Slashdot Mirror
AMD 90nm Evaluated
muyuubyou writes "The Tech Report has measured the new 90nm A64 3500+ against its 130nm counterpart and a Pentium 4 3.6Ghz 90nm.
AMD looks way ahead in the 90nm process especially when it comes to power consumption.
Note these are consumptions for the entire system including GeForce 6800 GTs and hefty PSUs. RTFineShortArticle for more detail on the configuration.
Leaving the PC on overnight is probably not a good idea with these new Pentium 4s."
That's impressive. Of course, since it's total system wattage, it'd be nice to have some information about disk usage over the period of time, etc.
I like, though, that the 130nm Athlon 64 is still better than the 90nm P4. It might just be time to buy another desktop.
It' snot going to make THAT much of a difference on your electric bill.
Now what I want to see is an analysis of the possible benefits to notebooks, specifically in extending battery life. Intel's Centrino seems to be doing fairly well in that department, but where is AMD's response?
There's a Mercedes gap too. I want one and can't afford one, but it's not government's job to do anything about it.
With every computing consuming more and more power, its looking like we will need a wind turbine or solar PV array for anyone to run a decent sized network of computers at home.
Anyone currently doing this? I'm thinking of installing a turbine, but unsure of where to start out.
IntechHosting - Free domain, 2GB, PHP, £4.95/$8.95
Can someone elighten me on this? Is there a reason why the SpeedStep and other power-saving methods that are used in most laptops can't be adapted to desktop systems?
The old joke is that all CPUs sleep at the same speed, but after seeing the power consumption graph on this site, it's obvious that "power-hungry CPU" doesn't just mean high heat during gaming. This suckers are hungry even while doing nothing at all
--
Free gmail invites
They found that the Prescott P4, with its emphasis on Mhz, puts out a lot of heat in spite of its 90 nm architecture. The new 90 nm AMD 64 is cooler and uses less energy than the 130 nm version. Great.
But what about performance? The new 90 nm Pentium M processors, the one with the funky names, aren't doing as well in terms of performance scalability because of electron leakage issues. Any such concerns here? How fast can the 90 nm Athlon 64 core go before it dies?
A NYC lawyer blogs. http://www.chuangblog.com/
As the proud owner of an old-school Duron, my computer isn't a problem. However, living in a bachelor pad which happens to be filled with geeks, we have a cable modem.
This makes our house faster than our friends' houses. So their computers migrate there also. And the bastards never remember to turn them off...
Having five or six power-hungry gaming systems around explains much about our recent power bills.
"We have to go forth and crush every world view that doesn't believe in tolerance and free speech." - David Brin
Buy an Intel Prescott based system if you live in the Artic Circle ...
Looking at the data in the article, would I be mad in assuming that a 90nm 3500+ uses around 23W in idle mode?
Assuming power supply is 75% efficient:
112W * 0.75 = 84W getting to system
179W * 0.75 = 134W (130nm under load, near TDP of 89W, let's assume 84W)
134W - 84W = 58W Mobo, Gfx, IDE, etc power consumption
84W - 58W = 26W
26W * 0.9 (motherboard VRM efficiency) = 23W
I suppose that system power usage also drops in idle mode though as well.
Yes, these figures are extremely dodgy and vague and aren't worth much more than the speculation they are. It looks like the 3.4GHz P4 uses over 100W under load though - that is shockingly high.
Well, in my book, power consumption is not a huge issue if there is proper cooling. Under normal and even high use conditions, the unit is designed to take the heat, and my server room needs a bit more heat anyway. Why shouldn't I leave it on? My units have good cooling, and since I run my boxes under normal server configuration, i'M not "overclocking".... Heat? No issue.
"Who are in control, they are not in control of anything - they don't even control themselves!" - Glen Beck
The ambient temperature in his office was 85 degrees F? I'm breaking a sweat at 72F. When the A/C turns off in our office over the weekend the ambient climbs to about 85 and all of my servers fans are on overdrive. I wonder if that had anything to do with the power consumption in this test, I'm curious to see what the diference is at a more normal operating temperature, say 69 degrees F.
(B) + (D) + (B) + (D) = (K) + (&)
they'd bill this as a "feature." Buy the processor and we'll bundle the radiator for free. Remember, supplies are limited, so hurry before winter approaches.
A friend of mine, an overclocking expert (inventor of the "Goldfinger devices" if anyone remembers those) said that the new shrunk cores overclock to around 3GHz if you can get your FSB high enough (though this won't be an issue with the FX chips, which aren't clock-locked).
To those paying attention, 2.2GHz in an Athlon64 can generally outperform a 3.4GHz PentiumIV, so this is a big deal.
Computer Science is no more about computers than astronomy is about telescopes. --E. W. Dijkstra
The ambient temperature in my office was about 85F/29C,
The *ambient* temp was 85F? Lord, I'd hate to think how much I'd be sweating in an 85 degree office with limited air movement...
This magazine writer works at a place that can't afford air conditioning? Or does he have so many computers in there that he's just cooking himself voluntarily?!?
