PS vs PC split
5 years ago
Germany

Since the PS1 and PC versions have (unsurprisingly) massive loading time differences, I decided to split the two categories for the moment. Ideally we'd have both versions on the same leaderboard with loadless times but we don't have a loadless timer and I dunno how to make one. If someone looks into it (should be simple because most PS1 runs are on emu) then let me know I guess.

Finland

Well it kinda depends on the computer and emulator your using to cut out the load times. Since for example people here have had load times of around 16-17 seconds on the WR runs. For some reason I would only be able to achieve that with 2x overclocking on the emulator but that causes audio screw ups. So i had to go with my like 19 second load times :D One way would also be just to remove the load times by retiming it and just basicly pausing the moment the victory animation starts and restarting it after the gameplay starts up again.

Finland

Oh and btw if we are doing categories at this point emulator and actual ps console should be a different thing too since emulator is way better on load times then the actual console.

København, Denmark

I spoke with Javasaurus about this some time ago. Tested some things and timed loads on some runs (compared my pb, Java’s and Mlie’s WR as we all use different emus). Not surprisingly loading times from emulator to emulator is very different. There is a ~6 seconds load difference between each level from "worst" to "best" emu. Luckily I found no in-game difference.

If fairness is the goal I can see two approaches. Either, as Joker suggested, we rank by loadless time where each runner has to retime his/her run before submission until loadless timer is made (if ever). This will involve retiming the ~30 PB-runs on the boards. If manpower would be an issue I’ll gladly help retiming.

OR we can require all future emu-runs to be on the same emulator. I don’t know how that would be checked or verified. I guess we would have to rely on the honor system. This would still split PS and PC runs, though.

Ofc there’s secret option number 3 to just leave the board as it is atm. That works too.

Finland

Eeeeh requiring a certain emulator is kinda bad as an idea but having the knowledge of the "best" emulator and settings for the chance to be even with everyone else would totally be fine. Still a huge shame that the actual console is slow as heck :/

København, Denmark

I assume it's only the PAL-PS that is slow. NTSC-version should be somewhat up to speed with emus. But since most running the game are European the chance of someone having an American Herc-game is low. But yeah, it sucks that PAL is this slow and painful to play.

Well, deciding on a "best" emulator would be nice and all but won't really do anything. To me it's ban or no ban, and no ban includes everything being fair game - excluding in-game manipulations and stuff. Especially now when 0 second loading time is the upper boundary.

That being said my vote is on loadless time. Everything appears to behave identically in-game, so it really is the loads that is the problem. There is a total of 19 loading screens with each submission (10 if we include the score-screen as "loading time" even though it isn't), so it might be an annoyance to submit in the future, but the leaderboards will at least depict the truely fastest time.

Sweden

Loadless does seem to be the best option, after reading this thread. That way, everyone can play on their preferred system and still make the run valid. I'd be happy to help with the re-timing of the runs!

Finland

Sooo we doing this then ? Or someone doing check ups on the performance of pc vs ps somemore? We could at least do the loadless times for the ps list ?

SpAceR20x and MiklMaar like this
Germany

Oh shit I didn't get any notifaction for any of these answers here, heh. If I don't respond after a long amount of time, try dropping me a message on Twitter or Discord, just for the future. I'm all open for merging the boards again for a loadless board. I would like someone to help for this though since this is mainly a game I moderate on the sideline. It's not the most active speedgame out there, clearly. But I don't have an assload of time on my hands for now (and the next few weeks at least) so another mod could be good. Any of you three would be suitable I'd say since you have all been running the game reasonably actively and didn't seem to be stirring shit anywhere (as far as I can tell). Do we wanna make everyone retime their runs manually? Or do we have spreadsheets with the PS1 loads in the rules so a set time can be subtracted? I dunno what's best but I feel like manual retiming is not too painful for this game either way.

Console and emu should be the same by the way. PAL is just naturally slow and there is nothing you can do about it other than switching to NTSC. There won't be any seperation for that kind of shit.

