Topic: 1GH/s, 20w, $500 — Butterflylabs, is it a scam? - page 16. (Read 123107 times)

Question to anyone that preordered: have you received any news or updates about the delivery schedule? Is there anyone who is past his chargeback deadline or who already did a chargeback and if so, how did BFL handle it?

I wake up expecting some proof and really thinking "this is 90% real right now" and instead wake up to find more delays ( initially it was supposed to be Friday then Sunday now next week LOL ). BFL fix your stuff please I wanna get rid of my noisy and hot GPUs ASAP.

Well, on the forum there has been much speculation as to the hardware used, optimization techniques, but they KNOW what devices are on the board.  My guess is they estimated what the device should be capable of producing.

They know the technical specs, we don't.  For us it is a black box that generates bitcoins.

We are paying attention to the demo because we want to see how well the widget can perform.  Until it measured, the number we are getting from BFL is/should be an educated extrapolation based on their knowledge of the hardware, and hopefully, they are spot on.  

However, it could also be that they are giving us measured result, but no one in the forum believes the numbers.

Until we see hashing measured by a trusted source, there is doubt.

Preorders were to fund the development, they were up front with this, I see no cause for complaint about preorder.  Preorders received $200 discount from the regular price.

Frankly, I can't imagine more problems NOT coming up.  Let's go one step at a time and see results for the demo.

If BFL is a scam, I'm sure they are already packing.  If a legit organization, they are working diligently to determine where the problem is.

We will see soon.

I have 3 day to cancel 5x 5970, maybe hurry up !...

so where did BFL get the performance figure from?

Technical specs? how can they advertise such performance and even have preorder?

 

I am sure bitcoin would REALLY take off it the Saudis required all petroleum purchases be conducted in bitcoin.

And then the Gov would take action.

Sure but there are cheaper ways to screw up bitcoin .. declare using it and mining it illegal and punishable by death.  Free attention from the press— say its a childpornterrorist tool—, a couple helicopters with flir, a few raids, some additional proposed legislation in a few more friendly countries, and a public execution or two and the bitcoin security is gone _real quick_ ... and not even a million bucks spent.  No fussy costly and highly public technical attacks that are going to result in civil litigation and might by thwarted counter measures that risk making the powers that be look incompetent.... just the core tools of authority: ink on paper and jackbooted thugs, tried and proven.  ("Goverment builds $10m building bitcoin busting ASIC farm, bitcoiner's change one value in their POW algorithm and do a network wide updated and the $10m farm is worthless", a counter measure we won't have forever but it's one we'll have up until the good guys are using asic/sasic in mass)

The long term security of bitcoin depends on it being to unimportant, uninteresting, and unconventional for that kind of adversary to bother with it.. until such a time as bitcoin is too important for them to afford and beloved to afford messing with it. If you look at how slowly most large institutions have adapted to technology this outcome doesn't seem unlikely.

Bitcoin isn't all negatives for the status-quo authorities— it should increase economic efficiency, and it removes the unfortunate attraction of printing money which all politicians hate (or at least hate when its for someone elses program) but almost universally can't resist when its for their own programs—... and if the public at large wants it, suppressing it would look bad indeed and wouldn't work out well in places that claim to have freedom.

Yes, which is why I am anxious to see real numbers.

Yes, the world is full of scammers.  They don't care a whit about anybody but themselves(Bernie Madoff).  I think we make a poor club to take advantage.  Fairly highly educated, skeptical, and fairly communicative. How long could a scam last?

Would pre-orders for the rig box take off if it is determined that the singles are fraudulent? Scam over, man.

All in all, software defects DO happen, and I am not happy it happened tonight, but I am sure the BFL people are even less happy.

So while we still have our life savings, let's wait a couple of days and see how it goes.

Sure, the Gov can always spend money at will. The US gov simply has to print more money. I am sure $10M is nothing since the gov has already printed trillions

Im sure Inaba could have put windows on there and im sure OS would have been discussed as they were planning a demo. Yes it could be a bug but I still find it awfully suspicious and convenient. How could they know precisely the MH/s and power consumption if it has never been running? Sometimes things dont work as well as planned you know

Agreed, that's why I suspect it is a software issue.  What good is the hardware without software to recognize it?

I was jazzed up to see performance numbers tonight, but I can wait to resolve the inevitable glitch.

Disappointed, but I understand.

This is just too friggin': (choose all that apply) unbelievable/ridiculous/scammy/unprofessional/amateur/scammy/scammy/scammy/disappointing.

They don't have one single prototype working to show Inaba? Goodness.

e: amendment to list to reflect sentiments below.

New hardware, new software,  Primary development was done on Windows. Inaba's laptop probably was running ubuntu(if the same as photo). I'm willing to give BFL the benefit of the doubt, we should have a clearer idea in a few days.

It just seems too convenient that they are having problems getting it to work and keep on pushing back a working demo.

True.

Legitimate software NEVER ships with bugs.  There can be NO DOUBT the Butterfly Single IS a SCAM!

DO NOT spend your money on the scammer widgits.

Surprise, surprise. Another delay in the demo. I will be surprised if they actually are able to show inaba it working. In the meantime I wonder if they got any 26k orders for the big box.

Exactly.  Imagine prior to the first GPU miner.  Network hashing power was <100GH/s.  An attacker who built a GPU farm could easily dominate the CPU based network.  GPU farming made the network more efficient (both in hashes/$ and hashes/W).  While an attacker can use GPU so can defenders and that makes it more of an even fight.

Eventually FGPA/sASICS will achieve >2MH/$ even if BFL production is just scammer garbage.  In time defenders will use them and the Bitcoin network will be even more efficient making an attack against it more difficult.

x86 CPU miner -> OpenCL CPU Miner ->  GPU Miner -> FPGA Miner -> sASIC Miner -> Cell based ASIC miner ->  Full Custom ASIC miner.

IF Bitcoin continues to grow the profit motive will lead to more and more efficient hashing hardware.  The motive is profits (both for research and for deployment) but the advantage is the network grows more secure.  Not just in nominal terms but in efficiency terms.

I'm sorry that this was surprising to you.

It's long been known to many other people that if someone wanted to sink a few million in NRE for a fully custom ASIC on a modern process (e.g. 45nm) they could churn out chips that do e.g. 10GH for a marginal cost of tens of dollars per chip (e.g. similar costs to a GPU chip, but massively more hash power. Actual performance figures are fuzzy only because the design of a custom asic would be thermally limited, so it's a question of how cheap you can get good packaging and how well can keep it cool).

This is why the introduction of improved efficiency miners (like the BFL, if it's real) to the public is a good and important thing. It's also a reason why GPU miners are good— while a custom asic would have a big advantage over GPUs their advantage over CPUs is even greater.

  Kinda expected the delay without any need for wacky conspiracies due to the nature of the product in its current form. ;p  Thanks for the update.

Indeed you are correct.  There are some technical issues with the timing between the board and the software that they are working out, apparently.  I understand this will probably bring all the conspiracy theorists to the fore, so have at it.  For everyone else, I'm told that they will be ready in a couple days or so for a live demo. 

My week is pretty open as far as being able to do any testing goes, so assuming they work out the problems I will be available any time this week.