ccMiner v1.2 is now in the github repo. It adds X13 and Diamond-Groestl
thanks to tsiv and Bombadil for these contributions.
I'll do a binary release tonight after our BBQ.
Christian
Has this version just got X13 added?
on a side note ive been doing some test mining jackpot on a pool as solo mining is now basically impossible, 3 days with nothing
and this is what i get from actually mining not from stats
Hash@Me - 25MH/s
1 Hour - 1422 JPC
1 Day - 34128 JPC
Doesn't seem too bad, but that is only ~7K per 750Ti, nowhere near the 22K and 10K estimates that were posted earlier. This is not a dig at those guys, just a realistic show of returnExactly what I'm getting...
Finished my first 24h at "jpc.hashatme.com" and mined 7236 JPCs with one 750.
My current calc says this (just adding the X13 as extra info
) :
TAG: NICEX13 | Name:NiceHashX13 | Algo: X13 | BTC/day: 0,005512250 | Coins/day: 0 | Best exchange: NiceHash | BTC price: 3,242500000 | BTC volume: 0 | Difficulty: 9,7697 | Blockreward: 3,2425
TAG: JPC | Name:Jackpotcoin | Algo: JHA | BTC/day: 0,003037224 | Coins/day: 7992,69483109522 | Best exchange: Bittrex | BTC price: 0,000000380 | BTC volume: 37,68271352 | Difficulty: 674,13401777 | Blockreward: 47829,69
So it's closer to your numbers now. Anyway, I think I might try making my profit calc keeping hourly averages and test that against my own setup.
Bombadil a heads up on what I found. Hourly averages don't work so well for many coins. Take for example Quazar coin (which you probably don't care about right now) that is CryptoNight based. The reward per block is all over the place. Real low to decently high. For a cryptonight coin it's probably 2nd to monero in overall profit as a whole. However by being able to watch the reward in real-time there are times where it's more profitable to mine it instead of monero. I'm not factoring in exchange rate differences but just reward per block. Cash it today, tomorrow or next week. But by watching the reward per block in real-time you can know when a particular coin is "ripe" for mining.
Then speaking of exchange prices. This is where you WILL WANT to use a bit of history and trending and sometimes overlook values or look at purchase volume. A person can very easily throw off your calc at present by placing a small buy order high in price. Your calc will pick this up and would switch to that coin because at that moment in time it's the most profitable. But really it is not. You need to take into consideration the "buying power/wall" that the coin presently has.
Also some coins like Jackpot can throw a monkey wrench into things. As an example lets say that TAC is ever so slightly more profitable then JPC. You might mine TAC because the "pure" numbers say it's more profitable. BUT because Jackpot has the superblocks worth a lot it depends on the pool you mine at. Some pools do a full split of the jackpot while others do a 50/50 split and some do no split at all (just like solo mining). Either way, if you are mining at a pool that shares some of the jackpot then the profit calc should probably push Jackpot profitability higher than TAC. It's one of those things where the "pure" numbers might not equal the most profitability over time.
Jackpot is really different from a strategy standpoint. For example there are at least 2 big pools that do a split. One big pool that does 50/50 split. So as a "human" we can easily see there are times where you might want to split your hash up between different pools to have a great shot at grabbing some of the jackpot. By pure numbers you might want to exclude the 50/50 pool and just split your hash with the two bigger pools that do a full split. Law of averages and probability say this is the better way to do it long term.
Also of course in the example above is the size of the current jackpot. The bigger the jackpot the more "influence" it should have on mining JPC over TAC or any other coin.
These are just a few examples but there are many more. These are really the only ones I think that matter for nVidia GPUs right now (with the exception of the CryptoNight coins). I just wanted to give you a heads up on maybe some other ways to think about things from a profitability standpoint. I'd share code with you but I've got stuff that reads deamons, does screeen scraping, does API and uses SQL server extensively so it's a bit all over the place and driven by the backend SQL server.
I looked threw your code and can see you know what you're doing. My style of coding would not work for you as I don't care for objects/classes extensively as they are slow in execution. I do a lot of stuff where speed is most important (think real-time acquisition of thousand of data points per second) for things like refinery operations, airline black-boxes, nasa, astro and NSA type stuff. Point being my coding style would not help you a lot and would look more like assembler then a modern language. You are quite capable judging by your code, so just getting some additional ideas is probably just as good as getting code, I'm sure if you like any of the ideas or thoughts I just threw out to you that you could implement them rather quickly.
Cheers, and I think your calc is a wonderful addition to the community,
Carlo
