Where did you stop and why?

[qpwo]

Do you think speed could be incrementally improved to the point of being practically useful?

[alexflint]

Well yes I think so. Happy to discuss!

…

On Thu, Jan 13, 2022 at 6:14 PM Luke Harold Miles ***@***.***> wrote: Do you think speed could be incrementally improved to the point of being practically useful? — Reply to this email directly, view it on GitHub <#1>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAE4J5326GKXNXHDK6QNFRTUV6BIHANCNFSM5L5Q5Y4A> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[qpwo]

I'll have to dig into it but first thing comes to mind is using a JIT interpreter would be a pretty large speedup. Which of the exponents if any in the algorithm itself do you think could be reduced?

[alexflint]

Right! The algorithm can look through traders for a long time before it finds one that characterizes any meaningful patterns in the current market, so a big improvement would be to more intelligently look for traders that exploit the current market.

…

On Sat, Jan 15, 2022 at 12:52 AM Luke Harold Miles ***@***.***> wrote: I'll have to dig into it but first thing comes to mind is using a JIT interpreter would be a pretty large speedup. Which of the exponents if any in the algorithm itself do you think could be reduced? — Reply to this email directly, view it on GitHub <#1 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AAE4J57KD3IYC45ERFSA5KLUWEYVNANCNFSM5L5Q5Y4A> . You are receiving this because you commented.Message ID: ***@***.***>