FPGA mining for fun and profit

[AntiVigilante]

See your private message.

Thx.

[1714101870]

1714101870

[greenlander]

Did anyone see this yet?
https://github.com/progranism/Open-Source-FPGA-Bitcoin-Miner

Lol, actually I did already see it, since the author announced it on this very forum.

http://forum.bitcoin.org/index.php?topic=9047.0

[Basiley]

32x bigger[8x ? didn't saw bigger chips/matrix] FPGA's, made on modern 0.045 waffer[mean working on clocks, lower(cuz arch limitation), but comparable to modern GPU's]can flush GPU's out of business and reinforce bitcoin or ruin it[cuz majority of inviduals can't afford it at 1st time].

[kjj]

The market for GPUs is millions of times the size of the high-end FPGA market, and gets MUCH better press.  As long as AMD doesn't do anything to make mining less efficient on their chips, we will have nothing to fear.

[Basiley]

The market for GPUs is millions of times the size of the high-end FPGA market, and gets MUCH better press.  As long as AMD doesn't do anything to make mining less efficient on their chips, we will have nothing to fear.

press ? what for ?
market ? how GPU's market created/grow ? have it anyway related with FPGA ? with or without BTC mining context ?
actually in newer AMD CPU chips[2 weeks before launch] dramatically improved sheduler/prefetcher[4-6 faster decoding/prediction/fetching of subs/micro/nano-ops/dataportion]as well as 128[finally !! hello real-quick SSE2 ! among other math thing ]FPS, but only four per 8 int exect pipes :/ so new AMD chips could compete with iNtel SB chips very well [in case of "8x core Bulldzoder vs 4x core SB"], esp if prices was competive. and they can be - recomeded price for low-end 8x core FX chip is $210, ie same as today 6x core Phenom II 1100T CPU. and behave more interesting in scientific computation :-P
for further detail - study a Software Optimization Guide for AMD Family 15h http://support.amd.com/us/Processor_TechDocs/47414.pdf very entertaining reading
<--initially im wanted just to get quick highlight on such chips, but when im start reading im can't stop, until finished it completely, then said "Wow" and smoke small cup of tea to combat stress/shock[hate/intolerate intoxication of any kind. including endorphines] and then check AMD shares cost on NYSE.
IMO only RRAM breach into market can make huger impact in IT advancement !! 0.3 ns swithing time !! isn't that cool ??! for about last 12 years memory subsystem become most serious and dramatic bottlenecks[thanks, Rambus to you contribution  into memory biz R&D stalemate !!].

[kjj]

The market for GPUs is millions of times the size of the high-end FPGA market, and gets MUCH better press.  As long as AMD doesn't do anything to make mining less efficient on their chips, we will have nothing to fear.

press ? what for ?
market ? how GPU's market created/grow ? have it anyway related with FPGA ? with or without BTC mining context ?

When a new video card comes out, hundreds of thousands of people know about it, and there are big events with reporters.  When a new FPGA comes out, twenty sweaty engineers know about it, and the press is nowhere to be seen.

[Basiley]

The market for GPUs is millions of times the size of the high-end FPGA market, and gets MUCH better press.  As long as AMD doesn't do anything to make mining less efficient on their chips, we will have nothing to fear.

press ? what for ?
market ? how GPU's market created/grow ? have it anyway related with FPGA ? with or without BTC mining context ?

When a new video card comes out, hundreds of thousands of people know about it, and there are big events with reporters.  When a new FPGA comes out, twenty sweaty engineers know about it, and the press is nowhere to be seen.

you don't get it.
sometimes press/PR noise don't help, but dammage effort to create something, bring it to market and etc.
both because competing for cooperation, conflict of interests[how many ppl buy GPU's for mining after this project rise to full-power ? how many NSA/CIA/FSB crunch-farms become deprecated waste of power/iron/silicon ? how quick nowdays chiper's become deprecated too ? and other examples[im intentionaly not enumerate rest very obvious examples of utilisation of such project fruits].
i mean if you plan grow its "'from ground" its okay ? both as BTC backend or "just for fun" or both, but PR support frequently more dammage than help for such start-ups. thats outsider opinion and strictly IMO, but i seen many very amazinng IT projects, collapsed just because big-biz sharks wanted no danger for their high-margins outdated-garbage-shipping "business".
sorry if im not understand something or explain inconsistent or do it wrong way, as well as for my "engrish".

[molecular]

Why is everyone getting upset over FPGAs? They're not a threat at all, and when they do become a "threat", they will be freely marketed to miners.

+1, exactly my thoughts

[Basiley]

Why is everyone getting upset over FPGAs? They're not a threat at all, and when they do become a "threat", they will be freely marketed to miners.

