Optiona is allways good. 16nm ASIC, 100w/THS 0.1J/GH on wall. MAX: 31THS/miner!!

[RichBC]

We are back with latest 16nm ASIC,supported by BITFURY.
1, 10THS/miner,
2,100W/THS, 0.25J/GH on wall
3, 4X 6pin PCI-E
4, Gross weight 5KG
5, size: 350x140x155(adjustable according the design)
the price is BTC0.42/THS, fob china.
miner ship at end of march or early April.

What the hell!!
What is that i am reading? the miner will have 0.25J/GH on wall??? the same eficience than a S7??  

100W /THS is 0.1J/GH and would be very good.  

Rich

Please, look at the first post, they add 0.25J/GH, not me...  

Ok sorry thought you had added that. But if they can't even get that right it just goes to show that they don't know what they are talking about...  

Rich

[J4bberwock]

We are back with latest 16nm ASIC,supported by BITFURY.
1, 10THS/miner,
2,100W/THS, 0.25J/GH on wall
3, 4X 6pin PCI-E
4, Gross weight 5KG
5, size: 350x140x155(adjustable according the design)
the price is BTC0.42/THS, fob china.
miner ship at end of march or early April.

What the hell!!
What is that i am reading? the miner will have 0.25J/GH on wall??? the same eficience than a S7??  

100W /THS is 0.1J/GH and would be very good.  

Rich

Please, look at the first post, they add 0.25J/GH, not me...  

Ok sorry thought you had added that. But if they can't even get that right it just goes to show that they don't know what they are talking about...  

Rich

Even the title of the thread have been updated, so maybe not a typo.
from

16nm ASIC, 10THS/MINER,100w/THS on wall.

to

16nm ASIC, 10THS/MINER,100w/THS 0.25J/GH on wall.

If those new numbers are real, there was no need to go to 16nm, and the price will be barely better than actual S7 pricing

[generalt]

If those new numbers are real, there was no need to go to 16nm, and the price will be barely better than actual S7 pricing

The price is worse if that's the case since you don't get it until April?

[xleejohnx]

Even the title of the thread have been updated, so maybe not a typo.
from

16nm ASIC, 10THS/MINER,100w/THS on wall.

to

16nm ASIC, 10THS/MINER,100w/THS 0.25J/GH on wall.

If those new numbers are real, there was no need to go to 16nm, and the price will be barely better than actual S7 pricing

yeah those numbers dont add up

[64dimensions]

From another thread this popped up: http://agenda.ge/news/48424/eng

It appears from this news that later in '15, Bitfury was still populating it's data centers with 28nm hashing chips.

[sidehack]

So they edit the thread title and contents to be self-contradictory? That doesn't make sense.

Also, of course Bitfury was filling their shelves with 28nm. I get no page from that link you posted, but they announced 0.2J/GH 28nm in February and just recently stated they had tested engineering samples of 16nm. Does anyone expect they went from tested engineering samples to complete miners in mass production in zero time?

[Mitak]

Yep they did in NEVER_LANd

[Brob12321]

Is there some sort of new technology for the board, its looks to be transparent and glowing blue ?  A transparent board would be pretty awesome if there's some material out there that could be used.  Honestly looks like the pics are completely fake. Is this thing even real ?

[Rabinovitch]

Is there some sort of new technology for the board, its looks to be transparent and glowing blue ?

It's obvious that it's just a nice picture. Just like in case of B-eleven miner.

[QuintLeo]

Probably a typo.
Because 4 PCIE plugs will already have some trouble for 1000w, but drawing 600w on each of them is impossible.
Even if it's a typo, I hope they will at least use 6 PCIE plugs for 1000w

 It would be nice if they would resolve their new inconsistancy - 100W/TH is 0.1 J/GH not 0.25

 On the other hand, just makes me more inclined to think "SCAM" when they can't even be consistant on basic specs.

 250 watts per PCI-E 6-pin connector is within the connector specs - the SP20E runs up to 288 watts/connector quite reliably. It's the morons at KnC that were trying to push over 300 watts/connector EXCEEDING the connector specs that had meltdown issues.

[asiabtc]

We are back with latest 16nm ASIC,supported by BITFURY.
1, 10THS/miner,
2,100W/THS, 0.25J/GH on wall
3, 4X 6pin PCI-E
4, Gross weight 5KG
5, size: 350x140x155(adjustable according the design)
the price is BTC0.42/THS, fob china.
miner ship at end of march or early April.

What the hell!!
What is that i am reading? the miner will have 0.25J/GH on wall??? the same eficience than a S7??  

100W /THS is 0.1J/GH and would be very good.  

Rich

Please, look at the first post, they add 0.25J/GH, not me...  

Ok sorry thought you had added that. But if they can't even get that right it just goes to show that they don't know what they are talking about...  

Rich

Even the title of the thread have been updated, so maybe not a typo.
from

16nm ASIC, 10THS/MINER,100w/THS on wall.

to

16nm ASIC, 10THS/MINER,100w/THS 0.25J/GH on wall.

If those new numbers are real, there was no need to go to 16nm, and the price will be barely better than actual S7 pricing

its mistake. has be corrected to 0.1J/GHS

[asiabtc]

We are back with latest 16nm ASIC,supported by BITFURY.
1, 10THS/miner,
2,100W/THS, 0.25J/GH on wall
3, 4X 6pin PCI-E
4, Gross weight 5KG
5, size: 350x140x155(adjustable according the design)
the price is BTC0.42/THS, fob china.
miner ship at end of march or early April.

