AMD Athlon 64 4000+ & FX-55: A Thorough Investigation
by Anand Lal Shimpi on October 19, 2004 1:04 AM EST- Posted in
- CPUs
Power Consumption
To measure power consumption we looked at overall system power consumption and tried to keep as many variables static. There are some basic differences which we cannot get around, mainly that the 925X uses lower voltage DDR-II while the nForce4 uses regular DDR, but for the most part our results were quite controlled. We also included power consumption figures from 130nm Socket-939 Athlon 64 3200+ and 3000+ chips, which as you may know, do not exist. The reason we did this was to show the sharp contrast to the power consumption figures of the 90nm 3500+ we've included in the charts below.
We measured power consumption in two states: idle sitting at the Windows desktop and under load while running our Windows Media Encoder 9 test, which proved to be one of the most strenuous CPU tests we ran as it pretty much isolated the CPU subsystem.
At idle, the 130nm Athlon 64s all consume just about the same amount of power, with the 90nm Pentium 4s doing a little more. But what's truly impressive is the 90nm Athlon 64 3500+, drawing less power than any of the other chips at idle - by a significant margin.
Under load the situation is no different; because of the fact that AMD didn't change anything architecturally (nothing major at least) with the 90nm shrink, the Athlon 64 got what the Pentium 4 should have received with the move down to 90nm - much lower power consumption. However, tack on 32-bit ALUs (as opposed to the 16-bit ALUs in the Northwood Pentium 4), a much longer and more power hungry pipeline as well as logic to support it, and you're looking at a much more power hungry chip. The 90nm Athlon 64 does not suffer this same fate and thus is almost a lower power chip compared to the competition here.
Facebook
Twitter
RSS
89 Comments
View All Comments
HardwareD00d - Tuesday, October 19, 2004 - link
Fantastic article, obviously very well thought out.I would have liked to see a comparison between the 4000+ and the "real" FX-53 to really back up your rebadging theory (yeah I know speed+cache+memory width are equal between the two, but just to make sure AMD isn't pulling some magic out of there butt somewhere else).
Marsumane - Tuesday, October 19, 2004 - link
Yes, thanks for the XP comparison. I find it interesting how its not performing as well as it used to in games. (doom 3, farcry, cs:s)Also, your ut2k4 benches seem off. How is doom 3 pulling 50% more frames at the same res? Maybe your ut is at 16x12? I pull similar frames on ut w/ my 9800p oced.
ksherman - Tuesday, October 19, 2004 - link
I like the ending... It sounds mysterious!alexruiz - Tuesday, October 19, 2004 - link
I will suggest again to include some Ulead Video Studio 8.0 benchmarks for video encoding. Ulead is by far the fastest consumer grade video editor / renderer, it is the most complete and one of the most popular. In fact, it is almost 50% faster than Pinnacle 9, and almost 100% faster than videowave.Roxio has really been working with Intel as all previous version of video wave ran better on AMD hardware. As reference, results video wave 6 or 7 would be interesting. Newer doesn't always mean better, as you can see from Adobe Premiere. Version 7.0 is quite slower than 6.5 doing the exact same thing in the same platform.
For DivX encoding, a run with virtualdub/virtualdubmod or DVD2AVI would be nice, as they are very fast and extensively used.
Just some comments
Alex
Araemo - Tuesday, October 19, 2004 - link
Thank you thank you thank you for including an Athlon XP.This allows me to better judge where my current Barton 2.4 Ghz sits. ;P So I know when an upgrade to the next cheap overclocker will give a good enough performance boost to be worth the money.
stephenbrooks - Tuesday, October 19, 2004 - link
Here's an idea to play with: how about some 2D scatter plots of Performance/£ and Performance/Watt? Obviously not on everything - that would clutter it - but perhaps on one or two key things it'd be nice to see.Zar0n - Tuesday, October 19, 2004 - link
EDIT #22 There is no 3400+ for SK 939 only 3500+Zar0n - Tuesday, October 19, 2004 - link
Nice article BUT:You should make C&C power consumption and temperature
Also some OC tests.
The Battle for Value is not correct:
1º WHAT about price of DDR1 VS DDR2?!
2º MB for INTEL are more expensive, ~40€ is a great difference in a MB price.
3º 0.09 AMD are just introduced so they are going to come down, not much but they are.
In order to be fair you should compare with AMD 3400+
AMD as a clear winner here.
mczak - Tuesday, October 19, 2004 - link
power consumption at idle - is this with or without cool 'n' quiet (I suspect without)?Uff - Tuesday, October 19, 2004 - link
I have to agree with #18 - it's not worth paying more than twice the price of a 3400+ just to get 3800+ on 939 platform.Many say 'OH! But s939 is more upgradable!', but if you think about it, by the time you upgrade next there are very likely going to be new motherboards available aswell and you end up upgrading that anyway. Not to mention motherboards cost virtually nothing compared to cpus.