+1, exactly my thoughts

they become threat, after SHA256 become really unusably unsafe[quicker than initially expected], cuz GPU's isn't very suited for majority of other hash'es crunching
while FPGA's stay afloat :-)

[molecular]

Why is everyone getting upset over FPGAs? They're not a threat at all, and when they do become a "threat", they will be freely marketed to miners.

+1, exactly my thoughts

they become threat, after SHA256 become really unusably unsafe[quicker than initially expected], cuz GPU's isn't very suited for majority of other hash'es crunching
while FPGA's stay afloat :-)

but by then, fpga code and hw (or ready-made mining stuff based on fpga) might be "freely marketed to miners", so anyone sitting on then-unsuitable GPU can switch.

Also, I'm not sure GPUs are unsuited for calculation of the majority of other hashes. Where do you take this info? It would be a great threat to people using ASICs, though.

Also: you don't seriously believe SHA256 will become "unsafe" soonish, do you?

[MoonShadow]

Even if there is some kind of shortcut found in SHA-256, this would simply require that the difficulty increase to compensate.  The only reason that we would need to migrate away from SHA-256 in short order is if a method of reversing the hash were found to be feasible, in which case SHA-256 would be broken.

[kjj]

The market for GPUs is millions of times the size of the high-end FPGA market, and gets MUCH better press.  As long as AMD doesn't do anything to make mining less efficient on their chips, we will have nothing to fear.

press ? what for ?
market ? how GPU's market created/grow ? have it anyway related with FPGA ? with or without BTC mining context ?

When a new video card comes out, hundreds of thousands of people know about it, and there are big events with reporters.  When a new FPGA comes out, twenty sweaty engineers know about it, and the press is nowhere to be seen.

you don't get it.
sometimes press/PR noise don't help, but dammage effort to create something, bring it to market and etc.
both because competing for cooperation, conflict of interests[how many ppl buy GPU's for mining after this project rise to full-power ? how many NSA/CIA/FSB crunch-farms become deprecated waste of power/iron/silicon ? how quick nowdays chiper's become deprecated too ? and other examples[im intentionaly not enumerate rest very obvious examples of utilisation of such project fruits].
i mean if you plan grow its "'from ground" its okay ? both as BTC backend or "just for fun" or both, but PR support frequently more dammage than help for such start-ups. thats outsider opinion and strictly IMO, but i seen many very amazinng IT projects, collapsed just because big-biz sharks wanted no danger for their high-margins outdated-garbage-shipping "business".
sorry if im not understand something or explain inconsistent or do it wrong way, as well as for my "engrish".

The whole point is that GPUs are likely to retain a huge advantage over FPGAs because they are a larger market.

[Zibbo]

What are the odds of getting the next block, and being able to prevent transactions that you don't approve of for ~10 minutes?  51%

You don't get it. If you have more than 50% of the mining power you can just continue searching for a valid block even after somebody else has found one before you. This way you can make sure that you control every block. This is listed in the weaknesses on the Wiki. Yes, you can technically still add transactions, but there's not much point if they'll never get confirmed.

No, he gets it, and you are still a little behind on the curve.  Having control of 51% of the hashing power of the whole network makes it possible to successfully attack the blockchain for a short period of time.  That period of time being one 10 minute interval.  The whitepaper doesn't go into detail about the odds of success of such an attack, other than to show how it's not really possible at all at less than 50%.  Having just over 50% of the network hashing power doesn't really give you very good odds of success past one block, and an attacker intending to deny transactions into the blockchain for longer than one block has to be able to be certain that no blocks can sneak in under him, for if one gets in and the next is built on top of that before he build one to overwrite that one and one on top to secure his false one then it become exponentially more difficult for him to overwrite two blocks back.  In practice, an attacker wishing to keep this up for an extended period of time needs at least double the hashing power of the network because it's like the attacker is trying to wade up river while the honest nodes are wading down river.  And even with double the rest of the network, some blocks are going to slip in and be covered up again anyway.  At which point the attacker has to choose between trying to overwrite two blocks and then write another before a third is made by the network or simply ignore the one that got away and overwrite the last one to take the network back.

If that's true, there must be some mechanism for preventing long chain splits (or rather reorganizations) that I don't know about. I mean if I have 51% of total hashing power, I can just start building my own chain in the dark, and wait for it to get longer than the "official" chain, which it always eventually will. Then I announce my chain to the rest of the network, and clients have no choice but to accept my chain as the official record of transactions. Rinse and repeat... Right? Tell me what I'm missing here. I mean sure, people will notice very quickly that something fishy is going on, but what can they do? Nobody has the authority to pick and choose from two valid chains for the rest of the network. Clients will automatically switch to the longest chain. Right??

