[Archive] BFL trolling museum

[Clipse]

Great way to get more preorders BFL by publishing absolute bullshit stats/price on your next products.

Coffeecup warmer at 3-5GH(thats a huge gap, so what is it? or did you thumbsuck the shit out of this) for $149 ? whatever, I expect 12-14weeks from now alot of pissed of people who havnt received jackshit.

[JWU42]

Great way to get more preorders BFL by publishing absolute bullshit stats/price on your next products.

Coffeecup warmer at 3-5GH(thats a huge gap, so what is it? or did you thumbsuck the shit out of this) for $149 ? whatever, I expect 12-14weeks from now alot of pissed of people who havnt received jackshit.

It is 3.5 not 3-5...

[ice_chill]

Great way to get more preorders BFL by publishing absolute bullshit stats/price on your next products.

Coffeecup warmer at 3-5GH(thats a huge gap, so what is it? or did you thumbsuck the shit out of this) for $149 ? whatever, I expect 12-14weeks from now alot of pissed of people who havnt received jackshit.

+1

[Syke]

Butterfly Labs (BF Labs Inc.), a market leader in microprocessor design

Lookout AMD and Intel, there's a new player in the house!

[punin]

PRWeb will publish any crap that is sent there (been there, done that). I'd just take this as 100% FUD until their own website clearly states the numbers and their rep on the forum confirms.

What comes to hardware, I have reason to believe this will be Altera IV hardcopy. And no way it's going to be powered by USB only. Huge upfront costs too, so they need to sell a lot of those singles and mini rigs. I beginning to think BFL's business conduct is downright outrageous. They are using FUD to scare miners into buying their equipment to get a monopoly of BTC mining. This development can kill bitcoin unless the developers will make a clear stance on this (ie. algo change or not). I have an idea I'm working on to counter this situation, stay tuned.

[kano]

3)    BitForce SC Mini Rig: a case & rack mount server providing 1 TH/s, priced at $29,899

If BFL made just one of these, they could earn $40k per week as 1/15th of the current network, so why sell them?

Something's not right with these equations unless the profit margins are ridiculously large...

Well.. thats not completely correct... i think they do mine, but there is no point in incresing the difficulty for them!
So its much more profitable to sell hundred of those rigs, and when the difficulty rises again, people will buy even morefrom them..
so this way its much more profitable...

Directly related to my comment on the previous page.
With mining they can make a absolute max (at the current price) of some small X% of the whole network.

What is X?

Well the issues are:
1) You can't make 100% - if one source makes 100% well that's OBVIOUSLY the end of BTC - I'd even suggest that 50% would be (nothing to do with the 51% issue)
2) How much can you mine and sell without killing the market?

I'd suggest that X is not even 10%

So assuming that 10% is even possible, your talking $790K in 6 months on some rather risky venture at trying to take a good portion of the BTC network without killing it.
Or they could sell hundreds or even thousands of these devices and get the money up front even before giving anyone anything and have that guaranteed up front profit.
So ... which one makes business sense?

[ice_chill]

PRWeb will publish any crap that is sent there (been there, done that). I'd just take this as 100% FUD until their own website clearly states the numbers and their rep on the forum confirms.

What comes to hardware, I have reason to believe this will be Altera IV hardcopy. And no way it's going to be powered by USB only. Huge upfront costs too, so they need to sell a lot of those singles and mini rigs. I beginning to think BFL's business conduct is downright outrageous. They are using FUD to scare miners into buying their equipment to get a monopoly of BTC mining. This development can kill bitcoin unless the developers will make a clear stance on this (ie. algo change or not). I have an idea I'm working on to counter this situation, stay tuned.

I don't know altera or not, but their website says:
The BitForce SC processor is the result of our long term development of a multipurpose SHA256 hashing engine in full custom ASIC

1. How can it be multipurpose if it is ASIC ?
2. We heard about custom ASIC in their original BFL Single.

[Clipse]

PRWeb will publish any crap that is sent there (been there, done that). I'd just take this as 100% FUD until their own website clearly states the numbers and their rep on the forum confirms.

What comes to hardware, I have reason to believe this will be Altera IV hardcopy. And no way it's going to be powered by USB only. Huge upfront costs too, so they need to sell a lot of those singles and mini rigs. I beginning to think BFL's business conduct is downright outrageous. They are using FUD to scare miners into buying their equipment to get a monopoly of BTC mining. This development can kill bitcoin unless the developers will make a clear stance on this (ie. algo change or not). I have an idea I'm working on to counter this situation, stay tuned.

I don't know altera or not, but their website says:
The BitForce SC processor is the result of our long term development of a multipurpose SHA256 hashing engine in full custom ASIC

1. How can it be multipurpose if it is ASIC ?
2. We heard about custom ASIC in their original BFL Single.

What is longterm? BFL had their first products out sometime end of last year/beginning this year.

Is 6months a longterm established development company where they are from? Boggles my mind.

[dave3]

It would be nice if they would hold off taking orders until they've got a product that's ready to ship.  I think it's still likely that they'd be able to sell and ship everything they can manufacture right away, even without preorders.  And that would save people from tying up their money for months on a preorder.  And probably prevent a lot of complaining here. 

Anyway, if they've got venture capital funding now, they shouldn't need to use money from preorders to build the machines, right?

[P4man]

If BFL made just one of these, they could earn $40k per week as 1/15th of the current network, so why sell them?

Something's not right with these equations unless the profit margins are ridiculously large...

Gross margins are almost infinite; a while ago I estimated a single 130nm wafer would yield between 1 and 10 TH worth of ASIC chips. Such a wafer costs around $1000.

To answer your question; if they mine themselves, they only have to produce a few wafers to reach their maximum profitability, while not being shielded from the risks (BTC value, another ASIC surfacing etc).  It makes more sense sell many more ASICs at very high (gross) profit margins. Which btw, still wouldnt prevent them from mining themselves later, it would only take a few more wafer runs to out-hash anything they sold.

THis is somewhat akin to a dollar auction; miners can only lose.

[ailikun]

With mining they can make a absolute max (at the current price) of some small X% of the whole network.

What is X?

Well the issues are:
1) You can't make 100% - if one source makes 100% well that's OBVIOUSLY the end of BTC - I'd even suggest that 50% would be (nothing to do with the 51% issue)
2) How much can you mine and sell without killing the market?

I'd suggest that X is not even 10%

So assuming that 10% is even possible, your talking $790K in 6 months on some rather risky venture at trying to take a good portion of the BTC network without killing it.
Or they could sell hundreds or even thousands of these devices and get the money up front even before giving anyone anything and have that guaranteed up front profit.
So ... which one makes business sense?

Yep. i totally agree with you...
Well honostly, i think that this whole thing with new asic that they developed is bullshit, theres no way it could be so efficient...

But anyway even if it was true, people that want to buy this stuff are totally reckless, cause its the same as giving your money away for charity to bfl company...
cause you aint gonna get nothing from this...

[bulanula]

If BFL made just one of these, they could earn $40k per week as 1/15th of the current network, so why sell them?

Something's not right with these equations unless the profit margins are ridiculously large...

Gross margins are almost infinite; a while ago I estimated a single 130nm wafer would yield between 1 and 10 TH worth of ASIC chips. Such a wafer costs around $1000.

To answer your question; if they mine themselves, they only have to produce a few wafers to reach their maximum profitability, while not being shielded from the risks.  It makes more sense sell many more ASICs at very high (gross) profit margins. Which btw, still wouldnt prevent them from mining themselves later, it would only take a few more wafer runs to out-hash anything they sold.

THis is somewhat akin to to a dollar auction; miners can only lose.

Dude, stop trying to warn miners about buying the wonderful ASIC.

What is your horse in this race ? Are you selling FPGAs ?

The foolish will get burned in any case so there is no stopping the awesomeness that is BFL Labs.

I see gigavps giving them all his savings and others like him ( Inaba ).

I want to see them get burned HARD ! Please don't ruin this for me P4man. Where would I get my entertainment otherwise

[P4man]

IM pretty sure both GigaVPS and Inaba would be smart enough to sell perpetual mining bonds on GBLSE. If they play their cards right, they will still make a nice profit, its the suckers buying those bonds that will be left holding the bag.

