socks
socks
  • Threads: 15
  • Posts: 364
Joined: Jul 13, 2011
October 23rd, 2016 at 3:28:01 PM permalink
I'm just throwing this question out in case anyone has a quick answer. I'm not expecting any deep thought.

I'm getting -5.23% for CStud value instead of -5.22%. Looking closer, I found that my total hand combinations for CStud calculations are low by 468055968 or .0024% of hands. Somewhere I'm dropping every 40000th hand or so?? What nook and/or cranny causes such small losses of hands?

Furthermore, apart from a net loss of hands, straights(incl royals) and pair/singletons counts are all low, while pair/set/house/quads are all high.

I've got my runs down to about 8 hrs, so the cost of trying things isn't prohibitive, but it's not 0, and I'm not sure where to look.
ThatDonGuy
ThatDonGuy
  • Threads: 122
  • Posts: 6679
Joined: Jun 22, 2011
October 23rd, 2016 at 5:10:27 PM permalink
At first glance, I don't see anything obvious. Note that 468066968 er, 468055968 = 4861 x 59 x 51 x 32.
Last edited by: ThatDonGuy on Oct 23, 2016
socks
socks
  • Threads: 15
  • Posts: 364
Joined: Jul 13, 2011
October 23rd, 2016 at 5:38:53 PM permalink
Quote: ThatDonGuy

At first glance, I don't see anything obvious. Note that 468066968 = 4861 x 59 x 51 x 32.


Thank you for the thought. Maybe it will come in handy.
socks
socks
  • Threads: 15
  • Posts: 364
Joined: Jul 13, 2011
October 28th, 2016 at 2:29:14 PM permalink
It looks like the problem might be with my threaded code. No amount of combinatorics was gonna find that. I have a fixed version running now to confirm.
  • Jump to: