Those hash/watt numbers are stupid, they're using rated TDP numbers to get at those values. Actual numbers would tell a very different story
Slomo4shO
Senior member
- Nov 17, 2008
- 586
- 0
- 71
The power use is so damn efficient it doesn't matter
Those figures are so wrong... Can easily hit 950-1000 khash on the 290X while undervolting with custom bios... Also, you won't be anywhere near 294 khash with the GTX 750 Ti with 60W power consumption...
On a side note, why doesn't anyone simplify Khash/s/watt to Khash/J
More slots = more hardware requirements = more costs.
Last edited:
Manufacturers should do separate mining and gaming cards. Little point in a dual purpose card if it may alienate one group over the other. It would also allow them to keep costs lower on both cards if they are not built for a multi-function purpose.
AMD or Nvidia could make a million different products around their GPU's. The limiting factor is still going to be the number of GPU's produced, not how many different kinds of cards they sell them as.
so going back to the original question it will be good if we keep it simple,
lets asume company A makes better hash per watt and another company lets call it company B that makes better hash per dollar, how do you compare both?
lets make some asumptions
you have two systems one with company A and another with company B, and have 4 months to pick a winner that its because there will be asics that make the same compute cheaper.
To simplify both systems cost the same: 100 dollar to buy the full system, and 100 hash gives you 50 dollars, the maintaneance cost is 1 dollar per kwh, witch one will give you beter profits if?:
System A
20 hashs per hour at 50 kwh
System B
39 hashs per hour at 100 kwh
now remember there is a cost of being late, at the 4 months period yuo make 1 hash per hour.
to clarify im drunk, i will try to run the number s and change the data, to benchmarks from sites, and try to investigate what is the performance of cards as time pass by.
in the following days, and if you have a better model post it and/or modify mine.
the idea is to have some type of benchmark or index that people could follow, to know what their profit will be at a 4 months period
lets asume company A makes better hash per watt and another company lets call it company B that makes better hash per dollar, how do you compare both?
lets make some asumptions
you have two systems one with company A and another with company B, and have 4 months to pick a winner that its because there will be asics that make the same compute cheaper.
To simplify both systems cost the same: 100 dollar to buy the full system, and 100 hash gives you 50 dollars, the maintaneance cost is 1 dollar per kwh, witch one will give you beter profits if?:
System A
20 hashs per hour at 50 kwh
System B
39 hashs per hour at 100 kwh
now remember there is a cost of being late, at the 4 months period yuo make 1 hash per hour.
to clarify im drunk, i will try to run the number s and change the data, to benchmarks from sites, and try to investigate what is the performance of cards as time pass by.
in the following days, and if you have a better model post it and/or modify mine.
the idea is to have some type of benchmark or index that people could follow, to know what their profit will be at a 4 months period
Well, I have one and it hits 290-300 kH/s without messing with the voltage at all. So if it's not 60W it can't be anything more than 75W.
Bitcoinwisdom has a great calculator that should be able to do your comparison:
http://bitcoinwisdom.com/litecoin/calculator
Not really truth. Here you can see that the card takes up to 140W while gaming.
Often taking 100Watts and more. We know that it doesn't downclock (throttle), so it is a matter of adjusting workload to get more than 75Watts constant consumption.
75W limit is as said previously, just a guidance for board manufacturers
thing is, the difference in power usage and cost is so huge, that you can probably get more boards and cheap PSUs with "sempron" style CPUs and still have a more efficient solution, which would cost less $ per day to run for the same hashrate
but someone should do the math, I'm to lazy.
BrightCandle
Diamond Member
- Mar 15, 2007
- 4,762
- 0
- 76
TDP != Peak usage
The TDP is the long term (think seconds) average of the power consumption for a component and allows the design of coolers. But momentarily a CPU/GPU/etc can consume quite a bit more current than the TDP would suggest.
So in essence TDP is an average not the peak figure, if you did this with an AMD card you would see a similar thing, as would you with a CPU. Its been known for quite a while that these components exceed the amps implied by TDP, but if you look at the electrical spec you will find peak consumption figures and how long that can be held that are very different to the TDP figures.
The TDP is the long term (think seconds) average of the power consumption for a component and allows the design of coolers. But momentarily a CPU/GPU/etc can consume quite a bit more current than the TDP would suggest.
So in essence TDP is an average not the peak figure, if you did this with an AMD card you would see a similar thing, as would you with a CPU. Its been known for quite a while that these components exceed the amps implied by TDP, but if you look at the electrical spec you will find peak consumption figures and how long that can be held that are very different to the TDP figures.
Well, they are almost there. But not quite.
Here is my 7870 optimized hashing some doge:
After a while it would settle around 403kh/s average. There is a amp reading in AIDA. 60 AMPS with 1,0 volts = 60Watts which is roughly what I see at the wall. around 70Watt delta between idle an mining.
400/60=6,66 kh/s/W.
Since the card takes less than 75Watts, can I possibly take those 2x6 pins out? Any idea what would happened?
