A new CPU power consumption comparison

Post by **NeilBlanchard** » Tue Oct 19, 2004 8:49 am

Hello:

I found this link over at Ace's Hardware:

http://www.goodwin.ee/sulo/Power2.htm

I have not "digested" this yet, but it looks interesting and informative.

Bitter Jitter · Post by **Bitter Jitter** » Tue Oct 19, 2004 10:25 am

That looks quite interesting. The sempron at 1.5v uses 8w more with load than the same chip at 1.4v. Which means that these 90nm chips are making about a 12w difference based on the manufacturing process. The other 8w is a result of the lower voltage.
That’s my guess anyway.

The only question is doesn't the Athlon 64 in lowest p state use 22w and not the 12w shown in this article. I wonder if this is true?

Still hats off to the author for trying, it clearly isn't easy to get 100% accurate results when measuring power consumption.

tof · Post by **tof** » Wed Oct 20, 2004 10:05 am

Excellent piece of information. This saves us lots of time(I had in mind doing it).

I believe that his assumption for the DC-DC converter supplying current to the CPU is pessimistic. If it was 84%, a P4 consumming 100W would mean 16W dissipated by the converter which does not have so big heatsinks... (remember the HS size required by a AMD K6-II spec'ed 18W!)

DanceMan · Post by **DanceMan** » Fri Oct 29, 2004 3:13 pm

K7 parts can be used in two different modes when idle. The blue bars represent what the chips were designed to consume when idle in Windows. In this mode the CPU would enter a “stop grant” state when Windows Idle process issues a hlt command. The state has sometimes been called “bus disconnect”. The problem seems to have been that most mobos were designed so that their CPU power regulator was fed from +5V PSU and most PSU-s and mobos couldn’t take the wild power consumption changes happening on this +5V rail (check out the max power consumption of a Palomino and see the huge difference). It was sometimes also said that earlier K7 chips had somehow “buggy” “bus disconnect” mode and mobos can only enable “bus disconnect” with Bartons. IMHO that was just a hogwash and when Bartons bacame commonplace, the bus disconnect was still not implemented on boards. I.e. I don’t think the “bug” was real, I think the proglem was only with wild fluctuations which plagued Bartons just as much. The red bars represent the K7 behaviour with the usual, default installations of Windows, Linux etc. in which case the CPU is not manually allowed to enter stop grant state at idle.

Does anyone have any more information or comment on this? Do any utilities exist that were written to address this issue, to override this and allow the cpu to enter the stop state? Or is that what Cool'nQuiet is?

JazzJackRabbit · Post by **JazzJackRabbit** » Fri Oct 29, 2004 4:30 pm

I kind of like the power consumption of K8-AX at 1.0Ghz. Anyone knows how well does it stack against A-XP-M at the same frequency?