BFL also made fools of themselves by trolling them.
Frankly, the only trolling I see is you stickying that post and putting that title on it. As for recommending purchasing fpgas at this point, particularly ones that cant be traded up for asics, and to someone who pays virtually nothing for his electricity.. well Im not (yet) a shareholder, but I would hope a CEO knows when to ignore advice from a shareholder. You're welcome to outvote me in the motion. |
|
|
|
OK shareholders......I think P4man is correct, and we need to decide sooner rather than later how we want to proceed.
I have created a motion on GLBSE to issue and additional 6,000 shares, the proceeds of which will be used to place an order for a BFL SC mini-rig.
I will take feedback here, and you can vote on GLBSE.
Currently BTCMC represents .3% of the mining network (1/300th). If we purchase a SC mini-rig, the network capacity could leap to 300 TH/s (25x) before our share of the block reward would go down.
I own 783 shares, I've voted no. What would be your preferred route? See four posts up. Enterpoint now, 28nm later, and BFL has probably lied about their SC minirig and its closer to 500gh for $60k so 22nm FPGAs should start rolling out about the time they come out with their SC gear. I have not seen the exact specs on Enterpoint.....what are they offering? A quad Spartan 6 board that has enough amps to run Tricone Mining's firmware. Seems to be the cheapest Spartan 6 solution per mh. They're a British company that specializes in custom computation gear, but the Cairnsmore is the first Bitcoin-related product they've made. https://bitcointalk.org/index.php?topic=78239.0BFL also made fools of themselves by trolling them. https://bitcointalk.org/index.php?topic=88363.0 |
|
|
|
OK shareholders......I think P4man is correct, and we need to decide sooner rather than later how we want to proceed.
I have created a motion on GLBSE to issue and additional 6,000 shares, the proceeds of which will be used to place an order for a BFL SC mini-rig.
I will take feedback here, and you can vote on GLBSE.
Currently BTCMC represents .3% of the mining network (1/300th). If we purchase a SC mini-rig, the network capacity could leap to 300 TH/s (25x) before our share of the block reward would go down.
I own 783 shares, I've voted no. What would be your preferred route? See four posts up. Enterpoint now, 28nm later, and BFL has probably lied about their SC minirig and its closer to 500gh for $60k so 22nm FPGAs should start rolling out about the time they come out with their SC gear. |
|
|
|
OK shareholders......I think P4man is correct, and we need to decide sooner rather than later how we want to proceed.
I have created a motion on GLBSE to issue and additional 6,000 shares, the proceeds of which will be used to place an order for a BFL SC mini-rig.
I will take feedback here, and you can vote on GLBSE.
Currently BTCMC represents .3% of the mining network (1/300th). If we purchase a SC mini-rig, the network capacity could leap to 300 TH/s (25x) before our share of the block reward would go down.
I own 783 shares, I've voted no. |
|
|
|
to asic or not to asic is basically the only important question for a mining operation right now, at least a mining operatoin which owns a substantial amount of BFL fpgas.
I see only two sensible approaches. One is betting the bank on getting your asic hardware sooner than most, and thus preorder as soon as you can, keep your fpgas and trade them in for asics when you can, and there is a slight chance of breaking even or even a nice profit. The window of opportunity will be short, and if BFL gives priority to pre orders that do not trade in FPGAs, you are most likely screwed.
The other approach is selling your FPGAs *asap* to someone who intends to make that bet.
Buying shares in your company now, is betting you as CEO will bet right. The rest is of little importance by comparison. Therefore, may I make a suggestion? Split the company in to gpus and fpgas. I might be tempted to buy shares in a gpu only operation with low electricity cost. Its a gamble, but if these asics get delayed and people start selling gpus in anticipation of asics and reward halving, there might be some good months ahead. Its a risk I would consider.
The FPGA/ASIC stuff, I dont want to touch that with a 10 foot pole.
Interesting thoughts.....I will definitly take it into consideration. There is a lot to think about in the coming months......I suppose rather than a split, I could put forth a motion to the shareholders to float another 5,000 shares for the purpose of purchasing a 1 TH/s rig and ensure we are viable going forward. Yeah, but you gotta worry, BFL might be lying out their ass. And if they aren't, thats a year and a half away. I say grow your farm with Enterpoint hardware until 28nm FPGAs come out at the end of this year. |
|
|
|
Hahaha Double SHA(256)posting Try writing a post on a Samsung Tab 10.1 and you will know what I have to endure !  Battery powered bluetooth micro-keyboard. They make them. |
|
|
|
|
DMC will probably end up buying from... well... anybody but BFL.
|
|
|
|
Does the trade in for value apply if we acquire singles on the used market? EG if I own 20 singles bought from another miner, will I get $599 each in trade-in?
