BFL ASIC specifications

[peterepeat]

BFL are showing some entrepreneurial initiative.
By selling chips, building on Avalon's approach by only asking half up front is an improvement, and shows a more attractive approach  for purchasing, than what Avalon offered.(they would have to, not being the first) 
Simple business model with good cashflow and margin built in with much less overhead - it makes sense.
It is difficult for a buisness to thrive with negative cashflow. This is exacerbated by negative customer sentiment, and can lead a buisness into a money losing spiral. To see BFL adapting to the market and making new offers is encouraging, as it provides competition to the encumbent monopoly of Avalon, and adds diversity to the emerging ASIC embedded platforms under development.
Hopefully it will allow BFL to survive longer, and grow as a business, and finally discover the benefits of looking after their customers.

[ecliptic]

Avalon is a monopoly like the GPL is a monopoly

[newmars]

BFL should release Reference documentation for people to acquire more information of the chip application.

[Schrankwand]

Funny. IF those are Global Foundries chips... why did they ever wait for shipments from China? GF has manufacturing plants in Germany, Singapore and Saratoga, NY. None in China.

[SLok]

Funny. IF those are Global Foundries chips... why did they ever wait for shipments from China?

They did not?

[dogie]

Funny. IF those are Global Foundries chips... why did they ever wait for shipments from China?

They did not?

To be honest, none of those places other than NY have particularly faster transit times for ocean freight due to the INSANE volume going from China to the US. Its like a mail box that gets collected by the mail man every 5 minutes.

[Schrankwand]

Funny. IF those are Global Foundries chips... why did they ever wait for shipments from China?

They did not?

I must have been mistaken then. Somehow, I have often read that they were waiting for shipments from China...

And Germany to US, with an express mailing service, insured, takes around 24 hours if you consider UPS or FEDEX. They are insanely fast, if you want it. And if I was BFL... I'd want that

[erk]

Funny. IF those are Global Foundries chips... why did they ever wait for shipments from China?

They did not?

I must have been mistaken then. Somehow, I have often read that they were waiting for shipments from China...

And Germany to US, with an express mailing service, insured, takes around 24 hours if you consider UPS or FEDEX. They are insanely fast, if you want it. And if I was BFL... I'd want that

I think some people were speculating that BFL used TSMC which is in Taiwan.

[mobodick]

   Design type: 100% Hand routed for performance density

Bullshit. They should change the supplier.

They must have found the shittiest design and verification engineers on the planet to get such low from a full custom 65nm asic.  Further, for a chip with such little logic on it their yields are crap.

My guess would be they used shitty engineers and global foundaries' standard cell library, the engineers didn't setup their simulations correctly and as such the cores fail under even the slightest real world heat.  This guess is additionally backed up by the high hardware error to accepted rates posted in screenshots by various Jalapeno recipients.   I would guess that the estimated hash rate on the pool side is much lower than the promised 5 gh/s.

12 posts and already trolling, what's the name of your other account?

No other account.  Just someone who is familiar with IC design and thinks such a small chip (7.5mm^2) consisting of no more than 16k 32-bit adders on a process that yields such low clock rates (for example see: http://ieeexplore.ieee.org/xpl/login.jsp?tp=&arnumber=4253311) and high defect rates (to quote from your original link: "A grade has 16 engines, B grade has 15 engines, C grade has 14 engines and D grade has no less than 12 engines.  All chips run at a minimum of 250 mhz.  Higher grade chips will run up to 294mhz.  The percentage distribution in each lot is 60% Grade A, 20% Grade B, 15% Grade C and 5% Grade D.").

If the error rate was this high, a single 300mm wafer would be luckly to yield one working consumer graphics card.

One mistake in your proposition is to assume that the delivered 'mix' is a fair representation of the yield.
I think BFL is perfectly capable of keeping a significant portion of the grade A chips for themselfs.
They (will) have many products to support and i can easily imagine they don't want to put lower grades in their devices. They also made promises to their customers about hashrtes and power so they actually need the higher grade chips.
So it would seem natural for them to want to sell off as much of the lower grades as possible.
So i don't think we can say anything about their actual yield of grade A chips.

