Luja wrote:I don´t agree with the removing of the confirmed inps. It's not an easy thing to make these records! :cry:
I hope to figure what´s the problem!
I've download the alphamame62.3 file, but I don´t use the command line version earlier. What exactly have to put to make an inp of the game Zwackery?
Thx
Luja, Remember this isn't *Entirely* a high score site, it's a replay site, if half of the people who try these recordings can't play them back for an unknown reason that's not fair. But i'm not suggesting that yet, since i think eventually we can find something to do to make these playback for everyone with who have the recorded settings. And sorry there isnt a mame32 version of 62, but hopefully the bat script will work at least for a test.
BM, by chance do you have an mame.ini on the Celery machine? And You don't have to check both machines for your recordings. Though be responsible, at least record with 62 if you know the game is one of the elite games that have this problem. So far it's
And You don't have to check both machines for your recordings.
I'm not talking about my own Inps. I mean new recordings that get upped to MARP every day. How are we gonna find out if those games have those problems or not. How many games have been upped already that have the problem and we simply don't know about it? This is really a mess...
yeah for others yeah, i think if you want to check something in playbackability (you're the lucky one who has both configurations) you proly should try both ways. (i was the lucky one when there were problems between xp and 98 since they both playback bad for me when the 98 and xp are on the same machine :).
do you still need me to record a game on AlphaMAME 0.62? Because I only have the DOS version and I'm not in a hurry to download the Windows version on dialup just to record one game. :P (not to mention messing with those controls again)
nah bbh, Luja prety much proved it since i can play his 62 back, when i couldn't do his tgmame69 but thanks though. There's definitley something wrong, it might have something to do with the rdtsc thing but selecting that specific option doesn't appear to help.
I think that the problem is more generic that only with a few games:
I'm almost sure (because I've problems in playback other player's inp) that the following games have the same problem:
* Angel kids
* Pairs
* Dungeon Magic
I add inps of those games and other one so you can test them!!! (witn alphamame32 .69)
Because we don´t known if this is a MAME 69 or ALPHAMAME 69 problem, I upload a Zwackery inp made in MAME69. If this is wrong too, we may report the bug to mametesters
good work bm, gotta get a right click menu thing for playback working to make it smoother, i get 900 on the standard mame one. Luja, try mame63 if you still have the time, maybe we can determine the exact version where this thing started. So far it's fine in mame62 and bad in mame69 (and 67).
***edit: with that post my hot air ownz 8% of all marpirc posts, heh, maybe i'll calm down for a while.***
Buttermaker wrote:Plays back to 80610 on the P4 and to 1610 on the Celery.
So mame62 is fine and mame63 is broken. Is that the case?
Someone should make one last recording in both mame62 and mame63 to verify that this is right. Then we can start looking to see what changed between 62 and 63.
mahlemiut wrote:One thing that I can think of that appeared in 0.63 was the change in timing system. Console version defaults to QueryPerformanceCounter(), MAME32 defaults to RDTSC. RDTSC can be forced with the -rdtsc switch. You can see which is being used by using the -log switch ("Use developers' log" in MAME32), run MAME, and opening error.log.
I think this was also the time they put in the mame_rand function, i would like to inspect this function to see if it has any unportable mechanisms? can anyone find the source of it for me? Just a file and line number would be cool.
