Again: dual mining is not possible for ZEC because it uses both GPU cores and memory intensively.
About new version: it takes more and more efforts to achieve more speed, so I cannot release new versions every day now. Also don't expect +50% speedup anymore, only 15-30% is possible in total.
Figured that might be the answer. Or simply that it's sort of pointless due to memory congestion to this point where mixing the algos to this point that dual use offer to profit increase over single algo mining.
Also don't expect jumps as precious. I would expect there is maybe a total 30-40% optimization left in Zcash total looking at power usage and Mem bandwidth. And that's probably only true for high end cards such as 290x390x over 470 due to hardware architecture.
I'm also not an idiot and realize that if you can make extra money pre-runninf a private farm for a while you will prob take the extra paycheck. I don't even think it's unethical.
Not to mention any further optimization will come at cost of power usage and CPU time.
I personally think the next "wave" will be if someone drops the some on equihash ASIC. But that's prob 10 months away and $5 mil. Possibly FPGA before that but not "life changing" from what I can figure with Xlinix and dram modules.
Thank you for your excellent work.
