Many games seem to require different levels of overcropping. There is, however, seemingly no standard level of overcropping. Several NES games need the overscan to be cropped to look proper. Note the blank blue area to the left and the green garbage on the right. Version 2.2.2 is promised to have brought improvements from Bizhawk, a more accurate emulator, but it is unknown how much is fixed.Įxample of faulty visuals that are exposed due to crop overscan not working. FCEUX scores fairly low in the same tests, despite being a recommended emulator on TASVideos. The New PPU is more accurate than the Old PPU, however.Nestopia has a high ranking in those same tests.However, according to the same tests, Mesen is actually slightly ahead of puNES, with a score of 98.1%.So just because an emulator has high ratings doesn't mean it is free of issues. These test ROMs used for determining accuracy in tests, unfortunately, don't test everything that matters for accurate emulation (and in some cases, test stuff that doesn't matter for emulating any games). PuNES is the most accurate NES emulator according to tests.That said Mesen does have a strong set feature in other areas, so if people want to try it they could definitely do a lot worse.*AoEX is based on NesterJ 1.12 Plus 0.61 RM, so it includes features like rewind mode, cheat codes support, rotated/mirrored screen, sepia palette, support to rare mappers (the pirate bootleg FF7 works on it), etc, but its compatibility is inferior to 1.13 beta 2. Things are a bit less exciting when it comes to a race as to who can pass the most test roms. I know you've put in a lot of work developing your emulator, but your post does kind of come across as a bit self serving. Higan had a bunch of those, like Speedy Gonzalez which failed on every other emulator because of some weird timing quirk. It would probably be more meaningful to people if you could demonstrate where these results show up as tangible differences in situations they're likely to encounter. Basically if I run them on my NES with just a standard controller connected to a LCD, my NES is going to fail a bunch of them as well. It's a more glaring problem in the TASvideos chart than yours, but still. I don't think some categories of tests should be included, like the ones that require specific controllers or a CRT since they tend to skew results so much. It's worth pointing out that there are loads of different configurations of PCBs and HDB will produce different results on every one of them. If you treat it as a test rom it becomes more of a "how good does this emulator implement the iNES specification" test than how accurate the base system itself is. I don't know how I feel about Holy Diver Batman, since it's more of a board diagnostic than a test rom. The TASvideos chart seems to reinforce that idea. How reliable are the tests? Are they all verified to consistently pass on a real NES? IIRC, blargg wrote most of them and while they all passed on his NES, some of them have sometimes been a bit erratic on others. These tests are more about emulating very specific details/quirks of the NES - the majority of which have absolutely no impact on any game) FCEUX, despite its relatively low score, is able to play virtually all games without any issue. For anyone that might be interested, here's a spreadsheet listing all the tests used & the results for each emulator: Įdit: As others have noted (and I should have mentioned this in my initial post), a good number of these tests do not necessarily have much impact when it comes to playing games. I may have made a few mistakes along the way - but there shouldn't be anything major wrong with the results. I tested 177 roms (almost everything on TASVideos + maybe 40-60 or so extra tests) and here are the overall test results: I have consulted the source code of every single emulator on this list while developing Mesen to help me fix countless bugs and am immensely grateful to their authors! Just to be clear: I'm not trying to hate on any other emulator. Wanting to see how Mesen really compares to other popular/highly accurate NES emulators, I just spent the last 3 or so hours running the majority of the tests I know of on 4 other NES emulators to see how they do. I've always been somewhat unsatisfied with the NES emulator accuracy comparison grid on TASVideos because it's missing a fair number of tests, uses some outdated versions of tests, etc. To mods: Please reject/delete this if it isn't appropriate in any sort of way.
0 Comments
Leave a Reply. |
Details
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |