1GH/s, 20w, $700 (was $500) — Butterflylabs, is it for real? (Part 2)

[btc_artist]

The guarantee was that should we not deliver at those specifications, the order could be canceled or the final resulting performance figures could be accepted.  We have kept to our guarantee.

meh, that might be legally true, but that's not how I (and the average person) read the product pre-order page.

[1713563656]

1713563656

[1713563656]

1713563656

[1713563656]

1713563656

[1713563656]

1713563656

[1713563656]

1713563656

[1713563656]

1713563656

[CubedRoot]

It's also interesting to note that the only negative comments are a virtual role call of the people who've staked their reputation on this product not existing at all.  I don't want to pick a fight, but at some point you may want to consider the long term implications of our company not being a scam.

The product doesn't exist.  The product debated was 1.05GH @ 19.8W for $500.  That product doesn't exist. 

You have another product which seems very much real and at much more realistic performance metrics.  I don't want to pick a fight but the only one who made a false claim was you.  You "guaranteed" delivery of a product in 4-6 weeks (long since passed) that would achieve 1.05GH/s @ 19.8W for $500.  A guarantee your very young company has already failed to live up to and likely never will.

DeathAndTaxes does make a very strong point here.   I'm sure BFL is indeed going to produce a good quality unit, and deliver said unit to its customers in a respectful time frame.  But, I do not think that product will be the product that they were first marketing.  The original claims of 1.05 GH @ 19.8 watts for $500 (or $599) just isnt what is going to hit the market.
I think they were a victim of their own optimism, and jumped the gun a bit on making claims that couldn't be substantiated.
I do think they will get the power system corrected, and that will help with consumption.  power run-aways are exponential by nature, and it appears they just had the circuit and chip underpowered in the demo.  This by nature will cause alot of heat, and power draw, thus the exponential run-away problem.
I will say this, the BFL guys have been very open with the community and they seem to want to deliver a solid stable product.  The first run of anything new will not always be the best.  Sometimes you just have to get it in the end-users hands in a large scale to find the gremlins.  Its one of the negatives about being an early adopter.

[Fiyasko]

I was fully on the "ITS A SCAM!" boat.
But, Now that Inaba has tested a unit (Inaba being one whom is not under BFL wing) and shown Real values,
Im impressed...

I didnt think one could get such high hashrates with such low power draw, So i screamed scam (20w's 1ghs)
But seeing this Overheating, Needs to be upgraded, OverANDunder powered unit...
I cant say that BFL is a scam, Perhaps, On the worst of ends, They will ship out those first 100 without fixing them.

But BFL has acknowledged that the power brick is insufficient... Hmmm.


What i want to know, Is how long this this is gonna last while running at full load, According to Inaba, The thing could've already "Sparked out" by now if he had it running 24/7 since he got it

Didnt he mention something about the rails themselves that carry the electricity are insuffecient, And therefore would Shortout overtime of use?

[bulanula]

I thought the only thing that was not guaranteed was the power and they said that the 1050 Mhash/s performance figure would be kept for the final product ?

[BFL]

I thought the only thing that was not guaranteed was the power and they said that the 1050 Mhash/s performance figure would be kept for the final product ?

To clarify, the guarantee was that if we didn't ship with those performance figures 1.05 gh/s + / - 10% @ 19.8 (power included), we would allow customers to cancel their order or accept delivery with the final performance figures.  At the moment, we can only guarantee the stable performance we have demonstrated at 832 mh/s but I think it's reasonable to assume when the production units are delivered with increased power capacity that they'll run faster.  I really want to avoid a situation of over promising after having got the power draw wrong, so we're only promising 832 mh/s at the moment and we're lowering our price accordingly until we have proven increased performance with the final delivery units.  In any case, ALL delivered units are being built with the larger power subsystem to compensate for the increased draw.

[Fiyasko]

Asking gain, warranty information?

Yes please

[BFL]

But BFL has acknowledged that the power brick is insufficient...

What i want to know, Is how long this this is gonna last while running at full load, According to Inaba, The thing could've already "Sparked out" by now if he had it running 24/7 since he got it

I'd like to make it clear that the power brick and the power regulator on the board are being beefed up to properly supply the power needed by the processors.  In fact, we will now have two independent power regulators to ensure a very stable and long life span.

