Bitfury: "16nm... sales to public start shortly" - page 54.

punin

hero member

Activity: 560

Merit: 500

Quote from: carlosmnk on January 17, 2016, 04:35:35 AM

Quote from: punin on January 17, 2016, 03:18:24 AM

Quote from: mavericklm on January 16, 2016, 09:38:49 PM

in the other thread it was April!

tomorrow it will be February!

Engineering chips are going to integrators already in February. So for a skilled and fast integrator I would say late march is possible. April is 100%.

... And what is the hashrate with heatsink and what's draining the chip?? Grin

See for yourself: https://www.youtube.com/watch?v=zZPum2zqGPE

Mitak

full member

Activity: 126

Merit: 100

Yes, so I said top heat sink is in the plan

el_rlee

legendary

Activity: 1600

Merit: 1014

Quote from: Mitak on January 19, 2016, 06:04:40 PM

Quote from: sidehack on January 19, 2016, 05:28:19 PM

Quote from: smracer on January 19, 2016, 03:44:11 PM

Quote from: Mitak on January 19, 2016, 03:00:07 AM

I am under NDA .
So just take my word for it.

How much do you think a board would cost to run 20 chips per board? Subtract the price of the BF chips so I am just talking about the board.

20 chips should run 2TH at .11w/gh so 220W per board. How big would the board need to be? 10" X 10"?

Qty 500 boards.

I am just curious as to the cost of the board in qty.

I don't have a ballpark price on hand but that's not too far from what we're looking to build. A pair of my boards would do 2TH/220W in approximately 6"x9" of PCB and screw onto an S1 heatsink.

What I am planing ( of course if ever) is 180x120 mm PCB for this kind of chip count with top(just in case) and bottom heat sink .
Any price range mentioned would be a big big speculation.
Just the bare PCB can be calculated for now and there is probability for change too.
For sure I would like to design a single USB stick type miner too

I think you would need the top heat sink because of the FC package.

Mitak

full member

Activity: 126

Merit: 100

Quote from: sidehack on January 19, 2016, 05:28:19 PM

Quote from: smracer on January 19, 2016, 03:44:11 PM

Quote from: Mitak on January 19, 2016, 03:00:07 AM

I am under NDA .
So just take my word for it.

How much do you think a board would cost to run 20 chips per board? Subtract the price of the BF chips so I am just talking about the board.

20 chips should run 2TH at .11w/gh so 220W per board. How big would the board need to be? 10" X 10"?

Qty 500 boards.

I am just curious as to the cost of the board in qty.

I don't have a ballpark price on hand but that's not too far from what we're looking to build. A pair of my boards would do 2TH/220W in approximately 6"x9" of PCB and screw onto an S1 heatsink.

What I am planing ( of course if ever) is 180x120 mm PCB for this kind of chip count with top(just in case) and bottom heat sink .
Any price range mentioned would be a big big speculation.
Just the bare PCB can be calculated for now and there is probability for change too.
For sure I would like to design a single USB stick type miner too

Mitak

full member

Activity: 126

Merit: 100

Quote from: el_rlee on January 19, 2016, 05:08:32 PM

Quote from: smracer on January 19, 2016, 03:44:11 PM

Quote from: Mitak on January 19, 2016, 03:00:07 AM

I am under NDA .
So just take my word for it.

How much do you think a board would cost to run 20 chips per board? Subtract the price of the BF chips so I am just talking about the board.

20 chips should run 2TH at .11w/gh so 220W per board. How big would the board need to be? 10" X 10"?

Qty 500 boards.

I am just curious as to the cost of the board in qty.

The number of chips doesn't make sense I guess... Chip count times chip core voltage should fit the DC voltage of the power supply.

In fact this is not true
You can always use some single DC-DC on the input and arrange it according to chip count

sidehack

legendary

Activity: 3374

Merit: 1859

Curmudgeonly hardware guy

Quote from: smracer on January 19, 2016, 03:44:11 PM

Quote from: Mitak on January 19, 2016, 03:00:07 AM

I am under NDA .
So just take my word for it.

How much do you think a board would cost to run 20 chips per board? Subtract the price of the BF chips so I am just talking about the board.

20 chips should run 2TH at .11w/gh so 220W per board. How big would the board need to be? 10" X 10"?

Qty 500 boards.

I am just curious as to the cost of the board in qty.

I don't have a ballpark price on hand but that's not too far from what we're looking to build. A pair of my boards would do 2TH/220W in approximately 6"x9" of PCB and screw onto an S1 heatsink.

el_rlee

legendary

Activity: 1600

Merit: 1014

Quote from: smracer on January 19, 2016, 03:44:11 PM

Quote from: Mitak on January 19, 2016, 03:00:07 AM

I am under NDA .
So just take my word for it.

How much do you think a board would cost to run 20 chips per board? Subtract the price of the BF chips so I am just talking about the board.

20 chips should run 2TH at .11w/gh so 220W per board. How big would the board need to be? 10" X 10"?

Qty 500 boards.

I am just curious as to the cost of the board in qty.

The number of chips doesn't make sense I guess... Chip count times chip core voltage should fit the DC voltage of the power supply.

smracer

donator

Activity: 1057

Merit: 1021

Quote from: Mitak on January 19, 2016, 03:00:07 AM

I am under NDA .
So just take my word for it.

How much do you think a board would cost to run 20 chips per board? Subtract the price of the BF chips so I am just talking about the board.

20 chips should run 2TH at .11w/gh so 220W per board. How big would the board need to be? 10" X 10"?

Qty 500 boards.

I am just curious as to the cost of the board in qty.

Mitak

full member

Activity: 126

Merit: 100

I am under NDA .
So just take my word for it.

kilo17

legendary

Activity: 980

Merit: 1001

aka "whocares"

Quote from: Mitak on January 19, 2016, 02:39:35 AM

I took part in team for about 3 different mining boards designs.
A lot of experience with electronic projects etc.
The software is not the problem just a drivers for open source .
Of course the firmware and the pcb is the key work here.

Nice, just curious. Care to elaborate a bit on the boards you have worked on, I am intrigued to say the least.

Mitak

full member

Activity: 126

Merit: 100

I took part in team for about 3 different mining boards designs.
A lot of experience with electronic projects etc.
The software is not the problem just a drivers for open source .
Of course the firmware and the pcb is the key work here.

kilo17

legendary

Activity: 980

Merit: 1001

aka "whocares"

Quote from: alh on January 18, 2016, 08:47:24 PM

Quote from: kilo17 on January 18, 2016, 08:26:31 PM

Quote from: Mitak on January 18, 2016, 05:29:23 PM

there are at least 3-4 teams(persons) , that have the needed experience and know how as an example:
sidehack , vs3 , me and a few others that did not write here lately I think

I am just curious about what qualifications you have to make an efficient miner? I am not asking to be an ass but out of genuine curiosity. I would go out on a limb and say that your average electronics guy cannot make an efficient miner, let alone the software to run it.

While I can't comment on Mitak, I can speak as a satisfied customer of sidehack. His Compac USB stick works just great. This includes a BM1384 (i.e. Antminer S5 class) part and has a very broad range of power speed options.

I would be willing to bet he could make a dandy USB stick based on their part that would be say 50GH or possibly more. If Bitfury would talk with him, and supply samples at a reasonable price, things would be swell.

Just my unsolicted $.02 on at least one members capabilities.

Sorry, I was asking about Mitak not Sidehack Et al.

sidehack

legendary

Activity: 3374

Merit: 1859

Curmudgeonly hardware guy

If we're looking at wrapping back around to our original TypeZero idea, which would be a quarter S1, probably ten chips to a board for about 4TH/400W stock setting on a four-board kit - which, with a decent heatsink and 80 or 120mm fan could be a standalone 100W pod miner. If I built a stickminer off it I'd want to get at least a 100GH top end capable. That's the main things I'd work on, probably not bother with anything bigger because someone else will probably do it cheaper and mass-produce it for big players. Y'all know Novak and I prefer to build for the people like us.

alh

legendary

Activity: 1846

Merit: 1052

Quote from: kilo17 on January 18, 2016, 08:26:31 PM

Quote from: Mitak on January 18, 2016, 05:29:23 PM

there are at least 3-4 teams(persons) , that have the needed experience and know how as an example:
sidehack , vs3 , me and a few others that did not write here lately I think

I am just curious about what qualifications you have to make an efficient miner? I am not asking to be an ass but out of genuine curiosity. I would go out on a limb and say that your average electronics guy cannot make an efficient miner, let alone the software to run it.

While I can't comment on Mitak, I can speak as a satisfied customer of sidehack. His Compac USB stick works just great. This includes a BM1384 (i.e. Antminer S5 class) part and has a very broad range of power speed options.

I would be willing to bet he could make a dandy USB stick based on their part that would be say 50GH or possibly more. If Bitfury would talk with him, and supply samples at a reasonable price, things would be swell.

Just my unsolicted $.02 on at least one members capabilities.

kilo17

legendary

Activity: 980

Merit: 1001

aka "whocares"

Quote from: Mitak on January 18, 2016, 05:29:23 PM

there are at least 3-4 teams(persons) , that have the needed experience and know how as an example:
sidehack , vs3 , me and a few others that did not write here lately I think

I am just curious about what qualifications you have to make an efficient miner? I am not asking to be an ass but out of genuine curiosity. I would go out on a limb and say that your average electronics guy cannot make an efficient miner, let alone the software to run it.

smracer

donator

Activity: 1057

Merit: 1021

Quote from: sidehack on January 18, 2016, 05:56:13 PM

The design Novak and I are working on should be flexible that we could port it to a different chip without a lot of problem. I'd love to get my hands on some new BF silicon.

How many chips would you put per board?

notlist3d

legendary

Activity: 1456

Merit: 1000

Quote from: wttbs on January 18, 2016, 09:55:44 AM

Quote from: adaseb on January 18, 2016, 09:46:28 AM

So since these chips are so efficient and powerful. Why sell them and not just mine themselves instead?

You are right about that, but I don't have $1M so I can't buy chips to build miners and go mining... Even if I want to buy rev3 chips for like $10000 or $50000, no chance how things look right now...

The reason is the MOQ is so high it helps out on r/d costs, costs of their own machines, etc. They sell in such big quanities that they will make a good profit.

So if they have the opportunity to sell a few bulk and make a few million easy quick ROI by selling chips. I predict they will do it, the trick is meeting that MOQ.

sidehack

legendary

Activity: 3374

Merit: 1859

Curmudgeonly hardware guy

The design Novak and I are working on should be flexible that we could port it to a different chip without a lot of problem. I'd love to get my hands on some new BF silicon.

Mitak

full member

Activity: 126

Merit: 100

there are at least 3-4 teams(persons) , that have the needed experience and know how as an example:
sidehack , vs3 , me and a few others that did not write here lately I think

smracer

donator

Activity: 1057

Merit: 1021

Quote from: tutorialevideo on January 18, 2016, 03:34:25 PM

Wee can make a group buy on these cips?

If we can get someone to design a board I would participate in a group buy.

Topic: Bitfury: "16nm... sales to public start shortly" - page 54. (Read 108551 times)