Bitcoin Forum

I was wondering if it makes any difference in which pool you join with your Asic as the MH/s will be high?

Any pool that has upgraded to Stratum or GBT based pool servers.

Stratum - BTC Guild & Slash
GBT - EclipseMC and Eligius

What is Stratum and GBT?

BitMinter will be ready for ASICs before ASICs ship.

In addition to GBT and Stratum, which are new mining protocols, using rollntime and variable difficulty should make ASIC mining over the old getwork protocol possible as well.

How would you use BFL equipment with Bitminter? Also what is "rollntime"?

Mining protocol, if I'm right!

I'll probably end up mining with HHTT - I actually like the idea of having 100Gh/s of hardware sat working on 2048-difficulty shares or some such thing  :D

You can use BitMinter client, an easy to use GUI-based miner that starts by clicking a button on the website, or any of the other miners that will support BFL ASICs. Our miner supports BFL FPGAs today, and will support BFL ASICs as soon as possible.


GBT (getblocktemplate) and Stratum are new mining protocols. Sadly, we have ended up with two competing standards.

rollntime is an improvement on the old getwork protocol (which most people are still using). It allows the miner to create new work by fiddling with the timestamp in the block, instead of constantly asking the server for more work.

Variable difficulty allows the server to set a higher target for proofs of work from miners, also reducing the constant chatter.

Without using these two features of the getwork protocol the new ASICs would create a lot of network traffic and extremely high load on the server.

Variable difficulty is also used by GBT and Stratum. Their advantage over getwork for ASIC mining is that they give the miner a template for a block which enables it to generate more work by itself than rollntime does.

It'd be awesome if Guns n' Roses lead guitarist Slash had made a mining pool... Welcome to the jungle...

/thread

For the love of network bandwidth, something that offers >1 difficulty shares.

I think Eclipse has difficulty 32 which is pretty neat.

I'm more concerned about stratum support than diff >1 shares.

to the OP.....
"why db of course!".... psh .. ROFL .. db... what a joke.

p2pool should be ready now too. Users can select the difficulty they work on for quite some time.

In general, the large pools are bad in terms of support  decentralization,
I recommend the use of medium-sized

such polmine.pl which supports   GBT

Maybe not the best example for a pool to use. Remember when they stole from their miners? https://bitcointalk.org/index.php?topic=50779.0

For some reason some people keep mining there.

So it looks like not many pools are ready for ASICS?

The ones that are will leave the rest behind..

if ASIC are getting delivered tomorrow, yes.
BUT as ASIC will not be here until December sometime (at earliest)... plenty of time :)
Personally I think this thread is rather premature.

According to BFL, shipping will begin in 2-3 weeks.. People need to be ready to milk those asics

That's in "BFL Time".  Add 6 months for the first batch.  Then assume 4-6 weeks delivery is 4-6 months for the first year.

"BFL TIME" LOL, that one gave me a tickle... Dont set your experiences in concrete maybe this time round they'll do a lil better... lol ;D

https://developer.valvesoftware.com/wiki/Valve_Time

valve time sounds alot like the place I come from...

So what are people preferring? Stratum or GBT?

+1
Also, don't forget the concept of "BFL Watts", where 1 BFL Watt is equal to about four real-world Watts once production actually begins.

very bad way  to promote your pool  mr. BitMinter.com Operator

1. polmine.pl NEVER stole anything
2. everything was described on forum.polmine.pl BEFORE

so dont LIE .. OK ?

polmine its the best pool, with new features,very profitable to miners ,  so keep your lies to you. OK ?

An please remeber that centralization is VERY BAD for BitCoin, it is beter to have a lot medium pool, than few big which use strange rules and promote by lies such a BitMinter.com

P2pool. Which has always been based on variable difficulty and fast local work generation.

P2Pool also protects your investment in Bitcoin (and Bitcoin hardware) by not degrading Bitcoin's decenteralization by blindly depending on a centralized pool which must be trusted to not attack bitcoin and not steal from their users. If the roll out of ASICs gives one of the early updated centralized pools 75% of hashpower and the ability to reverse and replace their own transactions at will, what should that rightfully do to the valuation of Bitcoin?

+1 to p2pool, while I haven't delved into it yet, I will probably once ASICs start arriving. Ultimately p2pool is best for te network as a whole.

what do you think about GBT pools? they should offer the same level of decentralisation as p2pool.
or am i missing something?

+2 p2pool!

They do not currently— Luke's very active protests to the contrary—.  A GBT pool is _transparent_ but unless you're running a local copy of Bitcoin and a running miner software that checks the work against the local copy, or against other pools— that transparency doesn't actually accomplish much. GBT pools also perform centralized payouts still, so even once someone develops software to sanity check check the GBT it still won't prevent a GBT pool operator from stealing from miners.

GBT is a good step in the right direction, when miner software evolves to add more checks (as BFGminer has been doing) it will prevent some kinds of abuses— I think miners should _refuse_ to use anything less transparent (though note: After discussing it with Slush he's agreed to add more transparency to Stratum)... but P2Pool is still the current gold standard relative to decentralization. It also has zero fees and pays miners the increasingly important transaction fees which few centeralized pools do.

