!
Yes, you can cherrypick low-powered CPUs of the time and argue that they should be compared to today's top-range. The top range these days, something like an i7 8700k, runs at 95W TDP. An i7 7700 is 65W TDP. Comparable P4s/PDs were around 115W-130W. I didn't know where to take "this shit" from, but you gave me a handy Wikipedia link, so I'm quoting it for you.
Perhaps shockingly, there is a correlation between TDP and normal operating temperatures between these time periods. If the cooling technology has improved, how come CPUs with comparable TDPs run at pretty much the same temperatures still?
-
Actually, it is you that is cherry picking. You are using 115W examples from the past ... those termed extreme editions. But today's equivalent would be the Kaby Lake X ... a 112W processor that runs every bit as hot*.
And you'll find that 70-80 degrees depends completely on your cooling solution. Intel chips can actually run up to around 95 degrees (today the same as yesteryear) before they thermally throttle and Intel's Tjunction is set at 100 degrees as stated on their website for any desktop processor you choose.
https://ark.intel.com/products/126686/Intel-Core-i7-8700-Processor-12M-Cache-up-to-4_60-GHz
So, yeah, there is no evidence at all they run cooler. The facts state they run as hot as they ever did. That's the desktop package and it hasn't changed much since they put the CPU onto the motherboard instead of using cartridges.
-
I can't help you, Thork. If you want to pretend that cooling technology has significantly improved, I won't stop you. I'll just add it to the pile of evidence of why nobody should ever listen to any of your claims on technology.
I will, however, invite you to consider the fact that you're running one of the coolest CPUs available for desktops in the past 2 decades. I'm sure it's a complete coincidonk and it was totally heat pipes that magick'd themselves into existence despite already being there. But perhaps if you have a moment of spare time, consider the correlation between TDP and standard operating temperatures, and how that hasn't changed in those 2 decades.
-
No, I'm not running one of the coolest CPUs in decades. It has a 65W TDP.
TDP!
Thermal Design Power.
That does not tell me how much power the chip consumes. If it boosts up it can be dragging 100W of power from the wall no problem. When it is running near idle it will sip just a few watts. TDP is a measure of the HEAT generated and a 65W TDP processor from 2001 will run the same temperature as a 65W TDP processor from 2018 if they have the exact same surface area and cooling solution.
The IPC will be improved in the 2018 version. It may have more cores. But if today's chip runs at 67 degrees in your system with a cooling solution and you could find brackets to fit the same solution to your 2001 chip, that too would sit at 67 degrees if both are under the same percentage load.
There is no way on earth you can say I'm runnig one of the coolest CPUs in decades. It just isn't true. And it doesn't matter that I am using AMD with a slightly worse IPC than the intel equivalent. It doesn't matter that the clock speed is lower. Both will get up to 95 degrees and then throttle and the cooler is going to dissipate the same heat from both at that 95 degrees max. If I have a better cooler, it will take more power to get to that 95 degrees so I can squeeze more performance out.
Thermal design power. Mine is 65W. That's been a desktop standard for a very long time. It aint run no cooler. Faster yes, cooler, no.
So how is it we can get these 65W TDP chips into laptops and AIOs these days when they make the same heat? Better frickin' heat pipes.
-
Actually Rushy's answer was what I needed. Heat pipes are not new for me. I've been a tech. I've seen a fuck ton of them. Hell, my own heatsink has them as does the heat sinks for my last several graphics cards.
See my thinking was that heat is generated not from the amount of actual wattage but when electrons impact a closed gate. So thus, the more transisters you have, the more calculations you can do, then the more electron impacts you'll have.
If they've managed to make that more efficient, cut down on the amperage or internal voltage or ... whatever, then yeah, less heat.
The idea of heat pipes was already assumed. After all, you need to move the heat from the CPU and the GPU to a spot away from both to dissipate it since you don't have the space to do it at the chips.
-
Electrons don't heat the chip. The switches move in the chip and that generates the heat. Up the clock speed, more cycles per second, more switches moving, more heat.
Changing the voltage doesn't change the the heat because more electrons, it is because you can up the clock speed making more switches move faster. Its a material bending causing heat thing. As the silicon flexes to shut or open a gate, more heat.
Yes, moving the heat away is the key. Heat pipes are very efficient at this as they take the heat to the exit in laptops etc. In a desktop, your CPU fan likely exhausts the heat into the case raising the ambient temperature. That is what heat pipes try to avoid by getting the heat directly to the exit. If you put a little fan on a cpu in a laptop, it'd cook itself and throttle in no time as the heat wouldn't escape well.
-
a 65W TDP processor from 2001 will run the same temperature as a 65W TDP processor from 2018 if they have the exact same surface area and cooling solution.
This is correct, and what currently happens.
So how is it we can get these 65W TDP chips into laptops and AIOs these days when they make the same heat? Better frickin' heat pipes.
We always shoved ~65W TDP chips into AIOs (though of course that restricts my timeframe to AIOs with LCD screens - CRT would be cheating on my part). You assert a change where there is none.
-
I kind of want to elaborate on my point to Lord Dave. I just dropped on him that it isn't electrical resistance that heats a chip. It is material bending, That's a really fundamental concept and I didn't give any references.
When I looked for a source, I saw many confirming Lord Dave's misconception and I guess because I have a degree in Engineering I assumed it would be easy to understand. I had an entire semester cocking about with ARM chips back in the day.
I knew this because the formulas for heating a chip don't have resistance in them. They are based on energy transfer. I found a good article here explaining why transitors heat up.
https://www.quora.com/Why-do-CPUs-get-so-hot-relative-to-the-other-components-in-a-computer
Recall that the amount of energy stored in a capacitor is CV2/2 where C is the capacitance (a measure of how much charge can be loaded into a capacitor to produce a given field) and V is the voltage across the capacitor. Now, each step of a computation inside a microprocessor is triggered by the edge of a "clock" of a certain frequency f. Each step of that computation will require many of those switches to change their state -- from on to off or off to on. So either energy has to move into those switches or it must be purged from the switches. The CV2/2 energy inside each of those switches that must be purged has to go somewhere. Consequently, it is released as heat. Thus, the power (energy per unit time) lost due to computation is roughly CV2f/2 because each nugget of energy is released each tick of the clock which moves with frequency f.
Notice that I have not discussed resistance, which surely is the mechanism for this energy transfer. You might expect that reducing the resistance of some of the wires in the microprocessor could somehow mitigate this problem. However, there is no variable for resistance in the formulas above.
Ok, on to Pete.
"We've always shoved 65W chips into AIOs". Actually not true. Apple's iMac was one of the first (2002) and their G4 processor in the first LCD iMac was just 18W TDP. They only moved to higher power chips with Intel's i platform but by then it was 2010 and they had great heatpipes and even then they nerfed the clockspeed for many years after choosing the S varients which didn't boost as high. We didn't get a 95W overclockable K series CPU in an imac until 2014.
There was a change. There are far more powerful small PCs and laptops than ever using higher TDP GPUs and CPUs. And the reason is better cooling.
-
I guess because I have a degree in Engineering I assumed it would be easy to understand.
Thanks for the sig.
-
Yeah, you can have that.
I guess when you have known something for 20 odd years, you expect it to be common knowledge, but you forget how you acquired that knowledge in the first place. Then you think back to an room with an oscilloscope pulling apart an ARM chip and you realise it probably isn't a normal experience for most people after all.
-
Apple's iMac was one of the first (2002) and their G4 processor in the first LCD iMac was just 18W TDP.
Again, you successfully cherry-picked an exception and are pretending for it to be the rule. But okay - even if we accept your terms, you agree that your "breakthrough" in iMacs happened about 9-10 years ago with Wolfdale and Lynnfield CPUs. Congratulations, even with layers upon layers of special pleading, you're proving yourself wrong.
-
I don't understand where you are going with this or what you are trying to pick me up on.
Desktop components in laptops, AIOs and sff PCs are made possible by heatpipes.
Heat pipes didn't become a thing you could buy without thousands of dollars (ie mainstream) until about 2003-2004. Since then the technology for this type of cooling has improved to the point you can now get gtx 1080s and k series processors in laptops and the like. That wasn't possible 10 years ago.
So when Dave asks, why is it that Thork's AIO has similar desktop components to mine and doesn't sound like a hairdryer, the answer is because heat pipe technology has improved in recent years to better deal with the heat. Thork paid more money for the cooling solution to make that possible.
Please, find the objection in this assertion so that we are on the same page.
-
So when Dave asks, why is it that Thork's AIO has similar desktop components to mine and doesn't sound like a hairdryer, the answer is because heat pipe technology has improved in recent years to better deal with the heat. Thork paid more money for the cooling solution to make that possible.
This assertion is entirely incorrect - it's not a matter of finding a single objection. Your components may be similar in performance to what Dave has (though I don't think he's actually said that), but they are 65W TDP, which is remarkable for their power. In the past, a powerful CPU was a very hot CPU. Nowadays, a powerful CPU is a lukewarm CPU, and the hot ones are rarely used in regular customer applications.
Obviously the use of heat pipes will help displace the heat to a fan, but the fan will still need to be there to eventually dissipate it into the air. So yes, while your 10-year-old breakthrough makes some of the logistics simple, it doesn't in any way reduce the need for heat dissipation. That's where a relatively lower power usage comes in.
-
Oh, I think I see what you are saying and I still disagree.
You are saying that my desktop CPU is giving me great performance and so it does more for the package it is in. In the past I would have needed a 140W processor to get that kind of functioanlity for gaming etc. And with that I agree.
But its still just a standard 65W desktop processor. Mainstream. And those were not put in laptops etc 10 years ago. We still have 95W higher end and 140W extreme processors today (skylake X is 140W i believe), but the fact is the mainstream has been 65W for donkeys years and you couldn't get those into a laptop before. Nor could you get high end graphic solutions into small devices. Now you can.
This isn't because the 65W processors push out less heat today. We discussed TDP and agreed they push out the same.
But today, Lord Dave can look at his tower and know for the first time in history (last 5 years or so), that there are many laptops out there that would kick its arse because they too can enjoy mainstream processors and high end GPU solutions. That wasn't the way it always was. You couldn't play Far Cry on high settings on a laptop when it first came out. The laptops of the day were too shit. Now they don't have lower power parts in the top end versions. They have a power brick the size of a breeze block and they are able to shift all that heat away, where before they couldn't.
-
But its still just a standard 65W desktop processor. Mainstream. And those were not put in laptops etc 10 years ago.
But they were put in AiOs 9-10 years ago.
-
And they were really noisy.
I had a Shuttle PC of the same era. They had got the tech in, but the heat was still a major issue and an AIO is much larger than a laptop. The first laptops to do this a while back used to burn your thighs. Today the same TDPs can be dealt with much more efficiently and make less noise. Because the heat pipe tech is better. And so now less noise and less burnt legs means more consumers want the power without the drawbacks and we see far more of that stuff going on.
-
Thanks, Thork. That article was helpful. It's not dissimilar from what I thought but different mechanics.
This means that the efficiency was increased either by having less need to switch the gates as much or lower carge amounts across the gates. Lower charge = less heat.
Also:
Heat pipes probably have become more efficient at thermal transfer but it's likely not as big of a deal as the chip efficiency.
-
Die shrinks are usually the best way. You normally see the biggest leaps when they shrink the architecture. So move from say 14nm to 10nm. This means more transistors on the same die (twice as many) SQRT[2], but those transistors are all smaller so you get the same heat. And so it takes less energy to open and close them, and they release less heat when they give up that energy. Twice as many transistors, same heat output. More grunt for same TDP.
Also:
Heat pipes probably have become more efficient at thermal transfer but it's likely not as big of a deal as the chip efficiency.
Please let's not go back there. They have the same TDP today, so same heat output. That didn't change. It is all about the cooling solution.
-
Please let's not go back there. They have the same TDP today, so same heat output. That didn't change. It is all about the cooling solution.
No, we have to go back there, because you're still hilariously wrong about this. Back in the day, a 65W TDP CPU was a relatively crap CPU. Now, a 65W TDP CPU is a relatively decent CPU. In both cases, we are comparing them to their contemporaries. That's the main factor that changed. You could have had a quiet AiO 10 years ago, but it would pale in comparison to a big-ass desktop. Now, you can have a quiet AiO that can somewhat compete with one.
Yes, hotter CPUs existed back then, and they exist today, but it is much, much easier to get some decent performance with the same low amount of heat that would previously come out of a low-mid-end machine.
-
I can't believe I have to spell this out.
Lets look at a game like Far Cry.
To get good frame rates (graphics aside), back in the day you needed a desktop CPU. And that would be fine because Far Cry would run on the mainstream desktop CPUs of the day.
Today there is Far Cry 5 or whatever they are up to. To get decent frame rates, you need a desktop CPU as you always did. Because spoiler ... not only are the CPUs better today, so are the games. And the operating systems, and productivity programs and evrything else.
Now, when developers optimize their products, they look at the tech of the day. They'll say "Hmmmm. What is the shittiest PCs people are likely to have right now?" Say an average PC from 5 years before. And they'll then make the game work for that on low settings and write down what those settings were and call them minimum requirements.
Then they'll say "Ok, so what is an average PC of today like?" and they'll make the medium settings of the game work for that hardware and call it recommended settings.
Then they'll say "We need to sell this game and so we want it to look boss. What are the best PCs enthusiast gamers have right now?" and they'll make sure their game with all settings at ultra match the best GPU and CPU combos of the day for the best gameplay experience possible.
Now, whether that was in 2001 or 2018 the recommended settings are going to be based on a 65W desktop processor of that day. Historically Intel's mainstream desktop offering.
So, if you want recommended settings in 2001 you needed a 65W desktop CPU, but those weren't in laptops back then because of heat. But today they are in the laptops and so ... you can play games with recommended settings and enjoy the top titles of today without turning everything down to low.
And so ... because you can get desktop CPUs and GPU chips into sff and laptop devices, you can now for the first time enjoy the same experience as a desktop user ... and your device will be filled with heat pipes to make that happen.
-
Now, whether that was in 2001 or 2018 the recommended settings are going to be based on a 65W desktop processor of that day. Historically Intel's mainstream desktop offering.
It doesn't matter how many times you spell it out. This continues to be incorrect. The "average" PC back then ran much hotter than an "average" PC does now.
-
Not if it had a 65W TDP processor in it ... which has been the mainstream offering from Intel since the P4 was released. Before that P3's were about 35W from what I remember.
the reason AMDs like the Bulldozer always ran hot is because they couldn't compete on IPC so they ramped up the power. They were hotter and had higher TDPs. The FX-8350 was 125W TDP! Those things were toasty, but not intel.
-
Actually Rushy's answer was what I needed. Heat pipes are not new for me. I've been a tech. I've seen a fuck ton of them. Hell, my own heatsink has them as does the heat sinks for my last several graphics cards.
See my thinking was that heat is generated not from the amount of actual wattage but when electrons impact a closed gate. So thus, the more transisters you have, the more calculations you can do, then the more electron impacts you'll have.
If they've managed to make that more efficient, cut down on the amperage or internal voltage or ... whatever, then yeah, less heat.
The idea of heat pipes was already assumed. After all, you need to move the heat from the CPU and the GPU to a spot away from both to dissipate it since you don't have the space to do it at the chips.
Watts = Voltage * Amperes. A modern CPU uses a vcore of about 1.2V (this highly depends on the CPU) so the current of a 65W TDP maxes out at around 54A. The current in a CPU is pretty wild and hard to monitor, so basically you should never bother wondering what it is. Also, the heat output of any given electronic circuit equals its wattage. A circuit that doesn't use energy doesn't output heat, and a circuit that uses energy outputs 100% of said energy as heat. A 1000W computer is also a 1000W spaceheater.
The key energy savings in a modern CPU isn't the TDP (the TDPs are actually about the same). The key is that the modern CPU does a lot more with the same energy, meaning tasks that used to work a CPU to death are now very, very easy. Things like running a busy OS (such as Wondows 10), outputting 4k video, and other seemingly heavy tasks are now pretty much nothing for a modern CPU. Unless you're purposefully stressing the CPU, or you doing video encoding or compiling or some other workstation task, then the CPU will never hit its TDP, whereas in the past, processors idled at their TDP because clock-gating and turbo boosting weren't feasible. An older 65W TDP CPU used 65W almost all the time, while a modern one only uses its TDP at max usage, which it will seldom experience under most circumstances.
Modern CPUs are just in general better at managing their energy usage. They're still fully capable of being loud, obnoxious power hogs, but thanks to modern power management techniques, they won't be 99% of the time, and basically 100% of the time for people who aren't power users.
-
Watts = Voltage * Amperes. A modern CPU uses a vcore of about 1.2V (this highly depends on the CPU) so the current of a 65W TDP maxes out at around 54A. The current in a CPU is pretty wild and hard to monitor, so basically you should never bother wondering what it is. Also, the heat output of any given electronic circuit equals its wattage. A circuit that doesn't use energy doesn't output heat, and a circuit that uses energy outputs 100% of said energy as heat. A 1000W computer is also a 1000W spaceheater.
TDP Watts isn't power consumption. It is a heat guide as in one watt = one joule per second. Pete made the same conflation earlier. A 65W processor is not using 65W from the wall, it is throwing out 65W of heat. TDP. Thermal Power design.
The key energy savings in a modern CPU isn't the TDP (the TDPs are actually about the same).
Great. hopefully Pete will read this.
The key is that the modern CPU does a lot more with the same energy, meaning tasks that used to work a CPU to death are now very, very easy. Things like running a busy OS (such as Wondows 10), outputting 4k video, and other seemingly heavy tasks are now pretty much nothing for a modern CPU. Unless you're purposefully stressing the CPU, or you doing video encoding or compiling or some other workstation task, then the CPU will never hit its TDP, whereas in the past, processors idled at their TDP because clock-gating and turbo boosting weren't feasible.
Modern CPUs are just in general better at managing their energy usage. They're still fully capable of being loud, obnoxious power hogs, but thanks to modern power management techniques, they won't be 99% of the time, and basically 100% of the time for people who aren't power users.
Agreed but when you do something like gaming, you are usually knocking close to the limit with a good GPU. CPU bottle necking is a thing on low resolutions with fast GPUs. A gtx 1080 running a game at 1080p will probably push enough frames to overwhelm most 65W TDP processors.
-
TDP Watts isn't power consumption. It is a heat guide as in one watt = one joule per second. Pete made the same conflation earlier. A 65W processor is not using 65W from the wall, it is throwing out 65W of heat. TDP. Thermal Power design.
Where the hell do you think that heat comes from? Never heard of the Law of Conservation of Energy?
-
TDP Watts isn't power consumption. It is a heat guide as in one watt = one joule per second. Pete made the same conflation earlier. A 65W processor is not using 65W from the wall, it is throwing out 65W of heat. TDP. Thermal Power design.
No one here cares about how much is coming out of the wall, Thork. That just depends on your PSU efficiency and has nothing to do with your CPU.
Also, TDP is your power consumption. If the CPU is using 65W, it's outputting 65W of heat. There's no getting around that.
Agreed but when you do something like gaming, you are usually knocking close to the limit with a good GPU. CPU bottle necking is a thing on low resolutions with fast GPUs. A gtx 1080 running a game at 1080p will probably push enough frames to overwhelm most 65W TDP processors.
...Okay? What consumer is buying a GTX 1080 and then running a 1080p monitor? Not a very smart one.
-
TDP Watts isn't power consumption. It is a heat guide as in one watt = one joule per second. Pete made the same conflation earlier. A 65W processor is not using 65W from the wall, it is throwing out 65W of heat. TDP. Thermal Power design.
Where the hell do you think that heat comes from? Never heard of the Law of Conservation of Energy?
???
The 65W TDP rating isn't watts= voltage x amps as Rushy asserted. 65W CPUs can pull way more power than that. 65W as in joules per second is what is thrown out as heat. And that is why after market fans are rated for the TDP of the processor. Got a 95W TDP fan, will dissipate 95W of heat thrown off a CPU. There are losses, some energy is dissipated in noise, vibration and some in useful work. That 65W is the shit left over. Like when you drive your car and only about 30% of the power from the gasoline gets to the wheels. The rest is lost to entropy via mechanical resistance, a warm engine, noise etc etc.
No one here cares about how much is coming out of the wall, Thork. That just depends on your PSU efficiency and has nothing to do with your CPU.
Also, TDP is your power consumption. If the CPU is using 65W, it's outputting 65W of heat. There's no getting around that.
Oh, Intel invented perpetual motion did they? A lossless machine? ffs.
...Okay? What consumer is buying a GTX 1080 and then running a 1080p monitor? Not a very smart one.
But by the same token, my RX580 is going to give my Ryzen 1400 a good workout. It'll throw a lot of frames but the system isn't going to give me 4k gaming, despite the screen being 4k.
-
The 65W TDP rating isn't watts= voltage x amps as Rushy asserted. 65W CPUs can pull way more power than that. 65W as in joules per second is what is thrown out as heat.
... it's the same unit. It just gets converted from electrical energy to thermal energy.
There are losses, some energy is dissipated in noise, vibration and some in useful work. That 65W is the shit left over. Like when you drive your car and only about 30% of the power from the gasoline gets to the wheels. The rest is lost to entropy via mechanical resistance, a warm engine, noise etc etc.
Ah yes, vibrating CPUs are always a problem for me as well.
Oh, Intel invented perpetual motion did they? A lossless machine? ffs.
Are you suggesting that heat is the useful product of a CPU? Because I fail to see how a CPU that produces 65W of heat is "lossless" otherwise.
-
The 65W TDP rating isn't watts= voltage x amps as Rushy asserted. 65W CPUs can pull way more power than that. 65W as in joules per second is what is thrown out as heat. And that is why after market fans are rated for the TDP of the processor. Got a 95W TDP fan, will dissipate 95W of heat thrown off a CPU. There are losses, some energy is dissipated in noise, vibration and some in useful work. That 65W is the shit left over. Like when you drive your car and only about 30% of the power from the gasoline gets to the wheels. The rest is lost to entropy via mechanical resistance, a warm engine, noise etc etc.
CPUs can pull more than their TDP, but it's unusual unless you're overclocking the CPU. At stock, the power usage of the processor at full tilt will be somewhere around their TDP. Usually lower, because the TDP is supposed to represent a theoretical maximum of heat dissipation.
