[BFL] List of Lies

[jimmothy]

Just thought it might be fun to make a thread to create an organized list of all the blatant lies from BFL. I am talking about things that can be factually proven to be inaccurate/misleading/bullshit. I would like this to be a civilized discussion so please attempt to keep the trolling to a minimum.

Updated last: May 4 2014



List of Lies
1. (monarch) The fastest and most power-efficient Bitcoin miner yet
2. Butterfly Labs has shipped more ASIC products than all competitors combined
3. Our facility in Kansas has the largest production capacity of all Bitcoin hardware manufacturers.
4. The Competition at least $ 17.50 /GH
5. 65nm ASIC chip is now powering the majority of the bitcoin network
6. The bottom line is that BFL is the only 28nm chip manufacturer on its 2nd generation ASIC
7. All manufacturers in this space have experienced some degree of delay with their first generation ASIC
8. 28nm products won't begin shipping til year end
9. (monarch) plan to begin shipments in February, 2014
10. Orders are shipped in order date priority so any order placed now should be expected to be delivered in March.
11. We are pioneers of the industry - having manufactured the first commercial Bitcoin mining hardware.
12. November / December    Initial Shipping
13. still on track for December/January
14. As we enter the 28nm era, we're the only competitor with a proven ASIC design in the field.
15. Tape out August 2013
16. Gen1 65nm asics will be around 1w/gh
17. BTC mining, being a zero sum activity, makes it viable and profitable to get as many people to cancel their orders as possible, so your position improves
18. We are so confident in our power consumption that we are offering up 1000 BTC to charity if we miss our power consumption targets by more than 10%
19. Everyone should be aware of the fact that you are allowed to upgrade 1, 2, 3 or 4 Jalepeno's to 1 Single and keep your place in line
20. Individual orders that are less than 6 months old and that paid for the reduced price 600GH or 300GH Monarch will likely not be delayed past the expected delivery date
21. We have developed the most technologically advanced, most power efficient mining chip on the planet by a factor of two.
22. KnC's new 20nm chip isn't even close to our 28nm chip in terms of performance.
23. I have spoken with Theymos, the operator of Bitcointalk, about this and I have his full knowledge and permission to perform this experiment
24. plan to begin shipments in March 2014
25. However, this is our second generation, so we have much greater clarity on the process
26. meaning the deployment of the Monarch will be delayed about 5 weeks from now
27. it is possible we will begin shipping limited quantities by the end of the week of the 21st(april), it's more likely we will begin shipping the following week, assuming no blocking issues arise.
28. our mining chip is more than 2x efficient than any other chip out
29. The Monarch product line is essentially 3x - 5x more efficient at any comparable hashrate than the competition
 
Forecast:
- (monarch 600GH) 350w (0.6w/GH conservative estimate)
- Orders are shipped in order date priority so any order placed now should be expected to be delivered in June.
- Refunds will be issued in 30 to 45 days from the date of request of refund. Refunds will be in USD.
- We plan to begin shipping within the next 10 - 12 days, but that timeline is still in flux as we adjust the board components to best utilize the chips.

Bonus:
- Josh/Inaba of BFL has now resorted to paying shills to boost his own trust rating and defaming critics by leave false negative trust.
- https://bitcointalk.org/index.php?topic=549403.msg5979745#msg5979745