YES ! Please ignore all of Bulanula's posts. He is correct though... I do realize that, but over 70% of the time he is trolling, I just wanted to say that. Who is that? His is profile says Scammer and he has all xxx's instead of coins?? Smart1985 overpaid Bulanula by 22.5 BTC for something, and Bulanula refuses to give the money back. That's why he got the "scammer tag" (X's instead of coins, and a label saying 'SCAMMER'.) And this is why Theymos won't let me be an admin. I would have banned him and requested that Smart1985 file a police report and send me a copy. |
|
|
|
errr.. don't they mean 2 pcie slots? LOL Single slot, double wide. |
|
|
|
so that's what it would look like if Intel made a GPU  F'n sexy! Yeah I love the industrial metal look without all the plastic. Helps heat dissipation too  This is not crappy AMD card with plastic cover  You do realize thats an air channel, right? Hot air channel will heat up top metal cover by convection ( hot air rises ) and conduction and the metal will get hot up there. Even the plastic on my AMD cards is quite hot to the touch and not normal room temp. even if it does not conduct heat as well as metal Ur an idiot. Not totally, a metal shroud would conduct more heat, resulting in different thermal dynamics. But I don't know how much of an actual temperature difference there would be. Technically you'd want a material that conducts zero heat so it is properly vented out of the case and not leaked into it. Plastic isn't entirely appropriate, but its cheap. Metal isn't appropriate. |
|
|
|
so that's what it would look like if Intel made a GPU F'n sexy! Yeah I love the industrial metal look without all the plastic. Helps heat dissipation too This is not crappy AMD card with plastic cover You do realize thats an air channel, right? |
|
|
|
so that's what it would look like if Intel made a GPU F'n sexy! Yeah I love the industrial metal look without all the plastic. If intel made a *working* GPU, I bet it would kick the shit out of AMD compute wise.. Then why don't they? |
|
|
|
|
Subbed, especially after BFL's stunt.
|
|
|
|
So what is it, exactly?
50-core x86 co-processor, code named Knights Corner, in a PCIe card. 50 normal cores? And 22nm trigate fab, so its probably some derivative of Sandy Bridge (but with Failabee-like ring busses). I wonder what sort of onboard memory it as and if its considered local memory and normally addressable. Well apparently they are x86 compatible, but I can't imagine that they are totally complete in the sense that they could be used as a main processor. I seem to recall reduced cache and a few other limitatoins. Cache isn't the issue, really. But if it still can perform just as well on highly branchy code, I might have a use for one of those. |
|
|
|
So what is it, exactly?
50-core x86 co-processor, code named Knights Corner, in a PCIe card. 50 normal cores? And 22nm trigate fab, so its probably some derivative of Sandy Bridge (but with Failabee-like ring busses). I wonder what sort of onboard memory it as and if its considered local memory and normally addressable. |
|
|
|
Oh boy, I'm just beginning to truly realize what a mess an algorithm change will be...
Even assuming a best case scenario - people don't have TENS OF THOUSANDS OF DOLLARS in ASICs to pay off yet - bitstreams and miners need to be changed and it will take MONTHS, it's going to be slow, the new fork (with the better algorithm) will generate mostly worthless coins for the moment (since very few are involved in it) so people will insist on staying on the old fork, with the vulnerability, arguments will arise not to mention the severe confidence hit Bitcoin will take. A lot of people will dump their coins when they hear Bitcoin is vulnerable to the very core and their money's security depends on using a fork that very few miners support. Not to mention that every merchant and service provider using Bitcoin will have to change their entire infrastructure... Holy shit, it's a nightmare. And that's the BEST case scenario...
The WORST case scenario is that ASICs have a good chunk of the hashing power. In that case that entire chunk is there to stay. Since ASICs can't be updated and the miners have lots of money invested in them, they would lose all of their investments if they would support the new fork. So they won't support it. Therefore the new fork will be reserved to a few percent of the hashing power and the old fork will carry on with the vulnerability. So the change would come at a dead slow pace and it could be enough to make Bitcoin dead in the water.
Either way, a hashing algorithm change is extremely likely to be disastrous for Bitcoin. The entire infrastructure would need updating, specialized hardware would need to be thrown out rendering tens, maybe hundreds of thousands of dollars lost. The dead slow updating and people insisting on keeping a vulnerable fork could be the final nail in the coffin.
Now that I managed to scare myself shitless regarding the future, can somebody tell me I'm wrong and how can all this be easily mitigated?
Maybe starting the updating procedure a year or so in advance, before any vulnerability is discovered, giving time to everyone to update slowly and be prepared. The new fork could start as a test net of sorts and be reset when it's decided to start the fork proper.
Perhaps start with the obvious: We don't need an algorithm change. We don't need one YET. But such an eventuality is being planned for. |
|
|
|
Well, because FPGAs are simply destroyed.
ASIC goes in here too. Should rename to FPGA/ASIC Take it up with Theymos. |
|
|
|
@rjk : I tried chaining pci-extenders and surprisingly it worked, but with a reduced hashrate
I lost ~6% on the card I connected in a 1x -> 1x -> 16x configuration with cablesaurus' extenders.
I checked several times that I didn't make a mistake (only inserting or removing the 1x -> 1x cable). Did you see anything like it (it was with a Saphire 5850 extreme) ?
6% is very high. It sounds like your -f/-I is too high/low. |
|
|
|
Well, because FPGAs are simply destroyed.
ASIC goes in here too. |
|
|
|
|
Show Posts
