help desk requested post here for developer

[Lobster]

ORIGINAL: fastfrank

Just had a reply from the help desk, they have asked the "producer" to try and contact the "developer" to review this string and try to debug. Not sure how this all works, not sure if the developer is also the programmer or whether the developer is the designer and will need to then contact the programmer.... I'm trying to keep the help desk in the loop since they are contactable and seem to accept some ownership to move this process forward. Help desk has asked the developer to contact us via this forum string.

Good luck with that. Ralph is the developer is a designer is the programmer. Bob is the production manager is a designer and is apparently the only one currently actively doing anything related to programming. That is the entire 'team'. Efforts by the community to expand this team have been fruitless.

[fastfrank]

Just what I feared. When I was young and very junior, I had responsibility to do on the fly s/w maintenance when stuff blew up, started looping, etc. Hated trying to debug someone else's code. I'm relieved Shadrach has identified a useful workaround. Is Matrix starved for human or financial resources, or are available assets drawn to the next project planned for release? I would expect most players would place major importance on reliable games, a buggy game wouldn't be worth much.

[Lobster]

We are hoping a patch will be done soon. It's basically the fault of the developer. We have people willing to help but as far as I know the developer needs to approve anything like that. I sometimes think we are the red headed stepchild.

[fastfrank]

I would be happy to volunteer

[Blond_Knight]

Well remember that even if people offer to help there might be more effort involved in that then fixing something themselves. There's NDA's to deal with, then youve got someone looking at the games entire code, which after all these years may look like a mess. Then questions to the developer like "Where's the section you want me to work on?", "How do I compile this to test?", or "Does bool PissOffTheLobster = true; do anything anymore?"
OK so that was just for fun but you get the idea.

On the other hand Ive reach out to Tamas to see if I can do anything and spent the last few days practicing C++.

[cathar1244]

If I turned off "Windows Fonts" in the game for the Huge fonts, the issue was gone, but the fixed-size fonts are basically unreadable...

Check, I turned those off and also saw the issue gone. You are correct, it is not a fun font to look at.

What bothers me about this is why some maps appear more affected than others. Even FITE and Campaign for South Vietnam have a fair amount of placenames on their maps, yet their memory use seems more bounded.

Cheers

[Shadrach]

ORIGINAL: cathar1244
What bothers me about this is why some maps appear more affected than others. Even FITE and Campaign for South Vietnam have a fair amount of placenames on their maps, yet their memory use seems more bounded.

I think maybe it's related to the concentration of place names in one map screen that gets loaded into memory as it scrolls into view. While other scenarios might have a lot of places, they are much farther apart. The France 1944 scenarios, especially the D-Day one, have such a mass of places on the map I'm asking myself why the designer even though it a good idea to have that amount of named urban tiles, even without knowing about any memory issues, it just looks a mess.

In other scenarios the memory growth might be so small as to be negligible, but it's still there. I just tested with "Sicily to Brenner Pass" and on the northern Italian plains with its many cities there was a noticeable rise. Not as big as France'44 of course, that one is special [:D]

Also would be interesting to know if anyone can reproduce this on Win7/8 or Vista. There have been big improvements to DPI scaling for instance in Windows 10 and so the problem might be limited to Windows 10 (then again, I doubt it).

[sPzAbt653]

would be interesting to know if anyone can reproduce this on Win7/8 or Vista.

I've never had this issue here with W7. I've even run 197 turns of CvC D21 nonstop with no crash.

[Shadrach]

ORIGINAL: sPzAbt653
I've never had this issue here with W7. I've even run 197 turns of CvC D21 nonstop with no crash.

OK, would you mind testing with France 1944 D-Day, with running Process Explorer and looking at the process Performance Graph for "Toaw 4.exe"? Just moving the map around the urban areas, you don't need to move units, it should show a huge increase in memory use.

https://docs.microsoft.com/en-us/sysint ... s-explorer

If it doesn't happen in 7 it would explain something, but not sure what [:D]

[sPzAbt653]

Does this little fella show us what we need, or do I have to get the Process Explorer?

[Shadrach]

ORIGINAL: sPzAbt653
Does this little fella show us what we need, or do I have to get the Process Explorer?

Ah the good ol' Windows 7 task manager, simple but powerful [:D]

Yeah, I guess - if that graph was made during running the game and moving the map around on the France 1944 D-Day scenario for a couple of minutes or until it crashes, whatever comes first.

If so it doesn't look like there's a leak in Windows 7, which is interesting.

It would be better of course if there was a way to show mem usage specifically for TOAW4 graphed, but it'll do.

[sPzAbt653]

Ok, I scrolled all over the map with that scenario and never had an issue. The CPU Usage hit 25% a couple times. I finally got tired of scrolling so took this screen shot, if it helps any.

[Shadrach]

Thanks! At least it show that Windows 7 doesn't have the leak, not sure how much that's going to help us but might help the developer, if they're even reading this [:D]

[sPzAbt653]

if they're even reading this

'They' read everything, nothing goes undetected.

[Curtis Lemay]

Can't the game be run under Win 10 with Win 7 emulation selected?

[Lobster]

January 14, 2020 Windows 7 individual support ends. Support will only be available at a cost for businesses and that cost will increase annually.

[Shadrach]

ORIGINAL: Curtis Lemay
Can't the game be run under Win 10 with Win 7 emulation selected?

I've tried setting "Opart 4.exe" to compatibility mode for Windows 7, and it didn't change anything.

Probably, if I could be bothered to install a virtual machine with Win7 it would work, but that would be kind of besides the point wouldn't it [:)]

[larryfulkerson]

I'm not having any problems with my WIN10 setup by running it as if it were WIN8.

[Shadrach]

ORIGINAL: larryfulkerson
I'm not having any problems with my WIN10 setup by running it as if it were WIN8.

This memory leak followed by a crash issue seems to happens only on certain scenarios where there's a lot of urban tiles, and hence, many place names on the map that gets loaded into memory on map scroll. Usually it will trigger a crash during the PO turn, as the map will move around a lot, and at start of a new turn, memory will be cleared again. So for a lot of cases, the user will never see this crash.

Out of interest, I've tried every single setting under the Compatibility tab. Even Windows '95 compatibility mode - game won't even start.

Are you able to run the same test as others have done above, preferably with Process Explorer (as I'm not sure Task Manager shows enough details)? Scroll around the France '44 map until the game crashes (or a couple of minutes) and paste a screenshot of the Performance Graph tab from the "Opart 4.exe" properties page?

In the end, a developer with Windows 10 needs to recreate this to have any hope of fixing it, anything else will be workarounds until the issue is fixed in code.
Unless of course, the bug is in Windows 10 itself... in which case we just have to wait for a fix from MS. But a developer should be able to answer this as well.

[Curtis Lemay]

ORIGINAL: Shadrach

In the end, a developer with Windows 10 needs to recreate this to have any hope of fixing it, anything else will be workarounds until the issue is fixed in code.

Not sure if we have one of those.