Bitcoin Forum
June 22, 2024, 05:55:16 AM *
News: Voting for pizza day contest
 
  Home Help Search Login Register More  
  Show Posts
Pages: [1]
1  Bitcoin / Mining / Re: [NEW POOL & NEW MINER] - BitcoinPool.com - Jump In! on: March 14, 2011, 04:00:37 AM
Hey bobR, my incredibly honed senses are telling me that you seem to be a tiny bit upset... To be honest, you are doing such a terrible job of asking a question, that it really doesn't deserve a response. Nevertheless, I will try to give you the information you are looking for, in the hope that it will put your torment mind to rest.

I will translate your great wall of flame into the following statement and try to answer that: "I am one of the users who were criticised by the operators of the pool for their lack of "efficiency". I do not understand why. Could someone explain what I am doing wrong, please?"

I believe the answer to your inferred question is in post #1 of this thread. For your convenience, I have read it and quoted the important section:


   Geebus and I raised a very important question in m0mchill's thread, and that question was “why doesn't the miner work through all (2^32 or) 4,294,967,296 hashes?”  The answer we got was not a satisfactory or a legitimate response to the original question, and we objected with real stats showing that decreasing your askrate for getwork request does NOT increase the probability that you will find a share or answer quicker.  In fact, you can find more than 1 answer/share per getwork, and that simply requesting a new getwork every 10 or 20 seconds is not the best route to finding shares.

The bottom line from our research was this:
- The same number of shares can be found in a 10 hour period using a askrate of 10 seconds OR our more efficient miner.  You will get about the same number of shares either way.

- Because the same number of shares are found in a given period, there is no reason why people should be using a decreased askrate.  The goal is to look for bitcoin's more efficiently, and when the miner's efficiency is closer to 100%, the amount of resources and bandwidth on the server is DRAMATICALLY reduced, and that means we can support MORE USERS.

- More user's means we as a pool can find blocks faster!


I would not be surprised if this issue has been explained several times since, and, as a person who is seeking an answer, I'm sure you have read them all. Thus you must have failed to understand the previous explanations and so I will try to give you my own understanding of the tale.

It all starts with the fact that, to do its thing, your miner has to request work from the server. By default, common miner programs keep requesting new work very often (e.g. every second). In fact, they request new work long before they have actually finished their old work (this can take minutes). This is particularly the case for comparatively slow mining hardware (e.g. a CPU miner, as opposed to a GPU miner), as this type of hardware takes much longer to find shares and to complete chunks of work.
What this means is that, generally, your miner is asking for new work long before it has even found any shares in its current work. This wouldn't be a problem for you directly, but, unfortunately, every work request produces load on the pool server.
So, as a "half way summary", your miner is putting as much load on the server as tens, or maybe even hundreds, of "efficient" users would (depending on your ask rate), but it is also turning in vastly fewer valuable results (shares), in the process.
This is precisely the type of inefficiency which geebus and FairUser are trying to combat on their server and which is why they have released their own edition of the poclbm miner. In the latest version, their modified miner uses both your resources and the server's resources efficiently. What they want you to do is to use their miner, because, that way, you can contribute to the pool without wasting disproportionately many server resources.
This is in your own interest too, for two reasons:
1: the pool operators don't get angry at you and
2: more people can use the pool at the same time and thus blocks are solved more frequently and thus you get paid more frequently and thus don't get the impression that this pool is paying you less, quite so easily.

To give you an idea, under the assumption that your user name on the pool is "RobertRibbeck", then, according to the latest public stats, your miner has requested work 74330 times, but only submitted 135 useful results. This is the issue. This is what you could address by using the latest version of poclbm-mod.

If you don't understand this explanation, or if you do, in fact, feel that this pool is paying you less for the work your miner submits, or, in particular, if you can't ask questions and explain your problems in a civilised manner, let alone read post #1, then I'm sorry to say that very few people would regret to see you go and leave both the pool and this thread behind.
2  Bitcoin / Mining / Re: [NEW POOL & NEW MINER] - BitcoinPool.com - Jump In! on: March 12, 2011, 10:17:36 AM
Common, guys, why do you think that ALL people arround you are idiots? Why do you think that m0mchil, diablo, jgarzik, puddinpop and others implemented frequent ask rate in their miners? Do you really think that they had not an idea to set it to something longer? They had, but all of them understand that it is bad.  You're reinventing wheel. Wheel with four sides.

I have to say that I find your criticism of geebus to be rather over the top and unjustified. I didn't see him calling anyone an idiot and, unless he was, you should not be accusing him of doing so. Are you after a flame war? This is not the place to have one. As the admin of another pool, I think you should be more respectful.

If you think that your example, that 5% of the submitted shares are stale, is realistic, then I would have liked to see an argument to back that up, as the example is irrelevant otherwise. Unfortunately, with your tone and your lack of thorough argumentation, you have rendered any actual point, that may have been contained in your post, pointless.

I would personally request that you refrain from posting inflammatory messages in this thread.
Pages: [1]
Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!