NVIDIA P102-100 Cryptomining Graphic Card Surfaces

[jabbadap]

That cooler looks like a carbon copy of the Zotac 1080Ti mini cooler, right down to the different sized fans. Weird.

Geh, both are PCPartner's brands so not so weird.

Just noticed there's a lot of mining products on inno3D side. Wonder how well they sell.

http://mining.inno3d.com/
http://www.inno3d.com/products.php?refid=398

 

[Valantar]

Geh, both are PCPartner's brands so not so weird.

Just noticed there's a lot of mining products on inno3D side. Wonder how well they sell.

http://mining.inno3d.com/
http://www.inno3d.com/products.php?refid=398

Ah, I didn't know that. Guess I was naive enough to assume Zotac was an independent company. Shouldn't be surprised, I guess - everything is a subsidiary of something else these days.

 

[bug]

Ah, I didn't know that. Guess I was naive enough to assume Zotac was an independent company. Shouldn't be surprised, I guess - everything is a subsidiary of something else these days.

Fwiw, they do have customized designs. I guess that counts more than who actually builds their goods.

 

[Casecutter]

but ether way at current payout levels mining is a waste of space.. i have 10 x 1070 cards running producing less than 15 dollars per day and its dropping.. roi at current hardware prices would be maybe two years but as i say the returns are dropping day by day..

Exactly my point for any Jonny-come-lately... As your hardware is probably paid for or getting close you're on the down hill slope, until that per/day production is close or less than the daily K/w usage. I'd see below $10 a day and it be time to divest of those 10 x 1070. Today some might pay $225 ($2250), in two/three weeks when everyone like you gets to the same realization I could very well see them worth like $125 apiece ($1125).

 

[cdawall]

I mean for the manufacturer (what on earth else would I mean? "In my view"? I don't even understand what that means.): how could it be cheaper for them to buy (very very expensive!) high-end GDDR5X chips and then partially disabling them through software, rather than, say, not populating all the channels? Now, my understanding of Eth performance is virtually nonexistent, but I get the impression (feel free to correct me!) that memory latency and VRAM size is what matters, not memory bandwidth? If so, wouldn't they save quite a bit on simply not populating those pads, and using the chips for other cards?

Memory bandwidth still plays a large roll. This is why a 192bit GP106 card can only hit 24-25mh/s while a 256bit GP104 card (or AMD polaris for that matter) can hit 31-32mh/s. Core configuration is not horridly important in this matter as the 470D, 470 and 480 (and their second gen counterparts) all maintain the same hashrate.

Amount of memory only matters as long as it is not full with the DAG size. So if you had a 4GB card and DAG was 2.37GB you are fine performance wont change if you have 8/12/16/10000000GB of memory. One of the things we have seen is issues with memory controller strength and "memory bashing" bringing down performance. That is why we saw the implementation of "compute mode" for polaris cards. This issue also existed in tahiti based cards that are especially weak when the memory controller starts getting maxed out. With a smaller DAG file the 79x0 and 280(x) can pull of 24mh/s, with a large DAG file that's down to 15mh/s.

Nvidia needs those channels populated for the bandwidth side of things, but the frame buffer limit from what I understand allows for some software trickery to get the latency down.

 

[trog100]

Exactly my point for any Jonny-come-lately... As your hardware is probably paid for or getting close you're on the down hill slope, until that per/day production is close or less than the daily K/w usage. I'd see below $10 a day and it be time to divest of those 10 x 1070. Today some might pay $225 ($2250), in two/three weeks when everyone like you gets to the same realization I could very well see them worth like $125 apiece ($1125).

yep similar to my own thoughts.. having said that i will probably see it through.. he he

i started off with the long term hodle approach and dont give up easy..

trog

 

[Casecutter]

yep similar to my own thoughts.. having said that i will probably see it through..

So at this point you probably paid those 10 cards off (twice over), so now you just run them till the power bill is more than you can hash. I suppose you can continue acquiring coin on the hope it see a run-up in value in the near term. Though with so much regulatory push, and then predatory theft/scams (wallets and exchanges etc.) it's hard to say how soon that might see any resurgence.

 

[trog100]

So at this point you probably paid those 10 cards off (twice over), so now you just run them till the power bill is more than you can hash. I suppose you can continue acquiring coin on the hope it see a run-up in value in the near term. Though with so much regulatory push, and then predatory theft/scams (wallets and exchanges etc.) it's hard to say how soon that might see any resurgence.

i dont just mine i have also bought coin.. i started off mining because i fancied building a mining machine.. this was last year.. buying coin was far more profitable than mining.. things have changed a little since then..

i am already down about 10K in US dollar terms from the peak.. i am still not showing a loss on my original investment but its getting close.. he he

but rightly or wrong i still expect crypto to recover and get back to its upward path.. currently its impossible to tell with any certainty what its gonna do..

trog