Okay, so it doesn't take long from reading posts like this to discern an obvious sentiment that I see from a lot of miners. Considering the mining world and comparisons of CPU/GPU, ASIC, FPGA... Of course, CPU/GPU comes first from the energetic people developing the technology in the first place. This is where we started: CPU/GPU.
So then, what credibility do coins being released on algorithms such as X11 as a starting point? Okay, not everything is judged from the algorithm a coin chooses. For example, coins like Polis and many others have viable business assets being developed and already in place which gives them value so the algorithm is less important. However, mining them still is. And what happens when no one mines it anymore because people stop making X11 miners and the existing miners breakdown after aging for too long?
The next 10 years will be very interesting and it will be interesting to see what major algorithm forks we see from coins who are serious about longevity.
ASICs are really only disruptive when the mining machine hashrate is very large relative to the overall network hashrate. If you look at something like the S9, when it was first released it was a very small spec of hashrate compared to the whole network. So it took a very large number of S9s to begin to have a noticeable effect on the overall profitability of BTC mining. X11 and Equihash and Ethash do not have several generations of ASICs in their history. When the limiting factor on the mining machine becomes the die size of the chips as it has with BTC ASICs, and the speed of the machines relative to the overall network hashrate is small, then ASICs will be far less disruptive.
At this point, it appears that cryptonight V7 ASICs probably already exist, a mere two months after the network forked to get rid of ASICs. With this in mind, it seems like the smarter move would be for coins to either stay on algorithms that have widespread ASICs or move to them. The BTC network is pretty well secured in terms of hashrate distribution despite the massive advantage that bitmain and antpool have there. Other algorithms will look more like that when successive generations of miners are built for them.
I think the only 100% effective way to escape ASICs is going to be to go full POS.
Of course, going full POS would escape any and all POW rather it be ASIC/GPU/CPU/FPGA. It would also create a situation where crypto currencies are no longer special in the sense that they would be attainable through some means other than purchasing them on some exchange rather that be coinbase or some discord chat room. lol... there definitely isn't enough negative sentiment against ASIC or any POW to make that happen.
But what I think is more interesting is to see how the lifecycle of bitcoin and litecoin will play out and what lessons may be learned by developers going forward. Now... "lifecycle" is very important as a concept to consider in this whole equation. Sure, bitcoin has a lot of support and infrastructure but other coins are emerging with completely different offerings. Smart contracts through ethereum has far more staying power than what bitcoin has right now. Major businesses have been spawned through ethereum in the past year. So again... think of the long-term bigger picture... the lifecycle that is to be played out and how that will guide future decisions and developments of new coins.
Any review/opinions I've seen for new coins that are on ASIC platforms are instantly 'thumbs-downed' while X16S and X16R are praised (Xevan was too until some recent flaws were found... it's still good but takes some tweaking to dodge ASICs). Okay, so already, we see the immediate demands from GPU/CPU miners demanding ASIC resistant coins and algo/coin developers know this and realize the importance.
So as bitcoin becomes less dominant in say 10+ years (maybe sooner depending on cryptocurrency adoption) this culture of ASIC will be fought against.
idk though... we've seen how ancient GPUs like the 7950 are still useful and people seem to always find a way to integrate old tech. I also wouldn't be surprised to see some odd implementation where ASICs that were thought dead all of a sudden are useful again. But for the most part I think we'll see emergence of radically improved ASIC resistant algorithms.
ASIC resistance ties into the lifecycle discussion. The very term "ASIC resistance" was derived of the lifecycle process and further evidence of "why" it is important. If crypto is to continue to grow it is the genesis "why" that will enable this and that means a ubiquitous mining capability that is NOT ASIC.
sorry... sort of all over the place but it is a very important and interesting discussion with some moving parts... I know the market is currently ASIC dominated (bitcoin) but I also believe that the long term lifecycle will result in going back to GPU/CPU mining availability based on community interest and community demand.