[TN]opyrt 2008-12-11 11:49:22 | I must have had the result 3.667213 about a quadrillion times now, and I've seen that this usually happens when a test gives the result 3.7-something which is higher than previous highest, it is quarantined, the four other tests are done, and BLAMMO! 3.667213. It seems to be the same 3.7-result every time. This happens on all three instances I have running. Any thoughts anyone? |
Stephen Brooks 2008-12-15 15:56:22 | Yes, I think this one's probably *really* converged now, though it had some interesting late-in-life surges. I notice that the "1" section variant topped out at 2.922755%, this 6-section variant got to 3.667213% as you are so bored of seeing To be scientific about it, as well as issuing a higher-number-of-sections version, I'm going to do a 3-section one. The thing is with this, increasing the number of sections means you can theoretically get to higher yields but the space becomes so vast the optimiser has trouble getting there. So there may well be an optimal number of sections. I'm going to try 3 and 12. So you can expect Linac900Removable3 to top out fairly quickly and simply, whereas "12" will probably get stuck all over the place but there might be some interesting results here. |
[TN]opyrt 2008-12-15 18:12:15 | Great news Stephen! It's going to be fun to see how these new variants perform! |
Universal Creations 2008-12-31 11:11:37 | I just made some progress with "the 6" (had some old results on my pc). |