P2Pool isn't free though: You have to effectively run a setup for solo mining, e.g. a copy of bitcoin-qt or bitcoind, and there is somewhat more network bandwidth usage. It has somewhat higher payout variance for smaller miners than a pool of the same size (due to the higher share difficulty), and if you don't want to get split off into a sub-pool of unupgraded miners you need to update from time to time (it tells you when a majority of hashpower is on a new version) .  OTOOH, it means you're always setup to solo mine, and the p2pool technical community tends to be very helpful.

Let's be honest now. You took money out of the accounts of your users because you wanted a buffer to run PPS. I agree PPS is very dangerous for a pool op and you need a large buffer to even attempt it. But you can't just take other people's money.

Can you understand that the people in this thread are unhappy that bitcoins suddenly disappeared from their accounts?

https://bitcointalk.org/index.php?topic=50779.0

The thread has a google translation of the explanation from your site. If it's not stealing, then what is it? I think you should apologize to those miners and give their money back.

Stratum - Slush is the best option

I'd recommend against using polmine.  They stole about 4BTC from me last year when they froze payouts and then retroactively switched payout methods.

LOL!  "BFL Time!" I love it.

I want to say BFL is the Valve of the bitcoin world... but Valve is too awesome so it would be an unfair comparison

Not quite... it actually accepts diff 1 shares and tests them all against its internal variable difficulty, and still operates off the getwork protocol. In its current form I'm pretty sure it would consume a lot of CPU trying to keep ASICs busy since it already consumes quite a lot of CPU without this added load.

A 60 GH/s rig would find a share in just under 1 minute. I find it hard to believe that it will cause anyone any problems.

I saw slash play live in Brisbane last year. 


That is how the song goes in my head now.
Welcome to the jungle...
We got FPGAs...

*continue awesome guitar riff*

http://hyperboleandahalf.blogspot.com/2010/04/alot-is-better-than-you-at-everything.html

When, someday, the ASIC may see the light, what will happen if a lot of this new computational power goes to the same pool? Which is quite feasible

Hope the pool supports GBT so the control remains with the miners.

Since you chimed in with this...does Eligius or EclipseMC actually support miners being able to build their own custom blocks yet?  Last I heard GBT pools were still sending miners the entire block to build, rather than expecting the miner to submit one of their own.  Just because GBT makes it possible doesn't mean it's being done.

Ozcoin + Stratum + VarDiff = FTW

+1!

hm... what happens if this is going to be p2pool ?

BitMinter is currently testing some goodness for ASICs:

• Stratum with variable difficulty
• GBT (getblocktemplate) with variable difficulty
• getwork with rollntime and variable difficulty

That's all the three viable options for ASIC mining. Join the test and help make sure everything is ready before your ASIC arrives.

See https://bitcointalk.org/index.php?topic=27062.msg1372864#msg1372864 for info.

Just join p2pool, the best pool.

BTCOxygen Mining Pool has support for ASICs

• We have Stratum support, so very few rejects.
• We pay PPS with a fee of only 2%. With PPS, your earnings are protected. We take the risk of bad luck.
• We have developed our own custom frontend. We can add features that you need. Just drop us a note at support@btcoxygen.com.

More info here: https://bitcointalk.org/index.php?topic=127815.0

Luke-jr? I eagerly await your response to this question.

I plan on going to p2pool - assuming the ASIC miners are as efficient there as on other pools.

Can somebody compare mining at p2pool and this BTCOxygen (or any other pool) ? Which is more profitable?

Not quite true.
Any pool that can handle the load will do.

You've got a point, but you're just nit-picking. I COULD technically re-roof my house with a hammer, or I could borrow my uncle's compressor and nail gun. Sure, a small pool using GW might be able to sustain a very small number of ASICs, but not that many, and why would you? A pool that is ASIC-ready needs to have the tools in place to handle the traffic.

Of course I agree that those new protocols are better for high loads, but standard getwork with ntime rolling is still good enough.

We'll see. It may work, but my analogy stands: there are better options. That's the whole point of this thread.

Answer the Question!

Before I allow miners to modify the transactions I hope bitcoind will get a JSON-RPC method that allows you to send it a block (or just prevblockhash + list of transactions) and get back whether this would be a valid or invalid block. I would like to not have to duplicate all those checks that bitcoind does.

If that happens, I'd be willing to support GBT transaction mutations in my pool.

Even without that, miners could implement checking for double-spend attempts across multiple pools. But the only reaction available would be to stop mining in all pools with conflicting transactions. Not a very good solution.

Either way, it would be interesting to know how many miners are interested in this. It makes more sense to spend time on this if miners actually care. Anyone want to set up a survey?

BIP 23 Proposals will hopefully be supported in bitcoind soon.

Or just compare with their own bitcoind and mine on the pool in the most agreement. :)