Open GoogleCodeExporter opened 8 years ago
I agree that "processors" has a non-intuitive meaning. However, I am not sold
on "threads" being the right word. What about:
Approaches
Populations
Attempts
Channels
Competing Build Orders
Original comment by qwerty10...@gmail.com
on 9 Nov 2010 at 1:53
Threads seemed like the logical choice to me, in-line with other programs which
use concurrent tasks (super-pi comes to mind).
In terms of other suggestions:
Approaches - unclear? Sounds like the total number of different "routes" to try.
Populations - Seems reasonable, but to people unfamiliar with EA would not be
clear what it meant
Attempts - misleading, implies more "the total number of tries" i.e. try 4
times then give up
Channels - also a tad unclear as to what it would do.
Competing Build Orders - Probably the clearest of the suggestions, but quite
verbose.
A variation on Threads could be Simultaneous Threads, but as with Competing
Build Orders it's a bit on the long side.
Original comment by kel.bolden@gmail.com
on 9 Nov 2010 at 2:02
I vote for "threads" or maybe even getting rid of it altogether. I mean if
your computer has 2 processors, then having more than 2 threads won't result in
much of a speed up, will it?
Original comment by mike.angstadt
on 10 Nov 2010 at 3:24
[deleted comment]
Earlier versions limited the number of threads to the number of cores, however,
having more competing builds or threads means greater variety and even though
each strain evolves slower, there are more strains (so more chances for it to
find a good build). I have 2 cores and I find 4 threads finds good solutions
faster than 2.
After thinking on it I'm not so sold on "Threads" either, I think Competing
{something} or Simultaneous {something}, or even [Max] Concurrent {something}
Original comment by kel.bolden@gmail.com
on 11 Nov 2010 at 4:19
Maybe 'Chambers' with a popup explaining that each working 'evolution chamber'
tries a different upgrade path and that's it's recommended to be around the
number of CPU cores on the machine?
Original comment by netpr...@gmail.com
on 22 Mar 2011 at 8:03
Original issue reported on code.google.com by
kel.bolden@gmail.com
on 9 Nov 2010 at 12:50