[LABCOIN] IPO [BTCT.CO] - Details/FAQ and Discussion (ASIC dev/sales/mining)

[bitwhizz]

Sorry if this has been asked & answered, just curious which is more likely to be true:

1.  Labcoin has no chips.
2.  The chips exist but are junk.
3.  The chips hash but don't clock up enough before thermal runaway / some other problem where they hash, but not well enough to be cost-effective.
4.  Soft/firmware tweakable problems -- will be hashing in no time.

Or

5.  Everything's working fine, the chips are hashing, Labcoin is already hashing & not releasing the news to cause panic & buy up their shares for nothing.
 

everything is a possibility but number 5

[1713990931]

1713990931

[1713990931]

1713990931

[BitThink]

Sorry if this has been asked & answered, just curious which is more likely to be true:

1.  Labcoin has no chips.
2.  The chips exist but are junk.
3.  The chips hash but don't clock up enough before thermal runaway / some other problem where they hash, but not well enough to be cost-effective.
4.  Soft/firmware tweakable problems -- will be hashing in no time.

Or

5.  Everything's working fine, the chips are hashing, Labcoin is already hashing & not releasing the news to cause panic & buy up their shares for nothing.
 

According to the theSeven's chat log posted couple of pages before, chips have a problem of keep crash and restarting. Not sure it's chip, board, or driver. He just said it seems fixable. Your can search and read it by yourself.

[BitThink]

TheSeven also mentioned that the testing chips were 'sneaked' into a universty fab to reduce cost (one of them works in a university). So it the chip has some problem, there's still a small chance it was due to the not so good quality of the product line.

[Puppet]

What are you talking about? I seriously doubt HashFast, Cointerra or KnC spent 9 months on their designs

Cointerra started in April and only promises hardware by late december. Are you giving their all star team less credit than Theswede?
KNC seems to have pulled it off a bit faster, although I dont know when they actually started. Just that they were already working on it in April, claim to have broken all speed records  and they are using orsoc, a large and experiienced team.

Note that neither of them have so far delivered a working asic.  A company that has is called BFL. Not small or underfunded either. How long did it take them again?

HashFast taped out like a week ago and expects first silicon by the middle of October.

They actually taped out end of august and say they promise shipping miners will start end of october.
But how long did it take them to reach tape out? And by them I mean Uniquify. Again a large and experienced team, and no working silicon yet, much less working miners, so lets just wait and see if they actually pull it off.

and they were 28nm.  Seems pretty obvious that the larger the feature size the shorter the development time.

If you're doing standard cell cmos it should be pretty easy to do a die shrink, you only have to design at the gate level.

I cant rhyme those two statements.

What exactly is it you think people have to do that would take, say, 3 months from when they determined the process node to when they taped out and ordered silicon from the fab?

I never mentioned tape out. I said working silicon. From tape out to working silicon is typically 2 months or more. If you believe they are close to tape out, but havent even selected a process yet, I got a bridge to sell.

[oxideNL]

The lab was located in Hong Kong right?
Than don't be alarmed if they turn silent in the next coming days
http://www.weather.com/news/weather-hurricanes/typhoon-usagi-threatens-taiwan-hong-kong-20130919

[crumbs]

Thanks all.  The only problem with that for me is their refusal to post pics.  Just seems unlikely that a group of people with access to a fab don't have access to a $100 digital cam, or are able to sneak into a fab & not able to sneak a pic out of China.  Just my take.

[BitThink]

Thanks all.  The only problem with that for me is their refusal to post pics.  Just seems unlikely that a group of people with access to a fab don't have access to a $100 digital cam, or are able to sneak into a fab & not able to sneak a pic out of China.  Just my take.

It is possible that they not only sneaked into the fab, but also the machine room and the racks. Disclosure of photos of the rack and the room may cause him lose the job. But still why they cannot provide an assembled board beyonds my understanding.

[BitThink]

The lab was located in Hong Kong right?
Than don't be alarmed if they turn silent in the next coming days
http://www.weather.com/news/weather-hurricanes/typhoon-usagi-threatens-taiwan-hong-kong-20130919

No, typhoon will not affect the Internet. Moreover, there's no lab, but just one manager and a couple of part time engineers having their own day jobs.

[Puppet]

From tape out to working silicon is typically 2 months or more. If you believe they are close to tape out, but havent even selected a process yet, I got a bridge to sell.

One more reference point. New entry: black arrow. They are not a huge company, and I dont know if they partnered with anyone else, but at least they do have some relevant bitcoin hardware experience and a good reputation. THey are now promising delivery of their new asics and miners a few days before March 2014:
http://www.blackarrowsoftware.com/store/minion.html

Im assuming they taped out, at the very least they picked a process and have finished the design and simulations. You think its likely labcoin will beat them to the punch? Im willing to bet with unreasonably uneven odds Labcoin wont have their own 65nm asic hashing  6 months from here.

