• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

587W at the Wall Plate!

Elfear

Elfear

Diamond Member
Wow. With my cards at 710/840 1.525/1.15/1.15/1.5V and my E6600@3.6GHz 1.42V and all system fans at full blast I measured 4.89A using a Fluke clamp meter. If my calculations are right that's 586.8W!

So you PSU gurus correct me if I'm wrong but that 587W should equate to ~440-470W (@75-80% efficiency) draw from the system components since some of the power is lost through the PSU. So does that mean at full capacity a good 600W PSU would be drawing 750W at 80% efficiency?

EDIT (11/15/06):

Just switched the Crossfire out for a 7600GS I'm borrowing until my G80 arrives. Everything else was kept the same except for the video card switch.

7600GS Results

2.15A
258W at the wallplate
~180W draw from the components

I can't believe how much less this thing consumes than my X1900XTs.

EDIT (11/20/06):

Testing procedure was exactly the same so the only variable is the gpu.

8800GTX Results

3.28A
394W at the wallplate
~292W draw from the components
 
My gaming rig with an overclocked Opteron and a 7900GTO draws less than 200W from the wall while looping 3DMark06 (measured using a Seasonic Power Angel)
 
That sounds like an unlikely power level. I'd guess that 400-450W was more likely.

Unfortuately, you can't estimate PC power consumption by using an ammeter or DMM. You need to use an AC power meter, otherwise you can get these hopelessly inflated readings.

If your PSU has active PFC, then the DMM readings/calculations will be pretty close to the real consumption. Without PFC, the calculations will be way, way off.
 
Does that include the monitor? My entire computer's power draw at load is just under 400W (measured by a UPS), but when the monitor's draw and PSU's efficiency are factored in, that comes down to only 230W.
 
That was power draw from only the computer. I cut up a power cord to expose the positive wire. I than used a Fluke Clamp meter to measure the power draw while running 3DMark06 and Orthos at the same time. I got a peak reading of 4.89A which equates to 587W from the wallplate or ~440-470W depending on efficiency.

EDIT: Looks like ~75% efficiency according to jonnyguru. Link
 
Originally posted by: Elfear
That was power draw from only the computer. I cut up a power cord to expose the positive wire. I than used a Fluke Clamp meter to measure the power draw while running 3DMark06 and Orthos at the same time. I got a peak reading of 4.89A which equates to 587W from the wallplate or ~440-470W depending on efficiency.
Click to expand...

But a measurement of rms current alone is useless. You can use the best ammeter available, but it won't give you a useful number, because you're measuring the wrong thing.

You can't calculate average power from rms current and rms voltage.

You have to calculate average power by the averaging the instantaneous power (instantaneous voltage x instantaneous current). This can only be done with an AC power meter.

What exact model of PSU do you have? In particular, does it have active PFC?

If it does have active PFC, then your estimate of power will be reasonably close. If it doesn't then your estimate will be a massive overestimate.
 
Based on his link the PSU has active PFC. Just knowing he's running Crossfire X1900 cards and such a highly overclocked C2D, we already know that at 75% efficiency and with loading the CPU & GPUs, he's probably looking at at least 400W from the wall already. His water cooling pump will draw some (dunno how much but a single fan in my system draws 5W difference between low and high RPM) and hard drives can draw 10W+ each (in case he's running a 6 drive array or something).

Sure 587W sounds like a lot. If it indeed is an overestimate, his system will probably be somewhere between 400-500W from the wall. Dunno. Elfear, you should buy a Power Angel just so we're all on the same page, so to speak. 😛
 
Well, if it does have active PFC - then 587 is going to be pretty close. Probably over 560 W in reality.

If that really is the case, then that is pretty scary, and I'm glad I'm not paying your electricity bill.

I don't know the efficiency of the OP's PSU - but if anyone else is considering building a system like that, for heavy usage, then buying the most efficient PSU you can get is likely to be a very good investment.

I've got a very crude script for comparing effects of PSU efficiency on running cost. here

 
Originally posted by: Zap
Based on his link the PSU has active PFC. Just knowing he's running Crossfire X1900 cards and such a highly overclocked C2D, we already know that at 75% efficiency and with loading the CPU & GPUs, he's probably looking at at least 400W from the wall already. His water cooling pump will draw some (dunno how much but a single fan in my system draws 5W difference between low and high RPM) and hard drives can draw 10W+ each (in case he's running a 6 drive array or something).

Sure 587W sounds like a lot. If it indeed is an overestimate, his system will probably be somewhere between 400-500W from the wall. Dunno. Elfear, you should buy a Power Angel just so we're all on the same page, so to speak. 😛
Click to expand...

I have been thinking about buying a Kill-O-Watt meter (same as Angel) but I haven't been able to convince my wife of the need yet. The best thing I could come up with was the Fluke meter. I don't pretend to know as much about EE stuff as Mark so I could be off. The PSU does indeed have Active PFC which, according to you guys, should make the measured amps pretty close to actual.

Besides the components in my original post, here is a breakdown of the rest of my system if that helps:

3x Sanyo Denki 120mmx38mm fans (6.24W each)
2x Thermaltake Smart Case Fan 80mm (8.4W each)
AQX50Z Pump (8W)
X-Fi Fatality (~20W??)
7200rpm hdd (13W)
DVD Burner (25W??)

That's not a lot of extra components so I suspect the big power draw is coming from the oced X1900XTs and C2D.
 
Originally posted by: Elfear
Besides the components in my original post, here is a breakdown of the rest of my system if that helps:

3x Sanyo Denki 120mmx38mm fans (6.24W each)
2x Thermaltake Smart Case Fan 80mm (8.4W each)
AQX50Z Pump (8W)
X-Fi Fatality (~20W??)
7200rpm hdd (13W)
DVD Burner (25W??)

That's not a lot of extra components so I suspect the big power draw is coming from the oced X1900XTs and C2D.
Click to expand...

Just goes to show how power hungry those X1900s are. The C2Ds aren't actually that bad - much better than the Pentiums.

Could be worse - you could have dual 8800GTXs 🙂

 
Originally posted by: WobbleWobble
My gaming rig with an overclocked Opteron and a 7900GTO draws less than 200W from the wall while looping 3DMark06 (measured using a Seasonic Power Angel)
Click to expand...
Heh!

I believed the other guy, but I don't believe you...

Sorry! 😉
 
Originally posted by: VinDSL
Originally posted by: WobbleWobble
My gaming rig with an overclocked Opteron and a 7900GTO draws less than 200W from the wall while looping 3DMark06 (measured using a Seasonic Power Angel)
Click to expand...
I believed the other guy, but I don't believe you...
Click to expand...

Want a photo?
 
Originally posted by: VinDSL
Originally posted by: WobbleWobble
My gaming rig with an overclocked Opteron and a 7900GTO draws less than 200W from the wall while looping 3DMark06 (measured using a Seasonic Power Angel)
Click to expand...
Heh!

I believed the other guy, but I don't believe you...

Sorry! 😉
Click to expand...

Tell me what proof you want. Like Zap suggested, I can provide you a photo. But you seem extremely suspicious of this, so if I have a photo of just the Power Angel, you won't believe it's the computer I'm talking about. If I take a photo with my computer right beside the Power Angel, you still won't believe it's the same machine.

I measured it with a Power Angel, which is just a rebadged Kill-A-Watt. I have nothing to gain with tell you a lie. 😛
 
Test added for 7600GS. I'll add the results here too.

7600GS Results
2.15A
258W at the wallplate
~180W draw from the components

Next up, 8800GTX
 
I jst tested my server. Peak is 139W and that's when it is first powered up, when I hear the HDDs spinning up. Idle is 74-76W. The PSU is an old Fortron 300W, the first model they made with a 120mm fan. IIRC the efficiency is somewhere around 65-70%. I'm thinking of replacing it with one of those mini-ITX PSUs like the PICO PSU or another model, since those have over 90% efficiency. This server is on 24/7 so the new PSU should pay for itself... eventually... like in 3 years or so. 😱

Server specs:
Mobile Pentium 4-M 1.6GHz (1.92GHz at 1.168v, overclocked and undervolted)
Shuttle AB60-R motherboard
Kingston HyperX PC3700 512MB single stick
Seagate 320GB SATA 7200.10 drive
Seagate 400GB EIDE 7200.9 drive
48X CDROM
PCI 3Com NIC
PCI Tseng Labs ET6000 video card

Just a side note. I have it on a small UPS (around 200VA) and even without the computer powered on, when I turn on the UPS it pulls 34W from the wall momentarily. I can see why LAN party organizers tell people to leave their UPS at home because I'd imagine larger ones coupled with systems turning on may draw a couple hundred extra watts.

Elfear, that's pretty crazy how just the video card change can make such a difference. How's the performance of the single 8800GTX versus Crossfire X1900XT? If it's anywhere close, then I'd go with the single card just for the near 200W savings at the wall!!!
 
Originally posted by: Zap

Elfear, that's pretty crazy how just the video card change can make such a difference. How's the performance of the single 8800GTX versus Crossfire X1900XT? If it's anywhere close, then I'd go with the single card just for the near 200W savings at the wall!!!
Click to expand...

I was pretty surprised as well at the difference made by just the GPU. I really wish I'd have measured power draw when the systems were idle as my computer spends most of its time in that state. I imagine the results between the X1900XT's and the 8800GTX would be a lot closer since the G80 runs at 3D clocks all the time.

I checked over a ton of reviews before I decided on switching to the 8800GTX. It should be equal to or a little faster than my Crossfire system. Sadly, school has kept me too busy to find out for sure.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top