What the hell!!
What is that i am reading? the miner will have 0.25J/GH on wall??? the same eficience than a S7?? 

its mistake. has been corrected to 0.1J/GH

[dogie]

250 watts per PCI-E 6-pin connector is within the connector specs

Its crazy how short everyone's memory is. When I suggested you could run the S7 off 4 PCI-E (~275W) everyone shit their pants, even though that's exactly what they'd already been doing for years.

[sidehack]

If you've got good 16AWG cables, sure. 18 gauge, forget about it. And when you get up close to capacity for your cables you want to check them evey now and then because if one pin isn't connecting properly you're overloading the other two which can cause heat, which can cause contact oxidation, which increases resistance which causes more heat. Bitmain likes to play it safe with power, which is why they put 4 connectors on the S3 (less than 100W each) - which makes the KFC Neptune hilarious in context with its single connector for an equal or larger power draw. A miner requesting more than about 250W per connector is a questionable design, and they darn better let you know what is and isn't an acceptable wire size to use on it.

[Finksy]

Its crazy how short everyone's memory is. When I suggested you could run the S7 off 4 PCI-E (~275W) everyone shit their pants, even though that's exactly what they'd already been doing for years.

It's crazy how poor your reading comprehension can be. Nobody was shitting their pants.  You suggested something that -although is within the guidelines for the wiring and connector- clearly violates Bitmain's warranty policy, and effectively voids it.  Something you have been very sure to inform others not to do (and to include disclaimers when they suggest doing such things to others) in the past. The hypocrisy was the issue, not the specs. And, your response to that was "Bitmain is wrong" in this case. 

Tell me, do you think Bitmain would hear the reasoning on PCIe and wiring specs if these boards would have only had 2x PCIe plugs melted to them instead of 3, regardless of whether or not it would have made a difference? (Hint: I'm aware it wouldn't have made a difference, except in the denial of warranty)

[dogie]

Tell me, do you think Bitmain would hear the reasoning on PCIe and wiring specs if these boards would have only had 2x PCIe plugs melted to them instead of 3, regardless of whether or not it would have made a difference? (Hint: I'm aware it wouldn't have made a difference, except in the denial of warranty)

What you've posted was clearly not caused by the PCI-E connectors so.................................................................. Yeah now I remember it was mostly you that was flipping your shit, because you wanted to sell more of your $60 breakout boards and $4 PCI-E cables.

Anyway, back to the actual topic of the thread....

Its crazy how short everyone's memory is. When I suggested you could run the S7 off 4 PCI-E (~275W) everyone shit their pants, even though that's exactly what they'd already been doing for years.

[sidehack]

The point that it voids Bitmain's warranty is still a valid concern. Not exactly valid for the actual topic of the thread, which is "16nm ASIC, 10THS/MINER,100w/THS 0.1J/GH on wall".

It would be nice to see at least six connectors on a 1KW miner. SP20s made me a bit leery when some clockings were asking for around 300W per connector. None of my hosted machines smoked out, but I did see one or two connectors starting to brown a bit.

[Finksy]

What you've posted was clearly not caused by the PCI-E connectors so.................................................................. Yeah now I remember it was mostly you that was flipping your shit, because you wanted to sell more of your $60 breakout boards and $4 PCI-E cables.

Good god you are truly helpless. You should really work on your reading comprehension if you still plan to go to University one of these days.

Yes, back to the topic at hand, dogie looking for re-assuring pats on the head to feel important.  Good dogie, you were right about something that absolutely nobody was questioning and is completely self-evident given published data. Have a cookie.

I'm not comfortable with over 250W on a PCIe cable personally.  I ran many S5's with single 1.35mm^2 PCIe cables per board (closer to 15awg cabling) and they all turned the connectors brown over time, and that is only about 275W.  If it's 1000W at the wall after PSU efficiency, it would probably be fine.  But if they want to leave room for over-clocking, or that is DC wattage then they would definitely be better off w/ 6 connectors.

[tutorialevideo]

Its crazy how short everyone's memory is. When I suggested you could run the S7 off 4 PCI-E (~275W) everyone shit their pants, even though that's exactly what they'd already been doing for years.

It's crazy how poor your reading comprehension can be. Nobody was shitting their pants.  You suggested something that -although is within the guidelines for the wiring and connector- clearly violates Bitmain's warranty policy, and effectively voids it.  Something you have been very sure to inform others not to do (and to include disclaimers when they suggest doing such things to others) in the past. The hypocrisy was the issue, not the specs. And, your response to that was "Bitmain is wrong" in this case. 

Tell me, do you think Bitmain would hear the reasoning on PCIe and wiring specs if these boards would have only had 2x PCIe plugs melted to them instead of 3, regardless of whether or not it would have made a difference? (Hint: I'm aware it wouldn't have made a difference, except in the denial of warranty)

This is crazy man is the first s7 on fire that i seen what is the main couse?

[0bit]

I would definitely need to see a demo or a review of this before I would even think to consider it. That price alone is a huge turn off for me on this product. Caveat emptor indeed.