Posts for CasualPokePlayer


Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
Go to RetroAchievements -> View Game Info and what does it show when you have the game loaded up.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
MAME emulation will not be modified, if there is an emulation issue that occurs for both the BizHawk MAME core and upstream MAME then there is nothing for us to do other than wait for MAME devs to fix it. Your suggestions on how to fix it may as well just come from you actually applying the fix, since none of the BizHawk devs would have any clue about the causes of these issues as we are not writing the emulation core, if you cannot do that then you cannot expect us to do so instead with even less knowledge.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
savestate.save and savestate.load do what you'd expect. Something going wrong is more likely just indicating a logic error on your part, not anything to blame on the savestate functions themselves.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
It is not like dual core. It is internal multithreading which does not cause any sync issues.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
I decided to take a look at this TAS and see how well it would sync for me. It did not go well: 1. Off the bat the game would crash on startup with
The type initializer for 'System.ConsoleDriver' threw an exception. ---> System.Exception: Magic number is wrong: 542
This is fixable (at least for me) by doing
export TERM=xterm
2. The game appears to softlock on startup. It appears to be time and thread related, if I set it to uncontrolled time, the game starts up as expected. This appears to happen regardless of the wait timeout choosen (using Infinite Waits as specified in the annotations doesn't help it seems). Starting the game up attached to gdb reveals many threads spawning, and more interestingly, two threads seem to be constantly waiting on each other (and the game appears to be softlocked at this point). Trying the game on the latest libTAS master does allow it to avoid this softlock, but it just desyncs quickly, and doesn't even appear to be consistent desyncs. I would fear this TAS is going to be rejected just due to sync issues.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
Thornstrom wrote:
it keeps shutting down without error message.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
Please try the latest dev build here: https://gitlab.com/TASVideos/BizHawk/-/jobs/4144670975/artifacts/download It's very likely these issues have already been resolved.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
LoadLibrary failure either means the dolphin dll in the dll folder is just missing (bad download?), or a dependency for that dll is missing. Dolphin requires the 2022 visual c++ redist package, which you can get here https://aka.ms/vs/17/release/vc_redist.x64.exe Note too that Dolphin dropped support for Windows 7-8.1 officially some time ago so the core might just not work for those OSes. Also, the firmware manager is intended for official firmware, not dolphin's free alternatives (which are packaged in hawk already). You do not need to set any firmware for 99.9% of games.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
If it's doing that, then that's what the controller itself is reporting. Emulators would not be intentionally doing that.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
I don't know what you are referring to exactly, but emulators should identify the direction for your *host* controller correctly. However, not all consoles actually follow this, so it may be the case that you have "inverted controls" due to this. Emulators may also allow inverting your axis to account for this.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
Closing BizHawk would automatically flush SaveRAM itself. If it doesn't, that's a bug. Also, you say "whenever the user saves in-game" yet also bring in Pokemon. If you consider the first and second generations which were on the GB/C, what is saving in-game in the perspective of the emulator? It can possibly see that SRAM was written to, but those games use SRAM as scratchpad too (along with many other games), and the perspective of the emulator cannot distinguish between that and actual saving. What you ask is simply not possible unless you want it to just write the current SRAM to disk potentially every single frame (or at the very least way too often).
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
YoshiRulz wrote:
All available options for a Libretro core are in Libretro > Settings.
Libretro settings are not actually implemented (they're forced to all default internally)
+RyeGuy+ wrote:
Basically, I want to make LibRetro core's controls to be the same as the built in Genesis Plus GX control scheme.
That's simply impossible for a libretro core unless you manually remap the controls. Is there a reason you're using libretro's gpgx instead of our own.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
Who cares that there's a 12 hour limit on videos, just split it up. That's been done before on publications, it wouldn't be the first. (Splitting also likely be inevitable for publishers regardless as you'd need a lot of space to store the entire encode in one sitting, the contents of the video don't mix well with video compression on that note)
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
melonDS does not have any GBA emulation period. It emulates connecting a GBA cart in the second slot, which BizHawk does support (you'd need to use multidisk bundler of course).
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
There is no guarentee at compatibility when loading a movie from an older BizHawk version.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
Citra is the only worthwhile 3DS emulator period. There are no alternatives.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
For mobile games, there is just no way to TAS them, outside of extremely specific per-game tools which are very unlikey to exist in any case.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
All TASing emulators are on PC. There are not and very likely never will be TASing emulators on mobile (using mobile would be an extremely bad environment for TASing anyways, so not exactly any motivation there).
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
Ah, I suspect you actually did use Octoshock. Nymashock has extra validation checks against incorrect BIOS files and thus rejects such BIOS files. Octoshock does not have these checks, so BizHawk recovers by falling back to Octoshock. Using the correct BIOS files should result in Nymashock being used. Also, your firmware is not bad per se, but it comes from an older revision of a US console than what the PSX cores target, so it is just "unacceptable." It's reported as unknown for the EU slot as this is a US bios not an EU bios.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
What version of BizHawk are you using?
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
Are you using the Nymashock core or the Octoshock core? The latter is known to have desyncs, the former should be desync proof.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
I think there's just some confusion on how the sync setting works, so I'll lay it out here. In 2.7, Use Real BIOS only cares about bios7/bios9, it does not care about firmware.bin. This is intentional (the wording is explicit about referring to *bios* files here intended to exclude not bios files from the scope), this is also ultimately going according towards upstream behavior on this setting. In 2.8, this was changed upstream a bit. Now, if Use Real BIOS is false, then a real firmware.bin will not be loaded, it will always loaded in the generated firmware. This I think was due to using real firmware not making sense without using real bios files, as you can't boot into the firmware without real bios files. If Use Real BIOS is true, then a real firmware.bin is loaded if it is available (i.e. set in the firmware manager), if it is unset, then it won't be loaded. So, the "issue" you brought up for 2.7 simply won't happen again due to this logic change, although 2.8's "issue" still exists. To be clear, melonDS is not the only core which has this "issue" of sync settings not hard mandating a real firmware file be used, various other cores have a similar "issue." This is ultimately "fixed" by just paying attention to set firmwares and the firmwares reported in the movie header (which is shown in the movie play dialog). This is as much of an issue ultimately as setting any kind of firmware file but it mismatches the movie, which is ultimately the user's fault and the user is given the needed information on how to solve this. Additionally, in 2.9, if you had a firmware.bin set and subsequently unset, the firmware hash shown in the play movie dialog will appear yellow, warning you of a mismatch (this logic also applies to changing the firmware.bin to some other file and subsequently having a mismatch with the one the movie was recorded with). For the case you brought up, such a warning won't be possible as a firmware hash simply won't be placed in the movie header, so there's no way to automatically detect a mismatch here, but here again it's the responsibility of the user to figure this out.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
2.9 may see some performance improvements, but even then you will need a beefy PC to have a chance at running at full speed maybe. The core is very intensive, and even more so with angrylion placed in.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
The sentence can work both ways, you might read out the sentence thinking the verb connects to games, but you might also interpret it as the verb connecting to "one of" as in any particular one of those games, therefore singular.
Emulator Coder, Experienced Forum User, Judge, Published Author, Experienced player (609)
Joined: 2/26/2020
Posts: 698
Location: California
No, and in the future if/when it does support it it will not be connecting to other mGBA processes, it will be like GB link cores which have both emulated consoles running side by side in the same process.