09541: kov, kovplus: The speech at the title screen cuts out

ID	Category [?]	Severity [?]	Reproducibility	Date Submitted	Last Update
09541	Sound	Minor	Always	3 days ago	13 hours ago

Tester	Robbbert	View Status	Public	Platform	MAME (Self-compiled)
Assigned To	jkm900	Resolution	Fixed	OS	Windows (x64)
Status [?]	Resolved			Driver	igs/pgm.cpp
Version	0.289	Fixed in Version	0.290GIT	Build	x64
		Fixed in Git Commit	40196c9	Github Pull Request #	#16036

Summary	09541: kov, kovplus: The speech at the title screen cuts out
Description	At the title screen, it should speak "The Knights of Valor", but now it just says "The Knights".
Steps To Reproduce	>mame kov (or mame kovplus) Let it run through, until the title screen appears and listen to what the voice says.
Additional Information	Original report: https://forum.mamedev.org/viewtopic.php?t=1167
Github Commit

Flags	Regression
Regression Version	0.289
Affected Sets / Systems	kov, kovplus

Attached Files

No.24661 Mamesick Senior Tester 3 days ago	Regression is clearly this commit: https://github.com/mamedev/mame/commit/60672585a12e6dab7816f50cceecf319248bc25c In details, the portion of code from line 436 to line 450. Reverting only that group of lines seems to fix the issue. Obviously, the original committer have to find the correct fix, reverting is not the solution.
No.24663 Robbbert Moderator 2 days ago edited on: 2 days ago	I noticed in hbmame, the game pgmfrog started out with sound then it just cut out and never came back. So I applied the suggested "fix" and the sound worked correctly. Not prepared to wait for who knows how long for the proper fix. If people actually tested stuff then this problem wouldn't have happened.
No.24664 JimCarlTay Senior Tester 1 day ago	As of commit 40196c9, the code that caused this bug has been commented out. I'm also not sure if an actual fix to this bug would be made, but I guess that would do for now.
No.24665 Robbbert Moderator 13 hours ago edited on: 13 hours ago	PGMFrog still cuts out sometimes, so there's still issues. Switching to 0.288 didn't fix it, but it works correctly in 0.245, so this looks like a separate regression from a long time ago. However, the subject of this report has been fixed which is all that matters here.