Hi
In 1.4.1 (including latest nightly), my keyboard media keys for volume control do weird things in RA. Afaict theyre not assigned to anything in RA, so they should let me control sound voluem right ? But vol up is taking a screenshot (key for that is set to “NULL” in cfg), vol down (NULL as well) select next shader preset…
Thanks
I am running latest RetroArch 1.4.1 on Windows 10 64-bit and I can’t load any content, it just says failed to load file: invalid argument. 1.3.6 simply crashes for me when loadng contents. Cores load and work just fine. Any ideas?
The current MAME mainline and MAME 2014 cores are immediately crashing for me after loading any rom, with 1.41.
All other cores I’ve tested are working correctly (FBA, Nestopia, Snes9x, Genesis GX, Mednafen PSX, PC Engine, etc.). Debug log looks normal to me.
Definitely using the correct romsets (as well as trying multiple others). Tried nightly1.41 and stable 1.40 with the same results… MAME cores used to work fine on this machine until recently.
Windows 7 32-bit, Core i5, Intel HD Graphics.
mame and mame 2014 should be working from last commits, since im using them a lot lately for some other tests. probably clear up you cfg files, some remaps etc.
else post logs so other can see what is going on…
(and i am using a lower gpu version than you so gpu should not be an issue with these cores)
Hi, I have downloaded a full fresh nightly 1.41 + current cores, using default .cfg, and still run into crashing with MAME (same roms load perfectly with FBA).
Windows error reporting shows the MAME core as the source of the fault (e.g. mame2014_libretro.dll)
Here is the RetroArch log:
my mistake… mame2014 seems to be broken… try using ume2014 instead if you really want to use the 2014 series of mame, else the newer mame_libretro(mame without number) which is version 0.182 should also work.
EDIT: turns out i was using an older build of mame2014 core from a previous test. you should probably redownload the core and make sure it did download correctly or just manually download it from buildbot repo.
@Wertz Thank you for testing on your side.
I have downloaded the latest MAME cores (2014 and 1.82) via updater and also manually from the repo, and I’m still experiencing a crash on rom loading.
Not sure what else I can try at this point, as I mentioned MAME was working fine on my machine until recently, and all the other cores are working fine for me 
test with ume2014, it should still work (thats mess+mame for 2014) if for some reasons that still will not work, something else is wrong
btw, what is the current processor(or cpu) you are using? the drivers are quite old( gawd i hope this is not pre-core series cpu)
else, try updating gpu drivers.
Tried UME2014, it crashes as well.
I wonder in what way the MAME related cores are different from all the other cores? Especially considering the latest FBA runs perfectly for me.
The cpu is core i5 2.67 Ghz, so there should be no issue there. Intel GPU drivers are the last released for my system; they are older, but have been running MAME just fine up until now.
ok if this is i5, then i dont think this is the latest driver version:
RetroArch [INFO] :: [GL]: Vendor: Intel, Renderer: Intel® HD Graphics. RetroArch [INFO] :: [GL]: Version: 2.1.0 - Build 8.15.10.2622.
get new ones directly from Intel and make sure its for the right microarchitecture (sandy bridge, ivy bridge, etc etc).
report back after driver update with new logs but use retroarch_debug instead of regular retroarch
@wertz I did try updating with drivers directly from Intel’s site which were slightly newer (but still 4 years old, my core i5 is 1st gen mobile), and still no luck.
Some users report. Crash on windows 32bit for mame 0.182 maybe the pb is also on 2014.But i don’t have a 32b versio to test and debug.so i need a proper debug log (i mean debuger backtrace) before i can look at.
I’m on Windows 7 32 bit, and I do get the crash in MAME 0.182 as well.
I see enough information here to believe that this is an actual issue and not just something wrong with my machine, so for now I’ll just wait to see if this gets resolved.
I’d be willing to help test if that would be useful.
Please, please change the Mouse-Cursor-behavior of retroarch! If I come from launchbox, the cursor is every time visible in the game. Only if I press the Hide-Cursor-Hot-Key (F11), the mouse cursor disappears. It annoys the shit out of me - every time.
I had the exact problem and made a thread about this some time ago, bumping it for some time. So i don’t think it’s a priority for the devs.
Your only chance to get rid of this is using RocketLauncher. Instead of loading the cores directly from launchbox, you can set them up in Rocketlauncher and then load that from Launchbox instead.
So instead of Launchbox - Cores, you add one more link in the chain so it becomes Launcbox - Rocketlauncher - Cores.
That’s because Rocketlauncher has a “hide mouse cursor” option. That’s how i did it and i got rid of the cursor.
Honestly, i don’t understand how this isn’t an issue for RetroArch itself. It should be easy to fix but then again, what do i know.
I’m not experiencing any mouse pointer issues on the current build for Windows 64-bit. There was an issue a while back with the system mouse pointer appearing in fullscreen but Twinaphex pushed a fix for that and I have not seen it happen since then.
I still have this problem but not with all cores. MESS 2015 and a MAME core from around the same period have this problem though. I’m not sure about newer or older versions. So it depends on the cores or their version. I got it fixed with Rocketlauncher.
Mouse pointer is visible only if you enable and use overlays.
Nope, it’s visible even when they are disabled in those particular cores.
I don’t use overlays…