This topic is for the purpose of discussing #6664: janus's Genesis Phantasy Star IV in 1:25:35.45

How does the math work? -1360 to skip a fight but +1530 to do the levelling that skipping that fight missed out on would be a 170 frame loss. Are the numbers correct?

Yes, the NET loss was 170 frames, but since I had saved 1360 after Igglanova, that truly equals 1530

That's what I mean - if the Igglanova skip loses 170 frames because you have to make it up by levelling later, wouldn't a TAS where you don't do Igglanova skip be 170 frames faster than this one? If these two events allow for later timesaves to be bigger, it's not clear from the description.

That delay would have been longer, since in the regular run I don't sleep in Kadary (I only stop by to warp later)

Where can I see an encode? Also is there any reason you use the 2nd controller for Z button presses?

It must me somehow linked to the C key I press, because I don't actively press it

Ok, because I was trying to see if I could get this synced on bizhawk, but because bizhawk doesn’t allow different combinations of controllers (like for example this movie uses 3 button for the first player and 6 button for the 2nd player), bizhawk isn’t able to import it.

Besides, I wanted to compare apples with apples. Bizhawk has so much lag that it ended up being significantly slower

As a note. Emulation differences are not accounted for when a run is compared to another emulator. Unless it's blatantly obvious your input is suboptimal and you need to use what you can just to show it. So the justification of "cause lag" isn't a justification.

BUt then how can I know how much faster/slower I am if I compare Gens 11a with Bizhawk?

I will admit that it can be tricky to know for sure how much time you’re saving with improved emulation, and it really depends on the game in most cases. One really simple way is to subtract the amount of lag frames in the game, this can be from taking the total lag frames and subtract it from the total frame count. However this may not always be accurate since some games may handle lag frames differently. A more consistent method would be by comparing each stage or area by their frame counts, and ignore loading areas. A good example from one of my submissions was the judgement from my Splatterhouse 2 submission last year: http://tasvideos.org/6338S.html. Here, slamo created a chart that had a standard start time for each stage. This could be something like the first instance of movement from the character, or as soon as the screen fully fades in. The other thing that can help is if you have a memory address that represents in game time, and it stays consistent even if the game has lag in it, that can usually be the best option to use as well. But that’s assuming the game you’re doing has that. One thing I personally like to do is to import movies into the emulator I work with, and see if I can get it to sync and compare my wips with that. However I cant guarantee this to work every time since some games may have a varying amount of emulation differences, especially if RNG changes because of it. So in general there are ways to compare movies that also removes emulation differences, and I’ll admit they aren’t always the most convenient to use. Personally I feel like I’m pretty experienced with movie comparisons in regards to emulation, so I’m used to it, but I understand why most people aren’t. If you are unsure if you’re saving time over an older movie because you aren’t sure if it’s just an emulation difference, there’s people on the site that are willing to help you figure it out. Sorry for the wall of text, I just have alot to say about this topic. EDIT: That being said, at the end of the day, Gens is still an accepted emulator. People are still free to use it if they wish. It's definitely not the worst emulator the site that's still being accepted, but that's also not saying very much. If you wish to use it, then that is your choice. People just generally don't like Gens due to how incredibly inaccurate and outdated it is.

Should I ever do an original run again, I will switch to Bizhawk since it seems to be the norm. But for reruns - unless they become obsolete - I will use the one I used before