The power adapter will also be different than the one used during the demonstration with Inaba.

Please be aware that we are not students building a project board.  This has been our collective profession for more than 10 years.  Even our competitors in these threads have commented on the quality of the design & components used.  We traditionally over design.  The company Butterfly Labs is a new brand created for the BitForce platform, but our individual manufacture experience has been well established.  I assure you that reliability of the delivered product will be of the highest levels.

[2112]

BTW where did the very exact performance numbers come from.

My guess is that they just used the estimates from the synthesis tool. It would appear that they didn't even simulate the chip before racing to design the PCB. Pre-simulation power estimates are known to be wildly inaccurate. I would tend to believe their claim that they've been 10 years in this business, which means that there are a lot of opportunities in the FPGA design business for vendors capable of better-organized work-flow.

The time it took for them to develop the software drivers is another piece of circumstantial evidence that they didn't create the simulation testbench.

[Turbor]

I'll stick with my German Überboard 

[RandyFolds]

It's also interesting to note that the only negative comments are a virtual role call of the people who've staked their reputation on this product not existing at all.  I don't want to pick a fight, but at some point you may want to consider the long term implications of our company not being a scam.

The product doesn't exist.  The product debated was 1.05GH @ 19.8W for $500.  That product doesn't exist. 

You have another product which seems very much real and at much more realistic performance metrics.  I don't want to pick a fight but the only one who made a false claim was you.  You "guaranteed" delivery of a product in 4-6 weeks (long since passed) that would achieve 1.05GH/s @ 19.8W for $500.  A guarantee your very young company has already failed to live up to and likely never will.

This times a million. The. Product. Advertised. Does. Not. Exist.

I thought the only thing that was not guaranteed was the power and they said that the 1050 Mhash/s performance figure would be kept for the final product ?

To clarify, the guarantee was that if we didn't ship with those performance figures 1.05 gh/s + / - 10% @ 19.8 (power included), we would allow customers to cancel their order or accept delivery with the final performance figures.  At the moment, we can only guarantee the stable performance we have demonstrated at 832 mh/s but I think it's reasonable to assume when the production units are delivered with increased power capacity that they'll run faster.  I really want to avoid a situation of over promising after having got the power draw wrong, so we're only promising 832 mh/s at the moment and we're lowering our price accordingly until we have proven increased performance with the final delivery units.  In any case, ALL delivered units are being built with the larger power subsystem to compensate for the increased draw.

Well, no shit. If this wasn't the case, you'd be committing theft and fraud.

What a disappointment...

[RoadStress]

BFL add a warranty and some legal info of your company and i might consider buying one (ofc after some units get shipped)!

[fizzisist]

It was never possible.

and by there "excellent" marketing communications, nearly all other unfirm FPGA developers scared and canceled their project.
good job.

From my point of view, this is a big part of the story here. This is what was so inethical/illegal. They almost scared us out of producing the X6500 in bulk. I'm glad we went with our guts and pulled the trigger, though.

"Bait-and-switch" is what this is called, and it's illegal. Yes, people can get refunds, but many probably won't. That's the point of bait-and-switch. Furthermore, many probably would have ordered FPGA mining boards from the competition weeks ago if this product hadn't been dangled in front of them like a carrot.

I'm not saying this was all intentional by BFL, but it's what happened whether they meant to or not.

[BFL]

Furthermore, many probably would have ordered FPGA mining boards from the competition weeks ago if this product hadn't been dangled in front of them like a carrot.

I'm not saying this was all intentional by BFL, but it's what happened whether they meant to or not.

Your motivations have been pretty clear all along.  You're having to deal with competition.  It's important to point out that you have not been wronged and I have no negative feelings towards you.  Our performance targets were posted in good faith and in the end we aren't all that far from those figures now aside from power which isn't the principle figure of importance for most people since it's still a small fraction of GPU power consumption.  Bait and switch would be accurate only if there was mal intent.  You can accuse us of being power optimistic, but that's as far as it goes.  

Previously you were shouting scam.  Now it's bait and switch.  You're just mad because someone is offering a more competitive product.  That's all.

Listen, I'm sorry if your plans were disrupted, but please don't resort to slander.