Sources:
1. See KNC, Avalon, Bitfury, Bitmain, Bitmine, Cointerra, Hashlast, Spondoolies, Dragon/lightning
2. Obviously not talking about hashrate because that would be absurd. I assume # of products which is still untrue considering the amount of asicminer/bitfury usb miners sold.
3. If largest production capacity means shipping out hardware in a timely manner, I would say they fall in dead last. (maybe hashfast is slower in shipping products)
4. Competition is actually as low as $3/GH. See bitmain
5. Obvious lie. Unless less than 5% of the network is now considered a majority.
6. See KNC.
7. See asicminer/bitmain
8. Already 3 months past years end and this statement is still not removed. *unless they mean this years end which would probably be accurate
9. Misleading. They have planned to ship every month since October of 2013.
10. Devices confirmed to not be shipping march.
11. See asicminer
12. Still have yet to ship anything.
13. See 12.
14. Was said literally months after KNC gave us a proven 28nm design.
15. Tape out September 2013
16. Actually around 3.5w/gh
17. Absolutely ridiculous to believe anyone would think this.
18. https://bitcointalk.org/index.php?topic=197050.0
19. Jalapenos intentionally sent out first.
20. March was expected delivery date for all orders before march.
21. Bfl 28nm chip is estimated to be 0.45 w/gh. Knc is estimated around 0.35 w/gh.
22. See 21
23. "He asked me whether he could demonstrate that the Trust system is broken. I said that he could, though I didn't know exactly what he was planning, and I don't like what he's doing" - theymos
24. March has officially passed. No products shipped.
25. Delays from gen1 were 8 months. Delays from gen2 are currently 6 months.
26. past 5 weeks since announcement.
27. past 2 weeks since announcement.
28. 2x better than bitfury would be below 0.2w/gh
29. see 28.

[Unacceptable]

Pretty much sums it up!!!

So....how you makin out there Ole Joshman Any news on the awesome power specs...& delivery dates

Didn't think so   

[mufa23]

C'mon guys, nearly three years later and everyone's still bitching about BFL?

What happen to your beef with Zhou? Pirate? Shtylman?

[Phinnaeus Gage]

C'mon guys, nearly three years later and everyone's still bitching about BFL?

What happen to your beef with Zhou? Pirate? Shtylman?

Add the middle vowel, and we'll be able to spell ZIPS. 

[PuertoLibre]

I swear that list is longer.

Where should I start and stop the list?
11. The only functioning 28nm implementation.

[bcp19]

There's one major thing missing from this list... a timeline as to when each comment was written.  The OP seriously seems to think that these are current claims, when several of them were made months ago.  Call it was it really is, a lack of updating the website with current information.

I challenge the OP with this:  Show me an ASIC company that has not told ANY lies.  Let's not focus on one, let's put them all on the table.  Otherwise, this is simply another hate thread intended to incite riot because the OP is just another troll.

[Bicknellski]

There is ample current and past evidence posted right here in this forum for BFL their shills and spokesmen lying. This is not the only thread for that. The news would be where they have told the truth. There seems to be a real disconnect here if anyone is denying that BFL has lied, continues to lie and will lie in the future.

https://www.scu.edu/ethics/publications/iie/v6n1/lying.html

Code:

Lies are morally wrong, then, for two reasons. First, lying corrupts 
the most important quality of my being human: my ability to make free, 
rational choices. Each lie I tell contradicts the part of me that gives me 
moral worth. Second, my lies rob others of their freedom to choose rationally. 
When my lie leads people to decide other than they would had they known 
the truth, I have harmed their human dignity and autonomy. Kant believed 
that to value ourselves and others as ends instead of means, we have perfect 
duties (i.e., no exceptions) to avoid damaging, interfering with, or misusing 
the ability to make free decisions; in other words - no lying.

[cedivad]

"List of Lies"
(so the grammar of the title of that page was fine )
Posted from Bitcointa.lk - #OolBzHSs2U6hoq0D

[jimmothy]

There's one major thing missing from this list... a timeline as to when each comment was written.  The OP seriously seems to think that these are current claims, when several of them were made months ago.  Call it was it really is, a lack of updating the website with current information.

A timeline is irrelevant because most of these statements were made after KNC shipped and began working on their next gen chip.

Please don't pretend the lack of updating the site with accurate information is anything but deceit. They seem to have no problem updating other parts of the site and are incredibly active on these forums and their own.

I challenge the OP with this:  Show me an ASIC company that has not told ANY lies.  Let's not focus on one, let's put them all on the table.  Otherwise, this is simply another hate thread intended to incite riot because the OP is just another troll.

