Measured at the wall, what's your Watt/MHps

[Fjordbit]

By my Killawatt, I'm running at 530 Watts for 800 MHps. Is this reasonable, or is this out of wack?

[1714066880]

1714066880

[1714066880]

1714066880

[1714066880]

1714066880

[1714066880]

1714066880

[1714066880]

1714066880

[Xephan]

By my Killawatt, I'm running at 530 Watts for 800 MHps. Is this reasonable, or is this out of wack?

Hard to say without knowing key components of your config esp GPUs, CPUs and PSU.

For example if you're doing 800MHps on 530W using purely CPU mining, it's astonishingly good

For what's it's worth I'm pulling 465W or so on a 80+ Silver PSU for 780MH/s, including 1 out of 3 core heavily loaded due to the bugged Catalyst 11.6B drivers.

And 470~474W for 800MH/s

Edit: Those numbers were off by a bit as I had my meter on VA. Actual watts is 433W and 439W.

[Smalleyster]

This is a great thread as it deals with the only two paramaters that really matter in the long run.

What is your input to output ratio?

The other factor is capital (what is your rig, what did it cost, how did you set it...) but that is realy just how you got started.

I hope this thread gets lots of responses.

[teukon]

I'm getting 723 MH/s from 300W (2.41 MH/J) but I spent extra on a good PSU and on aftermarket coolers for my two cards.

530W for 800 Mh/s seems a little high but not grossly so.  I have a friend with a 6990 managing 800 MH/s at 520W and he's perfectly happy with this.  I'd guess you are overvolting your cards and if so and you can stomach coming down to 750 MH/s you might save 50+ watts by lowering the clocks and voltages on the cards.

[cicada]

2 6950x2 rigs getting 720MH @ 410W each, so 1.4GH for ~820W

I've also got a pair of 5830's in my Opteron server, for 540MH @ 530W - obviously this is a poor ratio, but pre-mining I was living with the server pulling 200W on it's own so I'm OK with these numbers.

Looks like I've got some optimizing to do

[bmgjet]

280-290W at the wall.
530W PSU
Old 939 Duel Core Opteron overclocked at 3GHZ
4 HDDs
2 DVD burners
HD6850 overclocked at 1010mhz.
281Mhash as I speak.

[mikeo]

Current setup:

372 Watts, 614 MH/s

MSI 870-G45 M/B
AMD Sempron Sargus 140 CPU @ 2.7Ghz (not under volted/clocked)
Kingston 1GB DDR3 RAM
OCZ EliteXstream 800W 80 Certified PSU
4GB Verbatim USB 2.0 Flash Drive (diskless) LinuxCoin 0.2.1b w/2GB persistence file
2 x ATI Radeon HD 5830 GPUs (stock voltage 1.088, core: 975, mem: 300)
Antec 200 case

Replacement (now building):

Watts, MH/s (will post results as an edit later)

MSI 890FXA-GD65 M/B (2 PCIe x16, 4 PCIe x1)
CableSaurus.com PCIe extender cables
AMD Sempron Sargus 140 CPU @ 2.7Ghz (not under volted/clocked)
Kingston 1GB DDR3 RAM
TAGAN 900W PSU (may have to add 2nd PSU, we'll see)
4GB Verbatim USB 2.0 Flash Drive (diskless)
2 x ATI Radeon HD 5850 GPUs (probably at stock voltage, 900/300)
3 x ATI Radeon HD 5830 GPUs (stock voltage 1.088, core: 975, mem: 300)
HSPC Tech Station Kit, Standard Size (14x9.5) - This is proving to be VERY small, requiring mods to get the GPUs decently mounted

[Yannick]

I'm running at 574W for 824Mhash/sec (2x5870)

[Fjordbit]

Hard to say without knowing key components of your config esp GPUs, CPUs and PSU.

For example if you're doing 800MHps on 530W using purely CPU mining, it's astonishingly good

I think Smalleyster has the right idea on the spirit of this question. It's more of an idea of how efficient rigs can be or should be using these the two variables of input to output as a comparison. The point isn't to think of how good the hashrate is but how good the ratio is. It does seem like I'm on the low side, but not way out of range.

