@Pronssiparta - As a mod, I feel like you should be part of this discussion.

It was brought to my attention that RetroArch should be looked into as for providing an unfair advantage VS console runners.

Regrettably, I am not a subject matter expert but I do believe simple testing can yield results, so see below:

Using the same computer/settings, I've recorded 15 Loop Testing which is about 21 minutes of time (runs on this board should be between 15-20 minutes) to give us as realistic of an idea as possible over a real run.

Each loop is 5 of each character. Start and end of loop are when the Yin and Yang Logo fades out from the title screen. My conclusion is that RetroArch actually emulates the game SLOWER than console and therefore provide no advantage to runners, in this game.

See tests in the next post.

Video URL: CONSOLE TEST DETAILS: Console: Super Famicom Cart: SD2SNES Capture Card: GV-USB2 Capture Software: AmarecTV

Issues: Apologies on static in audio, I haven't had the time to fix it at the time of this request. Ironically, the person requesting this has been annoyed as of late with my audio, this confirms that I recorded this myself. :P Apologies on main timer at the bottom hidden for first 8 loops, crop resolved thereafter.

Results: Total Time: 21:00.94

(Splits averages for fun only) Rocky Loop 1: 1:29.65 Rocky Loop 2: 1.29.86 Rocky Loop 3: 1:29.90 Rocky Loop 4: 1:29.94 Rocky Loop 5: 1:29.81 Rocky Average: 1:29.832 (Safe to say I early split Loop 1 by about 0.2s but for sake of testing, kept it as is).

Colt Loop1: 1:38.90 Colt Loop 2: 1:38.88 Colt Loop 3: 1:1:38.89 Colt Loop 4: 1:38.84 Colt Loop 5: 1:38.86 Colt Average: 1:38.874

Tum Tum Loop 1: 1:03.52 Tum Tum Loop 2: 1:03.45 Tum Tum Loop 3: 1:03.89 Tum Tum Loop 4: 1:03.53 Tum Tum Loop 5: 1:03.43 Tum Tum Average: 1:03.564 (Safe to say I split late on Loop 3, but for sake of testing, kept it as is).

Video URL: EMULATOR TEST DETAILS: Emulator: RetroArch ver 1.8.9 Core: Snes9x - Current (1.60)

Issues: Apologies on the last split being named "Loop 5 Tum tum" when it should've been "Loop 15 Tum Tum"

Results: Total Time: 21:04.41

(Splits averages for fun only) Rocky Loop 1: 1:30.17 Rocky Loop 2: 1.30.13 Rocky Loop 3: 1:30.17 Rocky Loop 4: 1:30.21 Rocky Loop 5: 1:30.09 Rocky Average: 1:30.77

Colt Loop1: 1:39.30 Colt Loop 2: 1:39.10 Colt Loop 3: 1:1:39.10 Colt Loop 4: 1:39.14 Colt Loop 5: 1:39.16 Colt Average: 1:39.16 (Safe to say I split late on Loop 1, but for sake of testing, kept it as is).

Tum Tum Loop 1: 1:03.43 Tum Tum Loop 2: 1:03.62 Tum Tum Loop 3: 1:03.60 Tum Tum Loop 4: 1:03.48 Tum Tum Loop 5: 1:03.65 Tum Tum Average: 1:03.556

Video URL: Result: 21:04.30 I wanted to test this a little further(without providing all the above self calculated details, feel free to math yourself) . I wanted to know just how much of that was a result of RetroArch so I did the same test on Snes9x ver 1.60 standalone and 0.11. Considering human error during start and end time splits I attribute none of this to RetroArch.

I spent all morning on this so I am tired and hungry but when I find more time an energy I will test Snes9x v.1.53 whcih is the version that has the most rugged testing done to, and I'll test a completely different emulator with high benchmarks (such as Bizhawk + something core).