[bulanula]

IM pretty sure both GigaVPS and Inaba would be smart enough to sell perpetual mining bonds on GBLSE. If they play their cards right, they will still make a nice profit, its the suckers buying those bonds that will be left holding the bag.

Well somebody will still get burned in that case.

Othewise BFL would not be selling but just mining themselves using ASIC and leave the rest of us using FPGAs and GPUs in the virtual dust capturing like 90% of the hashrate and none would be the wiser. Unknown is already at 46% last time I checked ... they could use many pools to make it look like nothing is foul etc.

[DutchBrat]

I'm wondering if pools will be able to keep up with a sudden spike in the network speed

I doubt many pools will be able to handle 50 TH/s or 100 TH/s

Will mean a lot of additional hardware expenses on their side, without any additional income for them.... Back to solomining ?

[-ck]

I'm wondering if pools will be able to keep up with a sudden spike in the network speed

I doubt many pools will be able to handle 50 TH/s or 100 TH/s

Will mean a lot of additional hardware expenses on their side, without any additional income for them.... Back to solomining ?

No, there's an easy fix already for pools - shares at higher difficulty. The difficulty 1 share acceptance just needs to be scaled to the network size. Of course your mining software needs to support this, and cgminer is already ready for it 

[Cablez]

I am also in agreement that the numbers don't really make a lot of sense but this is BFL after all.

Two points I would like to see addressed are 1) Now that they have supposed VC funding, will they finally change their business model and not require prepayment for unrealized products with ridiculous wait times?

2) more an observation but why release a 1TH device at first anyway?  They could just as easily build up network difficulty in a controlled manner by having just the 3.5 and 40Gh/s units for sale for more reasonable prices (i.e. $500 for the 3.5 and $12500 for the 40). Its not like they won't sell hundreds of units either way. I am not a fan of forced upgrades so at least this won't cause as much chaos as their current plan.

Finally, without timelines or any real credible information this whole this is like reading the tabloids at the market, one big train wreck!  I am done with it.

[ice_chill]

Average CPU gets about 12Mhash, average GPU gets about 250Mhash, that's a 20 times increase.

Now the ASIC also gives a 20 times increase compared to their FPGAs in mhash/dollar value.

So I don't think this is as bad as people make it out to be, we already been through similar situation.

Remember if you have 400Mhash GPU then you get 33 times increase over CPU, and we survived that no problem. The the ASIC is the final say, so future increases should be predictable.

[Steve]

It's difficult to predict a lot of the future dynamics around bitcoin mining.  Regardless, ASIC appears to be the future of mining.

I've come to the conclusion that people are approaching mining with the wrong perspective.  A lot of people worry about the 51% attack on Bitcoin and ask themselves how easy is it to attack Bitcoin?  I believe this is the wrong question to ask.  What people should be asking is how easy is it to secure Bitcoin?  And the answer to this is what makes me very excited about Bitcoin's future.  Any companies investing in Bitcoin will be asking this question.  And eventually, I think governments will as well.  Theses entities will have a lot at stake and will want assurance that the network can be secured against attack.  They will invest in mining operations with 90% of their hardware sitting idle and ready to come online at a moments notice to thwart any attack.  They will negotiate with each other (or enter into treaties) regarding the amount of hashing power they apply to securing Bitcoin.  As the block reward dwindles and as the difficultly rises, it will become increasingly difficult to profit directly from mining.  Instead, these mining agreements will be about providing adequate security to the Bitcoin network to support reasonably low cost transactions to facilitate commerce while simultaneously keeping the electricity costs minimal.  

I think a lot of people might buy the $149 ASIC purely to contribute to the security of Bitcoin, without much regard for its profitability.  Companies might do the same with the $30k box, but view it more as insurance against an attack on the network.  They'll mine while it's profitable and shut down when it's not, but will always be on the ready to flip it back on if there's ever any kind of threat to the network.  Still, I think even with the difficulty rising and with increasing adoption pushing the exchange rate higher, a 25 BTC reward per block for the next 4 years is going to ensure that it's pretty profitable for a lot of people for quite a while yet.

[JWU42]

@steve - good post...