Watch the last mission and you see that there is a speed-up (over speed) in the video for the psx tas. You can see it when the elevator is activate or in long ways, it run faster than normal speed.
Any specific timestamps in the video to check out to observe this behaviour? Because I'm still not seeing it.
I will not argue/battle, this is a easy discussion.
The elevator is a simple way to check out the speed of the game. The elevators in the last mission, you can not manipulate the elevators, it runs always in the same time.
When we Only make a timing for the elevators than we have this result ->
(1 elevator go up and down) aquatiger ntsc, old sda run psx -> ~10 seconds current sda run pal psx -> ~10 seconds tas psx -> ~7 seconds
-
The psx tas is very good, because we can see all the tricks, where you need millions of restarts, this is very heavy to do this. But I can confirm, it works also when you play it real.
^ I think you've gotten closer than anyone else, but yeah, that guy is a beast.
I got into speedrunning through TASvideos, a perfect run of any given game was just a really appealing idea for some reason. I ran Syphon Filter in 2006 mostly because I was curious what a TAS of it would look like, and PSX emulators weren't up to the job at the time, so I figured a segmented run would be the next best thing. Working out route planning and strategies got me interested in speedrunning more broadly, so TAS can actually be a gateway into this community, not just a parallel interest.
(1 elevator go up and down) aquatiger ntsc, old sda run psx -> ~10 seconds current sda run pal psx -> ~10 seconds tas psx -> ~7 seconds
Like I asked, you posted the video on the previous page showing the comparisons between the various runs, so can you give a specific timestamp to look at, please? I had done timing myself of the videos going up the same elevator (the videos pause around the 33-minute mark so the runs are synced at around 33:25 and they go up an elevator about 5 seconds later). My timings of that elevator ride were a little over 8 seconds for the two NTSC runs and 10 seconds for the PAL run. When you take into consideration that PAL is 50Hz and NTSC 60Hz, that makes sense (if the one running at 50 FPS took 10 seconds, you'd expect the ones running at 60 FPS to take around 8.3 seconds, and my results were close enough to that) Likewise, the section 6 minutes in that I previously highlighted (the jumps across the rotating platforms, which is effectively another autoscroller), I timed just under 12 seconds for the NTSC runs and around 14.75 for the PAL run (again, about what would be expected given the 50/60Hz difference). You don't even need a timer program to see this, as it's abundantly clear: youtu.be/Bq1SJYnz5kE?t=369 And finally I timed the very last elevator before the runs are all synced again at 43:25. The NTSC runs both got up the final elevator in around 8.7 seconds while the PAL run was around 10.8 seconds (again, roughly as expected)
With that in mind, what you're saying doesn't make sense (and neglecting the fact that the old SDA run you've decided to mention is from a separate video I had to hunt for). You're trying to suggest that in an autoscroller section, there's an NTSC run that takes as long as a PAL run, and that two NTSC runs have vastly different times. I found the same sections in AquaTiger's run that I'd picked out of the Youtube vid and timed them manually as well, and they were in line with the times I'd got for the TAS runs in the Youtube video.
TAS is a jerk. He always beats my runs. I still don't know how he can be good at so many games. I really want to beat TAS at just one game some day.
I have always hated TAS, but only because I'm jealous.
I actually beat TAS once, but he came back and beat me handily once he found out.
Me too, except it doesn't really count because he blatantly eschewed optimization and focused on being stylish for the sake of "entertainment". What a tool