Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - Pieter Hulshoff

Pages: 1 2 [3] 4 5 ... 100
I'm a tad confused. You said it would close at random after launching arcade games, but I don't see you actually launching any games in this log file.
I do see RetroFE exiting normally after about 5 seconds; did you press any of the "Q,Escape,joyButton10" (as configured in controls.conf) buttons to do so? If not, might it be that the wiring is a bit wonky, and might be sending any of these signals to the computer?

CoinOPS does also contain a joy2key setup that might be causing this issue. Try running the RetroFE link instead to see if that's the case.

I've already fixed that problem locally, but I'm still looking into a control issue where pressing next/previous game just once (and briefly) will make it scrolls two games ... every time. The only way I can currently go the next game is by pressing next, next, previous, previous in quick succession (it will scroll 3 forward and 2 back...). As soon as I fix that I'll create a new exe for you to test.

Nevermind; it's a problem in the current code. I can reproduce the problem under Linux, so I'll have a look as soon as possible to fix this.

I'd like a small test setup, so I can reproduce the problem. It helps me to solve the problem.

Sent from my SM-G781B using Tapatalk

Can you zip up your test setup, and wetransfer (or something) it to me via PM? I'd like to see what you're seeing so I can analyze the problem.

OK, give this one a try and let me know. This should be used with the OLD core directory, so using the old releases of gstreamer and SDL!

Just post your log.txt; I'll have a look.

I'll see if I can install gstreamer 1.4, compile with that, and run the latest code in the old core directory with new SDL libraries. That should tell us something about the cause at least.

Hmmm, that might be a possible reason. I don't believe the old code stopped the video before deleting it, and the current code does that for over 70 videos in some of the themes. Sometimes it's just a matter of increasing chances when you do the same thing over and over again, each with a small chance of crashing the program. It's a rather old bug though; would that still be part of the current gstreamer code?

The problem doesn't exist on Linux; if it was I would have fixed it a long time ago. It may very well be an issue with one of the plugins actually. I noticed that 1.4.0 used a lot of internal video codecs while the latest releases use separate audio/video plugins.

I'd have to spend some time setting up a separate virtual machine to compile it that way. Gstreamer doesn't seem to be willing to have two versions installed on the system, and so I have to break down my current installation in order to compile it using the old gstreamer version (if I can remember which version I used back then). What doesn't help is that when I compiled this I was still on bitbucket; my old repository doesn't exist anymore.

OK, so so far we have several crashes within the gstreamer library itself, which apparently is an issue under Windows, but not under Linux. Very odd indeed.
We've also got a null point issue in the code; I'll give that a closer look as well. Apparently it managed to create a list without a selected item somehow; perhaps an issue with an empty list?

How are you debugging this btw? It's not like I don't have a Windows setup, but I'm not familiar with debugging under Windows.

Thanks. This appears to be a cross platform issue, but might also explain why some people with multiple windows using a single screen definition were having crash problems. I'll see if I can reproduce and debug the issue. Thanks again for the report.

Looking at your trace, it does appear to be a crash WITHIN the gstreamer library. I've tried upgrading the entire package to 1.20, but that didn't seem to have any impact. I'm not quite sure how to get this fixed though; the old library also had issues with memory leaks after all. I'll give this some more thought.

Pages: 1 2 [3] 4 5 ... 100