København, Denmark

No harm no foul. I do have multiple assloads of time, so I'll gladly use one of them to retime the runs.

The whole problem stems from different emus not having the same loading times, so subtracting the same times from each run will effectively do nothing. Manually, unfortunately, is the only viable method I see.

I would say every time "LOADING" appears on screen should be subtracted (so not the scorescreen, any accidental movie clips, the title screen before each level or fadeouts). We could make timing start when gaining control of Herc, to skip 1 loading screen, but there would still be 18 of the bad boys per submission, so that might just confuse more than do any good.

Sweden

I agree with Mik. And I can retime as well - should we decide who retimes which runs? Maybe create a Google doc so we don't retime runs twice.

Germany

A google doc wouldn't be necessary. There'll be a seperate column with loadless times so we'll see which ones have them and which ones don't when the loadless times are added.

Every frame that says "LOADING" would be great and all but that doesn't appear to be right. If you compare the two versions, PC even cuts out any and every fade inbetween levels and loads. There are essentially no loading screens on PC. There is a ton you'd have to cut out for PS. Hard to come up with a proper definition for that, that unifies the two versions.

Changing timing is pointless too btw, so scratch that. It'll only lead to people mistiming their runs and you having to retime them anyway. I know that from moderating Crash, which has similar rules.

Finland

Is it not just the simplest thing to cut out everything that happens between the moments you gain control of herc and beginning of the victory animation (fade out on the pegasus run since it has no "victory animation"). Its not that relevant really to time the score tables in my personal opinion.

København, Denmark

Hm, yeah, it's true. I just looked at Javasaurus', Joker's and Skullbot's PBs. If only "LOADING"-screens were taken off, Joker's PC-run would still have a ~3 secs advantage. Taking Skullbot's run that's another seconds faster (for whatever reason). So just eliminating loading-screens is obviously not enough.

What Joker suggests works fine for me. Imo timing should be paused at fadeout on all levels and not winning animation, but that's minor stuff. Alternatively we could unpause time during scorescreen-stuff, but the time you can waste for bad button-mashing is really not that significant, so it might be a big price to pay for a 0.1 seconds time difference at worst.

Germany

Personally I'm not a fan of counting the level results screens as loads. It's kinda janky but anything from when a fadeout BEGINS until the next fadein ENDS could work as a load. Would require to go frame-by-frame through every run though so the accurate amount is subtracted. I think that'll cover it at least. The boards aren't tooooo populated either so retiming the runs is generally not a big hassle, especially since the runs aren't even long for the most part. I'm assuming that the loads will stay the same for each person (unless they play on different computers or something) so it's easy to get rid of the loads if we go to other categories than the standard one.

General timing rules will still be the same however, just so we're clear on that.

So I'll mod MiklMaar then if everybody else is fine with that?

Mlie_Redfield likes this
Sweden

Counting fadeout to fadein does sound like the best idea, in my opinion.

I'm fine with Mik being a mod!

København, Denmark

Agreed, however it seems like you gain control of Herc at the start of fadein to each level. The scorescreen starts ticking at fadein start too. So I would argue that it should be from fadeout start to fadein start.

With that the stuff we define as loads would be fade-to-fades and whenever the game explicitly states it’s “LOADING”. This leaves 3 blocks of loads per PS-level (1 block per PC-level), which with the relative few levels in the game is tolerable.

With your Crash-experience in mind general timing should stay as it is now, yes.

Sweden

Love the look of the board! =D (So far)

Edited by the author 5 years ago
Germany

And as such all the retiming has been done. All runs without videos have received reasonable averages or loads based on runs from other categories by the same person instead. Rules also have a little note for how loads are defined.

Florida, USA

If you like it, go ahead to mantigames.com and play the game on your browser for free!

Game stats
Followers
115
Runs
135
Players
38
Latest threads
Posted 1 month ago
2 replies
Posted 4 years ago
23 replies
Posted 17 days ago
1 reply
Posted 9 months ago
2 replies
Posted 10 months ago
4 replies
Posted 2 years ago