[erk]

One mistake in your proposition is to assume that the delivered 'mix' is a fair representation of the yield.
I think BFL is perfectly capable of keeping a significant portion of the grade A chips for themselfs.
They (will) have many products to support and i can easily imagine they don't want to put lower grades in their devices. They also made promises to their customers about hashrtes and power so they actually need the higher grade chips.
So it would seem natural for them to want to sell off as much of the lower grades as possible.
So i don't think we can say anything about their actual yield of grade A chips.

Considering the ASIC's haven't even been ordered yet, as BFL are still ascertaining the final numbers before they pace the order, you have no idea what the yield will be. The process might have improved dramatically since the last wafers. They may not be able to deliver the low grade chips in volume.

[mobodick]

One mistake in your proposition is to assume that the delivered 'mix' is a fair representation of the yield.
I think BFL is perfectly capable of keeping a significant portion of the grade A chips for themselfs.
They (will) have many products to support and i can easily imagine they don't want to put lower grades in their devices. They also made promises to their customers about hashrtes and power so they actually need the higher grade chips.
So it would seem natural for them to want to sell off as much of the lower grades as possible.
So i don't think we can say anything about their actual yield of grade A chips.

Considering the ASIC's haven't even been ordered yet, as BFL are still ascertaining the final numbers before they pace the order, you have no idea what the yield will be. The process might have improved dramatically since the last wafers. They may not be able to deliver the low grade chips in volume.

I don't think that too little high grade chips will be a problem.

They may have a botched wafer that has a bugged design on it with almost no grade A chips from the yield.
They could mix that with a fixed wafer to somewhat recoup the costs of the bad one.

[Schrankwand]

Funny. IF those are Global Foundries chips... why did they ever wait for shipments from China?

They did not?

I must have been mistaken then. Somehow, I have often read that they were waiting for shipments from China...

And Germany to US, with an express mailing service, insured, takes around 24 hours if you consider UPS or FEDEX. They are insanely fast, if you want it. And if I was BFL... I'd want that

I think some people were speculating that BFL used TSMC which is in Taiwan.

It explains why they would have money problems though, GF isn't really the "cheapest" place. They are among the best, though.

[erk]

It explains why they would have money problems though, GF isn't really the "cheapest" place. They are among the best, though.

They got at least $7million in pre-order payments if you can believe http://bfl.ptz.ro/ which I don't.

Is that what you call money problems?

[J35st3r]

They got at least $7million in pre-order payments if you can believe http://bfl.ptz.ro/ which I don't.

Is that what you call money problems?

Then they should have hired professionals to do their design work for them in the first place. Actually they could do it right now, and maybe get a good second-gen product out of the door before they lose too many customers. Or perhaps Josh is just looking to buy an island using the pre-order stash?

[k9quaint]

It explains why they would have money problems though, GF isn't really the "cheapest" place. They are among the best, though.

They got at least $7million in pre-order payments if you can believe http://bfl.ptz.ro/ which I don't.

Is that what you call money problems?

That all depends on how much of that they have spent. According to BFL, they have spent none of the pre-order money they have.
Expensive chips eat into margins, which just doubled for BFL when they had to go from 1 chip Jalapeno's to 2 chips.

[erk]

They got at least $7million in pre-order payments if you can believe http://bfl.ptz.ro/ which I don't.

Is that what you call money problems?

Then they should have hired professionals to do their design work for them in the first place. Actually they could do it right now, and maybe get a good second-gen product out of the door before they lose too many customers. Or perhaps Josh is just looking to buy an island using the pre-order stash?

Loose customers yeah lol, like the 7million came from thin air did it?  I would say they have at least 7mill worth of customers.
The ASIC is designed, they work, and they are shipping fast. If you are trying to make a point about some other ASIC they may make in the future fair enough, if not then I suspect you are just trolling. The latter is more likely from the tone of your BS.