Challenge accepted and completed: Bitmain.

We don't need to put them all on the table because this is a thread to focus on BFL. I am not going to argue that every asic manufacturer is perfect in fact most are the complete opposite.

I think just about everyone would agree BFL is competing for worst asic manufacturer. (tied with hashfast currently)

I challenge you to argue that any specific point I made was inaccurate. Otherwise it is you who is the troll.

[jimmothy]

I swear that list is longer.

Where should I start and stop the list?
11. The only functioning 28nm implementation.

Source? Don't see that on their site.

[Bicknellski]

Check the quotes... surely you can find MORE "inconsistencies".

http://www.coindesk.com/butterfly-labs-coo-responds-to-detractors-amid-company-struggles/

http://www.youtube.com/watch?v=bLsT2jFxSsc  Josh Zerlan Interview... PLENTY of Points to PONDER here.

More stuff for you to reference.

http://www.reddit.com/r/Bitcoin/comments/1kkffk/bfl_announces_new_28nm_pci_mining_blade_600ghs/

Code:

BFL claims their new 28nm card will have a power draw of 350 Watts, 
which puts it squarely in AMD 6990 territory.

http://www.bit-tech.net/hardware/2011/03/08/amd-radeon-hd-6990-review/9

In case you're new to Bitcoin mining, 6990's run at pretty much the thermal 
limit for a PCI card. They come with scary warnings about damaging your 
system's motherboard and power supply, and have a dedicated "all bets are 
off if you flip this" switch on the back of the board used to enable "full power" 
mode. The 6990 was a difficult project even for AMD, and has had significant 
problems with reliability, especially in bitcoin mining applications. The fact 
that BFL is planning to build hardware that runs at this temperature and 
power density is, frankly, terrifying.

A PCI card is a very different environment than an external box that plugs in via 
USB. They'll be up against all sorts of nasty power, signal timing, and compatibility 
issues. If a card fails hard, it could easily take-out the motherboard, and destroy 
other cards plugged in to it.

Its highly likely the finished product won't have anywhere near the hashing power 
BFL claims it will, and its almost guaranteed that BFL will run into epic delays 
during design and manufacturing, just like last time.

[PuertoLibre]

I swear that list is longer.

Where should I start and stop the list?
11. The only functioning 28nm implementation.

Source? Don't see that on their site.

Here you go:

More marketing lies:

http://www.butterflylabs.com/monarch/

As we enter the 28nm era, we're the only competitor with a proven ASIC design in the field.

KnC, HashFast, Cointerra, Bitmine all have 28nm designs in the field.

It all depends on how you interpret the quoted statement, but in either interpretation, it is still false.

[Bicknellski]

http://www.reddit.com/r/Bitcoin/comments/1kkffk/bfl_announces_new_28nm_pci_mining_blade_600ghs/

[–]ericools 12 points 7 months ago

I thought the FPGA seemed nice, I'm actually upset that they talked me out of ordering their FPGA rig by telling me the ASIC was about to ship and would make it unprofitable. Still waiting on my ASIC.

[PuertoLibre]

Here is a highlighted sample of various lies [in red], half-truths, debatable white lies, or just misinformation [in yellow] (on just one page):



Posted with the intentions of facilitating discussions and under FAIR USE.

Please click the thumbnail above to see the full image.

[tom_o]

What about the 1000BTC bet about power efficiency that was supposed to go to charity that Josh conveniently forgot about after the BTC value spiked last year.

[cedivad]

What about the 1000BTC bet about power efficiency that was supposed to go to charity that Josh conveniently forgot about after the BTC value spiked last year.

They didn't forgot about it, they created an ad-hoc charity for the goal of receiving the donation.
Posted from Bitcointa.lk - #fJtMAvQG9VRqL7gJ

[bcp19]

There's one major thing missing from this list... a timeline as to when each comment was written.  The OP seriously seems to think that these are current claims, when several of them were made months ago.  Call it was it really is, a lack of updating the website with current information.

