[OLD] Eligius: ASIC, no registration, no fee CPPSRB BTC + 105% PPS NMC, 877 #

[Bobnova]

If I hadn't already left Eligius when I realized his wiki flat out lied about his payout methods (may have been fixed now, I don't know, I'm not coming back) I'd certainly leave now.

[phelix]

Luke, could you please shed some light on the CLC events related to eligius or point me to where I can find this info.

Specifically: were eligius pool hashes used to merge mine CLC or any other chains besides btc and nmc?

Thanks.

[jake262144]

Angry neighborhood bastard mod here: Jenkins has been permbanned. Do not repeat his mistake. Continue this thread in a straight forward and orderly manner.

Jolly good.


Luke, I have a technical question totally non-related to the recent snafu.
I get great ping times to your pool (50 to 70 ms) but somehow I often manage a stale or two during LP notifications.
What algorithm does the pool use when sending LP messages? Biggest miners first? Random?
There must be some heavy magic being used at Bitminter, they have higher pings but hardly any stales.

[Luke-Jr]

What algorithm does the pool use when sending LP messages? Biggest miners first? Random?

Not random, but might as well be. At one point, I modified pushpool to prioritize the more efficient clients (ie, rollntime+noncerange), but I'm not sure if that's live right now. A few months ago, I began writing Eloipool, a very-fast modular Python pool server, and the first with internal work generation (even before PSJ), but I had to turn it off when I enabled NMC merged mining (because Eloipool doesn't support it yet), and the JSON-RPC Server module is very quickly thrown together and cannot handle more than a handful of connected clients. I'd love to get it finished, which would probably solve most of the stales and related issues, but I haven't been able to justify the time as of late. If anyone wants to help, I have the code up on Gitorious...

[Luke-Jr]

BTW, bonus 0.4% shares for cgminer users if you apply my patch: https://github.com/ckolivas/cgminer/pull/68

[jake262144]

BTW, bonus 0.4% shares for cgminer users if you apply my patch: https://github.com/ckolivas/cgminer/pull/68

Thank you kindly, sir. That loop optimization looks great with one less comparison to make.

...and before you guys post, no the patch doesn't contain a list of alt-coin targets ...although it could increase the attack rate by 0.4% 

[Luke-Jr]

BTW, bonus 0.4% shares for cgminer users if you apply my patch: https://github.com/ckolivas/cgminer/pull/68

Thank you kindly, sir. That loop optimization looks great with one less comparison to make.

Actually, the optimization there is trivial and has no net effect. What's important is the bugfix so it doesn't discard all shares starting with FF in the nonce

[jake262144]

That's what I get for not reading the whole file.
I stand corrected. And 0.4% faster.

BTW, that while(1) kinda bugs me. I've always preferred a macroed while(TRUE) instead.

[makomk]

Luke, could you please shed some light on the CLC events related to eligius or point me to where I can find this info.

Specifically: were eligius pool hashes used to merge mine CLC or any other chains besides btc and nmc?

Thanks.

Well, he's pointing well over 250 GHash/sec of mining power at CLC, and he's also been begging in IRC for remote access to computers with OpenCL-capable GPUs for unrelated software develoment purposes due to not having any of his own, so it'd be rather... surprising if he didn't use Eligius users' hash power for this.

Edit: Coiledcoin block #6948 corresponds to Bitcoin block #161234, the 3rd most recent block mined by Eligius, so yeah...

[rjk]

Luke, could you please shed some light on the CLC events related to eligius or point me to where I can find this info.

Specifically: were eligius pool hashes used to merge mine CLC or any other chains besides btc and nmc?

Thanks.

Well, he's pointing well over 250 GHash/sec of mining power at CLC, and he's also been begging in IRC for remote access to computers with OpenCL-capable GPUs for unrelated software develoment purposes due to not having any of his own, so it'd be rather... surprising if he didn't use Eligius users' hash power for this.

Edit: Coiledcoin block #6948 corresponds to Bitcoin block #161234, the 3rd most recent block mined by Eligius, so yeah...

So in other words, the following quote is indeed a lie:

While there are personal attacks in this thread which are bad, most of the discussion of what was done with the pool and coilcoin (or whatever it is) are relevant to the topic of the pool and should not be deleted.

Why does it seem like people don't read anything? The Coiledcoin nonsense is NOT RELATED TO THE POOL.

Shameful. Utterly shameful.

[makomk]

Well, he's pointing well over 250 GHash/sec of mining power at CLC, and he's also been begging in IRC for remote access to computers with OpenCL-capable GPUs for unrelated software develoment purposes due to not having any of his own, so it'd be rather... surprising if he didn't use Eligius users' hash power for this.

Edit: Coiledcoin block #6948 corresponds to Bitcoin block #161234, the 3rd most recent block mined by Eligius, so yeah...

So in other words, the following quote is indeed a lie:

While there are personal attacks in this thread which are bad, most of the discussion of what was done with the pool and coilcoin (or whatever it is) are relevant to the topic of the pool and should not be deleted.

Why does it seem like people don't read anything? The Coiledcoin nonsense is NOT RELATED TO THE POOL.

Shameful. Utterly shameful.

Well, it's certainly rather less than honest, but he genuinely does seem to believe that whatever other things he does with his pool are irrelevant to miners so long as they get their Bitcoins.

(By the way, the same is true of the Eligius-mined Bitcoin block two earlier, block #161142, and Coiledcoin block #6433. Also, by "corresponds" I mean that they share the exact same proof-of-work and were therefore merged-mined together. There are almost certainly more too, I just haven't looked further back.)

[jake262144]

...he genuinely does seem to believe that whatever other things he does with his pool are irrelevant to miners so long as they get their Bitcoins.