For example:
(https://images.anandtech.com/graphs/graph11859/91808.png)
I also don't see how that suddenly invalidates Watts = Voltage * Amperes...
-
(https://www.gamersnexus.net/images/media/2015/features/watt-draw-bench-cpus.png)
Take the i7 6700. It is a 65W TDP processor.
At idle it uses 71W. At load it uses 132 watts. Where does 65W come into that? The 65W TDP is the THERMAL POWER DESIGN ... how much heat it'll kick out at its 95 degree throttling limit on a CPU die that size. Its the waste. 132 watts goes in, 65 watts is expelled as heat at that design limit. You need a fan that will deal with 65W of expelled heat. No less. If you have a 45W TDP fan, its going to throttle below its advertised boost clocks and not perform as you expected.
Watts = joules per second.
65W is the max output waste in heat. Not the volts x amps the CPU needs to run.
-
132 watts goes in, 65 watts is expelled as heat at that design limit.
And where does the rest go? Does your CPU serenade you by vibrating out last year's Eurovision tunes like a 67W speaker?
-
>Full system power draw
>Full system
>FULL SYSTEM
Thork this is the turning point where you either tell me you're trolling or you should really just stop. The graph you just posted is the output of the ENTIRE COMPUTER not the CPU.
-
132 watts goes in, 65 watts is expelled as heat at that design limit.
And where does the rest go? Does your CPU serenade you by vibrating out last year's Eurovision tunes like a 67W speaker?
I already explained. Useful work ... moving switches, moving air and other damper issues, vibration, noise, ... you can't expect to put in 132W of energy, use every last watt and get that all out perfectly as heat. Machines don't convert energy that efficiently. Otherwise homes would be heated by CPUs.
>Full system power draw
>Full system
>FULL SYSTEM
Thork this is the turning point where you either tell me you're trolling or you should really just stop. The graph you just posted is the output of the ENTIRE COMPUTER not the CPU.
Don't be a muppet. A GPU can pull 300W. i7 computers don't pull just 132W at full load. You know that. That's why your PC hasn't got a 150W power supply.
Those charts are CPU only.
-
Thork I highly suggest you actually read the commentary regarding that picture on the website you pulled it from.
-
(https://www.gamersnexus.net/images/media/2015/features/watt-draw-bench-gpus.png)
Maybe you should. Same page. And here is the GPU draw.
-
I already explained. Useful work ... moving switches, moving air and other damper issues, vibration, noise, ... you can't expect to put in 132W of energy, use every last watt and get that all out perfectly as heat. Machines don't convert energy that efficiently. Otherwise homes would be heated by CPUs.
Heat is the waste product. CPUs do negligible amounts of useful work. Their job is to flip bits in memory, which uses almost no energy by itself, but produces a huge amount of waste heat as a byproduct.
CPUs aren't 100% efficient, they are (almost) 100% inefficient.
As for the other forms of waste you mentioned, have you ever heard your CPU producing tens of watts of noise or vibration?
-
Yeah, no. Just google it. I can't explain it any simpler.
Maybe this guy explaining it will help?
https://www.youtube.com/watch?v=dfxjOwDDnKc
TDP ... the shit heat expelled ... not the power draw.
-
Yeah, no. Just google it. I can't explain it any simpler.
The reason you can't explain it any simpler is that it's bullshit.
Maybe this guy explaining it will help?
This guy seems to be just as confused as you are. He's also talking about power draw "from the wall", which is not individual component power usage because of inefficiencies in the PSU.
You still haven't explained where the extra energy goes. If your point had any substance, you would be able to do that. Stop waving your hands and saying "useful work" and "my CPU is a vibrator" and explain, specifically, what form the extra energy takes.
-
Both me and the guy in the video explain where the energy goes. Stop being obtuse.
Losses. Entropy. You can't convert 100% of electrical energy into 100% heat energy. Especially in something not designed to turn energy into heat. A CPU is designed so you can use it to troll the hell out of me on this forum from a far off land. >:(
-
Just a quick snide remark for Thork: Rushy and I are saying exactly the same thing, we're just approaching it from different angles. Telling me to read his take on what I already described differently is unlikely to change my mind.
-
Losses. Entropy.
Neither of these is a specific form of energy. Heat is a specific form of energy which falls into both of these categories. Try again.
You can't convert 100% of electrical energy into 100% heat energy.
There is a question of quantity here which you're conveniently ignoring. It's not as if you're saying that a CPU with 65W TDP is consuming 66W of power. You're saying it consumes twice as much power.
Stop using vague, hand-wavey "explanations" and tell me where the extra 67W of energy ends up.
-
Oh for fuck's sake, not another Thork thread where he says stupid things about technology and tricks everyone into responding for weeks.
Good game. Thanks for playing. ::)
None of these are hard concepts. You can use google. you could learn. You'd ratehr tell me I'm wrong ... only I'm not.
Next week we should discuss why Intel CPUs are better at games than AMD, even though the IPC is very comparable for productivity. Explaining the difference between mesh, infinity fabric and ringbus will take me weeks at the rate you lot learn.
Losses. Entropy.
Neither of these is a specific form of energy. Heat is a specific form of energy which falls into both of these categories. Try again.
You can't convert 100% of electrical energy into 100% heat energy.
There is a question of quantity here which you're conveniently ignoring. It's not as if you're saying that a CPU with 65W TDP is consuming 66W of power. You're saying it consumes twice as much power.
Stop using vague, hand-wavey "explanations" and tell me where the extra 67W of energy ends up.
Well, some of the power is passed on to other components. CPU needs to talk to the ram? Needs to send electrical signals to it. Wants to talk to the GPU? More signals needed. Motherboard, resistors, signals to drives and other chips.
Some is lost to electro-magnetic radiation.
Some is noise, you can hear a CPU.
Some is vibration. If a CPU operates at 4GHz ... that's a lot of very shallow amplitude but high frequency oscillation. Imperceptible to you as a human, but guzzling power all the same.
Electrostatic losses when not all power sent to capacitors/transistors reaches them.
These days even quantum tunneling is a loss. Not all your electrons do the thing they were sent to do.
Working around broken transistors. Your CPU isn't perfect. Lots of the bits are broken and it has to reroute around this. This is to do with yields and the silicon lottery - that's a conversation for another day ... but another inefficiency all the same.
There are loads. A CPU isn't just made to generate heat. In fact it is designed to limit the heat as much as possible as heat will be the limiting factor on performance for that CPU. Otherwise you could just push it to 50GHz.
-
Well, some of the power is passed on to other components. CPU needs to talk to the ram? Needs to send electrical signals to it. Wants to talk to the GPU? More signals needed. Motherboard, resistors, signals to drives and other chips.
This is negligible, since the purpose of these signals is to transmit information and not power, so there isn't a lot of current involved. But more importantly, this will just be turned into waste heat when it reaches those components anyway, so it makes no difference to the total heat the system generates.
Some is lost to electro-magnetic radiation.
Which is a form of heat, albeit not one that needs to be extracted by a cooling system.
Some is noise, you can hear a CPU.
Not over a 10W speaker. Also negligible.
Some is vibration. If a CPU operates at 4GHz ... that's a lot of very shallow amplitude but high frequency oscillation. Imperceptible to you as a human, but guzzling power all the same.
4 GHz waves cannot permeate effectively in air. A 4 GHz sound wave in air has a wavelength of about 80 nm, which is short enough that it spans about 20 nitrogen molecules. You're playing pool with molecules at this point. This means the vibrations are never going to leave the computer, and therefore do not qualify as a form of loss.
Guess what happens to vibrations when they impart energy but don't travel very far as waves. Go on, you can think about this one for a bit.
Electrostatic losses when not all power send to capacitors/transistors reaches them.
Another form of heat...
These days even quantum tunneling is a loss. Not all your electrons do the thing they were sent to do.
"These days"? Are you aware that quantum tunneling didn't just spawn into existence when it was discovered?
Nevertheless, this is both negligible and irrelevant, since electrons are matter and not energy.
There are loads. A CPU isn't just made to generate heat.
Actually, that is literally what a CPU is designed to do. It's an entropy machine.
Are you familiar with the Second Law of Thermodynamics? It states that the total entropy of a closed system must always increase. Now consider what a computer is supposed to do. The job of a computer is to take disorganised data -- with high entropy -- and turn it into a form that can be easily accessed by humans -- with low entropy.
Now we have a problem. To do its job, a computer needs to reduce the total entropy in its memory banks. Since the entropy in a closed system cannot decrease, this means that the computer cannot be a closed system, and it must generate more entropy than it removes. The more entropy it generates, the better it performs.
This is the technical reason for the common sense correlation between higher TDP and better performance. When it comes to computers, better performance is more waste heat.
-
Good game. Thanks for playing. ::)
From IRC:
<&Parsifal> sexdroid: I have an engineering degree so I assumed it would be easy to understand
<@sexdroid> Yes
<@sexdroid> I regret losing my restraint in that thread
<@sexdroid> I got thoroughly Thork'd
-
Heat pipes are relatively new technology
-
It sounds like y'all are arguing over different parts of the same thing.
Parsifal, Pete, and Rushy sound like they're saying "TDP is electrical power that's being converted to heat" which is accurate. The CPU won't generate any watts of heat without electrical power and it will need to draw at least the amount of power needed to produce the heat it outputs.
Thork is saying that the CPU is pulling more than the heat generated (which is also true), but where he's wrong is the energy it's pulling isn't stored in the CPU. It just passes through. Like, unless I'm mistaken, a CPU isn't going to store a charge inside itself then release that charge to the motherboard to send a signal to read RAM. It'll simply have a path of electrical charge that leads from the motherboard's BUS through the CPU and out to the RAM.
I would personally ony count the electrical power the CPU uses for computation within itself, not what essentially passes through after a calculation.
Of course, having almost no background in microprocessor architecture and engineering, I could be wrong about the above.
-
@Parsifal I told you that about half of the power going in comes out as heat.
You asked, where does the rest go.
I gave you a bunch of possible places including communication, losses and mechanical work of physically moving the transistors which have a mechanical resistance, not all of which becomes heat.
Why is this hard to understand? You know you can't put 100% electrical energy in and get 100% heat out. There will be a difference between those two numbers.
Power draw (watts) = work done (watts) + other losses (watts) + heat losses TDP (watts).
You know TDP can't be power draw.
Also some of your objections are weird for a scientifically minded person.
Well, some of the power is passed on to other components.
This is negligible, since the purpose of these signals is to transmit information and not power, so there isn't a lot of current involved. But more importantly, this will just be turned into waste heat when it reaches those components anyway, so it makes no difference to the total heat the system generates.
Well CPUs do send 12v communications to some components so ... but mostly if the energy arrives at the RAM stick and loses heat there, that's not part of the TDP ... the heat coming off the processor die. You know that. That's heat elsewhere.
Some is lost to electro-magnetic radiation.
Which is a form of heat, albeit not one that needs to be extracted by a cooling system.
No, that's generating a magnetic field which causes resistance of its own as it permeates materials such as your copper heatsink. Pushing something metallic into a magnetic field uses energy. Like wise pushing a field out through metals must therefore also use energy because of newton's second law. Sure some heat, but not all heat. Some is going to be left hand rule force in Newtons.
Some is noise, you can hear a CPU.
Not over a 10W speaker. Also negligible.
Its all negligible. it takes 80,000 watts to power a Nissan Leaf. you are asking me where just 67 watts goes. The average human fart produces around 60 watts of power.
Some is vibration. If a CPU operates at 4GHz ... that's a lot of very shallow amplitude but high frequency oscillation. Imperceptible to you as a human, but guzzling power all the same.
4 GHz waves cannot permeate effectively in air. A 4 GHz sound wave in air has a wavelength of about 80 nm, which is short enough that it spans about 20 nitrogen molecules. You're playing pool with molecules at this point. This means the vibrations are never going to leave the computer, and therefore do not qualify as a form of loss.
Guess what happens to vibrations when they impart energy but don't travel very far as waves. Go on, you can think about this one for a bit.
That pool of moelcules is still being moved around. That's a loss. Billions of transistors, all moving gates, shifting molecules millions of times a second, it adds up.
Electrostatic losses when not all power send to capacitors/transistors reaches them.
Another form of heat...
No. That isn't heat. That's electrons that were stuck to something sodding off and having to be replaced by new electrons. Bit like a balloon you rubbed on your head losing its charge after a while.
These days even quantum tunneling is a loss. Not all your electrons do the thing they were sent to do.
"These days"? Are you aware that quantum tunneling didn't just spawn into existence when it was discovered?
Nevertheless, this is both negligible and irrelevant, since electrons are matter and not energy.
Yes these days as the gates are only 14nm and the electrons can tunnel the gap. Much less a problem at 80nm gates. Its a problem that will only get worse as transistor sizes get smaller.
There are loads. A CPU isn't just made to generate heat.
Actually, that is literally what a CPU is designed to do. It's an entropy machine.
No. No, consider how silly that is. It is a counting machine. It happens like all machines to have losses.
Are you familiar with the Second Law of Thermodynamics? It states that the total entropy of a closed system must always increase. Now consider what a computer is supposed to do. The job of a computer is to take disorganised data -- with high entropy -- and turn it into a form that can be easily accessed by humans -- with low entropy.
Now we have a problem. To do its job, a computer needs to reduce the total entropy in its memory banks. Since the entropy in a closed system cannot decrease, this means that the computer cannot be a closed system, and it must generate more entropy than it removes. The more entropy it generates, the better it performs.
This is the technical reason for the common sense correlation between higher TDP and better performance. When it comes to computers, better performance is more waste heat.
No. I can take a 65W TDP processor and have 8 cores like the Ryzen 1700. I can also take a 65W TDP processor like the ryzen 1400. The former performs better because it has more cores. Not because it has a higher TDP. They have the same TDP. I could take a 7th gen i5 and give it hyperthreading. Still the same TDP but it will perform better.
TDP is the amount of heat it will throw off. Not the power draw. Why is this hard? A hyper threaded i7 guzzles a lot more than a non-threaded i5. They still have the same TDP. they don't have the same power draw or the same performance. The i7 does more work. It needs more power. But its heat loss is comparable because the energy from power draw is used in other ways. It is more efficient, hence the reason it performs better and Intel want another $100 out of you for the privilege.
It sounds like y'all are arguing over different parts of the same thing.
Parsifal, Pete, and Rushy sound like they're saying "TDP is electrical power that's being converted to heat" which is accurate. The CPU won't generate any watts of heat without electrical power and it will need to draw at least the amount of power needed to produce the heat it outputs.
Thork is saying that the CPU is pulling more than the heat generated (which is also true), but where he's wrong is the energy it's pulling isn't stored in the CPU. It just passes through. Like, unless I'm mistaken, a CPU isn't going to store a charge inside itself then release that charge to the motherboard to send a signal to read RAM. It'll simply have a path of electrical charge that leads from the motherboard's BUS through the CPU and out to the RAM.
I would personally ony count the electrical power the CPU uses for computation within itself, not what essentially passes through after a calculation.
Of course, having almost no background in microprocessor architecture and engineering, I could be wrong about the above.
The point is, not all the power drawn in, comes out as heat. Its very simple. So TDP is a lower number than power draw. Your chip always pulls more power than it expels in heat. Kinda obvious. Heat is a loss. The chip is doing useful work as well.
total power = useful work - losses (of which one loss is TDP and must therefore be a lower number than total power drawn)
-
(https://i.imgur.com/SslYRs3.png)
-
Like, unless I'm mistaken, a CPU isn't going to store a charge inside itself then release that charge to the motherboard to send a signal to read RAM. It'll simply have a path of electrical charge that leads from the motherboard's BUS through the CPU and out to the RAM.
Yeah, you are mistaken. That's exactly how a CPU works. They have capacitors that store electrical energy and then release it on demand. Its not all transistors in a CPU. You need diodes, capacitors and resistors to make all the various types of logic gates.
-
Holy goddamn this thread.
-
Like, unless I'm mistaken, a CPU isn't going to store a charge inside itself then release that charge to the motherboard to send a signal to read RAM. It'll simply have a path of electrical charge that leads from the motherboard's BUS through the CPU and out to the RAM.
Yeah, you are mistaken. That's exactly how a CPU works. They have capacitors that store electrical energy and then release it on demand. Its not all transistors in a CPU. You need diodes, capacitors and resistors to make all the various types of logic gates.
So a logic gate will output a pre-stored charge on the output? Why wouldn't it just use the input signals? Unless it's a NOT and needs to output from nothing to something.
-
Holy goddamn this thread.
Hey, I'm learning a lot. I'm happy.
-
So a logic gate will output a pre-stored charge on the output? Why wouldn't it just use the input signals? Unless it's a NOT and needs to output from nothing to something.
Because it might be waiting for something else to complete a calculation, and it needs to put that info somewhere.
-
Holy goddamn this thread.
Hey, I'm learning a lot. I'm happy.
You're learning a lot of extremely wrong things.
-
Holy goddamn this thread.
Hey, I'm learning a lot. I'm happy.
You're learning a lot of extremely wrong things.
Well stop posting wrong things and make it easier for him.
-
While I'm gonna side with Thork that the electrical energy used must be more than the wattage of heat being outputted (based on what I'm reading) I'm also gonna not suggest the total energy it draws is somehow double that.
-
While I'm gonna side with Thork that the electrical energy used must be more than the wattage of heat being outputted (based on what I'm reading) I'm also gonna not suggest the total energy it draws is somehow double that.
Reading this thread or some outside source?
Just seems a quick way to disagree with everyone in one go.
-
NERD FiGHT!!!!!
-
Baby Thork,
You are wrong here. Nearly 100% of the energy that goes into a computer is released as heat eventually (in a short timeframe, like a day). The CPU's TDP is a measure of essentially how much energy it turns into heat. Sure, there's a little bit that's transferred around the motherboard for signalling, but mind you, these communication lines have relatively high impedances compared to a power supply; probably no more than a quarter of a watt goes into signalling. Why? Well, it turns out that such energy would be used for ... heating other components, particularly in sensitive areas. There's just no need for any signalling pathway to use a lot of energy.
Switching transistors does not store useful energy. Switching losses happen because there is a voltage across the source and drain (or collector and emitter with bipolar transistors) when the gate (base) is driven on/off. During the switching delay, current leaks across a voltage drop. Transistors also have leakage losses. All of these losses turn to heat. This is why CPUs need lots of energy.
DRAM stands for dynamic random access memory. Why is it called dynamic? Because it needs to be "refreshed" as its capacitors leak current. That leakage current obviously doesn't go toward charging batteries (otherwise you could decrease the refresh rate by ditching the batteries). It goes to heat.
As for outside signalling, the energy use is minimal, and those devices primarily convert energy to heat as well. The one exception is if you decide to use your computer ports that contain supply rails to charge batteries. Yeah, I haven't seen that in common use.
Another way to look at a computer is to find all of the places it can store energy:
1. Capacitors (very small amount of energy)
2. RTC battery
3. Extremely small magnetic fields (and these will re-release the energy as soon as the computer is turned off).
The rest of the energy must go somewhere, and the only place for it to go is heat.
And you are wrong about CPU capacitors. They store minimal energy. Even the best capacitors today in terms of energy density store little compared to batteries, and the ones in the CPU are not made for energy storage. We are talking about micro/nanojoules that they can store.
Most of the CPU is transistor-transistor logic, so it is indeed mostly transistors. From transistors, gates are built. Then we use an HDL to describe electronic circuits, which is then synthesized into transistor layout.
The TDP quoted in the CPU manual is lower than power draw because of inefficiencies in the power supply and use in other parts of the computer (GPUs, spinning hard drives, lasers), which convert the electrical energy to -you guessed it- heat. If we sum up the power dissipation for all of the components and factor in the inefficiency of the power supply, we get essentially how much power is drawn.
The main takeaway is computers are essentially heaters that perform useful computation. They are, however, far less efficient than your air conditioner at heating.
-
There are a number of reasons why this cannot be.
The first is that electricity runs on a circuit. And to do that, some of the electrons have to make their way around the circuit AND BACK TO POWER SUPPLY. Not all of the energy drawn by a CPU is used. Much returns to the power supply. Otherwise you only need one wire to make a circuit, not two. That is energy dissipated not as heat on the die.
So electrical watts in, already doesn't equal watts of heat out.
Next is Landauer's principle stating that electrical energy used is proportional to ln2 of the heat energy out. Not 1:1.
https://en.wikipedia.org/wiki/Landauer%27s_principle
Next is assuming the transistor contains all the resistance in the circuit, and therefore it should have to dissipate all the power in that circuit. But that is not the case.
Take a simple one transistor switch for example, make Vcc 12 Volts and the load resistor 10 ohms, now assume you want 100 watts from the load (it could be a light globe a motor or inputs to other transistors.
So you need to have 10 volts across the load resistor, you have 12 Volts as VCC, so the transistor (collector/emitter) from one let of a two resistor voltage divider.
So you have 10 volts across the load resistor, and 2volts across the transistor, but still only 10 amps into the load.
The transistor is only dropping 2 volts, (not 10 volts) (12volts - 2 Volts = 10 volts). But the transistor is dissipating 120 watts so 100watts goes into the load and 20 watts goes into heat in the transistor.
Now if the transistors collector/emitter resistance was the same as the load resistance, you would require a VCC of 20 volts and half the power would go to the load and half to the transistor.
The efficiency of a transistor is that it has a low on resistance and therefore low voltage (and power) consumption.
Finally assuming a processor is all transistors when 30% of its power used gets taken by the cache (a storage device) is semi-retarted. A processor has lots of parts, not just cores.
(https://www.cs.uaf.edu/2009/fall/cs441/proj1/russell/images/df94g6m6_4gxrw53gk_b.png)
Also I think reducing the PCIe lanes to "probably no more than a quarter of a watt goes into signalling" is a reach.
TDP is Watts in joules and a guesstimate of how many heat watts you need to dissipate. Power draw is volts x amps to give you electrical watts required. The two are not the same.
-
TDP is Watts in joules and a guesstimate of how many heat watts you need to dissipate. Power draw is volts x amps to give you electrical watts required. The two are not the same.
