Would it make more sense if some black frames didn't count as load frames? Because during a normal level transition, there's nine black frames between the end of the level and the GUI appearing (or whatever you call the "WORLD X-X" and Mario's lives in the middle of the screen), so it would make more sense to take nine frames off of the current load times, right? So these would be the new numbers (copied and edited from https://www.speedrun.com/smbtll/guide/jj2r0):
FDS loading frames: varies framerate: 60.0988139 load time: varies
Everdrive N8 (os 1.16 or earlier) loading frames: 255 framerate: 60.0988139 load time: 4.243
Everdrive N8 (os 1.20 or later) loading frames: 292 framerate: 60.0988139 load time: 4.859
Everdrive N8 Pro loading frames: 310 framerate: 60.0988139 load time: 5.158
Powerpak loading frames: 97 or 98? framerate: 60.0988139 load time: 1.614 or 1.631?
FCEUX loading frames: 361 framerate: 60.0988139 load time: 6.007
Nestopia loading frames: 341 framerate: 60.0988139 load time: 5.674
Bizhawk loading frames: 368 framerate: 60.0988139 load time: 6.123
Mesen loading frames: 336 framerate: 60.0988139 load time: 5.591
loopy's romhack loading frames: 1 framerate: 60.0988139 load time: 0.017
Reproduction cartridge loading frames: ? framerate: 60.0988139 load time: ?
Wii VC loading frames: ? framerate: 59.824 load time: ?
Wii U VC loading frames: 220 framerate: 59.94 load time: 3.670
3DS VC loading frames: ? framerate: ? load time: ?
GBA loading frames: ? framerate: 59.7275006 load time: ?
Switch loading frames: 310 or 311 framerate: 60 load time: 5.167 or 5.183 ... please manually check the load time
Edit: Level transitions, in which the level before the transition starts has a black background (this includes castle transitions), are slightly different from level transitions, in which the level before the transition starts has a blue background. Level transitions, in which the level before the transition starts has a black background (again, this includes castle transitions), have ten standard black frames, and level transitions, in which the level before the transition starts has a blue background, have one standard blue frame and nine standard black frames. These are the correct numbers (once again, copied and edited from https://www.speedrun.com/smbtll/guide/jj2r0):
FDS loading frames: varies framerate: 60.0988139 load time: varies
Everdrive N8 (os 1.16 or earlier) loading frames: 254 framerate: 60.0988139 load time: 4.226
Everdrive N8 (os 1.20 or later) loading frames: 291 framerate: 60.0988139 load time: 4.842
Everdrive N8 Pro loading frames: 309 framerate: 60.0988139 load time: 5.142
Powerpak loading frames: 96 or 97? framerate: 60.0988139 load time: 1.597 or 1.614?
FCEUX loading frames: 360 framerate: 60.0988139 load time: 5.990
Nestopia loading frames: 340 framerate: 60.0988139 load time: 5.657
Bizhawk loading frames: 367 framerate: 60.0988139 load time: 6.107
Mesen loading frames: 335 framerate: 60.0988139 load time: 5.574
loopy's romhack loading frames: 0 framerate: 60.0988139 load time: 0
Reproduction cartridge loading frames: ? framerate: 60.0988139 load time: ?
Wii VC loading frames: ? framerate: 59.824 load time: ?
Wii U VC loading frames: 219 framerate: 59.94 load time: 3.654
3DS VC loading frames: ? framerate: ? load time: ?
GBA loading frames: ? framerate: 59.7275006 load time: ?
Switch loading frames: 309 or 310 framerate: 60 load time: 5.150 or 5.167 ... please manually check the load time
While it's true that not all black frames between 4-4 and 5-1 are frames where the game is actively loading data from the disk, I'm not sure if anyone wants to actually go through the trouble of retiming hundreds of runs just to account for the ten standard black frames between the 4-4 and 5-1 transition? It's not entirely accurate, but it's a lot simpler to just count all black frames between 4-4 and 5-1 as load frames, especially since all black frames are visually identical and that method of timing has been used for so long.
I remember coming to this realization a small while back and I was too scared to bring it up. In all honesty I think it should be changed because "technically" we are removing time that should count towards the final time. If there was a button we could press that would change all the times and add +0.166 I would say we press the button, but unfortunately it's not that easy. Definitely an interesting topic to bring up though.
i personally dont think this would be a good change, as in a way its 'changing history' of times. a lot of past times have worked under the impression that certain barriers were where they were, and having that possibly be changed by 9 frames could easily make a lot of those turn out to not mean anything in the end. since so much has happened with the requirements for second barriers or similar thought to be where they are, it just makes sense to me to leave it as-is even if it technically isn't the true accurate time without loads.
If there was a button we could press that would change all the times and add +0.166 I would say we press the button, but unfortunately it's not that easy.
If it's about having a magic button - it should be possible to automate the task with selenium
We have added Any% D-4 (No S+Q) as a category, to allow runners to speedrun Any% D-4 continuously from 1-1 to D-4.
Enjoy!