I can't quite agree with him (EDIT:: Luke), that's a blatant breach of agreement.
Miners mine at Eligius with the assumption that their hashing power is being exclusively used for Bitcoin (and since merged mining was introduced, Namecoin) mining.
Any other usage of the hashing power is simply unauthorized.

Please think the situation over once more Luke.

[makomk]

I can't quite agree with him, that's a blatant breach of agreement.
Miners mine at Eligius with the assumption that their hashing power is being exclusively used for Bitcoin (and since merged mining was introduced, Namecoin) mining.
Any other usage of the hashing power is simply unauthorized.

Please think the situation over once more Luke.

Errm, good luck with that. Apparently he's already made his position on this quite clear:

There's an ongoing discussion in #bitcoin-otc about this right now.

Direct quote:

Code:

<btc_novice> luke-jr also, it was unethical to take your users' hashing power and use it for other purposes
<+luke-jr> btc_novice: I didn't.
<btc_novice> luke-jr did you use your hashing power, or eligius combined hashing power?
<+luke-jr> btc_novice: all hashing involving CLC was done by me

Again, both technically true and intentionally misleading. Because of the way merged mining works hashing involving CLC was indeed done by him - even if they were helping to mine CLC the miners just hashed normal Bitcoin blocks which unbeknownst to them were constructed to be used as proof-of-work for CLC blocks. Even if their hash power was being utilized to help shut down CLC - and it looks like it was - they weren't directly hashing anything CLC-related.

[k9quaint]

...he genuinely does seem to believe that whatever other things he does with his pool are irrelevant to miners so long as they get their Bitcoins.

I can't quite agree with him, that's a blatant breach of agreement.
Miners mine at Eligius with the assumption that their hashing power is being exclusively used for Bitcoin (and since merged mining was introduced, Namecoin) mining.
Any other usage of the hashing power is simply unauthorized.

Please think the situation over once more Luke.

Incorrect. There was and still is no agreement to provide Coiledcoins. There is no agreement promising not to use the completed POWs for other calculations besides those promised to the users.

[makomk]

OK, so since Luke Jr doesn't seem to think he can explain this usefully, a brief summary merged mining for miners:

As far as miners are concerned, it looks to your mining software as though you're just mining Bitcoins normally with all your mining power. Your mining software gets normal Bitcoin-mining work and submits normal shares at the same rate as it would if you weren't merged mining, if the pool is well-written it'll credit your shares towards Bitcoins at the same rate as if it wasn't merged mining, and the pool will have the same chance of finding blocks too. From a purely Bitcoin-mining perspective it doesn't have any effect whatsoever.

However, because of a clever trick with the work you're sent that you have no way to detect, the shares you submit can also be used as shares towards every other merged-mining-supporting altchain that the pool chooses to mine and can be used to generate blocks for that chain. Again, they don't stop being valid as Bitcoin shares or blocks because they're used them in this way - in theory all the chains gets the benefit of all the shares you submit, though in practice this is only entirely true for Bitcoin. This is actually how I can prove that Eligius is merged mining CLC. A large proportion of the Bitcoin blocks found by Eligius were created from the same share as a CLC block.

In summary, whilst this doesn't affect your income from Bitcoin or Namecoin mining in any way it does mean that you're helping to shut down CLC by mining at Eligius. How important this is, is something for you to judge.

Edit: as k9quant in IRC points out, there's one small inaccuracy. It's more accurate to say that you have no easy way to detect that you're merged mining and no reliable way to tell which chains you're helping to mine on. If you personally find a Bitcoin block you can tell from the block that merged mining was used but not which chains, and in theory you could also detect that you'd mined a block on a particular altchain if you knew to look out for it.

[Luke-Jr]

The relation between Eligius miners and CLC, is the same relationship as between Eligius miners and Joe Drugdealer's money laundering (that he is abusing Bitcoin to accomplish).

[jake262144]

...I can't quite agree with him, that's a blatant breach of agreement...

Incorrect. There was and still is no agreement to provide Coiledcoins. There is no agreement promising not to use the completed POWs for other calculations besides those promised to the users.

By HIM, I meant Luke.   I'll try to be very precise here:

EDIT:: I edited a part of this post out as I felt it was out of place. Not wrong or incorrect, just in the wrong place here as this is a technical forum.
EDIT:: Legal discussion does not belong here and is certainly not had with individuals trying to raise a row.

[Luke-Jr]

Testing welcome: https://github.com/ckolivas/cgminer/pull/69

This one isn't simple, and probably doesn't bring much hashrate improvement. It does, however, prepare the way for mining with ButterFly Labs's BitForce Single FPGA miner.

[MacT]

 How many real $$ has Luke mined in I0coin, ixCoin, and CLC, using Eligius pool power, that has not been shared back to
the Eligius Pool members ?

[jake262144]

 How many real $$ has Luke mined in I0coin, ixCoin, and CLC, using Eligius pool power, that has not been shared back to
the Eligius Pool members ?

Is this in any way relevant? Those coins are nearly worthless and there is a much bigger matter at hand.

By running a pool he governs his miner's resources (hash power). Allegedly, he also uses these resources for purposes other than mining Bitcoins which he wasn't entitled to do.

An analogy is you desperately need one:
You have access to your employer's computer in your work. You are allowed to use the machine's resources as far as this usage is work-related.
If you are caught playing Quake on that machine not only can you get fired but also be sued for unauthorized use of your employer's resources.

I'm going out of my way here to warn Luke that this is not necessarily child's play. He is a very good coder and I wouldn't like him to get in serious trouble.

I've said all I have to say, this is the last time I'm joining this particular discussion.