stephenbrooks.orgForumMuon1GeneralProgress towards v4.22
Username: Password:
Search site:
Subscribe to thread via RSS
Stephen Brooks
2002-08-21 13:03:27
Today (Wednesday) I implemented the following features.

  • Changing CWD to containing-dir (by parsing commandline), for running as service etc.
  • Put all the checksumming code and FTP passwords into one module "security.c" which I do not release with the source-code distributions
  • Checksums for individual records of results.txt
  • Results pre-send reparser removes repeated and bad-checksum results
  • Muon1 also auto-removes results with "rank" in them (i.e. from bestXXX)
  • Message box for version notification is asynchronous (threaded)


More to come tomorrow.  Will be next week or later when I upload the alpha for this.  I don't know what definitions other people use, but for me,

"alpha" becomes "beta" when people have run it for 24 hours without bugs
"beta" becomes release when people have run it for 3 days to 1 week without bugs.


"As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline"
px3
2002-08-21 13:33:13
Hi Stephen,

does it really make sense to port the client to *nix while you're updating the code ?

i still can port the current client to *nix, but imho it's wasted time, while your working on a newer version.

Hope to get a statement about your point of view...

Kind Regards,

Rolf
px3
Stephen Brooks
2002-08-21 14:02:27
That thought had occurred to me as well.  What I've been wondering about is whether I can port it to linux myself, and add #ifdefs to switch between the two environments.  There's nothing really wrong with v4.21b if you think you can port that to linux, but there may be another version out before you finish.
px3
2002-08-21 14:18:04
Oki, i keep my hands off that port.. (at the moment)

as mentioned in another thread: it's your project, your code.. i only can offer help.

but i think we both thought about the same problem.. big grin

if you need any help in porting, feel free to contact me..


Regards

Rolf
PX3
Stephen Brooks
2002-08-22 10:37:08
Today (Thursday) I did the following things.

  • Checksum now takes into account the record _score_ as well as the design itself
  • Manualsend.exe now gives a more sensible error when there is no results file to send


Still working on it, however.  And today I did some other simulations to test out the reliability of the results and other stuff.


"As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline"
Orbi-tel
2002-08-25 02:07:35
Stephen,

I have tried to port both 4.21b and the command line version to Linux and there are still major issues:

You are using some lcc includes (in rnd.c) which gcc does not have
You are still using non-portable structure defs which gcc does not like

Rather than try to muddle through and produce something that might work I think you are the only one who can safely do the *nix porting so I have given up (again) for now.
Pascal
2002-08-25 02:54:33
Hey Stephen, congrats to a yield of 2.454 percent. 
What did you do to achieve such a high yield?

___________________________
Member of www.rechenkraft.net - German Website about Distributed Computing Projects

1: Athlon TB-C, 1.2 GC/s, 256 MB DDR-RAM, Erazor x², ADSL-Flatrate, NIC Intel, Win 98 SE Mainboard MSI-6380 Rev.  1
2: Pentium III, 600 MC/s, 256 MB RAM, NIC Intel, Win 98 SE
Stephen Brooks
2002-08-25 05:43:52
Orbi-tel: if it's any help, I'll be compiling using g++ on UNIX, as g++ is the closer match to LCC out of the two.  The code reduces to more efficient plain C in the innermost functions (although I should look into inlining a few of them + unrolling some loops too).

Pascal: yes, that was got by doing a rather simple thing: making all the solenoids as large, long and powerful as possible and making all the gaps as short as possible.  If this turns out to be the optimal configuration after all then this project will have been rather boring, but I'm hoping maybe we can find a different design with even higher yield than this.  If it does turn out to be "boring", version 5 will simulate the third and final section of this design too, and the optimisation for that ought to be non-trivial.


"As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline"
Orbi-tel
2002-08-25 13:33:46
Stephen,

I already used g++ (gcc uses g++ what compiling c++ code).  Note, you need to name the main file muon1.cc (or .cpp) so gcc knows it's a c++ source file.  It's easier to run gcc than g++.

Good Luck!!!
Pascal
2002-09-02 03:28:09
As I see in the stats and in my personally results, the yield is still increasing continuously.  So I think it was generally a good thing to put Stephen's manually edited results into the TOP.  Or how do yout think about that?
So everything is going on and working fine. 

It would be nice to know when there will be any version 4.22 and perhaps, when there is a client in version 5 available. 

Will we use the best results of version 4.2x for the client version 5?

I would like to tell something more about the future of the project.  wink

Thanks a lot.

___________________________
Member of www.rechenkraft.net - German Website about Distributed Computing Projects

1: Athlon TB-C, 1.2 GC/s, 256 MB DDR-RAM, Erazor x², ADSL-Flatrate, NIC Intel, Win 98 SE Mainboard MSI-6380 Rev.  1
2: Pentium III, 600 MC/s, 256 MB RAM, NIC Intel, Win 98 SE
Stephen Brooks
2002-09-03 07:18:53
I think the addition of my manual result to the project certainly helped, and it appears now as if it helped in a non-trivial way.  I wouldn't like to say anything too conclusive as-yet, but the scores appear to be rather higher than I would expect as just statistical variations of my "best" result.  So it probably isn't over yet.

Version 4.22 has a lot of bugs fixed in it already.  I'm just stuck in a rather difficult piece of programming right now, which I think I more-or-less _have_ to do in order to make the program general enough to do more complex simulations.


"As every 11-year-old kid knows, if you concentrate enough Van-der-Graff generators and expensive special effects in one place, you create a spiral space-time whirly thing, AND an interesting plotline"
: contact : - - -
E-mail: sbstrudel characterstephenbrooks.org
Yahoo: scrutney_mallard
Jabber: stephenbrooksstrudel characterjabber.org
Twitter: stephenjbrooks

Site has had 15715942 accesses.