A timeline is irrelevant because most of these statements were made after KNC shipped and began working on their next gen chip.

Please don't pretend the lack of updating the site with accurate information is anything but deceit. They seem to have no problem updating other parts of the site and are incredibly active on these forums and their own.

I challenge the OP with this:  Show me an ASIC company that has not told ANY lies.  Let's not focus on one, let's put them all on the table.  Otherwise, this is simply another hate thread intended to incite riot because the OP is just another troll.

Challenge accepted and completed: Bitmain.

We don't need to put them all on the table because this is a thread to focus on BFL. I am not going to argue that every asic manufacturer is perfect in fact most are the complete opposite.

I think just about everyone would agree BFL is competing for worst asic manufacturer. (tied with hashfast currently)

So, you're saying KnC shipped in mid-August?  Damn, I must have missed that in all the Avalon whining in the KnC thread.

Deceit is intentional, meaning it was originally posted with the intent to deceive.  Procrastination/laziness is a more accurate description and is hardly deceit.  It is a bad business model, but you and so many others seem to think everything needs to be perfect.

Yes, I can see it's 'purpose' is to focus on BFL, couldn't use one of the 50+ threads already out there, had to create another to add to the mess.

As to the trolling part, #1 This one is a tough call, as your examples didn't have existing product (except for Avalon) when it was made, but the problem comes about with companies underpowering chips to reach those power specs.  Until there is a fully working Monarch in hand running at the stated 600GH to see actualy power specs, you technically cannot claim this a lie.  They were shooting for .7W/GH at the time of conception and now it looks like they will beat that, but you nitpick other statements to call it a lie.

#2, I can technically give this to you as ASICMiner had thousands of those little USB stick toys they sold.  Considering they were pfft for hashing power and vastly overpriced (2BTC back in May and would have earned less than .7BTC as of today.)  Discounting AM though, the statement was true when made.

#3, On one of my visits to KC, I noticed their board noting a record 1 day shipping of over 1000 units and the weekly record was over 4000.  At that point in time, there was NO other ASIC manufacturer that had that much production capability, and unless you again count the little USB sticks, which wouldn't be put together in a workplace like BFL has but in a much bigger/vastly automated place, it's doubtful any of the current companies could match this output.

#4.  Your failing to account for time means that you think this is still accurate to this day, but Bitmain was not around posting about a $3/GH claim back in August.  What was true in August is no longer true today, if the statement had read: "The Competition at least $ 17.50 /GH, AND WILL NEVER GET LOWER", then you could indeed say it was a lie.

#5, back in August, 65nm technology WAS powering the majority of the bitcoin network.

#6, So, back in August KnC had a working 1st gen ASIC was was developing a 2nd Gen?

#7, AM, really?!?  Wow.  You love to throw Bitmain out there, but you can only point to Nov when they started posting on the forums.  Considering they were 'founded in 1Q 2013' and nothing was heard until November, you cannot in truth say they had no delay, you can only state that they worked out their delays BEFORE announcing anything.

#8, Technically a lie because they were overly optimistic about their foundry/testing timeline.

#9, You missed the word plan.  True, they suck at timelines, but that isn't lying.

#10, Until you have a fully working 600GH monarch in hand, you cannot call this a lie.  While you pulled the .6 number from their product page, their forum posting says:

Power & Performance better than expected
The initial power consumption numbers for Monarch are in and they're about 20% better than anticipated, meaning we expect the Monarch to consume 0.45W/GH at the chip level. This translates to about 300W per 600GH at the wall. Chip performance of all systems is as good as or better than predicted. To put this in perspective, this makes the Monarch chip nearly twice as power efficient as compared to our 28nm competition whose products operate between 0.9 and 1.0w per GH at the wall.

.45 at chip level = 270 watts which leaves 80 watts available to reach their 350W listing on the product page.  So much for your 'chips alone are .6w' statement.

