Hello there!

I'm the mod on another Super FX game called Vortex and was running some specific emulation tests and attract mode tests for it. Attract Mode tests consist of running a few common emulators and methods of play and comparing them by syncing up a portion of the title screen and allowing them to run over the course of an hour.

For those that don't know, "Attract Mode" is a piece of the title that shows gameplay or impressive elements of the game to "attract" folks. Commonly it was used to entice folks to play arcade games. but most early console games and SNES games have them too.

For kicks, I decided to run the same Attract Mode tests with the rest of the Super FX games I have access to. One of them is Star Fox.

Here's the raw footage:

Basically I'm seeing how far out of sync the different options are after an hour.

Keep in mind, while the attract mode tests are a good ballpark as to what might be accurate, there's a lot of other factors too. It might give folks an idea for other emulation options to put through paces and eventually allow though.

Based on the video, it appears there's a:

15 second difference between BSNES, the most accurate emulator in terms of timing.

Super NT and SNES classic are slower, but considering the clock difference,(HDMI is 60fps, not the SNES 60.0988fps) it makes sense.

SD2SNES/FXPAK seems rather accurate to hardware for Star Fox, or at least a bit more accurate than the Switch.

Switch appears to be a statistically insignificant bit faster than hardware.

I will be following up on some other emulation options including Ares, Higan, and MiSTER options some time in the future and will update then.

I've received a few private messages asking if I tested the differences between console revisions. I only own two and can't really be very scientific about it, but I do have a 1chip and non-1chip.

Here's the footage:

It's statistically insignificant, but there appears to be an incredibly slight speed difference for Revision 0, or 1.0 carts (the ones where the Star Fox logo is noticeably more red) on non-1chip. No difference on the later revision in attract tests. That doesn't mean there couldn't be differences later on, but I suspect the console revision wouldn't matter enough either way.