Hi. Today i was doing blsc runs and i tried to beat my flamar time. I did many times in a row but not quite so much to beat Siegel time. At this point i checked his run to see what he was doing better than me and it felt a tiny bit slower compared to mine.. so i put both runs in a video editor and i noticed that when both runs starts at the same time ( https://prnt.sc/22ya2jm ) they desync midway ( https://prnt.sc/22ya70s ) with his run being 0.55 seconds behind on that timing (but on the same section and jump as you can see from the screen) and in the end i finish the run 2 seconds faster than him but his time is still better. With that said i am not sure if i have NTSC or PAL but does it change the timing of time trials too? I would like to try myself on an emulator but i dont know from where to start x)

EDIT: I can also record and post the desync and stuff if you want

Update: Out of curiosity i downloaded a ps1 emulator and used a ps1 iso of buzz i had. Before you say anything i have both a ps1 and i had the original game x) but anyway I tried playing the game.. FPS is capped at 30 or around that and it's quite harder to play but this is my best attempt ( https://prnt.sc/22z5my0 )... yeah it's 0,6s faster than my PC Game best time. I tried to do the level 5 times and my times were all low 57 to 56. I switched right after that to the pc game and my first 5 plays were all around the same 57.60 except this one ( https://prnt.sc/22z6jl9 ) that was a personal best there. Has anybody tried the same thing?

I'd imagine it'd have to do with the fact that PAL versions of PS1 games run at 50 FPS while NTSC versions run at 60 FPS. Is the timer perhaps running slightly slower for Siegel, allowing them to get better in-game times?

Haven't really delved much myself into in-game timers since I've been running mostly Any% full game, which is timed with RTA.

I remember that Mario Kart 64 (i think?) had a similiar thing with times that had to be converted from NTSC to PAL or something like that. At this point i might try all the other levels and see if my times are better overall or just flamar. Will post results soon ^^

Update: I've done some runs on Jo Ad TT and Canis Lunis TT on the PS1 Emulator (All recorded). The results are: Jo Ad ( https://prnt.sc/23184r3 ) and Canis Lunis ( https://prnt.sc/2319b58 )

I've beaten both my PB and almost world record in just few tries even tho on Jo Ad i did a slow start + i bumped on a few turns and on Canis Lunis i hit a rock by mistake. I dont know yet if my pc version is pal or ntsc but if it's pal there are some time difference for sure x)

I don't know about NTSC/PAL (just checked and my pc disc says "Version 1.0/UK"), but I think it might be to do with lag? In my flamar pb (on pc) I was using the Microsoft WARP software rendering, and it was pretty laggy (does that desync line up with a lag spike?) - I wasn't able to beat it once I switched to GPU rendering.

It might be that with lower fps the time goes slower than the gameplay and so you end the level with a faster time even tho it should be around the same

I've been taking a look at the differences between NTSC and PAL on emulator/hardware, and it seems as though everything (except audio) is locked to the framerate - the NTSC version runs 20% faster in real time (compare (PAL, I think it's probably on hardware as the channel only plays PAL games), (PAL/PS3) and (NSTC)) - the same happens on PCSXR. This seems to be how it works for Toy Story 2 also. It looks (I have some recordings of running to the bridge in Jo-Ad 2 for PC/PAL/NTSC comparison) like the internal timer is consistent between NTSC and PAL (just running slower in real time on PAL), so I think the only difference is between PC and PS1 (not sure about dreamcast version but looks like it runs like PS1 from some videos I looked up).

Sorry it took so long, but I finally got it working with the Microsoft WARP driver (using dgvoodoo) with fraps running, and made some recordings.

It seems like normally (for me) the game switches between 60/30/20 FPS and stays stable at those framerates - it seems like the timer is consistent with real time for these framerates (presumably the game runs at the correct speed also). Video:

I haven't been able to record a big lag spike like in some of my time trials yet, but I would assume that when the game drops below 19-20 FPS the desync can happen, giving an apparently faster time.

I did record a bit of lag happening at the start of the level (when you assume control of Buzz), and although most of the lag was I think before the timer started, the audio definitely desynced a bit - seems 1-2 frames behind the no-lag recording relative to the timer, and Buzz seems about 1 frame ahead of the no-lag recording relative to the timer.