I checked now and my game run at 60 fps
Last month the game was updated and there are now 96 levels in total. I suggest adding a "Level 81-96" category. I also suggest changing the "Full Game (Extended)" category to include the new levels. The only run in this category is mine and it's okay for me if that run becomes obsolete.
I'm uploading the script on the Resources page. I modified a small thing compared to the script that I uploaded on Mediafire, so I suggest to download the new one.
Yes, I studied computer engineering. Speaking of the timing method, the main thing to do is to define a standard for non-real time, whatever it is between the "Time without loads" that I have proposed or the sum of the in-game times. Summing the in-game times might be more accurate, but I haven't found a way to automate it, so it would require a manual sum. Not only that, for some Crash Courses there is no timer and we need a frame count for those levels.
You have to split manually. I only implemented the in-game timer
I found a variable equal to 0 when you're playing or if you are in the main menu and it's different from 0 if the game is loading. I used it to create a basic in-game timer which pauses when the game is loading (it's different from the in-game timer proposed by you which was an all-time level sum). Before I upload this to the Resources page, I think it's better if you test it on you computer. https://www.mediafire.com/file/ys6ll3gazshj279/MM2_-_ASL.rar/file In the rar file there is the script and a layout for LiveSplit that shows the real time and the in-game time. If you don't know how to to set up the script, read here: https://github.com/zoton2/LiveSplit.Scripts/blob/master/LiveSplit.GTAVC-README.md Here is an example video of the script:
For "Turn it Up" I calculated the time using the frames. I did it in my video editor but, if you can't do it in that way, you can use Virtualdub ( http://www.virtualdub.org/ ) which is a video player that shows the frames. You can simply take the starting and the ending points and do a subtraction to get the time. Speaking of the main argument of this thread, I was trying to see if there a way to calculate the in-game time automatically, so I started looking a way to do it. Basically I need to find a variable that tells if that game is loading or not, if you can control the car or not, or something similar. So, it could be that, instead of in-game time and real time distincion, we could automatically calculate time with loads and time without loads (as for example in GTA Vice City: https://www.speedrun.com/gtavc#Any_No_SSU ). If you want, I could continue to search for a way to automate the time calculation.
It is strange, on my computer the game works even without dgvoodoo (even if it is rather slow).
I'm sorry I only answered now, but I was busy in the last few days
Uhm, it may depend on the version of the game or even on the operating system. What is your os? I'll send you (as private message) a link to download another version of the game. I can't put it here due to forum rules. Can you tell me if the traffic is the same as mine or if nothing changes?
Can you take a screenshot of the level using PRS? I want to see if also that one is different
https://imgur.com/a/nKuzkkz I took some screenshots of the differences. I also tried a density that is in between ARS and PRS and it seems to be the case that you get. I can't explain why there's a white car in grkmkprl's run, I tried different densities but I never get the same traffic. Maybe the traffic depends also on something else like the version of the game but, without looking into the code, it's difficult to have a right answer.
I think traffic patterns are generated using the traffic density bar, which is why there are always the same cars at the beginning of the level. Once the level starts, I think every car makes turns on every crossroad based on a random number. I'm pretty sure that the number is chosen using a function that, given a seed, will give a "random" number each time the function is used. Usually, the seed is set using the time, so I think that if you set the time in your computer to a specific date and time, cars will always follow the same path (while normally they tend to do different things even though in the same level). But traffic patterns don't use the random function to be generated, that's why there's always the same traffic when you start a level. About "Amateur Race Settings" and "Professional Race Settings". Making a distinction using ARS and PRS would be better than just using "Default race settings" but only for a few levels. For example, the only difference in "The Knowledge" between ARS and PRS is the time given to complete the level, which is irrelevant in a speedrun. I don't know how many levels have substantial differences between the two settings, so I don't think we really need to distinguish them
Like in TLA1, there are 80 levels in TLA2 now. I suggest to add 2 new categories: "Level 65-80" & "Full Game (Extended)".
MRS is not so different from DRS in "All Races" and in "All Checkpoint" because you unlock the GTR-1 very soon. "All Crash Course" is a lot different because you're not forced to drive the Mustang Fastback and the London Cab. Also "All Blitz" and "All Circuit" are very different because, if you start from a new player, you can't drive the GTR-1. I think that the MRS is useful for some category, so I think we shouldn't eliminate it.
I think that the "Default race settings" and "Modified race settings" are good for individual levels but they are not good for full-game runs. The problem is that, in MRS, there's written "You must start from a new player" but you can modify the settings of a race only when you finished it at least once. So I suggest to remove that rule in order to make the category possible. This will also make possible the use of the GTR-1 (or any other car) since the beginning of the game