(https://i.imgur.com/bxeAwbN.png)
-
Stop talking about electrons. When it comes to the behaviour of electrical energy in a circuit, the individual charge carriers are irrelevant. What you are saying is equivalent to claiming that pollution in rush hour doesn't happen because the cars go back home at the end of the day.
You obviously don't have the basic foundational knowledge required to understand how electricity works. I'm far from an expert myself, of course, but I'm amazed you can get an engineering degree with so little understanding of... well, anything. Please just stop before you sap what remains of everyone else's IQ.
-
Stop talking about electrons. When it comes to the behaviour of electrical energy in a circuit, the individual charge carriers are irrelevant. What you are saying is equivalent to claiming that pollution in rush hour doesn't happen because the cars go back home at the end of the day.
And you are claiming that all of the petrol goes into heating the engine and none of the energy does into driving you anywhere. You haven't even considered the kinetic energy of moving the switches.
This so unbelievably simple. Not all of a processors drawn power ends up as heat on the die. I think anyone with half a brain cell could work that out. There is no such thing as a perfect machine.
-
And you are claiming that all of the petrol goes into heating the engine and none of the energy does into driving you anywhere.
If the CPU's job was to impart kinetic energy unto the computer, this might be a valid analogy to make. Unfortunately for you, a CPU does a negligible amount of useful work, for the physical definition of "work".
-
You haven't even considered the kinetic energy of moving the switches.
This is the part where Thork tells us he thinks that transistors are literally a series of switches that get moved back and forth in a CPU, isn't it?
-
And you are claiming that all of the petrol goes into heating the engine and none of the energy does into driving you anywhere.
If the CPU's job was to impart kinetic energy unto the computer, this might be a valid analogy to make. Unfortunately for you, a CPU does a negligible amount of useful work, for the physical definition of "work".
Maybe your PC does a negligible amount of useful work. That I could believe.
-
This so unbelievably simple. Not all of a processors drawn power ends up as heat on the die. I think anyone with half a brain cell could work that out. There is no such thing as a perfect machine.
This continues to be your failing. The fact that CPUs dissipate virtually all their power as heat is a sign of their extreme imperfection. To say that they're not perfect hurts your argument.
This is the part where Thork tells us he thinks that transistors are literally a series of switches that get moved back and forth in a CPU, isn't it?
Transistors haven't been in computers all that long. It is a relatively new technology.
-
I'm still waiting for Thork to explain what he really meant by "You haven't even considered the kinetic energy of moving the switches."
-
This so unbelievably simple. Not all of a processors drawn power ends up as heat on the die. I think anyone with half a brain cell could work that out. There is no such thing as a perfect machine.
This continues to be your failing. The fact that CPUs dissipate virtually all their power as heat is a sign of their extreme imperfection. To say that they're not perfect hurts your argument.
This is the part where Thork tells us he thinks that transistors are literally a series of switches that get moved back and forth in a CPU, isn't it?
Transistors haven't been in computers all that long. It is a relatively new technology.
No, it would make them perfect heaters. They aren't.
I'm still waiting for Thork to explain what he really meant by "You haven't even considered the kinetic energy of moving the switches."
You're still going around in circles.
-
So I'm going to go ahead and guess that you:
1. hurriedly googled what a transistor actually is.
2. realized it doesn't actually move.
3. realized you now have to brush that comment under the rug.
Now Thork, this may come as a shock to you, but everything you said in your previous comment was just as wrong as saying that transistors are a series of moving switches. Why you seem to think your software engineering degree or whatever suddenly makes you a computer hardware engineer is beyond me, but this thread is making you look like an absolute fool who understands less about electricity than a elementary school student with a my-first-circuit kit.
I would also like to point out that a 100% efficient heater is of course possible. Attach a battery to a resistor. You now have a 100% efficient heater, and 100% of the battery's stored chemical energy will become heat eventually. You're welcome.
-
I pasted the wrong image. ...
I will tell you what.
I have posted tons of links through out this thread and you just keep saying ... nah, you are wrong. That's just your shitty uneducated opinion. How about you link to something that backs YOUR position. Just one link. Cos I've seen fuck all out of any of you.
Show me TDP = power draw.
-
Show me TDP = power draw.
Nobody is saying that it does. That doesn't make your fantasy world in which CPUs absorb energy and violate fundamental laws of physics real.
-
I pasted the wrong image. ...
I will tell you what.
I have posted tons of links through out this thread and you just keep saying ... nah, you are wrong. That's just your shitty uneducated opinion. How about you link to something that backs YOUR position. Just one link. Cos I've seen fuck all out of any of you.
Show me TDP = power draw.
Heat dissipated in a circuit always equals the power draw of that circuit. Literally an introductory concept of electronic physics. TDP is just a theoretical amount, Thork. No CPU actually sticks to its TDP 100% of the time.
-
Show me TDP = power draw.
Nobody is saying that it does. That doesn't make your fantasy world in which CPUs absorb energy and violate fundamental laws of physics real.
Yes they were. Pete said
I will, however, invite you to consider the fact that you're running one of the coolest CPUs available for desktops in the past 2 decades.
And I rebutted that by saying my processor had TDP of 65W and therefore was no cooler than 65W desktop processors of 20 years ago.
And then you all didn't know the difference between TDP and power draw.
Back on topic... my PC is not running a cool CPU. Its just got better cooling than was possible 20 years ago because of heat pipe tech and so AIOs pushing desktop performance is now a thing.
don't just start lying in the thread.
-
I pasted the wrong image. ...
I will tell you what.
I have posted tons of links through out this thread and you just keep saying ... nah, you are wrong. That's just your shitty uneducated opinion. How about you link to something that backs YOUR position. Just one link. Cos I've seen fuck all out of any of you.
Show me TDP = power draw.
Heat dissipated in a circuit always equals the power draw of that circuit. Literally an introductory concept of electronic physics. TDP is just a theoretical amount, Thork. No CPU actually sticks to its TDP 100% of the time.
Not on the CPU die it doesn't for all the reasons I gave earlier. You can't say every watt that goes into a cpu comes out as a perfect full watt of heat. That's just bollocks. SHOW ME A LINK!
-
And I rebutted that by saying my processor had TDP of 65W and therefore was no cooler than 65W desktop processors of 20 years ago.
Indeed, your high-end CPU is (probably) no cooler than a very-low-end CPU of yesteryear. The fact that you keep trying to conflate them is hilarious.
Its just got better cooling than was possible 20 years ago because of heat pipe tech and so AIOs pushing desktop performance is now a thing.
This continues to be factually incorrect. Heat pipes are not a new invention, nor have they seen any significant developments in the past decade. If you're so desperate as to compare it to stuff from 1998, well, you're still wrong, but slightly less wrong.
-
And I rebutted that by saying my processor had TDP of 65W and therefore was no cooler than 65W desktop processors of 20 years ago.
Which is not necessarily correct. Have you considered the problem may be that you don't know the difference between TDP and heat output?
-
And I rebutted that by saying my processor had TDP of 65W and therefore was no cooler than 65W desktop processors of 20 years ago.
Yes and that's still wrong because you don't understand that a CPU today doesn't have to use its max energy rating all the time unlike older CPUs. A modern 4 GHz CPU can downclock itself to 1 GHz while idling or performing easy tasks, which saves a tremendous amount of energy, and thereby heat.
And then you all didn't know the difference between TDP and power draw.
Heat dissipated and power used are the same thing, Thork. The problem here is that you insisted on using the power drawn by the wall, in which we all had to explain to you that is heat dissipated by your PSU, not your CPU. If your CPU draws 120W from the wall, but dissipates 65W of heat, then that means the other 55W is heat coming from some other components (namely, the PSU).
Back on topic... my PC is not running a cool CPU. Its just got better cooling than was possible 20 years ago because of heat pipe tech and so AIOs pushing desktop performance is now a thing.
don't just start lying in the thread.
Your PC is most definitely running a cool CPU, and it's definitely not thanks to the """"relatively new"""" discovery that copper can transfer heat from one place to another...
-
thanks to """"relatively new"""" discovery that copper can transfer heat from one place to another...
Now, Rushy, don't lie. It's a copper tube filled with a fluid.
-
And I rebutted that by saying my processor had TDP of 65W and therefore was no cooler than 65W desktop processors of 20 years ago.
Which is not necessarily correct. Have you considered the problem may be that you don't know the difference between TDP and heat output?
This is mind boggling nonsense. If my PC has a TDP of 65W, I can expect it to warm my PC as much as any other 65W TDP processor on any given day. I shouldn't expect it to run cooler than a 35W TDP processor of yesteryear.
And I rebutted that by saying my processor had TDP of 65W and therefore was no cooler than 65W desktop processors of 20 years ago.
Yes and that's still wrong because you don't understand that a CPU today doesn't have to use its max energy rating all the time unlike older CPUs. A modern 4 GHz CPU can downclock itself to 1 GHz while idling or performing easy tasks, which saves a tremendous amount of energy, and thereby heat.
Incorrect. I'd love you to find me a 4GHz processor that drops to 1Ghz. They don't. 2.2GHz from 4 GHz is still a hell of a drop these days. Also as someone who probably overclocks their potato, you should know that locks your clock speed all the time.
And then you all didn't know the difference between TDP and power draw.
Heat dissipated and power used are the same thing, Thork. The problem here is that you insisted on using the power drawn by the wall, in which we all had to explain to you that is heat dissipated by your PSU, not your CPU. If your CPU draws 120W from the wall, but dissipates 65W of heat, then that means the other 55W is heat coming from some other components (namely, the PSU).
No. You guys are the ones who quoted power draw figures from CPUs and made out they were how much heat needed to be dissipated.
Back on topic... my PC is not running a cool CPU. Its just got better cooling than was possible 20 years ago because of heat pipe tech and so AIOs pushing desktop performance is now a thing.
don't just start lying in the thread.
Your PC is most definitely running a cool CPU, and it's definitely not thanks to """"relatively new"""" discovery that copper can transfer heat from one place to another...
Yeah, I'd like a link out of you. I know you never back anything you say with evidence, but you write so much shit. Show me that the ryzen 1400 is a cool chip please.
thanks to """"relatively new"""" discovery that copper can transfer heat from one place to another...
Now, Rushy, don't lie. It's a copper tube filled with a fluid.
Kind of important to note that. That used to be an expensive thing to manufacture.
-
Incorrect. I'd love you to find me a 4GHz processor that drops to 1Ghz. They don't. 2.2GHz from 4 GHz is still a hell of a drop these days. Also as someone who probably overclocks their potato, you should know that locks your clock speed all the time.
Thork, my current processor, an 8700k, drops from 5GHz to 800MHz when idle. Could you really, just really please stop making shit up...
-
Incorrect. I'd love you to find me a 4GHz processor that drops to 1Ghz. They don't. 2.2GHz from 4 GHz is still a hell of a drop these days. Also as someone who probably overclocks their potato, you should know that locks your clock speed all the time.
Thork, my current processor, an 8700k, drops from 5GHz to 800MHz when idle. Could you really, just really please stop making shit up...
Prove that.
(http://i68.tinypic.com/2dv7o9j.jpg)
Here's my 1.8GHz base clock PC at 4%. Never gets below 1.2 GHz.
Show me your 5Ghz processor at 800Mhz please.
