[ANN] Bitfury ASIC sales in EU and Europe

[sidehack]

If I'm thinking right about how they described the J/GH relationship at higher hashrates, 6-7W per chip is well below the expected upper range (potentially over 20W?). I'd guess the footprint is changed to something a fair bit larger.

[klondike_bar]

If I'm thinking right about how they described the J/GH relationship at higher hashrates, 6-7W per chip is well below the expected upper range (potentially over 20W?). I'd guess the footprint is changed to something a fair bit larger.

The Antminer S7 is about 10w/chip. Its likely these would be a similar footprint, providing ~200GH

[KNK]

6-7W per chip is at the guaranteed speed of 100GH/s
~10W is the maximum for air cooling and 140GH
~13W for 184GH/s and immersion cooling
the minimum for 50-60GH is ~4W ... it seems it is not impossible to have the same QFN48 after all ... let the dreams come true

[sidehack]

Are those posted numbers, or assumed numbers?

So, a bit of a question.

The design target was 40 gigahash per second with power efficiency of 0.06 joules per gigahash. On average, the measured power efficiency of tested engineering samples of the new BitFury Chip ranges from 0.055 joules per gigahash to 0.07 joules per gigahash
...
J / Gh metric in working modes starting from 55 Gh/s up to 180 Gh/s follows an almost linear relationship of 0.0011, while at 40 – 55 Gh/s measured slope converges to plateau.

So, let's see if I follow what they're saying. At 40-55GH the efficiency is pretty flat around 0.06J/GH (with deviation as noted above), and above this the J/GH increased with a slope of 0.0011

So the J/GH could be estimated by .0011(G-55) + 0.06 for any given hashrage "G" (greater than 55).

Which means at 55GH/0.06J they're drawing 3.3W and 100GH gets us about 0.11J/GH for an 11W chip. The 140GH air-cooling max sees 0.15J/GH and 21.5W dissipation. That's a pretty hot chip; I wonder how big it is to handle 22W under air cooling. That's getting into the A1 neighborhood, about where the A3222 ran if I'm remembering right. Quite a bit hotter than the BM1384, whose practical air-cooling limit was about 13W. Also the consideration that if it's doing that 21.5W at a tidy low voltage, say 0.6V, that's 36A per ASIC to get top clock.

I could be wrong, but that's how I understood the press release. How comfortable is QFN48 at 13W? BM1384 in a larger QFN56 aren't too uncomfortable with it, but if I'm figuring right your 13W would only get about 110GH and the 140GH actually clocks in at something like 20W. They only state the 0.06J/GH as the efficiency in the lowerbound 40-55GH operating, not across the whole range up to 180GH.

[Mitak]

If I'm thinking right about how they described the J/GH relationship at higher hashrates, 6-7W per chip is well below the expected upper range (potentially over 20W?). I'd guess the footprint is changed to something a fair bit larger.

I tottaly agree on this.
There is simply no way to cool  about 20W from small QFN ( 8x8 if I remember  correctly)

[KNK]

If I'm thinking right about how they described the J/GH relationship at higher hashrates, 6-7W per chip is well below the expected upper range (potentially over 20W?). I'd guess the footprint is changed to something a fair bit larger.

I tottaly agree on this.
There is simply no way to cool  about 20W from small QFN ( 8x8 if I remember  correctly)

How you got that 20W?
At that power the chip should provide 284GH/s not 184GH/s as stated ... maybe you mistyped 1 with 2 in your calculations

[Mitak]

20W would be based on worst case scenario.
Do have in mind that chip suplier numbers are usually a little bit "optimistic"

[KNK]

Your worst case scenario is probably at the wall with low grade ( 60% efficiency) PSU. 20W at the chip is close to 0.11J/GH too far from 0.07J/GH
The last time, the numbers provided where correct and this information was provided from real samples tests, not supplier expectations.
Let's just wait for the final chip specs.

[Mitak]

Let us hope your speculation is closer to the real numbers than mine

[Gyrsur]

is this the press release? http://www.businesswire.com/news/home/20151216005453/en/BitFury-Announces-Mass-Production-Fastest-Effective-16nm

and this the sales site? http://www.bitfurystrikesback.com/

[sidehack]

Your worst case scenario is probably at the wall with low grade ( 60% efficiency) PSU. 20W at the chip is close to 0.11J/GH too far from 0.07J/GH
The last time, the numbers provided where correct and this information was provided from real samples tests, not supplier expectations.
Let's just wait for the final chip specs.

Like I quoted directly from their press release, they stated 0.055-0.07 as the 40-55GH range and the J/GH increased from there with a slope of 0.0011. If I'm understanding what that means, then 0.07J/GH is not flat performance but it increases with operating frequency - which makes sense when you consider how everything else has ever worked. If I understand the formula correct based on what they said, then at 140GH the J/Gh is about 0.154, not 0.07, which puts you at over 21W per ASIC.

How I got 20W per chip, why it's far from 0.07J/GH, and other questions like that are actually not very helpful because they're the questions I'm trying to ask myself. I showed my work, and the rationale behind it, and am asking for a sanity check. So, I say again, are your numbers based entirely off about 0.07J/GH flat performance taken from any posted test data, or is it an assumption? Because my numbers are taken from how I interpret some fairly awkwardly handled posted information and I'd like to know (based on facts, not assumptions) if I'm wrong.

[KNK]

My bad, sorry!
I have calculated 0.0011 slope for the power not for J/Gh

[sidehack]

Does it look like I'm reading it right? More information would be nice, with a graph or posted data points for the range, but until we get that it's kinda guesswork.

[Mitak]

I think that if there is live prototype and some clue of engineering samples ETA , they can share the technical  specs , pin out etc.
This will give me at least something to do during the coming holidays

[KNK]

Does it look like I'm reading it right?

Yes, You are right:

Code:

 J/GH - 0.07 = 0.0011 * (GH - 55) 

so for 140GH: J/GH = 0.07 + 0.0011 * (140-55) = 0.07 + 0.0011 * 85 = 0.07 + 0.0935 = 0.1635 and then 140 * 0.1635 = 22.89W
if the chip is at 0.055 J/GH it is still 20.79W

[sidehack]

That's neighborhood of what I was thinking, yeah. 22W is pretty hot for air cooling unless it's probably at least a 10x10mm package, possibly double-side heatsinked? If that power range is right, it's very likely not footprint-compatible with the old chips.

[dogie]

That's neighborhood of what I was thinking, yeah. 22W is pretty hot for air cooling unless it's probably at least a 10x10mm package, possibly double-side heatsinked? If that power range is right, it's very likely not footprint-compatible with the old chips.

I don't know why everyone keeps quoting "for air cooling", they'll design their systems to run with their ridiculous immersion cooling first and foremost. What is possible with air cooling is an afterthought and doesn't need to hit the same specs.

[RoadStress]

Impressive work! Good job Bitfury!

[sidehack]

I quote air cooling stats because I have no rodent hindquarters to donate regarding immersion cooling. I want to play with these chips and I design for practical air cooling because that covers myself and all my target customers. I do not care at all what Bitfury builds for themselves or their millionaire buddies. That's my reason; can't speak for anyone else.

[rammy2k2]

Impressive work! Good job Bitfury!

*image removed*

good job for who ? for sure not for us small miners ...