[RandyFolds]

Our performance targets were posted in good faith and in the end we aren't all that far from those figures now aside from power which isn't the principle figure of importance for most people since it's still a small fraction of GPU power consumption.

400% isn't all that far? You should work for the government. fizzisist has not been here screaming 'BFL is a scam', he has been contributing (as an expert in FPGA technology) to the thread that you didn't have time for until a few days ago when you failed to deliver on a number of promises. Unfortunately, calling a spade a spade is not slander.

...aside from power which isn't the principle figure of importance for most people.

Interesting...having spent more than ten minutes on these forums, I would beg to differ.

[Turbor]

I think you have a good product. But you advertised it 3 months too early (with specs you guessed) imo.

[bulanula]

Furthermore, many probably would have ordered FPGA mining boards from the competition weeks ago if this product hadn't been dangled in front of them like a carrot.

I'm not saying this was all intentional by BFL, but it's what happened whether they meant to or not.

Your motivations have been pretty clear all along.  You're having to deal with competition.  It's important to point out that you have not been wronged and I have no negative feelings towards you.  Our performance targets were posted in good faith and in the end we aren't all that far from those figures now aside from power which isn't the principle figure of importance for most people since it's still a small fraction of GPU power consumption.  Bait and switch would be accurate only if there was mal intent.  You can accuse us of being power optimistic, but that's as far as it goes.  

Previously you were shouting scam.  Now it's bait and switch.  You're just mad because someone is offering a more competitive product.  That's all.

Listen, I'm sorry if your plans were disrupted, but please don't resort to slander.

Haters gonna hate on you BFL. Now all the FPGA designers are screwed by your product, they are fighting back with more BS and bad mouthing your company after they saw they are screwed and that this may be indeed, real after all. If I were you ( FPGA designers ), I would move to Russia because you seem to hate competition. Here in Russia they love monopolies ! Welcome. Free market FTW. I guess after about some time the people who buy BFL products now are screwed too as 28nm products come out in 2012 ?  

Anyway,

warranty ?

what is the tech behind your product ?

tell us more about this "10 year experience" thingy ?

Thanks !

[Fiyasko]

./facepalm

He did not say he had 10yrs of BitForce related experiance.
He said he (he being who he works for/as/with) has about ten years of experiance with hardware building boards

[fizzisist]

Haters gonna hate on you BFL. Now all the FPGA designers are screwed by your product, they are fighting back with more BS and bad mouthing your company after they saw they are screwed and that this may be indeed, real after all. If I were you ( FPGA designers ), I would move to Russia because you seem to hate competition. Here in Russia they love monopolies ! Welcome. Free market FTW. I guess after about some time the people who buy BFL products now are screwed too as 28nm products come out in 2012 ?  

Actually, competition is one thing, unfair competition is another. There's a reason bait-and-switch is illegal.

Also, as RandyFolds nicely pointed out, I never "shouted scam" about BFL. If you review my posts made about this, the most direct address I ever made to the "scam" question was in this post: https://bitcointalk.org/index.php?topic=48863.msg633227#msg633227

Where I said the following:

I'm certainly not claiming that I know BFL is a scam, but I do find it very suspicious. To summarize, here are the things that strike me as at least a little bit odd about them:

And then went on to summarize some of the outstanding questions I and others had about BFL. Most of those questions are still outstanding, such as the alleged "other applications" claim. Judging by the test results, it does appear to be FPGA based, but they claimed before that it couldn't do anything other than SHA256. Still waiting for clarification on that. If it truly can only do two-step SHA256, claiming that it has other applications is seriously misleading, and is another way that they are attempting to compete unfairly.

For the most part, I intentionally tried to avoid the discussion because of my obvious conflict-of-interest.

[bulanula]

Care to give us a breakdown of your product vs BFL Labs ( maybe in your own thread ) so the average Joe can see what the differences are etc. ? How scalable is your solution ( in terms of units / area of space ) etc. ?

Thanks !

[Fiyasko]

Care to give us a breakdown of your product vs BFL Labs ( maybe in your own thread ) so the average Joe can see what the differences are etc. ? How scalable is your solution ( in terms of units / area of space ) etc. ?

Thanks !

Shouldnt something like that be on the BFL webpage?