-
(https://i.imgur.com/OryTcFB.jpg)
(https://i.imgur.com/0X9zhoA.jpg)
-
Kind of important to note that. That used to be an expensive thing to manufacture.
Yes, in the 1940s it was. Keep in mind that you're arguing that powerful AiOs (not to be confused with laptops - you like to make that conflation too) were not recently possible, even at the time when the technology you credit for this was commonplace in computing.
You also remarked that a significant amount of energy in CPUs is converted to kinetic energy. I'll just let that sink in.
Since you like links, perhaps others can explain this more simply: https://electronics.stackexchange.com/questions/79166/where-does-all-the-power-consumed-by-a-cpu-go - because the Law of Conservation of Energy is a thing, and because your computer doesn't power many things outside of its case, virtually all energy drawn will turn into heat inside the case.
-
(https://i.imgur.com/OryTcFB.jpg)
(https://i.imgur.com/0X9zhoA.jpg)
Task manager please. real time.
-
This is mind boggling nonsense. If my PC has a TDP of 65W, I can expect it to warm my PC as much as any other 65W TDP processor on any given day. I shouldn't expect it to run cooler than a 35W TDP processor of yesteryear.
So you actually don't know the difference between TDP and heat output. Colour me surprised.
-
This is mind boggling nonsense. If my PC has a TDP of 65W, I can expect it to warm my PC as much as any other 65W TDP processor on any given day. I shouldn't expect it to run cooler than a 35W TDP processor of yesteryear.
So you actually don't know the difference between TDP and heat output. Colour me surprised.
No one knows the heat output at any give moment you muppet. TDP is the design specification saying put a cooler on this that dissipates 65W of heat for this type of processor. Ergo, you need a cooling solution deal with 65W. Same as you did 20 years ago with a 65W processor.
-
No one knows the heat output at any give moment you muppet.
Of course you do. You can measure power drawn with a decent power supply, and, since physics applies, you'll get the heat output of your machine, within a very reasonable margin of error.
TDP is the design specification saying put a cooler on this that dissipates 65W of heat for this type of processor. Ergo, you need a cooling solution deal with 65W. Same as you did 20 years ago with a 65W processor.
I'm glad you've finally admitted that the cooling solutions didn't change much.
-
Thork, how am I supposed to record a video of my processor idling? If I begin recording my desktop, then by definition my computer is no longer idling.
Also, never use the task manager to look up your clocks. It's just asking Windows what it thinks the clockrate is, it doesn't actually ping any hardware. It is absolute trash when it comes to measuring any kind of performance metric. Use CPU-Z or PassMark.
-
No one knows the heat output at any give moment you muppet.
Of course you do. You can measure power drawn with a decent power supply, and, since physics applies, you'll get the heat output of your machine, within a very reasonable margin of error.
Nice try. Watts in aren't perfectly transferred to watts of heat out on the die. Does your processor not have any PCIe lanes? Does it not send any of its power to the board, other components or back to the power supply? Is it just there to heat the computer up?
TDP is the design specification saying put a cooler on this that dissipates 65W of heat for this type of processor. Ergo, you need a cooling solution deal with 65W. Same as you did 20 years ago with a 65W processor.
I'm glad you've finally admitted that the cooling solutions didn't change much.
They very fact you can now have desktop processors in small devices proves how wrong you are.
Thork, how am I supposed to record a video of my processor idling? If I begin recording my desktop, then by definition my computer is no longer idling.
Also, never use the task manager to look up your clocks. It's just asking Windows what it thinks the clockrate is, it doesn't actually ping any hardware. It is absolute trash when it comes to measuring any kind of performance metric. Use CPU-Z or PassMark.
Well I managed it on the page before. I'll take that as you not succeeding and being proved wrong. All I wanted was
(http://i68.tinypic.com/2dv7o9j.jpg)
Here's my 1.8GHz base clock PC at 4%. Never gets below 1.2 GHz.
Show me your 5Ghz processor at 800Mhz please.
-
Nice try. Watts in aren't perfectly transferred to watts of heat out on the die.
I don't solely care about your die, and neither do you. Your AiO is tightly packed - unless you're arguing that this heat somehow escapes the computer's case, it still needs to be dissipated. Otherwise, your case is effectively a nice, small greenhouse.
They very fact you can now have desktop processors in small devices proves how wrong you are.
But this was true 10-15 years ago, too.
Well I managed it on the page before. I'll take that as you not succeeding and being proved wrong. All I wanted was
(http://i68.tinypic.com/2dv7o9j.jpg)
Here's my 1.8GHz base clock PC at 4%. Never gets below 1.2 GHz.
Show me your 5Ghz processor at 800Mhz please.
You do realise that this is entirely configurable, right?
-
Here's my computer using intel extreme tuning utilily.
(http://i65.tinypic.com/von04i.jpg)
Same results in real time. 5% used. doesn't get below 1.2Ghz.
Show me your 5Ghz processor at 800Mhz please.
-
2.2GHz from 4 GHz is still a hell of a drop these days.
mine idles at <1.4 ghz
(https://i.imgur.com/fhneWAb.png)
-
Thork. Take a deep breath. Now another one.
You can adjust your CPU's multiplier stepping. The fact that you chose to run with the defaults is your prerogative, but it proves absolutely zilch.
Also, stop using Intel's bloatware, you're hurting your already underpowered PC.
-
Yeah, gonna need to see some real time cpu usage. Rushy has been a while, hasn't he? Takes this long to make a screenshot?
-
Yeah, gonna need to see some real time cpu usage.
Define "real time cpu usage" - all you've given us so far is screenshots.
Do you want to come to my house and play with my BIOS settings? I can show you multipliers you haven't imagined before. If we try hard enough, my CPU might even vibrate.
-
TDP is Watts in joules and a guesstimate of how many heat watts you need to dissipate. Power draw is volts x amps to give you electrical watts required. The two are not the same.
(https://i.imgur.com/bxeAwbN.png)
Yeah.... Either bait or he's operating on a different level of logic where Joules =/= Joules if you have different types of energy.
-
Yeah, gonna need to see some real time cpu usage. Rushy has been a while, hasn't he? Takes this long to make a screenshot?
i did. see my screenshot.
-
Yeah.... Either bait or he's operating on a different level of logic where Joules =/= Joules if you have different types of energy.
Thork's entire gimmick here is to assert that conservation of energy does not apply because heat pipes. It's great and I never want it to end.
I'm sorry I ever doubted you, Thork.
-
Just a screen shot of something actively monitoring cpu speed. Not a screenshot of a default setting which is unattainable in the real world.
This isn't hard. Post a screenshot of an 8700k @5GHz doing 800mhz. Very simple request. I'll accept anything with a graph on it that plots performance over time.
Yeah, gonna need to see some real time cpu usage. Rushy has been a while, hasn't he? Takes this long to make a screenshot?
i did. see my screenshot.
and you don't have a 5Ghz processor not did it get as low as 800Mhz.
I want to see the silicon wonder Rushy has.
-
I want to see the silicon wonder Rushy has.
just to be clear, your original claim was that 4.0ghz to 2.2ghz was basically the limit of currentyear cpus. not sure why we need to wait for rushy to do anything since that clearly isn't close to true.
btw my cpu is a 4-year-old fx-6300.
-
and you don't have a 5Ghz processor not did it get as low as 800Mhz.
I want to see the silicon wonder Rushy has.
Don't worry - you're not the only person confused by this.
https://communities.intel.com/thread/124867
Seems to be a common issue among non-technical people.
-
I want to see the silicon wonder Rushy has.
just to be clear, your original claim was that 4.0ghz to 2.2ghz was basically the limit of currentyear cpus. not sure why we need to wait for rushy to do anything since that clearly isn't close to true.
btw my cpu is a 4-year-old fx-6300.
And I'm surprised to see it as low as 1.4Ghz.
But I'd really like to see the 5Ghz processor that goes down to 800Mhz.
-
But I'd really like to see the 5Ghz processor that goes down to 800Mhz.
Is the account of a confused father not enough for you? You heartless bastard.
-
But I'd really like to see the 5Ghz processor that goes down to 800Mhz.
Is the account of a confused father not enough for you? You heartless bastard.
No. Fuck him and his child. They probably got that number from CPU-Z not realising that isn't what the processor is actually doing.
Where is Rushy? I could have entered my bios and underclocked my PC in this time. How hard is a screenshot?
-
No. Fuck him and his child. They probably got that number from CPU-Z not realising that isn't what the processor is actually doing.
CPU-Z is de facto the industry standard for these measurements. You're really reaching here.
Where is Rushy? I could have entered my bios and underclocked my PC in this time. How hard is a screenshot?
Don't worry, he's been working hard with the BBC to produce a top-notch vidya for you.
-
Yeah, gonna need to see something in real time. Something with a graph. Not just a theoretical calculation with a bus speed and multiplier.
-
Yeah, gonna need to see something in real time.
You still haven't clarified what you mean by that. Do you want a video? A live stream with zappers for participants to answer questions? A visit to my house laced with sexual innuendos?
-
https://www.youtube.com/watch?v=VHGQsmCz8A4
This video isn't really for Thork, but for everyone who isn't Thork to learn something about how a computer actually works.
-
So ... your PC isn't at 800Mhz.
Whilst your "idle cores" go to 800Mhz, your active cores are still nailed at 2.2Ghz+. Hence the reason you can't get a real time graph showing 800Mhz. Your PC is always at 2.2Ghz or more.
So thanks for showing me I'm right. Cheers.
-
So ... your PC isn't at 800Mhz.
Whilst your "idle cores" go to 800Mhz, your active cores are still nailed at 2.2Ghz+. Hence the reason you can't get a real time graph showing 800Mhz. Your PC is always at 2.2Ghz or more.
So thanks for showing me I'm right. Cheers.
*This is what Thork actually believes*
-
So ... your PC isn't at 800Mhz.
What makes you think so? You wanted something that's not based on a theoretical calculation, and Rushy provided exactly that.
-
So ... your PC isn't at 800Mhz.
What makes you think so? You wanted something that's not based on a theoretical calculation, and Rushy provided exactly that.
No. I wanted something real world and not theoretical. I asked as much about 8 times.
So ... your PC isn't at 800Mhz.
Whilst your "idle cores" go to 800Mhz, your active cores are still nailed at 2.2Ghz+. Hence the reason you can't get a real time graph showing 800Mhz. Your PC is always at 2.2Ghz or more.
So thanks for showing me I'm right. Cheers.
*This is what Thork actually believes*
Yup. Do you have trouble reading numbers from a screen?
-
No. I wanted something real world and not theoretical. I asked as much about 8 times.
Well, exactly. CPU-Z produces its numbers by querying the hardware. taskmgr uses weighted averages and simplified maths. You keep stating what you want and complaining that people aren't showing you what you explicitly don't want.
-
An easier way to look at it would be through what governs the clock. The clock is always a multiple of the bus speed (it's quite literally impossible for it to be anything else in current CPU architectures). When you see things like task manager outputting 3.27, 1.92, 2.25, etc. these are quite frankly impossible clock numbers. The clock isn't actually at any of these numbers, but rather this is just an average of the clock speeds over time. It also doesn't help that a modern CPU can have multiple clock speeds thanks to advanced core architecture. Core 0, 1, 2 might be at 800 MHz while Core 3 is at 5000 MHz. Task manager would think this translates to 1.8 GHz clockspeed for the entire processor, which is why you should never trust what it says. It's gibberish.
