Viewing Issue Advanced Details
ID Category [?] Severity [?] Reproducibility Date Submitted Last Update
01992 Crash/Freeze Critical (emulation) Always Jul 8, 2008, 21:22 Apr 23, 2009, 18:49
Tester t800 View Status Public Platform MAME (Official Binary)
Assigned To robiza Resolution Fixed OS
Status [?] Resolved Driver
Version 0.126 Fixed in Version 0.127 Build I686
Fixed in Git Commit Github Pull Request #
Summary 01992: timecrisa, timecris: SubCPU start wait fails test
Description When starting this game, the SUBCPU START WAIT test fails every time (as shown in screenshot) therefore not allowing the game to start.

This behaviour is not shown in MAME 0.125 using the same romset, where the game starts as expected.
Steps To Reproduce
Additional Information
Github Commit
Flags
Regression Version
Affected Sets / Systems timecrisa, timecris
Attached Files
jpg file icon Error.jpg (70,902 bytes) Jul 8, 2008, 21:22
Relationships
related to 02654Resolvedrobiza  timecris, timecrisa: Unable to start game due to improper initialization 
Notes
11
User avatar
No.01553
Tafoid
Administrator
Jul 8, 2008, 21:36
Confirmed and added actual regression version.
User avatar
No.01844
john_iv
Senior Tester
Jul 30, 2008, 06:37
Doesn't happen on first run, but happens once *.nv file has been written.
User avatar
No.02075
tititou
Tester
Aug 15, 2008, 18:39
edited on: Aug 16, 2008, 07:10
The regression seems to be caused by the patch against emu/cpuexec.c introduced in 125u3.

In this patch the value in cpu[cpunum].iloops can be forced to 0.

In several places in namcos22.c, a test is made against 0 is made via cpu_getiloops() which reads this particular variable. I guess that the change of behaviour introduced in cpuexec.c doesn't match well with these uses.

I have made some hack to namcos22.c. I have something that works on 125u3 but not on 126u5. If interested, ask me.
User avatar
No.02088
tititou
Tester
Aug 16, 2008, 21:35
edited on: Aug 16, 2008, 21:38
A made some additional testing with my ugly patch on namcos22.c

The results are :
   - 0126, 0126u1, 0126u2 : seems also to be OK
    - 0126u3, 0126u4 : games seems to run (at least, sound is there) but there is no more graphics. Then only display is the blue cross where you shoot. More over, when moving this cross, the former display does not disappear
   - 0126u5 : nothing anymore. No graphics, no music, no blue cross

Hope this helps...
User avatar
No.02098
robiza
Developer
Aug 17, 2008, 18:52
the regression version i'snt 0.125u3

with patch for cpu_exec in 125u3 the game works if nvram is not present
but with no patch the game not works completly

yet now is fixed
No.02446
user413
Sep 13, 2008, 21:45
.127 fixes this problem, the first two times I played it were ok but all future runs (even after deleting the .nv file several times) fail the test explained in this bug. Why does this happen?
User avatar
No.02447
Tafoid
Administrator
Sep 13, 2008, 22:02
Tap F2 (Service Mode) button when you get the error on screen, then hit F3 (Reset) - the testing will restart then you will be put into Service Menu. simply hit F2 again when there and a proper NVRAM will be written. Subsequent MAME starts should work fine.
No.02449
user413
Sep 14, 2008, 01:40
edited on: Sep 14, 2008, 01:48
Thanks! Hope that works! I appreciate your help.
User avatar
No.03057
machrider
Tester
Nov 17, 2008, 22:41
I'm getting this issue once again in version 0.128u3 of Mame. I tried F2 and F3 yet it does nothing.
User avatar
No.03058
tititou
Tester
Nov 18, 2008, 07:43
Same for me, removing the .nv file doesn't seem to help either.
User avatar
No.03059
Tafoid
Administrator
Nov 18, 2008, 16:52
I'll set this up as a new bug - the results appears the same, the bug I believe is different.