[Basiley]

The market for GPUs is millions of times the size of the high-end FPGA market, and gets MUCH better press.  As long as AMD doesn't do anything to make mining less efficient on their chips, we will have nothing to fear.

press ? what for ?
market ? how GPU's market created/grow ? have it anyway related with FPGA ? with or without BTC mining context ?

When a new video card comes out, hundreds of thousands of people know about it, and there are big events with reporters.  When a new FPGA comes out, twenty sweaty engineers know about it, and the press is nowhere to be seen.

you don't get it.
sometimes press/PR noise don't help, but dammage effort to create something, bring it to market and etc.
both because competing for cooperation, conflict of interests[how many ppl buy GPU's for mining after this project rise to full-power ? how many NSA/CIA/FSB crunch-farms become deprecated waste of power/iron/silicon ? how quick nowdays chiper's become deprecated too ? and other examples[im intentionaly not enumerate rest very obvious examples of utilisation of such project fruits].
i mean if you plan grow its "'from ground" its okay ? both as BTC backend or "just for fun" or both, but PR support frequently more dammage than help for such start-ups. thats outsider opinion and strictly IMO, but i seen many very amazinng IT projects, collapsed just because big-biz sharks wanted no danger for their high-margins outdated-garbage-shipping "business".
sorry if im not understand something or explain inconsistent or do it wrong way, as well as for my "engrish".

The whole point is that GPUs are likely to retain a huge advantage over FPGAs because they are a larger market.

underdeveloped/underscaled/slow/not-mass-produced things fail anyway.
and FPGA hardly make any specific there.

[MoonShadow]

What are the odds of getting the next block, and being able to prevent transactions that you don't approve of for ~10 minutes?  51%

You don't get it. If you have more than 50% of the mining power you can just continue searching for a valid block even after somebody else has found one before you. This way you can make sure that you control every block. This is listed in the weaknesses on the Wiki. Yes, you can technically still add transactions, but there's not much point if they'll never get confirmed.

No, he gets it, and you are still a little behind on the curve.  Having control of 51% of the hashing power of the whole network makes it possible to successfully attack the blockchain for a short period of time.  That period of time being one 10 minute interval.  The whitepaper doesn't go into detail about the odds of success of such an attack, other than to show how it's not really possible at all at less than 50%.  Having just over 50% of the network hashing power doesn't really give you very good odds of success past one block, and an attacker intending to deny transactions into the blockchain for longer than one block has to be able to be certain that no blocks can sneak in under him, for if one gets in and the next is built on top of that before he build one to overwrite that one and one on top to secure his false one then it become exponentially more difficult for him to overwrite two blocks back.  In practice, an attacker wishing to keep this up for an extended period of time needs at least double the hashing power of the network because it's like the attacker is trying to wade up river while the honest nodes are wading down river.  And even with double the rest of the network, some blocks are going to slip in and be covered up again anyway.  At which point the attacker has to choose between trying to overwrite two blocks and then write another before a third is made by the network or simply ignore the one that got away and overwrite the last one to take the network back.

If that's true, there must be some mechanism for preventing long chain splits (or rather reorganizations) that I don't know about. I mean if I have 51% of total hashing power, I can just start building my own chain in the dark, and wait for it to get longer than the "official" chain, which it always eventually will. Then I announce my chain to the rest of the network, and clients have no choice but to accept my chain as the official record of transactions. Rinse and repeat... Right? Tell me what I'm missing here. I mean sure, people will notice very quickly that something fishy is going on, but what can they do? Nobody has the authority to pick and choose from two valid chains for the rest of the network. Clients will automatically switch to the longest chain. Right??

What you are describing is a slightly different attack.  This is the double spend attack in a nutshell, and not a denial of service attack.  The DOS requires live participation in the network.

Still, it's not about the longest blockchain, it's about the one with the greatest proof of work.  But yes, hashing a false chain in the dark, so long as you have a least as much power as the remainder of the network, will eventually get you ahead of the true chain.  51% of the network still doesn't cut it, though.  This would only give you 2% more power than the rest of the network, meaning that you would only be able to get one block ahead every 50 blocks or so, and if the rest of the network continues to grow as it has for the past two years, your still a loser in less than a day.

[Basiley]

hm, start investigating/counter-intel mining pools creators ?
bloody KGB hired Tycho to ruin BitCoin mastermind plans ? :-)
how to fix/resolve it ? insisting/force pool creators/maintainers give control/move physically pool-related infrastructure to bitcoin foundation or affilated entity[mining-dedicated sub-division ?]
or just scrap pools support ?