-
...
Where and how the fuck did we go from "A CPU's Energy Draw" to "Look I can underclock it to 800 mhz!"? Did I skip a page? I think I skipped a page.
-
...
Okay so...
Isn't the CPU downgrading strictly as a power save feature and people like me who turn that off don't see it downstep? ( downloaded CPU-Z and checked)
And Thork's system is likely the same? Not conserving power by downgrading the clock speed of some cores when idle?
Also: Clock mulipliers are a thing, right? And they can be changed on the fly can't they? So a CPU core could run at a multiplier of x1, which would be whatever the FSB speed is. Yes?
-
...
Okay so...
Isn't the CPU downgrading strictly as a power save feature and people like me who turn that off don't see it downstep? ( downloaded CPU-Z and checked)
And Thork's system is likely the same? Not conserving power by downgrading the clock speed of some cores when idle?
Also: Clock mulipliers are a thing, right? And they can be changed on the fly can't they? So a CPU core could run at a multiplier of x1, which would be whatever the FSB speed is. Yes?
Many modern cpus will downclock when they idle to conserve energy. I highly suggest you don't turn this feature off, as it makes an enormous difference in how much energy your computer is using while idling, or even while doing simple tasks. And yes, you could run a CPU with a x1 multiplier, though I imagine you'll get some undesirable performance issues if that happens.
This makes a big difference because every time the clock signal goes from low-to-high or high-to-low then it consumes energy. By downclocking, you're allowing the computer to save energy when you don't need high clocks to keep up with whatever you're doing. The processor doesn't really need to check its current state, say, at a rate of 3 GHz, when it really isn't doing anything that requires that kind of throughput.
For a more in-depth approach to all of this, you can read about clock-gating here: https://www.allaboutcircuits.com/technical-articles/use-of-clock-gating-to-reduce-power-consumption/ (keep in mind this article is talking more about turning the clock on/off in embedded systems, not necessarily talking about clocks used by a x86 processor.)
The key takeaway from all of this is that a clock signal wastes energy, even when the circuit its governing isn't actually doing anything. The higher the clockrate, the more energy you waste.
-
Isn't the CPU downgrading strictly as a power save feature and people like me who turn that off don't see it downstep?
Why would you turn that off? You don't need your CPU's full power 99% of the time, because most programs are very rarely CPU-bound. Operating systems are very good at increasing the clock rate when needed, so there's no (or negligible) performance impact. You're just wasting electricity, as Rushy says, and possibly also making your fan louder than it needs to be.
-
This thread is as good as Infinity War
-
This thread is as good as Infinity War
Not until half the people die
-
Out of curiosity, Baby Thork, from what university did you get your degree? What engineering is it in? Software? Certainly not electrical.
You must certainly understand the conservation of energy, do you? There's nowhere in a computer that can store the energy that gets put in (say, 30A @ 12V = 360W = 360J/s). So it has to be released as heat. Even the vibrations and fan movement get converted to heat (otherwise the fans would move faster and the vibrations would get bigger over time).
Landauer's principle deals with entropy and how energy must be converted to heat to manipulate information. It also deals with energy amounts far smaller than are relevant.
I can also show you a CPU that downclocks from ~20 MHz to 1 Hz when I set it to do that. Just get me an FPGA and I'll program one in for you. No virtual memory, minimal I/O, no exceptions, and no context switching.
Now, as for your assertion that 1 joule of electricity is not equal to 1 joule of heat. This is false. There is nowhere in the computer that will store this joule. Any computer that didn't produce 1 joule of heat from the 1 joule of electricity would simply accumulate energy. Do you mean to tell me my laptop is a bomb? There is a set amount of energy that any computer can store before it must convert the rest to heat (or simply just draw less).
Of course, if I charge my phone from my computer, then obviously 1 watt of electricity in does not produce 1 watt of heat. But that's nitpicking and not touching on the fundamental misunderstanding you have of ... physics.
-
I'm not sure why you have a problem with this.
Some of the power goes to other components via PCIe and other i/o. Therefore it isn't going to be heat on the die, is it? The power was transferred elsewhere. Also there are other ways to dissipate energy, such as electromagentic force, kinetic energy (https://books.google.co.uk/books?id=AN77CwAAQBAJ&pg=PA115&lpg=PA115&dq=kinetic+energy+in+transistor&source=bl&ots=UEZcRdqevx&sig=r3sXyA29iPBFNwdkUWhqpAlfhIA&hl=en&sa=X&ved=0ahUKEwia386f_uHaAhUpL8AKHbkIB9A4ChDoAQgoMAA#v=onepage&q=kinetic%20energy%20in%20transistor&f=false) etc.
Not all the power into a CPU ends up as heat on the die. Its very simple.
-
I'm not sure why you have a problem with this.
Some of the power goes to other components via PCIe and other i/o. Therefore it isn't going to be heat on the die, is it? The power was transferred elsewhere. Also there are other ways to dissipate energy, such as electromagentic force, kinetic energy (https://books.google.co.uk/books?id=AN77CwAAQBAJ&pg=PA115&lpg=PA115&dq=kinetic+energy+in+transistor&source=bl&ots=UEZcRdqevx&sig=r3sXyA29iPBFNwdkUWhqpAlfhIA&hl=en&sa=X&ved=0ahUKEwia386f_uHaAhUpL8AKHbkIB9A4ChDoAQgoMAA#v=onepage&q=kinetic%20energy%20in%20transistor&f=false) etc.
Not all the power into a CPU ends up as heat on the die. Its very simple.
You're talking about heat, EM waves, and kinetic energy as if those things are actually different when they are actually all forms of heat. If heat isn't being dissipated by the CPU, then power isn't, either. I don't care what other components are consuming power on the board. The power consumed by the CPU equals its heat output. What you're telling us is the the CPU consumes 100W, only dissipating 60W? That's not possible. You're erroneously including the power consumption of other parts of the computer in the power consumption of the CPU. Basically this is exactly what your problem was when you posted those two pictures earlier and claimed they proved your point when they actually did the opposite.
-
I think you're all being too hard on poor Thork. If he wants to be wrong, that's his prerogative.
-
I'm not sure why you have a problem with this.
Some of the power goes to other components via PCIe and other i/o. Therefore it isn't going to be heat on the die, is it? The power was transferred elsewhere. Also there are other ways to dissipate energy, such as electromagentic force, kinetic energy (https://books.google.co.uk/books?id=AN77CwAAQBAJ&pg=PA115&lpg=PA115&dq=kinetic+energy+in+transistor&source=bl&ots=UEZcRdqevx&sig=r3sXyA29iPBFNwdkUWhqpAlfhIA&hl=en&sa=X&ved=0ahUKEwia386f_uHaAhUpL8AKHbkIB9A4ChDoAQgoMAA#v=onepage&q=kinetic%20energy%20in%20transistor&f=false) etc.
Not all the power into a CPU ends up as heat on the die. Its very simple.
Ok, seriously, how much power do you think a CPU sends when it passes instructions and data between itself and RAM or the ALU or whatever component its talking to?
And given that data comes back TO the CPU, how much more data does a CPU send out than it takes it? Cause it should be more in than out. We don't take tiny data and make them bigger, we take big data and use it then tell something what to do.
So give us that, will ya?
-
Many modern cpus will downclock when they idle to conserve energy. I highly suggest you don't turn this feature off, as it makes an enormous difference in how much energy your computer is using while idling, or even while doing simple tasks. And yes, you could run a CPU with a x1 multiplier, though I imagine you'll get some undesirable performance issues if that happens.
This makes a big difference because every time the clock signal goes from low-to-high or high-to-low then it consumes energy. By downclocking, you're allowing the computer to save energy when you don't need high clocks to keep up with whatever you're doing. The processor doesn't really need to check its current state, say, at a rate of 3 GHz, when it really isn't doing anything that requires that kind of throughput.
For a more in-depth approach to all of this, you can read about clock-gating here: https://www.allaboutcircuits.com/technical-articles/use-of-clock-gating-to-reduce-power-consumption/ (https://www.allaboutcircuits.com/technical-articles/use-of-clock-gating-to-reduce-power-consumption/) (keep in mind this article is talking more about turning the clock on/off in embedded systems, not necessarily talking about clocks used by a x86 processor.)
The key takeaway from all of this is that a clock signal wastes energy, even when the circuit its governing isn't actually doing anything. The higher the clockrate, the more energy you waste.
Isn't the CPU downgrading strictly as a power save feature and people like me who turn that off don't see it downstep?
Why would you turn that off? You don't need your CPU's full power 99% of the time, because most programs are very rarely CPU-bound. Operating systems are very good at increasing the clock rate when needed, so there's no (or negligible) performance impact. You're just wasting electricity, as Rushy says, and possibly also making your fan louder than it needs to be.
Welp... I turned it back on. Or what I thought was it. (Green Power!)
Turns out... it doesn't seem to do much. I'm still idling at full power according to Cpu-Z.
-
Not all the power into a CPU ends up as heat on the die.
The die, by itself, is irrelevant. Stop backpedalling and take your failure like a man.
Heat which ends up somewhere else inside your tightly-packed AiO still needs to find its way out of the case. This, by the way, is another fantastic reason for why your "heat pipes" ramble is completely misplaced.
-
Not all the power into a CPU ends up as heat on the die.
The die, by itself, is irrelevant. Stop backpedalling and take your failure like a man.
Heat which ends up somewhere else inside your tightly-packed AiO still needs to find its way out of the case. This, by the way, is another fantastic reason for why your "heat pipes" ramble is completely misplaced.
How is the die irrelevant when talking about TDP of a processor? You don't even know what you are arguing any more.
Also an AIO isn't tighly packed. Its has a 27inch screen and is 2 inches deep. Its got about 10 times as much space inside as a laptop, yet uses laptop components. The problem is the CPU and GPU (desktop components). And those have high TDPs that are dealt with by the heat pipes.
-
How is the die irrelevant when talking about TDP of a processor? You don't even know what you are arguing any more.
The die by itself is irrelevant. Thork, you're gonna struggle in Brexit Britain if you don't learn our language.
Also an AIO isn't tighly packed. Its has a 27inch screen and is 3 inches deep. Its got about 10 times as much space inside as a laptop
Which is why I repeatedly told you to fuck off when you tried bringing up laptops.
yet uses laptop components.
No, your argument is the opposite of that. If you meant to say the opposite of what you said - well, that would explain why you're so horrendously wrong, but it doesn't make you look any better.
-
How is the die irrelevant when talking about TDP of a processor? You don't even know what you are arguing any more.
The die by itself is irrelevant. Thork, you're gonna struggle in Brexit Britain if you don't learn our language.
Don't you tell me how to speak English you dirty immigrant! >o< It is not irrelevant when that 65W TDP of heat is the thing you have to deal with. <removed, R4> -junker
Also an AIO isn't tighly packed. Its has a 27inch screen and is 3 inches deep. Its got about 10 times as much space inside as a laptop
Which is why I repeatedly told you to fuck off when you tried bringing up laptops.