What *does* roast-geek smell like?
the relative difference between the 90nm processors (defined as the difference divided by the average) in power consumptions is huge, and pretty consistent: 30% at idle, and then 43%, 45% and 44% for the other tests. These are huge numbers !
151W (idle) * 12 hours * 360 days * 15c/unit = $100 a year extra on your electricity bill BEFORE you factor in the power used in your A/C to remove that heat.
If you are nice and do Folding or SETI or RC72 or whatever it is now, then you're looking at $150 at least.
If you are in an office, you can see how the costs could rapidly ramp up!
The P4 system he was running was running at about 150 watts at idle.
Now, if you are running an A/C unit then you will not only have to consider the 150W your computer is using, but also the power that your A/C is using to fight the heat that it produces.
100% of the power used in the PC becomes heat (I think). So that is 150 W of heat. Your A/C, however is not 100% efficient. I really have no idea what the numbers are there. But it can't be more than 100% efficient so that is another 150 Watts (at least)
So your 150W computer is costing you 300W at the least.
Now, if you on the other hand live up north, then it looks much better. The heat produced will actually help your heating system, so that it doesn't have to run as much. My physics knowledge is a bit rusty, but I think you can say that if your heating system is based on electricity then it will cost you nothing extra to run your PC.
Please let me know if/where I'm wrong.
The Internet is full. Go Away!!!
The problem is, it doesn't use 100 watts more than AMD. The difference at idle is 40 watts.
The 6800GT in their test-bed sucks more power than any CPU they're testing.
And who knows how the various motherboards being tested are affecting the measurements. They aren't even from the same maker (Asus vs Abit). Odd that they take a top tier manufacturer for the AMD tests (Asus), and a manufacturer known for shit (Abit) for the Intel.
Not only that, they have the intel running DDR2 533 vs DDR 400 on the AMD - the memory alone counts for A LOT of the power usage, especially in the mpeg encoding tests.
The AMD rigs use an AGP card, the Intel rigs PCI-E. How does the faster Intel bus affect power usage?
The AMD probably does use less power. I just hate lies and horseshit motivated by some hardon people have for the "underdog".
What was preventing them from taking some accurate measurements?
I don't need no instructions to know how to rock!!!!
The one place where your figures aren't quite right is in the air conditioning department. An air conditioner, being a heat pump, just needs to move the heat from one spot to another, and the "typical" phase-change A/C unit is fairly efficient at it.
To put some figures on it, an air conditioner with an EER of 12 means that it can move 12,000 BTUs with 1000 watt-hours of electricity.
Now, 12,000 BTUs is equivalent to 3516 watt-hours of heat. So for every 3,516 watts of heat generation, you'll be expending 1,000 watts to move that heat to the outside of your building. And that's with an EER of 12, some units exist with EERs as high as 17.
So, for every 150 watts of power your computer is using, figure 40 to 60 watts for your A/C.
On the other hand, were you using a peltier device for cooling, you'd be in bad shape. If the EER figure were applied to them, it would be less than 1. For example, to move 30 watts of heat across a peltier, you'd need to apply approximately 45 watts of heat to it - meaning you'd be removing 30 watts from the cold side, but you'd need to remove 75(!) watts from the hot side.
steve
Oh, you're not stuck, you're just unable to let go of the onion rings.
The power saving methods are designed to cut the ACTIVE power use of the chip - the power that is dissipated by the transistors flipping from 0 to 1 (and 1 to 0). The challenge, as we shrink geometries (I work next to a semi fab) is that the LEAKAGE power (the power being dissipated to heat just from putting power on the chip) is growing relatively fast (in % of total power).
This is one reason why IBM and others have moved to a Silicon-on-Insulator process - it reduces the amount of leakage power. Other methods for doing this include actually powering down certain portions of the chip so that they aren't drawing power even if they aren't switching. IBM calls them voltage islands. The challenge is the tradeoff to time in powering those sections of the chip back up again.
These developments are part of a natural evolution in the semiconductor industry to deal with the heat creation of semiconductors. I'm not saying we're winning, but it's getting some focus and there are methods of keeping it from getting totally out of control (at least in the near future)
This seems a poor comparison between the AMD CPUs. Given they have taken a 130 nm chip and underclocked it, that means the chip is capable of higher clock speeds and therefore has "hotter" (from a speed sense) transistors as we used to say at AMD (used to work there). Since the transistors can deliver more current when on (leading to the higher clock speeds), by definition (subthreshold slope is limited by physics to ~60 mV/dec of current) they will "leak" more in the off state than transistors that don't supply so much current (and therefore run slower). I wish they had had equally rated (by AMD) chips to remove this uncertainty, although everyone seems to be focusing on the difference between the Intel and AMD boxes (which opens up a world of concerns....is it the motherboard under load increasing its demand, they have different memory systems which could contribute when stressed, is the PCI-E bus not as efficient as the (assumed) AGP, etc.).
Even if their chips eventually become cheaper and better, again? That sounds like a stupid strategy. In general, all strategies of the form "I will never/always do X" are stupid.
Breakfast served all day!
From what I've read, PCIe isn't going to noticably boost your performance over AGP 8x. So its not worth upgrading, if your system is already pretty current.
Of course, if your going dual video card or your current system is out-of-date then...