Hello,
The less crowded a pool is, the more shares per block you will get but the more crowded a pool is the less it takes to find a block. So can we make a mathematical equation taking into account the hash rates, number of miners and current and estimated difficulties in two or more pools to judge which pool is more profitable and mine in that pool?
Secondly, i know this is an extremely newbie question but is there anyone who would mind telling me how and why the difficulty changes, rises or decreases with time?
Thirdly, I have 3 Asus R9 280x on my rig. I mined different coins, although it was one week only, before. All of them was scrypt ones and I got ~750Kh/s per card. Why does my cards mine at ~530~550 Kh/s per card on my rig when mining BDC? As Cloudpost has the same cards I know this is normal but want to learn why.
Lastly, I want to thank to and congratulate Nikolas and colleagues for first thinking and than coding a coin as different as BDC. I hope it will be very beneficial for all who mine it.
Thanks for your feedback!
1. About pools. Mining is based on luck and you cannot predict when you'll find a block. Pool effictiveness is the same, the only difference that bigger pools find blocks more often. But it's much better for coin security to have several pools with a stable hashrate. I think that it's good to choose a pool that is phisically closer to your cards (check location and ping pools). I don't like the fact that most miners try to move to biggest pool. I myself mined in
https://www.minep.it/pool/bdc for several days, and will move my workers there soon.
2. The difficulty adjusts to keep block time close to 2 minuites. Net hashtare increases, blocks are getting solved faster, difficulty rises so blocks will be solved at normal speed. Difficulty retarget is a formula that calculates difficulty based on previous blocks parameters.
3. BDC is scrypt-jane algo. It's different to scrypt, it also has got a flexible N-factor (for now we have Nfactor=9), which affects the gpu performance (for both scrypt-jane and scrypt-n algos).
