the "example" is showing SCRYPT capabilities, not scrypt-n. The difference being scrypt isn't adaptive (it doesn't change over time) and coins like litecoin are forever stuck at nfactor 10 (unlike scrypt-n's nfactor 11 with automatic increases over time)
also, the speeds they show in that article are very clearly indicative of nfactor 10 mining (as the knc was showing to do nfactor 11 at about half the rate it can do nfactor 10, which is how pretty much all devices react to nfactor bumps as long as they have the memory capacity)
I am NOT doubting that the knc titan can do nfactor 10 or nfactor 11. I AM doubting it can do nfactor 12 or above. There aren't any coins out right now that even run nfactor 12 so there would be no way to currently test such capabilites. (and there weren't any when the knc titan was in production as vertcoin forked away from scrpyt-n before doing an nfactor bump.
Vertcoin already experimented with different nfactors and basically the consensus was to switch to a different hashing method.
https://bitcointalk.org/index.php?topic=404364It's not the argument if an ASIC exists now, its if an ASIC
could be designed to mine scrypt:n and the overwhelming idea is that its possible. So if you're looking for long term success, no scryptn is not the way to go. There's a reason most scryptn, x11, x13 coins are dead or on their way there.
>inb4 litecoin is still alive and wellThere's no economic analysis to say ASICs are a bad thing for cryptocurrencies. Yes it 'tends' to lead to centralization and away from the original intended model. But for all intesive purposes the coins that are doing really well are getting pounded by ASICs.
So basically you have 2 choices, not claim ASIC resistance, or not give a fuck. Unfortunately Garlicoin already claimed ASIC resistance.