[Ytterbium]

Cointerra started in April and only promises hardware by late december. Are you giving their all star team less credit than Theswede?
KNC seems to have pulled it off a bit faster, although I dont know when they actually started. Just that they were already working on it in April, claim to have broken all speed records  and they are using orsoc, a large and experiienced team.

It's TheSeven, not TheSwede who's working on the 65nm design.  And obviously I realize it takes longer and costs more the smaller the feature size. But this is 65nm, not 28nm.

Also, you didn't answer my question, which is:

What exactly is it you think people have to do that would take, say, 3 months from when they determined the process node to when they taped out and ordered silicon from the fab?

If you don't know what the actual steps are, how can you say you know how long it takes to do them? If you know, you should be able to answer. If you don't, then you really can't say.

Note that neither of them have so far delivered a working asic.  A company that has is called BFL. Not small or underfunded either. How long did it take them again?

BFL is run by morons. How long did it take between the time bitfury settled on a 55nm design for them to have working silicon? The question isn't about when they started, but rather when exactly they were locked in to a process node.

They actually taped out end of august and say they promise shipping miners will start end of october.
But how long did it take them to reach tape out? And by them I mean Uniquify. Again a large and experienced team, and no working silicon yet, much less working miners, so lets just wait and see if they actually pull it off.

Again, the question isn't when they started on the design, but when they were locked into a process node.

and they were 28nm.  Seems pretty obvious that the larger the feature size the shorter the development time.

If you're doing standard cell cmos it should be pretty easy to do a die shrink, you only have to design at the gate level.

I cant rhyme those two statements.

There are two steps in developing a chip, doing the logic and doing the physical layout.  The smaller the process node, the more expensive it is to do the layout. More expense implies more man-hours, but not necessarily more actual calender  time.

What exactly is it you think people have to do that would take, say, 3 months from when they determined the process node to when they taped out and ordered silicon from the fab?

I never mentioned tape out. I said working silicon. From tape out to working silicon is typically 2 months or more. If you believe they are close to tape out, but haven't even selected a process yet, I got a bridge to sell.

Why can't you actually answer that question.  What are the steps that need to be done between when the process node is fixed, and tapeout occurs.  And how much time do those steps take to do? (and how does that timescale differ at different process nodes)

It doesn't make much sense for you to claim you know how long it takes to do something if you can't even say what you think the steps actually are.

You could very well be right, I've never taped out an IC. But unless you can explain how you're deriving your time estimates it's not clear why I should just assume they're correct.

[Ytterbium]

From tape out to working silicon is typically 2 months or more. If you believe they are close to tape out, but havent even selected a process yet, I got a bridge to sell.

One more reference point. New entry: black arrow. They are not a huge company, and I dont know if they partnered with anyone else, but at least they do have some relevant bitcoin hardware experience and a good reputation. THey are now promising delivery of their new asics and miners a few days before March 2014:
http://www.blackarrowsoftware.com/store/minion.html

Im assuming they taped out, at the very least they picked a process and have finished the design and simulations. You think its likely labcoin will beat them to the punch? Im willing to bet with unreasonably uneven odds Labcoin wont have their own 65nm asic hashing  6 months from here.

Those are 28nm, not 65nm. And again, what exactly is it that you think they have to do between when the process node is fixed and when they tape out? What are the actual steps?  How do we know blackarrow couldn't drop down to a 32 or 45nm node between now and march? If the price of bitcoin went up to $1k and they could afford it, could they switch to a 22nm node? When would the deadline be for that?

The question isn't "how long is it between starting an ASIC design and having finished silicon" but rather "how late in the process can they change process nodes?" - you say it has to be 9 months before finished silicon, but you haven't presented any evidence that that's the case.

[dopamine]

the public notice announcement to spread FUD is hilarious

[Puppet]

I have never taped out a thing in my life either, and my experience with the industry is image sensors, not asics, and not even from design POV. Most of the steps will be similar, but lets not waste time going over that.   I am not a rocket scientist either, and cant tell you all the steps involved in launching a rocket and docking it with the space station. I can tell you you almost certainly cant do it 6 weeks if you start from scratch. Simply deducting from common sense how long it took others.

How long did it take between the time bitfury settled on a 55nm design for them to have working silicon?

I dont know, but I do know bitfury raised the idea of developing his own asic back in may of 2012, I know they taped out in March this year, had prototypes hashing in July  and have only recently been shipping in volume.
I might also add, if Bitfury was involved in this project, Id probably have bought in.