You still can't get desktop fans inside them ... you have to use heatpipes like on A LAPTOP!
yet uses laptop components.
No, your argument is the opposite of that. If you meant to say the opposite of what you said - well, that would explain why you're so horrendously wrong, but it doesn't make you look any better.
The sky is blue. Would you like to make an 8 page thread disputing that as well?
-
It is not irrelevant when that 65W TDP of heat is the thing you have to deal with.
Yes, now you're back from irrelevant to incorrect. 65W TDP is what the cooling solution directly interfacing with the CPU has to handle. It is not the full story when it comes to heat dissipation inside an AiO.
But it is how the AIO manages to deal with a desktop CPU.
You still can't get desktop fans inside them ... you have to use heatpipes like on A LAPTOP!
And? It's not like heat pipes are particularly new or exciting.
They are both new and exciting. They are what make time travel small form factor PCs possible. Show me a sff PC from 1995 with a full blown desktop CPU in it.
The sky is blue. Would you like to make an 8 page thread disputing that as well?
Only if you open it by saying "the sun is a peculiar shade of lime green, which is caused by a recent invention called the windmill" and then act very surprised when no one agrees with you. As you can see, I'm a sucker for these.
Bonus points if you get LD to believe you for a minute or two.
I'm happy to state the world is flat, and you never have any issue argeeing with me on that point.
-
A chip that isn't 'the coolest in decades' as you earlier bungled
That was you, darling.
No, I'm not running one of the coolest CPUs in decades.
I said this:
you're running one of the coolest CPUs available for desktops in the past 2 decades
It is not irrelevant when that 65W TDP of heat is the thing you have to deal with.
Yes, now you're back from irrelevant to incorrect. 65W TDP is what the cooling solution directly interfacing with the CPU has to handle. It is not the full story when it comes to heat dissipation inside an AiO.
You still can't get desktop fans inside them ... you have to use heatpipes like on A LAPTOP!
And? It's not like heat pipes are particularly new or exciting.
The sky is blue. Would you like to make an 8 page thread disputing that as well?
Only if you open it by saying "the sun is a peculiar shade of lime green, which is caused by a recent invention called the windmill" and then act very surprised when no one agrees with you. As you can see, I'm a sucker for these.
Bonus points if you get LD to believe you for a minute or two.
-
Well now you are deleting posts. You can see above my responses.
-
Well now you are deleting posts.
A minor technical difficulty, I wouldn't want to bother you with details, lest you might start vibrating. My post is still there in its entirety, with some additions for your enjoyment. You just happened to be wrong on a couple more things, so I added them for you :)
-
They are both new and exciting. They are what make time travel small form factor PCs possible. Show me a sff PC from 1995 with a full blown desktop CPU in it.
Thork, let me present to you the Unisys CWD-4002, a small form factor PC from 1995 with a full blown i486 desktop CPU. :)
https://www.youtube.com/watch?v=ddgmzmw6_qE
-
They are both new and exciting. They are what make time travel small form factor PCs possible. Show me a sff PC from 1995 with a full blown desktop CPU in it.
Thork, let me present to you the Unisys CWD-4002, a small form factor PC from 1995 with a full blown i486 desktop CPU. :)
https://www.youtube.com/watch?v=ddgmzmw6_qE
Had a TDP of 10-12W. Doesn't need heatpipes like the 65W monster in my AIO. ;)
-
Had a TDP of 10-12W. Doesn't need heatpipes like the 65W monster in my AIO. ;)
That's not what you asked for though is it.
-
I suppose that depends on what you hope to get out of a debate with someone online. If you just want to win, don't ever engage me. I'll never admit it even if I'm losing.
-
He realized he's wrong about five pages ago. This is just an effort to convince us he's trolling instead of just accepting that he's an idiot.
-
He realized he's wrong about five pages ago. This is just an effort to convince us he's trolling instead of just accepting that he's an idiot.
>implying doubt on the subject
-
So back to the subject of me getting a new PC ...
What I have ordered comes with an m.2 ssd. Reads and write around 500Mb/s.
Having currently got an NVMe in my NUC, I'm not keen on this backward step.
I'm thinking of dropping a Samsung 970 250GB card in ... this means new windows installation and voiding my dell warrenty... but it also means no Dell crapware. So its two birds with one stone. Think that one reads at 3200 Mb/s. 6 times faster.
(http://images.samsung.com/is/image/samsung/sg-970-evo-nvme-m2-ssd-mz-v7e250bw-lperspectiveblack-98249876?$PD_GALLERY_JPG$)
I don't want to take the one out my NUC. My girlfriend is having this PC when I'm done with it. Seems a hassle to give her the ssd in the new PC and then load everything back on for her. The new NVMe should last about 3 years (I usually get a new computer every 3 years or less).
-
Don't you tell me how to speak English you dirty immigrant! >o< It is not irrelevant when that 65W TDP of heat is the thing you have to deal with. <removed, R4> -junker
Cool it with the personal attacks, even if it is "just a joke." And don't put any information about people that can identify them unless they have explicitly permitted it. Warned.
-
I don't think when there are around 1 million immigrants from said nation you can't say that information is 'personally identifiable'. Posting images of my face, as you have done with no reprimand ... that's different. I'm the only one with my face.
Yes, the 'personal attack' was a joke. I'm sure Pete didn't view it any other way, but I'll respect feelings and all that other PC nonsense for the future.
You heartless bastard.
One might argue this is a personal attack as it is aimed squarely at me and calls into question my legitimacy ... but its also a joke so ... :-\
-
I don't think when there are around 1 million immigrants from said nation you can't say that information is 'personally identifiable'. Posting images of my face, as you have done with no reprimand ... that's different. I'm the only one with my face.
You had more specific information than "immigrant," which is why I edited it out (and won't repeat it here). Also, I have re-posted your previously posted pictures on a rare occasion (I wonder who put them up in the first place). And I have taken down any post you have requested. It is one of the reasons I rarely bother engaging with you, as you are emotionally immature (at least here), as well as wildly inconsistent and hypocritical.
Yes, the 'personal attack' was a joke. I'm sure Pete didn't view it any other way, but I'll respect feelings and all that other PC nonsense for the future.
The report I was responding to was specifically about rule 4, but I figured I may as well cover the other rules while I was here.
-
this means [...] voiding my dell warrenty
That's not strictly right, and I'm not just referring to your inability to spell things. While the UK has valiantly resisted sensible consumer rights, thanks to pressure from the evil bureaucrats of the EU, your computer will still be covered by warranty for any failures that aren't directly linked to you shoving a new SSD in it.
https://www.mirror.co.uk/money/what-really-invalidates-warranty-you-10180230
An even stricter standard has recently been adopted in the US
http://www.bbc.co.uk/news/technology-43724348
It seems that Dell's general practice falls short of UK standards, but if you suddenly decide that you don't fancy a fiery dispute in case of a fault, you can just swap the disks back.
https://www.dell.com/community/Customer-Care/Hard-Drive-Warranty-Question/td-p/4564611
-
I don't think when there are around 1 million immigrants from said nation you can't say that information is 'personally identifiable'. Posting images of my face, as you have done with no reprimand ... that's different. I'm the only one with my face.
You had more specific information than "immigrant," which is why I edited it out (and won't repeat it here).
Yes, I mentioned which nation he was from. That narrows him down to one in a million.
Also, I have re-posted your previously posted pictures on a rare occasion
From an account I closed over a year before.
It is one of the reasons I rarely bother engaging with you, as you are emotionally immature (at least here), as well as wildly inconsistent and hypocritical.
That suits me. You break more rules here than most of the regulars and there is no comeback for the rest of us as you are a mod. Hypocrite, thy name is Junker.
this means [...] voiding my dell warrenty
That's not strictly right, and I'm not just referring to your inability to spell things. While the UK has valiantly resisted sensible consumer rights, thanks to pressure from the evil bureaucrats of the EU, your computer will still be covered by warranty for any failures that aren't directly linked to you shoving a new SSD in it.
https://www.mirror.co.uk/money/what-really-invalidates-warranty-you-10180230
An even stricter standard has recently been adopted in the US
http://www.bbc.co.uk/news/technology-43724348
It seems that Dell's general practice falls short of UK standards, but if you suddenly decide that you don't fancy a fiery dispute in case of a fault, you can just swap the disks back.
https://www.dell.com/community/Customer-Care/Hard-Drive-Warranty-Question/td-p/4564611
I think it does break the warranty because I have to open the back up. Then wipe all their 'special' software that keeps the PC working oh so well.But yeah, I was thinking if I kept the other disk as they sent it, how could they know if I swapped it back?
-
I think it does break the warranty because I have to open the back up. Then wipe all their 'special' software that keeps the PC working oh so well.But yeah, I was thinking if I kept the other disk as they sent it, how could they know if I swapped it back?
Yup. Opening the back used to be controversial, but there is now ample precedent for it. Manufacturers and vendors might try to bullshit you in the hopes you'll back down, but ultimately you're in the right. Removing their crapware doesn't affect the hardware, so they can't say anything about that, and swapping the SSD should still mean all your other parts are covered (though, again, they might try throwing a hissy)
-
additional whining and ranting
If you have an issue, take it to S&C (or delete your account again as usual). Arguing moderation over a simple warning just derails the thread further. So have another warning, and keep your posts on topic, or take it to the appropriate forum.
-
I think it does break the warranty because I have to open the back up. Then wipe all their 'special' software that keeps the PC working oh so well.But yeah, I was thinking if I kept the other disk as they sent it, how could they know if I swapped it back?
Yup. Opening the back used to be controversial, but there is now ample precedent for it. Manufacturers and vendors might try to bullshit you in the hopes you'll back down, but ultimately you're in the right. Removing their crapware doesn't affect the hardware, so they can't say anything about that, and swapping the SSD should still mean all your other parts are covered (though, again, they might try throwing a hissy)
I thought they'll be more picky over me using manufacturer drivers instead of their special sauce relabeled ones, but they've made it pretty easy for me to get the new drivers. They seem to be mostly direct from the manufacturer.
http://www.dell.com/support/home/uk/en/ukdhs1/product-support/product/inspiron-27-7775-aio/drivers
Just found out it has been shipped. Squeeeeeeee! :D
-
"Hmm, how can I make this Dell product even shittier? Oh, I know, I'll install Dell's drivers on it!"
-
"Hmm, how can I make this Dell product even shittier? Oh, I know, I'll install Dell's drivers on it!"
Having worked on Dell for years, Dell drivers are mostly manufacturer driver. I never had an issue getting the drivers from their website. You can update later if they're out of date.
-
Opted for a Dell Inspiron 27 7775.
Ryzen 1400
RX 580
8 GB DDR4 2400 Ram
128 Gb SSD
1TB HDD
Wireless card
Wondows 10
27" UHD (3840x2160) display
That is quite a monstrosity you've got there. Also, I was playing some games just now and my CPU almost vibrated out of its socket. I'm already rocking about 10 heat pipes atm. How many more you reckon I'll need to get this baby under control?
It looks like you use Brave, though. That's p dope.
-
God this thread sucks.
I recently bought myself a laptop that folds all the way back to become a tablet. It's awesome even though I never use it as a tablet. I only do it in front of people so they know I'm on the cutting edge of technology.