[Archive] BFL trolling museum

[btharper]

Not cool, switching from your "business hat" to your "asshat" to respond to criticisms in the same thread.

It is, however, consistent with BFL's FUD flooding, just with multiple alt accounts.

BFL just doesn't _get_ it.

Given some of the crap that's come out in response, I'm surprised he hasn't switched more. Some of these people need to be torn a new asshole for spewing out that BFL is an evil, lying, horeshit company when they aren't even doing what they're being accused of.

The 1/3 plan was out a while ago, this isn't news, and now that it's old BFL shouldn't change it or they will piss people off (I expect this did help keep sales up through the first month at least).

All Josh came on to ask was if the community had any suggestions about how to help add some transparency to the random lottery system (which still only counts as 1/3 of the first shipment). To me however, it sounds like they can't release enough information to make it transparent as even the number of orders is guarded information.

Now on to my opinions:
As for demo units, definitely stick to jalapenos only if they'll be released ready to mine, and I'd suggest keeping it to a VERY small number (say 3-5). If however, you can get agreements signed to mine exclusively somewhere for show you could easily demonstrate your full line of products, even to professional mining operations. For example, send the unit off to whoever under the agreement they can only use it on the testnet until one month after everyone else gets theirs and require them to meet a proof of work quota for that whole month (please do clear it with the devs before you drop 1TH/s on the testnet though, this is just an example. You may need to use a different, possibly custom, chain to keep everything from interfering or getting onto the main chain. If you create your own chain you should release proof that it was not pre-mined to forge the timestamp.)

As for the random shipments. If you want to do things transparently you'll have to publish some proprietary information, I can't see any way around this; you do not need to release any personal information however. My suggestion is to release a list of SHA256(SHA256(Order number + email + order date)) so that anyone can verify the hash of their order but everything else is hidden, then just use a random hash in each consecutive block starting from a block you choose at least a week ahead of time (random means the block hash probably wouldn't work, I'd suggest the merkle root or the hash of the generate transaction, or both XOR'd together).

As for how many units to stock for the first shipment. I'd definitely suggest waiting until you can fill the first day's worth of orders with the 1/3 dedicated to order date priority, and probably until you can do the same for the first 4-7 day's worth of orders.

[1714109052]

1714109052

[1714109052]

1714109052

[1714109052]

1714109052

[imsaguy]

Now on to my opinions:
As for demo units, definitely stick to jalapenos only if they'll be released ready to mine, and I'd suggest keeping it to a VERY small number (say 3-5). If however, you can get agreements signed to mine exclusively somewhere for show you could easily demonstrate your full line of products, even to professional mining operations. For example, send the unit off to whoever under the agreement they can only use it on the testnet until one month after everyone else gets theirs and require them to meet a proof of work quota for that whole month (please do clear it with the devs before you drop 1TH/s on the testnet though, this is just an example. You may need to use a different, possibly custom, chain to keep everything from interfering or getting onto the main chain. If you create your own chain you should release proof that it was not pre-mined to forge the timestamp.)

I think the testnet idea is a good idea.  I'm against a select few getting units to 'prove' BFL and getting to whore the low difficult on mainnet, thereby getting a jump on the 1/3 people.  I'm certainly not against devs getting the hardware before hand to actually make sure the mining software works well with it.

[davidspitzer]

I think he's saying that you're wrong about them needing someone to do the free software development.

Correct.  We do have a little thing that's been in development called EasyMiner.  It's languished on the back burner for a bit whilest we try to get equipment out the door, but it's not been entirely idle and we are going to be focusing more resources on it as we get into gear.  This is one of the reasons I have taken over the customer service and shipping processes, to free up some people who's resources can be better utilized elsewhere.

Yes please - I hope you have mining turned on and ready to go on easyminer before the first SC hits the ground. One of my concerns is spending hours and hours trying to get some software to mine with the SC gear. A complete turn-key solution would work for me. Tune and Mine  - either solo or pool would be awesome.

