1GH/s, 20w, $500 — Butterflylabs, is it a scam?

[sadpandatech]

Forgive to newb, please.

I keep getting distracted by Mt Gox Live.

  bad, newbie, bad! ;p

  All kidding aside, I know right!? Friggin down almost a $buck one 24h run. That's nuts. Poor lil tradin' bot was working overtime trying to keep up.


  BFL, we need more geek paraphernalia!

[1714129500]

1714129500

[1714129500]

1714129500

[1714129500]

1714129500

[1714129500]

1714129500

[fred0]

  BFL, we need more geek paraphernalia!

I'm optimistic about the single.  My understanding is the mining software that will be shipping with the unit is windows only -- but will run in a virtual box under linux. 

Crossing fingers.

[sadpandatech]

  BFL, we need more geek paraphernalia!

I'm optimistic about the single.  My understanding is the mining software that will be shipping with the unit is windows only -- but will run in a virtual box under linux. 

Crossing fingers.

  I've grown cynical in my years but I am watching it closely. I'm not holding my breath though and have drawn my plans as if it does not exist until such a time that it is proven otherwise.

 hmm, virtual box but not directly under wine? Wonder if that helps narrow down the language it's in.?

[fred0]

  BFL, we need more geek paraphernalia!

I'm optimistic about the single.  My understanding is the mining software that will be shipping with the unit is windows only -- but will run in a virtual box under linux. 

Crossing fingers.

  I've grown cynical in my years but I am watching it closely. I'm not holding my breath though and have drawn my plans as if it does not exist until such a time that it is proven otherwise.

 hmm, virtual box but not directly under wine? Wonder if that helps narrow down the language it's in.?

I have a canoscan 90 scanner that does not work under ubuntu (no sane driver), but works like a charm in virtual box.  Also, it is not that there isn't a linux version of the mining software, but the shipping version will be windows.

I certainly am skeptical, but scam is not my primary reaction.  While not out of the question, I just don't see a scammer saying, "I think I'll try to pull a fast one on a small group of smart people."  I think it would have to be a numbers game.  But I really don't know since I'm not a con-man, and really haven't given "pulling the perfect con-job" a good study.

[rjk]

I certainly am skeptical, but scam is not my primary reaction.  While not out of the question, I just don't see a scammer saying, "I think I'll try to pull a fast one on a small group of smart people."  I think it would have to be a numbers game.  But I really don't know since I'm not a con-man, and really haven't given "pulling the perfect con-job" a good study.

This is my view, and indeed the time and effort invested so far in this project is much greater than in any other scam I have seen. Thus my gut feel that this is going to be legit.

[rjk]

Can someone explain to me how the "Rig Box" with 32 cards gets a higher output than the "Single" x32?

Are the cards in the "Rig Box" different?  Could 32 "singles even fit in the "Rig Box"?


1.05 for the "Single" and 54.4 for the "Rig box"

Thanks.

I presume they won't be using the same card as the single, because the single does not look like it is designed to be plugged into a system. If the Rig Box is modular, it must use different boards (likely yet-to-be-developed)

[sadpandatech]

How could they know that it will have the output they claim if they have yet to developed it?

  same way any pcb/chip dev knows. They design it and emulate its function before they build it. presumably in this case, of course.

[legolouman]

I was watching this thread a while ago. Have we come to any definitive-ish conclusion? Have any of the people that ordered one reported back?

[fred0]

I was watching this thread a while ago. Have we come to any definitive-ish conclusion? Have any of the people that ordered one reported back?

No, nothing yet. We can not even see the Lab or meet the guy in KC.



I do hope this works as the biggest problem I have with bitcoin is that the use of electricity is massive and bad for the environment. I do hope this works out, however I doubt it will in the time stated.

Let's see what happens before the 25th.  I don't believe that two pre-orders from the same company will fly from a sales POV.  My hunch is that they will end the pre-orders for the singles before they start pre-orders for the rig box. Not to mean the singles will be shipping, only that the pre-orders will end and maybe some kind of public demo of the singles. I don't KNOW. My HUNCH!

[legolouman]

Okay guys, thanks. I had read a while ago, that the business also wasn't registered legitimately, anything on that?

[BitcoinPappi]

THis looks interesting, wouldnt this be a nice Christmas present if it is really true

[fred0]

Can someone explain to me how the "Rig Box" with 32 cards gets a higher output than the "Single" x32?

Are the cards in the "Rig Box" different?  Could 32 "singles even fit in the "Rig Box"?


1.05 for the "Single" and 54.4 for the "Rig box"

Thanks.

This observation is interesting to me.  The performance "guarantee" as stated on the pre-order form is listed as 1.05 GH/s +- 10% variance.  However, it is always better to under promise guaranteed performance levels.  The BFL rep who remarked about the previous betting hinted that the performance might be better than described.  54.4/32 = 1.7 GH/s.  I think that if they had advertised that number they wouldn't get any orders.  "Obvious scam is a scam" being the operative thought.

That said, a total piece of high risk speculation on my part.

[DeathAndTaxes]

+/- 10% is one thing +/- 70% is another.  