Anyway, your argument seems to be they may be further ahead in the design and closer to tapeout that Im willing to believe, and that may be so. But you should be asking Theswede for evidence of that rather than asking me to prove the negative. Everything I have seen so far points towards little more than vague ideas and not being anywhere near tape out. In fact, Ive not seen anything that leads me to believe there will ever be one "rev2".  TheSeven has recently been contracted to do "some work". Fine, but  my guess is that he has been asked to come up with some numbers, rough projections.  If that had already been done, wouldnt we have heard them? So he is probably not doing more a preliminary feasibility study,  One that should lead to canceling this whole project, if there ever was one. And if it doesnt, you can go by any of the other bitcoin asics to get an idea of when it could realistically be deployed. Not this year, that much Im very confident off. Next summer seems a lot more plausible and financially unfeasible.

[JohnyBigs]

Again all you guys miss the point,  the development was already funded by Labcoin itself, ipo money is just used to purchase chips hardware ect.

So they have been working on these chips for probably over a yesr.

[Ytterbium]

I have never taped out a thing in my life either, and my experience with the industry is image sensors, not asics, and not even from design POV. Most of the steps will be similar, but lets not waste time going over that.   I am not a rocket scientist either, and cant tell you all the steps involved in launching a rocket and docking it with the space station. I can tell you you almost certainly cant do it 6 weeks if you start from scratch. Simply deducting from common sense how long it took others.

First of all, common sense is actually a pretty terrible way to figure things out. You either do research and find out, or admit you don't know and don't care enough to bother finding out.

Only NASA, The Russians, and Space X have ever built rockets that carried capsules that have docked with the space station.  I wouldn't even try to guess how many people have designed integrated circuits.

Either way, you didn't claim it would take a long time to design a chip.  You said it would take nine months between when the process node size was fixed and couldn't be changed, and when the chips were ready for use.  

Now, apparently you're telling me you have no idea why that would be the case, other then "Common sense".  

Switching the process node would be more like adjusting the payload capacity of a rocket. While it's still in the blueprint phase before construction begins.  Could you decide you need 5 engines instead of 4 six weeks before construction starts? Or even after construction has begun?

The design for the Saturn V rocket was changed from four engines to five - It didn't take six weeks, but this was all done with pencil, paper and slide rules, no CAD, no computers at all.

TheSeven said he was using a hardware description language to generate the chips.  I don't exactly know what it is that would make it take longer to generate, for example, a 40nm chip design compared to 65nm chip. Or why it would take 6 months to do a die shrink. As far as I know the only difference is how much the intermediary companies and fab charge you to actually do the tapeout.

How long did it take between the time bitfury settled on a 55nm design for them to have working silicon?

I dont know, but I do know bitfury raised the idea of developing his own asic back in may of 2012, I know they taped out in March this year, had prototypes hashing in July  and have only recently been shipping in volume.
I might also add, if Bitfury was involved in this project, Id probably have bought in.

But the question is, what was the latest they could have switched to a different feature size.

Anyway, your argument seems to be they may be further ahead in the design and closer to tapeout that Im willing to believe, and that may be so. But you should be asking Theswede for evidence of that rather than asking me to prove the negative. Everything I have seen so far points towards little more than vague ideas and not being anywhere near tape out. In fact, Ive not seen anything that leads me to believe there will ever be one "rev2".  TheSeven has recently been contracted to do "some work". Fine, but  my guess is that he has been asked to come up with some numbers, rough projections.  If that had already been done, wouldnt we have heard them? So he is probably not doing more a preliminary feasibility study,  One that should lead to canceling this whole project, if there ever was one. And if it doesnt, you can go by any of the other bitcoin asics to get an idea of when it could realistically be deployed. Not this year, that much Im very confident off. Next summer seems a lot more plausible and financially unfeasible.

I don't know where they are in their design. I don't know what steps they need to go through between having having a verilog codebase they can simulate and having a GDSII file or mask set that they can use to produce chips, how long those steps take, how much they cost, etc for various process nodes.

But there doesn't seem to be much basis for saying that the process node needs to be fixed 9 months before chips finish.

[Ytterbium]

Been a while since the last deposit. No sign of the promised update. 

It just came in.  1.7BTC over 3 hours, 30 minutes.

[Puppet]

But there doesn't seem to be much basis for saying that the process node needs to be fixed 9 months before chips finish.

All the evidence points in that direction, that especially for a small team it does take that much time.
Need one more ? Asicminer started their design in june  2012, selected the process and published preliminary specs in July,  taped out end of september, started hashing in February.
https://bitcointalk.org/index.php?topic=91173.msg1003326#msg1003326

Friedcat is no idiot, wasnt underfunded, used an old "simple" process node,  but it took  8 months between selecting their process node and hashing. What basis do you have to assume labcoin could go so much faster?

[JohnyBigs]

Been a while since the last deposit. No sign of the promised update.  

It just came in.  1.7BTC over 3 hours, 30 minutes.

Another .17btc came in

[utens]

After all there could be something good goin on, seems like improving..

[KCBitcoin]

so they started mining with their Avalons again?