I think Avalon wins as worst myself, but that's a judgement call due to the 800k chip fiasco.  

I challenge you to argue that any specific point I made was inaccurate. Otherwise it is you who is the troll.

 
Challenge accepted and completed.

[bcp19]

In addition, if you want to be technical (and I know you guys love being technical), Bitmain initially promised 0.68 J/GH on chip, which is meaningless.  Joules is an energy measurement and is usually referenced over time, and there's no time measuement included.  ANY manufacturer out there could make the same claim and it would not be a lie.  While it might take 1 manufacturer .5 sec to reach .68J/GH, it might take another manufacturer 3 seconds.  But that's only misleading, so you'd probably discount it being an untruth.

[jimmothy]

So, you're saying KnC shipped in mid-August?  Damn, I must have missed that in all the Avalon whining in the KnC thread.

KNC prototype working at the time of statement. And since they were comparing preorders you need to include hashfast/cointerra.

Deceit is intentional, meaning it was originally posted with the intent to deceive.  Procrastination/laziness is a more accurate description and is hardly deceit.  It is a bad business model, but you and so many others seem to think everything needs to be perfect.

It is clearly intentional when the monarch page is updated multiple times yet they intentionally chose not to update the things that would make them look bad (being realistic).

#2, I can technically give this to you as ASICMiner had thousands of those little USB stick toys they sold.  Considering they were pfft for hashing power and vastly overpriced (2BTC back in May and would have earned less than .7BTC as of today.)  Discounting AM though, the statement was true when made.

Asicminer also sold something like 75,000 8GH blades along with those little overpriced usbs.

#6, So, back in August KnC had a working 1st gen ASIC was was developing a 2nd Gen?

Correct.

Your failing to account for time means that you think this is still accurate to this day, but Bitmain was not around posting about a $3/GH claim back in August.  What was true in August is no longer true today, if the statement had read: "The Competition at least $ 17.50 /GH, AND WILL NEVER GET LOWER", then you could indeed say it was a lie.

Do you understand false advertising? That is like saying the ford model T is the fastest car in the world because at one point it may have been. You cannot use the argument "they never touched it since announcement" because they clearly edited the price of their monarch but forgot to update the price of the competition literally pixels apart.

#10, Until you have a fully working 600GH monarch in hand, you cannot call this a lie.  While you pulled the .6 number from their product page, their forum posting says:

Power & Performance better than expected
The initial power consumption numbers for Monarch are in and they're about 20% better than anticipated, meaning we expect the Monarch to consume 0.45W/GH at the chip level. This translates to about 300W per 600GH at the wall. Chip performance of all systems is as good as or better than predicted. To put this in perspective, this makes the Monarch chip nearly twice as power efficient as compared to our 28nm competition whose products operate between 0.9 and 1.0w per GH at the wall.

.45 at chip level = 270 watts which leaves 80 watts available to reach their 350W listing on the product page.  So much for your 'chips alone are .6w' statement.

You are right that I cannot guarantee it is a lie but I would be willing to bet on it. There is a 1% chance in my mind they meet 350W @ 600GH. I will add it to a separate section.

I think Avalon wins as worst myself, but that's a judgement call due to the 800k chip fiasco.  

Fun fact: Avalon gave full btc refunds for a 2? month delay. BFL refused all refunds for an 8+ month delay along with failure to meet advertised specs.

[jimmothy]

In addition, if you want to be technical (and I know you guys love being technical), Bitmain initially promised 0.68 J/GH on chip, which is meaningless.  Joules is an energy measurement and is usually referenced over time, and there's no time measuement included.  ANY manufacturer out there could make the same claim and it would not be a lie.  While it might take 1 manufacturer .5 sec to reach .68J/GH, it might take another manufacturer 3 seconds.  But that's only misleading, so you'd probably discount it being an untruth.

That is simply not how it works.

1 Joule = 1 Watt/second