There are so many expertly crafted software solutions, but many of them require very complex setup. I just want to mine, I don't want to compile binaries. That is why I use bitminter. Double click... Push start and voila Mining.

[SgtSpike]

I apologize for being vague, but some of the numbers I can't disclose, others we don't want to disclose because if we have to change them, it will cause more problems and yet more of those numbers we don't have a solid figure for yet.  It's not a matter of wanting to be vague, but a matter of having to be vague to give you any information at all.  It's either vagueness or nothing on some points unfortunately.  People don't like radio silence, as evidenced by this thread and I'm trying to provide as much information as I can up to this point.

Right now, we plan on shipping a mix of all three types of units and yes, 1/3 from each amount we plan on shipping.  Each class of mining hardware will be treated on it's own 1/3's balance sheet, not all lumped together. Does that make sense?

As far as what's "early" and what's "late" is going to depend on how many units we ship in the first batch.  Like I said, we do not want to ship out just a handful of orders in the first batch and trickle them out.  We want to ship a large quantity in the first batch.  

Let me ask you (and by extension everyone else): Do you want us to ship out a small quantity of units first and then a larger quantity later (thereby giving the first shipments a major advantage) or hold on to the units until we have enough (I am still going over what the "enough" is, specifically) to satisfy a large portion of the first set of orders at once?

It was my impression that the overwhelming majority wanted the playing field to be as level as possible, thus holding on to the orders until there's enough to ship a reasonable quantity would be preferred over shipping out a handful of each right away?  Is that thinking incorrect?

I would consider enough units to cover the first two days' orders to be sufficient for a first shipment (or maybe less), but without talking on what a "small quantity" and a "large quantity" means, it's hard to say what my opinion on that would be in direct relation to the way you asked the question.  I know you have to be vague with the information you can release, so I tried to give a specific answer.

A new question for you Josh - and BTW, I do very much appreciate you addressing questions and concerns promptly on the forum here - what will be the order for the 1/3 of converted orders?  If I requested that my 5/25 order be changed to SC singles on 6/25, but then someone else requests their order from 4/25 be changed to SC singles on 7/25, who would be in front of who?

[freeAgent]

It looks like orders are going strong at BFL...~100/day based on my recent orders 4 days apart.

[c_k]

Let me ask you (and by extension everyone else): Do you want us to ship out a small quantity of units first and then a larger quantity later (thereby giving the first shipments a major advantage) or hold on to the units until we have enough (I am still going over what the "enough" is, specifically) to satisfy a large portion of the first set of orders at once?

It was my impression that the overwhelming majority wanted the playing field to be as level as possible, thus holding on to the orders until there's enough to ship a reasonable quantity would be preferred over shipping out a handful of each right away?  Is that thinking incorrect?

Ship all per-orders at once and anything after the pre-order cutoff (I am assuming that might be October 1st) in a FIFO fashion.

Surely it isn't too hard to do considering ASICs tend to get manufactured in larger quantities, right?

[kano]

...

Ship all per-orders at once and anything after the pre-order cutoff (I am assuming that might be October 1st) in a FIFO fashion.

Surely it isn't too hard to do considering ASICs tend to get manufactured in larger quantities, right?

Don't forget the old lady with the soldering iron ... she has to put them together and QA them all ... that takes a while for one person
(also why the QA up until now has not been very good)

[DutchBrat]

Let me ask you (and by extension everyone else): Do you want us to ship out a small quantity of units first and then a larger quantity later (thereby giving the first shipments a major advantage) or hold on to the units until we have enough (I am still going over what the "enough" is, specifically) to satisfy a large portion of the first set of orders at once?

It was my impression that the overwhelming majority wanted the playing field to be as level as possible, thus holding on to the orders until there's enough to ship a reasonable quantity would be preferred over shipping out a handful of each right away?  Is that thinking incorrect?

Ship all per-orders at once and anything after the pre-order cutoff (I am assuming that might be October 1st) in a FIFO fashion.

Surely it isn't too hard to do considering ASICs tend to get manufactured in larger quantities, right?

