Headrush69

Since that "weird" score was changed, dig dug no longer is adding hi-scores for me now. Beat it 3 times now and no upload. Other games seems to be working ok. (I'm on beta 3 now)

Isn't that the point of the functionality in LB, so we can see it there? A web thread with hi scores is going to get unruly fast, IMHO.

Ahh. I don't think it's a big issue as these defaults will certainly be overwritten fairly quickly after everyone enables this feature and plays a few times.

Testing some games, and in Burgertime you had the only hiscore so far. My score got added, but it copied your 28,000 score, not the 3500 I had.

Do the nvram or the hi files store whether the cheat plugin is enabled? I've never used cheats but it was enabled and even though I have disabled it in mame.ini, my LB logs say cheat enabled. Wondering if old hi score files will need to be flushed?

For what it's worth, I just use MAME for everything and no issues, but It will depend on your hardware. MAME strives for accuracy and FB Neo (FinalBurn) may perform better on lesser hardware. There is some lengthy debates as to whether FB Neo has the advantages over MAME that it did in the past. I'd just say use MAME and if you find a game that absolutely needs FB Neo, just set that game to launch with it instead.

I don't think having a keyboard attached matters as Retroarch doesn't multi-keyboard support by default. (I have one attached with my iPac Ultimate and no issues ever) The auto configs shouldn't be changing keyboard mappings you set in retroarch.cfg So the question is, if you start a system in Retroarch, bring up the Quick Menu and go to Controls and then the Port 2 Controls, are the keyboard mappings you set in retroarch.cfg still showing there or are they blank?

I normally use Altirra for 5200 and can use the XBOX sticks for true analog input, but that hack works with my cabinet controls as well and frankly plays better anyways. Thanks for the heads up. (Just need a hack for centering input for MAME starwars now ? )

Sort of got pushed to the back burner do to other issues, but I haven’t forgotten. You can check out the Mame tester web site and see if anyone else has noticed yet. I will post the issue eventually if no one else has. Edit: Do you know any other games that show the problem? I'm trying to narrow down where the issue is before submitting a bug report, but so far my testing the results are very inconsistent. Edit2: I'm not 100% convinced this is a MAME joystick issue. I was able to get the same issue under specific situations with Kegs32 and GSplus emulators as well. All GS joystick test programs I tried showed analog controls being reported properly. Additionally, the latest MAME versions show real time output when configuring analog controls and they report correct outputs as well, The interesting thing is in Gauntlet, after character selection, there is no joystick drift in game. I tried to find a video of this being played on a real GS for comparison but was unsuccessful. Might fire up Sweet16 (a great GS emu on older OS X versions) to see how it runs there.

Are you indeed using MAME software lists are is your NeoCD media just standalone images?

I've read that sometimes when updating video drivers, the refresh rate can get changed. Maybe try checking that out following this first post: Libretro forum post about syncing

I would suggest installing a clean copy of Retroarch to a different location and see what the default behavior does. At the very least this will tell us if it's a configuration issue somewhere with your current setup, or a fault with core Retroarch. From there we will know what to check first.

Just tested KOF Neowave as well and it runs perfect as well. I only have an i3-7100 and a GTX970 and this is upscaled and with per-pixel alpha sorting too. I'm using the MAME rom set not LST/BIN format. (MAME set is preferred set now) Edit: What options do you have on for this core? Threaded Rendering?

Have no idea what hardware you are running on, shaders you are using, etc, but I use flycast core for all my Sammy games and they run great. Dolphin Blue runs perfectly. Doesn't give us much to go on. Does it start? If not likely a rom issue. If it runs, what happens?

Maybe I'm reading too much into your comment, but I wasn't in anyway making any judgement on your responses or your knowledge of the issue. I just thought it might be helpful to point out with that emulator filenames matter and it wasn't a location issue. Are you suggesting I shouldn't have added anything once solved?

There are inconsistencies in the roms pointed to in his/her screenshots. I don't know but, somehow in the process duplicates of the roms were created (the -1 variants) and that's what LB pointed to. Those will never start using m2emulator as it uses game names, not file names. I doubt it was the emulator location that fixed the issue, but looking at the final screen shots you see the games in LB are linked to the correct game names now. Hopefully that's helpful for someone else in the future.

I notice your file name for this rom doesn't match the normal filename. (daytona-1.zip instead of daytona.zip) Are all your filenames like that? Edit: This emulator uses filenames without extensions, to know which game to start, so if named different you will have issues.

I'm trying to understand everything before posting about the issue on MAME Testers site. Maybe you can verify something else I see. When using -gameio joy and going into Configure Inputs (This game), in the gameio section, the button labels are off. For example player one buttons are labelled P2 B1, P2 B2, for player two they are labelled P3 B1, P3 B2, etc. What I find interesting is I would assume the code for this port is the same when using the apple2ee machine, but it displays correctly there.

Are you testing with more than Gauntlet for the IIGS? I can reproduce that issue using standalone MAME with Gauntlet, but other games seem to be fine. Edit: I ran a joystick test program within the MAME using both an analog joystick and the XBOX 360 analog controller, and values are read properly and consistently with a stick center analog value of 128 with no or little to no drift. So I don't think the issue is a dead zone problem but appears more specific to Gauntlet itself with MAME. (This is using -gameio joy slot option) Edit 2: Tried the same Gauntlet image using GSport emulator and no joystick drift.

Just due to the limitations of using LED light sources, I don't think there is a good substitution for black. I started using the colors.ini file but have since customized every MAME game I have in my system. If a game originally had black buttons, I switched to something else. The colors.ini tries to match the original arcade controls, which I think is OK if you have a single game purpose cabinet, but I personally prefer having different colors per action for a multi game cabinet. If 2 buttons are the same color, the user knows they do the same action. This is real handy when you have lots of buttons and want to give the user many options whether left or right handed. Additionally, I keep coin insert and player starts button colors consistent across games.

You need the neogeo.zip bios file as well. Any time you get a message like this from MAME, the filename in the parentheses is telling you the file where it is expecting those files. So just add that set as well.

Best Low Budget low profile GPUs I had a GT 1030 which I started with that was fine for MAME. Only reason I upgraded was to use with emulators for more advanced systems like CEMU,

What are the specs if the Dell? If using MAME is your main focus, single core speed and and SSD are going to give you the best improvement. In MAME games, my NVidia GT 1030 ran games as well as my NVidia 970 GTX, which is much more powerful. If you plan to use some high end shaders, or higher platform emulators like Cemu, Dolphin, etc, than the GPU will become more of a factor.

It could be specific to my system, I don't use fast-forwarding normally, but when I turned off vsync and tested FF in other cores, I didn't have it working on them either. Edit: Needed to turn Threaded Video on and now it works. I searched for that commit again and it was run-ahead, not fast-forwarding, so maybe it does work. Apparently save states do work according to github page.