[J35st3r]

They got at least $7million in pre-order payments if you can believe http://bfl.ptz.ro/ which I don't.

Is that what you call money problems?

Then they should have hired professionals to do their design work for them in the first place. Actually they could do it right now, and maybe get a good second-gen product out of the door before they lose too many customers. Or perhaps Josh is just looking to buy an island using the pre-order stash?

Loose customers yeah lol, like the 7million came from thin air did it?  I would say they have at least 7mill worth of customers.
The ASIC is designed, they work, and they are shipping fast. If you are trying to make a point about some other ASIC they may make in the future fair enough, if not then I suspect you are just trolling. The latter is more likely from the tone of your BS.

Yup, trolling. But then you're shilling. I'm new here and I thought it was about time I lost the Mr Nice Guy facade and tried out my darker instincts. But it is my honest opinion of BFL. Nice to rattle your cage  BTW its lose, not loose (though some of their customers might be getting a bit loose in the bowel department given all the delays and shenanigans).

[Schrankwand]

It explains why they would have money problems though, GF isn't really the "cheapest" place. They are among the best, though.

They got at least $7million in pre-order payments if you can believe http://bfl.ptz.ro/ which I don't.

Is that what you call money problems?

Yes. With 10 estimated employees, 4-5 million dollars in costs towards development through subcontractors, I call it money problems.


10 employees at around 25-30k if they are not so well paid, makes for another 360.000. Lets say they burnt the first batches of chips, blew a lot of cases.... blew lots of chips, have to pay health insurance. Another 300.000.

I have calculated an ASICs business, since I have no idea what to do there. I called up contractors and specialists for everything, since I thought if BFL makes millionaires, why not become one myself?

As it turned out, some of the more prestigious development companies asked between 1.5 and 2 million EURO in advance for initial development and estimated a timeframe of one to two years until serial production could start.


 

[k9quaint]

It explains why they would have money problems though, GF isn't really the "cheapest" place. They are among the best, though.

They got at least $7million in pre-order payments if you can believe http://bfl.ptz.ro/ which I don't.

Is that what you call money problems?

Yes. With 10 estimated employees, 4-5 million dollars in costs towards development through subcontractors, I call it money problems.


10 employees at around 25-30k if they are not so well paid, makes for another 360.000. Lets say they burnt the first batches of chips, blew a lot of cases.... blew lots of chips, have to pay health insurance. Another 300.000.

I have calculated an ASICs business, since I have no idea what to do there. I called up contractors and specialists for everything, since I thought if BFL makes millionaires, why not become one myself?

As it turned out, some of the more prestigious development companies asked between 1.5 and 2 million EURO in advance for initial development and estimated a timeframe of one to two years until serial production could start.
 

^^This^^
is precisely why I am concerned about BFL's pre-order money.
If they have used the pre-order money to fund their ASIC development (which has run over budget on both time and money), then BFL customers have to hope that they don't get hit by a wave of refunds and that the pre-order cash keeps rolling in.
If they have not used the pre-order money (unlikely imo) then how did they fund the $4-5 million in operating costs up to this point?

[erk]

Yes. With 10 estimated employees, 4-5 million dollars in costs towards development through subcontractors, I call it money problems.


10 employees at around 25-30k if they are not so well paid, makes for another 360.000. Lets say they burnt the first batches of chips, blew a lot of cases.... blew lots of chips, have to pay health insurance. Another 300.000.

I have calculated an ASICs business, since I have no idea what to do there. I called up contractors and specialists for everything, since I thought if BFL makes millionaires, why not become one myself?

As it turned out, some of the more prestigious development companies asked between 1.5 and 2 million EURO in advance for initial development and estimated a timeframe of one to two years until serial production could start.


 

Of course you wouldn't have a clue. This is utter BS. you don't know about investors or revenue from their FPGA sales or nothing about them, it's just a really obvious smear attempt.

Most of the pre-orders were placed after the the wafers and design were done, which kind of blows your revenue wild guesses.