Stock settings gives 330kh/s and takes almost 85 Amps@1,2V which is 100Watts
Shame it is not 270X - those get 25% more khash/s ;/
Here is my 7870 optimized hashing some doge:
After a while it would settle around 403kh/s average. There is a amp reading in AIDA. 60 AMPS with 1,0 volts = 60Watts which is roughly what I see at the wall. around 70Watt delta between idle an mining.
400/60=6,66 kh/s/W.
Since the card takes less than 75Watts, can I possibly take those 2x6 pins out? Any idea what would happened?
Stock settings gives 330kh/s and takes almost 85 Amps@1,2V which is 100Watts
Shame it is not 270X - those get 25% more khash/s ;/
BrightCandle
Diamond Member
- Mar 15, 2007
- 4,762
- 0
- 76
I have been trying to find the PCI-E electrical specification. The high level 75W is what everyone knows but the true electrical spec probably allows 150W or so for 1ms or something like that, just like the CPU electrical specs do.
The other possibility is that Tomshardware's equipment is wrong and that it doesn't pull that amount of power at all, its a factor of the equipment used and the frequency being too high and inaccurate.
The other possibility is that Tomshardware's equipment is wrong and that it doesn't pull that amount of power at all, its a factor of the equipment used and the frequency being too high and inaccurate.
I think it could be other way around. When frequency is too low, you know only half of the truth. You can't see what is happening between the readings. With high frequency readings you know exactly what is going on.
Wait a sec... you actually believe your 7870 is hashing at ~70 Watt :ninja:
:biggrin:
That, or they are sending me wrong power bills.
170Watts total system power consumption at the wall while hashing, 100 watts idle.
Keysplayr
Elite Member
- Jan 16, 2003
- 21,219
- 56
- 91
If he does, then I don't know how he made post 33 making it seem like the 750Ti was running games at a buck forty.
Lulz, by that fail logic my 1195 core 1475 GDDR5 290 is only drawing 250-260w.
Hand me the kool-aid.
Hand me the kool-aid.
Honestly, given the insane efficiency improvements with Maxwell - the real interesting aspect of this story is what happens when GM200 and GM204 are released and the Maxwell uarch is scaled with tons of CUDA cores. Kepler from what I remember scaled very linearly in this respect and there is no reason to expect otherwise from the large die Maxwell parts - given GM107s relative hashing performance, big Maxwell should outright beat AMD across the board in terms of hashrate. Ignoring the implications of the hash per watt of GM107 in terms of how it will directly affect the hashrate of big Maxwell, one would have to essentially wear blinders. GM200 and GM204 will be hashing monsters. It should be mentioned that the directly comparable AMD counterpart to the 750ti in terms of TDP is the R7-250 (65W TDP). The hashrate of that part is far lower than the 750ti.
Now with that stated - I am not a fan of this whatsoever. Personally I don't really care for mining and never have, as funny as this entire situation is i'd rather people just used these GPUs for gaming. I don't have any ill will toward those who do mine, it just isn't my thing. Anyone who's into it, cool. But as for me.....I want a Maxwell based GPU for PC gaming, which is what consumer level Radeons and Geforce cards are designed and marketed for in the first place.
Now with that stated - I am not a fan of this whatsoever. Personally I don't really care for mining and never have, as funny as this entire situation is i'd rather people just used these GPUs for gaming. I don't have any ill will toward those who do mine, it just isn't my thing. Anyone who's into it, cool. But as for me.....I want a Maxwell based GPU for PC gaming, which is what consumer level Radeons and Geforce cards are designed and marketed for in the first place.
FTFY
If it makes you a bit happier, it takes 130Watts at stock voltage and 1200MHz OC,
Also, I measured it with killawatt
If it makes you a bit happier, it takes 130Watts at stock voltage and 1200MHz OC,
Also, I measured it with killawatt
[/B]
Not really truth. Here you can see that the card takes up to 140W while gaming.
Often taking 100Watts and more. We know that it doesn't downclock (throttle), so it is a matter of adjusting workload to get more than 75Watts constant consumption.
75W limit is as said previously, just a guidance for board manufacturers
hahahaha
Serious?
You see the graph and you spin it how ever you like. You say whatever you like, often taking 100watts? haha. Just as often only taking 25 watts. On average its 64watts. That is how you read the data. 64 watts, not 140.
This is a very demanding game. Puts the GPU to a total crawl, and the consumption is 64watts average.
Hmm. That's interesting when the 270X reviewed here:
http://www.anandtech.com/show/7400/the-radeon-r9-280x-review-feat-asus-xfx/20
Is getting 240W during gaming as the total system draw, and 284W in Furmark. So you're getting 130W total system draw. Alrighty... Apparently your Pitcairn card based system is drawing half the power of the Pitcairn system tested by AT, that's quite the accomplishment considering your 7870 is also overclocked. Do you have a 5W CPU with no HD or SSD? No fans at all? I wonder what would account for the huge power consumption difference of the pitcairn systems as measured by reviews and your apparent 130W pitcairn system.
Last edited:
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes + WCL Discussion Threads
- Started by Tigerick
- Replies: 24K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter