Cool article...From a power consumption point of view, seems like the plain GF 6800 has good performance to power consumption. No wonder some companies are starting to announce passively cooled versions! Thanks for confirming my suspicions! :)

Will we be able to see some old GF4 cards just compare?

How is it that the GeForce 6800GT draws less power than the Ultra when overclocked past Ultra speeds? They are based on the same core, they have the same amount of pipelines, they have the same amount and type of RAM, and I'm almost sure the benchmarks of the OC'ed GT will surpass those of the stock Ultra! So how is the power consumption still lower? Am I missing something, perhaps in a cut-down core, or maybe some additional power-savings?

XbitLabs has done a great job on measuring the real power consumption of graphic cards, superior than any other web site!!!

:-)

This is the best article done on power consumption of graphics cards. I must applaud Xbit editors. I look forward to 6600 GT fanless as replacement for my fanless Ti4200 (Quadro4 700 XGL). Perhaps the 6800 could even be used if the voltage was reduced a bit.

awesome article although now i'm confused as to how I got my measured power consumption figures for my Asus 6800U heh

Excellent article, very interested to read the results.

Are you a retard or just ignorant:

"The card eats no more power than the topmost competitor model, the ATI RADEON X800 XT Platinum Edition."

So now in Tim Tscheblockov's world, 72.09 is no more than 63.23? What kind of statements are you trying to make, nVidia malaligned I'm sure.

Nice Article, i really hope you can keep it up to date as new chips get released, like the new mainstream ones.
One chip's power consumption i'd be interested in too is the one of the 5900XT, though i guess you cant test everything..
Anyways, nice job!

Mike

Great article,

Have you tried to measure the consumption of underclock boards? With some external tools (e.g. ATITool) it's possible to lower the clock speed in Idle mode.

Frank

Interesting choice of drivers...;)

Try measuring the 6800 series with SM3.0 enabled driver sets (61.77+) & DX9.0c installed, with FarCry "SM3" patch. FarCry pipeline use may not be indicative of all power consumption scenarios. Unfortunately, I know of no way to activate all transistors of a graphics ASIC for measuring Pmax...:)

P.S. Shunt will be using power too.

Thank you for those nice power consumption review. This is very important news for me, since I am planning to buy a new videocard.

Do you plan to keep posting power consumption data in your reviews? it would be really nice to see how Nvidia and ATi behaves in this issue, as we can discover some flaws - the Radeon 9800 PRO eg. has a too high consumption IMHO.

Also, is it so hard to make those measurements? seems that there are very few power consumption reviews on the net.

Again, thank you. This review showed me not to buy many videocards.

Best Regards

RV410 Is known as X700 now..

Will a 6800GT work on a Shuttles Silent X 240W?! I've heard people using 6800GTs in them without problem.

The method is sound enough....... but surely, a "SHUNT" is a PARALLEL connection...... isn't it? (not SERIES connection, as we have here)

Thank you for an excellent article on Power Consumption of Contemporary Graphics Accelerators.

I am building a small game PC (that use a 150w DC-DC converter board) and found the information very usefull.

Based on your article I have decided to use a 9600 XT graphics card due to the low power consumption, however I am uncertain whether to use a 128 or 256 MB version.

Will the 256 MB version generate any significant increase in the current on the 3,3V or 5V power rails compared to the 128 MB version (appr. how much) ?

Regards,
Ole

Hello!

You guys did a great Job, but you have just forgot about a little Thing:

The Power-Consuption of an such Chips is in Square to the delivered Voltage.
If you place a Shunt in Front you can calculate the failure from the Voltage-Drop back.

You used: P = V5 * (dV5/R) + V12 * (dV12/R) and so on for the AGP-Voltages...

You should use: P = V5 (dV5/R) * ((V5+dV5)/V5)² + V12 * (dV12/R) * ((V12+dV12)/V12)² and so on for the AGP-Voltages

This elliminates (not exactly, but better then without it) the Calulating-failure from the Voltage-drops on the Shunts.

If I calculate the Powerconsumption from the overclocked X800 on Load:

You get: 55.8871
You would get: 67.6696

Its not much, but this are about nearly 2W or 2,7% more ;-)

If you want you can use this an calculate your Results new!

Regards, Easy

Yes the 9600XT is a good card but you might need to run it at 4X AGP. I personally installed one in an IDEQ system with 200w psu and it wouldn't run at 8X. I also had problems running it in any resolution above 1024x768. I would go for the 128MB version as it usually has better quality memory. Note 256's are usually only a tenner more - crap memory.

Don't wanna seem that I did not read the article, however, I have a XW4100 Hewlet Packard workstation with a 280W PSU, my back-up software shows me that my computer is using 140 W (with a Riva TNT2) and 200W with a Geforce 5900XT, will it be enough this power to use a 6800GT?
P.S. some data about my psu
HIPRO HP-D2808F3P
AC INPUT 200-240V/4A
DC OUTPUT +5V / 16A; +12V / 17A
+3.3v / 23A -12V / 0.8A
+5V stb / 2A

+5V & 3.3V shall not exceed 156W
+5V & 12V shall not exceed 232W
Max Output Power 280W

I am now considering using a Radeon 9700pro (which are still very common in the US), but started to wonder if the Radeon 9700pro use considerably less (or more) power than the Radeon 9800pro (which use a bit to much power for my purpose) ? Before buying a 9700pro to test this I wondered what your opinion is ?