1) Is it possible to build a GPU machine capable of reaching ASIC levels with the same energy costs? So what I am saying is it possible to convert a GPU into an ASIC miner with software? I don't understand the concept of ASIC miners and don't understand why an energy efficient GPU modified with software can't do the same job. The cost of a couple high end video cards is much higher or the same as a good ASIC miner, but if I lost interest in the hobby I would still have a high end GPU to show for it. If thats not possible and someone can explain that would be interesting as it is my profession.
2) Bitcoin is out of the reach of most single users. Even if I joined a pool the money generated would be low. I don't really want to buy an ASIC machine because it serves only one function. Which are the best alternatives to bitcoin?
3) Are the alternatives to bitcoin going to go to ASIC miners?
I would really enjoy setting up a system that could be used for gaming and bitcoin mining. I have so much spare equipment including old video cards from old Nvidia GTX cards from 2008 to an R270 in a spare computer that might function mining coins though I think they are closer to fire hazards with PCI express expander/add on cards. So if anyone can help me out on where to get started and answer my questions. I would jump right in, but an ASIC miner is going to end up as a doorstop once the 21 millionth bitcoin is mined so I would like to hear about alternatives if there are any.
1) No. ASIC are specifically designed to do one thing well, with NO excess circuitry for doing anything else (except perhaps chip temp and power usage monitoring). GPUs are designed to be a lot more flexable, which needs a lot more circuitry to do the same thing and limits their performance on that "one thing" compared to an ASIC.
The only reason GPUs are viable on most altcoins is that nobody thinks those particular altcoins have enough of a market to support an ASIC (or in the case of ETH the question of "when it will go POS and no longer be mineable at all" seems to have killed any interest in making an ASIC for that algo, expecially coupled with the rather high amount of RAM needed to mine it at all which would lead to an ASIC for it being somewhat less of a performance jump vs GPUs than on algos that currently DO have ASIC options).
2) Bitcoin isn't really out of the reach of most single users, but it IS somewhat less expensive to built a low-end GPU-based (or dual purpose) rig that can be effective at mining Altcoins at this time.
3) It depends on the altcoin - if one of them gets big enough long enough (ZEC looks like the best current candidate, then probably XMR) someone MIGHT build an ASIC for it eventually.
I expect to see someone build a ZEC ASIC at some point, and I'll be a little supprised if one DOESN'T show up before the end of the year.