[Zibbo]

What are the odds of getting the next block, and being able to prevent transactions that you don't approve of for ~10 minutes?  51%

You don't get it. If you have more than 50% of the mining power you can just continue searching for a valid block even after somebody else has found one before you. This way you can make sure that you control every block. This is listed in the weaknesses on the Wiki. Yes, you can technically still add transactions, but there's not much point if they'll never get confirmed.

No, he gets it, and you are still a little behind on the curve.  Having control of 51% of the hashing power of the whole network makes it possible to successfully attack the blockchain for a short period of time.  That period of time being one 10 minute interval.  The whitepaper doesn't go into detail about the odds of success of such an attack, other than to show how it's not really possible at all at less than 50%.  Having just over 50% of the network hashing power doesn't really give you very good odds of success past one block, and an attacker intending to deny transactions into the blockchain for longer than one block has to be able to be certain that no blocks can sneak in under him, for if one gets in and the next is built on top of that before he build one to overwrite that one and one on top to secure his false one then it become exponentially more difficult for him to overwrite two blocks back.  In practice, an attacker wishing to keep this up for an extended period of time needs at least double the hashing power of the network because it's like the attacker is trying to wade up river while the honest nodes are wading down river.  And even with double the rest of the network, some blocks are going to slip in and be covered up again anyway.  At which point the attacker has to choose between trying to overwrite two blocks and then write another before a third is made by the network or simply ignore the one that got away and overwrite the last one to take the network back.

If that's true, there must be some mechanism for preventing long chain splits (or rather reorganizations) that I don't know about. I mean if I have 51% of total hashing power, I can just start building my own chain in the dark, and wait for it to get longer than the "official" chain, which it always eventually will. Then I announce my chain to the rest of the network, and clients have no choice but to accept my chain as the official record of transactions. Rinse and repeat... Right? Tell me what I'm missing here. I mean sure, people will notice very quickly that something fishy is going on, but what can they do? Nobody has the authority to pick and choose from two valid chains for the rest of the network. Clients will automatically switch to the longest chain. Right??

What you are describing is a slightly different attack.  This is the double spend attack in a nutshell, and not a denial of service attack.  The DOS requires live participation in the network.

Still, it's not about the longest blockchain, it's about the one with the greatest proof of work.  But yes, hashing a false chain in the dark, so long as you have a least as much power as the remainder of the network, will eventually get you ahead of the true chain.  51% of the network still doesn't cut it, though.  This would only give you 2% more power than the rest of the network, meaning that you would only be able to get one block ahead every 50 blocks or so, and if the rest of the network continues to grow as it has for the past two years, your still a loser in less than a day.

Double spending should be easy with 51% (meaning of course on average 51% of total hashing power during the attack) as well. Buy something with 10000BTC, and announce the transfer normally. Start building a blockchain in the dark with that transaction going to your own address instead. Wait until whatever you bought is delivered to you, and your dark blockchain is longer (greater proof of work if you insist), and dump your longer fake chain to the rest of the network, invalidating the original transfer. Yes, with just 1% edge, if you get a bit unlucky, you might have to wait for a long time to catch up with the official chain, but theoretically it doesn't matter if it takes days or even weeks. Practically you would want to have a bit more than 51% like 55-60% to make it easier, or maybe DOS some of the larger pools to give yourself some edge. But still, it's very much doable.

It's of course possible that I have misunderstood something.

[Basiley]

should/can http://forum.bitcoin.org/index.php?topic=9137.0 trick/approach solve this problem ?
or complete pools decomposition/takeover really necessarily ?

[molecular]

should/can http://forum.bitcoin.org/index.php?topic=9137.0 trick/approach solve this problem ?
or complete pools decomposition/takeover really necessarily ?

I think the solution you referenced will work to resolve the issue.

There are some caveats and details to take care of, however, and it will probably take a while before someone implements it and gets a pool going.

For now, my suggestion is to let no pool go >50%. I never mine in biggest pool.

[Syke]

based on the huge difficulty jump today.... I think someone has succeeded in fpga/asic bitcoin flooding

*bell ring*
We have a winner!

Two FPGA clusters went live.

Let me toss out some rough numbers. Please correct me if I'm too far off.

You have about 300 chips doing 200 MHash/s? That's 60 GHash/s you brought online? While that probably puts you in as one of the top miners, that's not a significant cause for the recent network growth.

For someone without piles of existing fpgas sitting around idle: If purchased new, that's $250k in hardware costs bringing in $10k per week in bitcoin value. Half a year to pay off the initial investment (assuming price/difficulty growth remains steady). GPUs would break even in less than half that time.