I suggested that back in July : adjust the 1st production run to the number of pre orders so you can fill them all

Thats what pre orders are for

But that idea was not deemed possible back then, a lottery made more sense

Hence I did not order ;-)

[imsaguy]

Like I said, we do not want to ship out just a handful of orders in the first batch and trickle them out.  We want to ship a large quantity in the first batch.  

Let me ask you (and by extension everyone else): Do you want us to ship out a small quantity of units first and then a larger quantity later (thereby giving the first shipments a major advantage) or hold on to the units until we have enough (I am still going over what the "enough" is, specifically) to satisfy a large portion of the first set of orders at once?

Its funny, because you make a statement indicating you don't want to ship small orders, but then turn around and ask if the community wants you to ship small quantities.  That's what is upsetting.

[silverbox]

Let me ask you (and by extension everyone else): Do you want us to ship out a small quantity of units first and then a larger quantity later (thereby giving the first shipments a major advantage) or hold on to the units until we have enough (I am still going over what the "enough" is, specifically) to satisfy a large portion of the first set of orders at once?
It was my impression that the overwhelming majority wanted the playing field to be as level as possible, thus holding on to the orders until there's enough to ship a reasonable quantity would be preferred over shipping out a handful of each right away?  Is that thinking incorrect?

A larger quantity later, but this is only if BFL will not be "test" mining on this equipment the whole time, driving difficulty skyhigh, while none of the paying customers get any benefit

[nedbert9]

So, if testnet exists for testing purposes wouldn't it be safer/more fair to use it to test the ASIC units.

The rise in difficulty alone on testnet would be a validation of the design for all to observe.

Releasing test units prior to batch delivery is going to piss people off even with the best of intentions - particularly if it's unnecessary.

[streblo]

Like I said, we do not want to ship out just a handful of orders in the first batch and trickle them out.  We want to ship a large quantity in the first batch.  

Let me ask you (and by extension everyone else): Do you want us to ship out a small quantity of units first and then a larger quantity later (thereby giving the first shipments a major advantage) or hold on to the units until we have enough (I am still going over what the "enough" is, specifically) to satisfy a large portion of the first set of orders at once?

My vote is for BFL to deliver on what it has already stated:

1) Orders will be shipped according to the 1/3 FIFO, 1/3 Trade-ins, 1/3 random schedule
2) Orders will be shipped in October

From the sounds of things, it seems like you won't have a long enough time to build up a large shipment. Hence, my vote is for a handful of orders to be sent out (according to the "1/3rd schedule") and a trickle out, thereafter.

Any way you slice it, someone is going to have the advantage of receiving hardware first, whether they get shipped the units faster by living in the midwest USA; or placed their order the quickest; or are buddy-buddy with BFL. Let's stick to the plan you, BFL, have already set in place and give that advantage to those who placed orders first -- it's the fairest option!

[tbcoin]

Without asking how many orders exist until now, or which model ... 

Can I ask how much is the hashrate total sum of the orders? I think it's an interesting info.

[btharper]

Like I said, we do not want to ship out just a handful of orders in the first batch and trickle them out.  We want to ship a large quantity in the first batch.  

Let me ask you (and by extension everyone else): Do you want us to ship out a small quantity of units first and then a larger quantity later (thereby giving the first shipments a major advantage) or hold on to the units until we have enough (I am still going over what the "enough" is, specifically) to satisfy a large portion of the first set of orders at once?

Its funny, because you make a statement indicating you don't want to ship small orders, but then turn around and ask if the community wants you to ship small quantities.  That's what is upsetting.

He's trying to see what we want, Josh is willing to change policy to make the forums happy. You could just play along and say you would prefer that a larger first shipment be kept. Problem solved.

For me personally, a small starting shipment (that still includes the first day or two) makes sense. Everyone is worried about driving up the difficulty in a spike. Letting them trickle out lets the difficulty trickle up more slowly. Although to say anything we need about ten times as much information; When would the second shipment be? How large would the second shipment be? Will the production rate be ramped up as time goes on? Will you be able to handle order volume after clearing the pre-orders? and on and on and on.

Without asking how many orders exist until now, or which model ... 

Can I ask how much is the hashrate total sum of the orders? I think it's an interesting info.

This would be very interesting to me as well, even if it's just a ballpark answer (down to 0.1 TH maybe, that would take 28 Jalapenos).

[tbcoin]

Given that we currently only have 17TH, we have to know what to expect once are all orders out , 20, 50,100,1000TH more? really do not know how many big miners can afford a mini rig SC, but not it takes a lot for a big impact, in the case of single SC any miner would be willing to pay the price.

Are 3 multiplications and 1 sum that would give us much information.

[kano]

Given that we currently only have 17TH, we have to know what to expect once are all orders out , 20, 50,100,1000TH more? really do not know how many big miners can afford a mini rig SC, but not it takes a lot for a big impact, in the case of single SC any miner would be willing to pay the price.

Are 3 multiplications and 1 sum that would give us much information.

Actually ... no one NEEDS to know what to expect since people have sent BFL their money already.

The only reason anyone would need to know, would be to help them decide to give their money to BFL.
Too late for that

You will find out when it happens.

[Askit2]

I guess I should have asked here.
Why did the upgrade to 8 Jalapeno's disappear?
Will it come back?
Thanks for the help.

[tbcoin]

Given that we currently only have 17TH, we have to know what to expect once are all orders out , 20, 50,100,1000TH more? really do not know how many big miners can afford a mini rig SC, but not it takes a lot for a big impact, in the case of single SC any miner would be willing to pay the price.

Are 3 multiplications and 1 sum that would give us much information.

Actually ... no one NEEDS to know what to expect since people have sent BFL their money already.

The only reason anyone would need to know, would be to help them decide to give their money to BFL.
Too late for that

You will find out when it happens.

I've already sent my money, but would not be bad know if a week after receiving my order be equivalent to mine today at 100MH.
Maybe I want to increase my order, what is good for BFL.

[fcmatt]

by the time these show up the diff will be 3.2 million and btc price will be 2 bucks again..

[KIDC]

Ok. So I'm late to this discussion but seeing that Josh is already responding to community concerns I'd like to chime in with my ideas/concerns also, I think I have worthwhile comments. Full disclosure - I have 240 GH of BitForce SC Singles on order since day one. I have a tendency to make some very long posts on here so skip to bottom for TL;DR. Some of the following I'm sure has already been suggested by others - regardless I'm going to make my points here also.

My main concern regarding demo units shipped to high profile miners is the network difficulty impact. With such a large leap in efficiency and hashrate density, network difficulty will obviously explode - and those lucky few who have demo units are going to make very large profits while the rest of the network is comprised of relatively mundane GPUs and FPGAs. Even the almighty rigbox will collapse in the face of the Single SC. If BFL can assure that they will not be used for the user's profiteering, and that they will be limited to a test network (of which they receive no BTC), then I am much more comfortable with this idea. I also, however, consider it to be unnecessary anyway. It should be clear just from the order tracking thread that BFL has ample orders and will have ample orders up until the network is so saturated with ASIC that it is no longer cost effective for new purchases for mining. One other option in this regard would be to release videos on Youtube or other sites displaying the product working and clearly showing it's size, power draw, features, design, etc. This would be simple to perform and would put the community at ease I'm sure. I'm also sure that somebody of Inaba's caliber could easily accomplish this. Perhaps also/instead, a livestream could be setup, showing a webcam of the unit in question, and a desktop display of the unit's output. If BFL does decide to release demo units and allow their operators to profit from their extreme mining power, then I would like it to be a handful of Jalapenos - both to limit network impact and profitability. A Jalapeno SC demonstration alone would prove that the technology works as advertised.

Now, for reasons that should be obvious to everybody involved in "ASICgate", delivery time is of critical importance. I did myself order on the first day, while knowing that the intended policy was 1/3 trade-in, 1/3 order date, and 1/3 random access for first month purchases. I was not comfortable with the random order picks, however I paid in the hopes that BFL will ship my product in order. I made considerable effort and sacrifices in order to pay for these units immediately and in full, and I personally feel strongly that those who have done the same and paid up front, literally right from the beginning and have put their trust in BFL should be taken care of and have their units first. However I also appreciate that the stated policy was to have a third of the orders randomized. Now I don't know what BFL's intention with this is. My personal opinion is that a portion of the early order customers - perhaps just the first day if possible - should be protected from what you might call "order sniping". These people have paid up front in full and have had their money in BFLs hands for the longest and should have their orders either shipped out all together on the same day or should have them shipped out in order before any other orders or "random lottery winners" have their orders fulfilled. This seems to be the only fair way to handle things. I do feel that it would be bad for BFL to now renege on their original statement regarding the random orders. However I feel it would be very unfair for somebody such as me, who is one of BFLs best customers, to have their order put on hold by somebody who ordered weeks later and only took my position due to a random lottery system. So I feel the only way to do this would be to protect the first order customers (up to a week perhaps) from the random lottery. I think also that perhaps bundling the orders into 1 week blocks after that would be a good idea. I feel that shipping the entire first month as one huge block order would be logistically difficult and far too large an impact on the network at once. So maybe 1 to 7 days of the first orders get processed and shipped as one large batch together one the same day, with immunity from random order sniping. After that weekly ship outs for the remaining three weeks of orders, with random orders mixed in. Obviously those with low order numbers have a exponentially higher interest in their orders being shipped first rather than being grouped into a batch shipment. I myself will probably be at a disadvantage if orders are grouped into a batch, but I'm not too concerned about that as long as my product is shipped relative the when I ordered (first day). I will feel burned if somebody else gets their order in front of me due to random placement - this shouldn't happen for the first orders - nor for people such as myself who have now spent many thousands of dollars with BFL.

I also noticed that Josh stated that the Minirig SC and Single SC will have the same power consumption as the currently existing FPGA units. This is great news to me and is something I was concerned about. 40 GH/sec for 82 watts is absolutely fantastic. I am trusting in Josh that this is correct. I feel that I can trust him, since he is "one of us" so to speak. I appreciate your time Josh, and know that you are just trying to merge business requirements with community desires. I feel that you're going to make a solid liaison between us and BFL, one that was long overdue, and one that can be trusted due to your solid standing in the community.

On another quick note - if anybody is going to get any demo units, from a strategy and ethical standpoint I feel it needs to be the main developers of the mining software that these will be running primarily ckolivas, kano, and luke-jr of CGMiner/BFGMiner. Even with that in mind, the incredible performance of the Minirig SC is so ridiculous that no more than one demo unit should be allowed on the network at one time.

That's all I have for now. (ALL I have? lol)

TL;DR:

1. No units of larger network impact than Jalapenos should be sent as demo units. Demo units aren't even necessary due to so many orders already. If BFL feel they must provide a demo for whatever reason, then I suggest a livestream or youtube video demo rather than sending superprofitable product to people and blowing up the difficulty for everybody else.
2. First order purchasers should not be penalized for their faith in BFL. They should be rewarded by being isolated from the 1/3 random order lottery. This should be done for up to 7 days of the first orders. They could be sent in order or as one large batch together to be shipped all on the same day (probably hundreds of units). Other orders after that include the random placement possibility, and could be shipped FIFO or by bulk. First entire month of all pre-orders would avoid both previous problems, but isn't practical due to the huge number of devices shipped and an absolutely epic network impact - so I strongly recommend against doing that.
3. I am trusting Josh/Inaba because I feel he represents the community and BFL. While he may have "gone pro" or whatever, I feel he still can empathize with us and understand our needs better than any other BFL liaison here on bitcointalk.
4. ASIC hardware demos, if BFL feels they must be released, need to go to developers who then commit to shipping reliable, solid code - CKolivas, Kano, Luke-jr, etc. No more than one Minirig should be on the network at once, period.

- KIDC