One thing that this ratio will give you is resilience to difficulty rises and/or btc value drops. The higher the ratio, the longer you can mine and remain profitable. Of course electrical rates matter here too.

[Xephan]

I think Smalleyster has the right idea on the spirit of this question. It's more of an idea of how efficient rigs can be or should be using these the two variables of input to output as a comparison. The point isn't to think of how good the hashrate is but how good the ratio is. It does seem like I'm on the low side, but not way out of range.

One thing that this ratio will give you is resilience to difficulty rises and/or btc value drops. The higher the ratio, the longer you can mine and remain profitable. Of course electrical rates matter here too.

Agreed, but knowing this ratio is not as helpful if we did post the config. Like in my absurd example, ratio might seem low but is actually extremely efficient for the kind of equipment (purely CPU) mining. In your real case for example, while it may be low, the important thing is why? Imagine if somebody decided it was not worthwhile doing this because his efficiency was just too poor for long term... and the problem might actually just be the fact he's using an old crap $30 PSU with 60% efficiency which could had been fixed by a simple $50 purchase.

With the additional details, everybody will get a better idea where they stand

[Smalleyster]

Ok so the quick and dirty analysis so far:

Watts   Mhps       Watts/Mhps   Mhps/Watt
530            800       0.663          1.509
433            780       0.555          1.801
439            800       0.549          1.822
300            723       0.415          2.410
520            800       0.650          1.538
820           1400       0.586          1.707
285            281       1.014          0.986
372            614       0.606          1.651
574            824       0.697          1.436
         Average:        0.637          1.651

I think the most telling ratio here is that one can generally expect about 1.7 Mhps per 1 Watt of power input.

The 2.4 and .9 seem to be "Flyers" or "Outliers".

Removing them gives us an average of 1.638 which is not a significant change to a rounded 1.6.

Rule of thumb: expect just over 1 1/2 Mhps for every 1 Watt of power used.

Thanks for the data guys, that was fun. 8^)

[cicada]

Hooray, I'm average!

[piim]

1135 MH/s and 563 W.

Rig:
Corsair 750TX 750W
2x XFX 5870 @ 930/300, 78C and 80C
1x VTX3D 5830 @ 950/300, 70C
Intel Core i3-2100 3.1GHz
Corsair XMS3 DDR3 1333MHz 2x2GB CL8
LinuxCoin running on some random 4GB USB stick

[CYPER]

800W for 1760Mh/s = 2.2Mh/s for every Watt (the bigger the better)

I believe that is pretty good

[teukon]

The latest new entries:

Watts   Mhps       Watts/Mhps   Mhps/Watt
1135          563       0.496          2.016
800           1760       0.455          2.2

This makes my claim of 2.41 MH/J more plausible and I still get to be an anomylous result!  Good work guys .

[fitty]

701 watts at the wall
3x 6950 unlocked shaders
OCed to 905
1200 mhs.

[CYPER]

The latest new entries:

Watts   Mhps       Watts/Mhps   Mhps/Watt
1135          563       0.496          2.016
800           1760       0.455          2.2

This makes my claim of 2.41 MH/J more plausible and I still get to be an anomylous result!  Good work guys .

You got exactly the same figures as me = 1760

What is the configuration?

[Smalleyster]

Watts   .   Mhps   .   Watts/Mhps   .   Mhps/Watt
530   .   800   .   0.663   .   1.509
433   .   780   .   0.555   .   1.801
439   .   800   .   0.549   .   1.822
300   .   723   .   0.415   .   2.410
520   .   800   .   0.650   .   1.538
820   .   1400   .   0.586   .   1.707
285   .   281   .   1.014   .   0.986
372   .   614   .   0.606   .   1.651
574   .   824   .   0.697   .   1.436
563   .   1135   .   0.496   .   2.016
800   .   1760   .   0.455   .   2.200
701   .   1200   .   0.584   .   1.712
....   .   Average:   .   0.606   .   1.732

Ah Frig the formatting debacle! Y'all can see what matters! 8^)

[fitty]

4x5830
OC 950
785 at the wall
1200 mhash

[Kermee]

Majority of my rigs are 4x5830 at 998/300 (317 MH/s) with 640W at the wall (Kill-a-Watt).

~1.98 MH/watt.

Cheers,
Kermee