It seems illogical that they would design multiple different boards with different FPGA given how small the market is.   The "single" would be fine for use in a cluster, maybe having different connectors, and communication methods but roughly the same design.  Having a different board makes little sense.  I mean 1.05GH using 2 FPGA for one product and then 1.7GH per board for the second product.

If the rig consisted of similar dual FPGA designs one could expect it to be roughly 1.05 * 32 = 33.6GH/s.  Even +10% would be 36.96GH which makes the 54.4 number illogical.  Not the first thing w/ this company.

[fred0]

+/- 10% is one thing +/- 70% is another.  

It seems illogical that they would design multiple different boards with different FPGA given how small the market is.   The "single" would be fine for use in a cluster, maybe having different connectors, and communication methods but roughly the same design.  Having a different board makes little sense.  I mean 1.05GH using 2 FPGA for one product and then 1.7GH per board for the second product.

If the rig consisted of similar dual FPGA designs one could expect it to be roughly 1.05 * 32 = 33.6GH/s.  Even +10% would be 36.96GH which makes the 54.4 number illogical.  Not the first thing w/ this company.

It may be that the single is capable of performing 1.7 GH/s, when in a properly efficient case.  In that little box with the 80mm fan and tiny heat sink, and small power requirement, it may only be capable of 1.05.  The same unit in a properly vented, higher powered case it may be able to reach 1.7.  Like I said high risk speculation on my part.

In any case I am hoping we'll have a better grasp of the situation in 2 or 3 weeks.  Let's see.

[DeathAndTaxes]

+/- 10% is one thing +/- 70% is another. 

It seems illogical that they would design multiple different boards with different FPGA given how small the market is.   The "single" would be fine for use in a cluster, maybe having different connectors, and communication methods but roughly the same design.  Having a different board makes little sense.  I mean 1.05GH using 2 FPGA for one product and then 1.7GH per board for the second product.

If the rig consisted of similar dual FPGA designs one could expect it to be roughly 1.05 * 32 = 33.6GH/s.  Even +10% would be 36.96GH which makes the 54.4 number illogical.  Not the first thing w/ this company.

It may be that the single is capable of performing 1.7 GH/s, when in a properly efficient case.  In that little box with the 80mm fan and tiny heat sink, and small power requirement, it may only be capable of 1.05.  The same unit in a properly vented, higher powered case it may be able to reach 1.7.  Like I said high risk speculation on my part.

In any case I am hoping we'll have a better grasp of the situation in 2 or 3 weeks.  Let's see.

Given the high cost of FPGA that would be beyond stupid.  It would be like car company selling 8 cylinder cars and then making a compact model which has 4 cylinders by just turning off 4 of the cylinders (but keeping all the cost of 8 cylinder engine) and putting lackluster cooling in it. 

FPGA is very very very expensive.  Every clock cycle is worth something.  You don't build a product which could run at 1.7 GH and say "well 1 GH is good enough because we fucked up the cooling and power requirements."  It just doesn't make any sense. 

[RandyFolds]

+/- 10% is one thing +/- 70% is another.  

It seems illogical that they would design multiple different boards with different FPGA given how small the market is.   The "single" would be fine for use in a cluster, maybe having different connectors, and communication methods but roughly the same design.  Having a different board makes little sense.  I mean 1.05GH using 2 FPGA for one product and then 1.7GH per board for the second product.

If the rig consisted of similar dual FPGA designs one could expect it to be roughly 1.05 * 32 = 33.6GH/s.  Even +10% would be 36.96GH which makes the 54.4 number illogical.  Not the first thing w/ this company.

It may be that the single is capable of performing 1.7 GH/s, when in a properly efficient case.  In that little box with the 80mm fan and tiny heat sink, and small power requirement, it may only be capable of 1.05.  The same unit in a properly vented, higher powered case it may be able to reach 1.7.  Like I said high risk speculation on my part.

In any case I am hoping we'll have a better grasp of the situation in 2 or 3 weeks.  Let's see.

You are far too optimistic. Better cooling will not cause a 70% leap in efficiency.

[rjk]

Given the high cost of FPGA that would be beyond stupid.  It would be like car company selling 8 cylinder cars and then making a compact model which has 4 cylinders by just turning off 4 of the cylinders (but keeping all the cost of 8 cylinder engine) and putting lackluster cooling in it. 

FPGA is very very very expensive.  Every clock cycle is worth something.  You don't build a product which could run at 1.7 GH and say "well 1 GH is good enough because we fucked up the cooling and power requirements."  It just doesn't make any sense. 

However, if they were ASICs, they could be designed with fuses and then have no redesign cost when they want to increase the output. The idea was Gmaxwell's, not mine

[fred0]

Well the thread was started 18 Oct. BFL has a 4 to 8 week delivery time for pre-order units.  So we are approaching the delivery time zone.  Some time (unless we hear otherwise) on or around (hopefully before) 18 Dec, someone will receive a single.

Let's see what happens.

[rph]

4-8 weeks probably means 8+ weeks. If they meant 4 weeks, they would have said 2-4 weeks.
Marketing 101!

-rph

[bulanula]

I am sure BFL guy will explain why it has been delayed either on the website or in this thread

Waiting for the scam to arrive ?