I've went ahead and added DuckStation as a viable emulator. Be sure to check out the updated rules if you plan on running with DuckStation especially, as that requires a little more from the runner to ensure legitimacy.
Regarding accuracy, it's identical to ePSXe 1.9.0 with no enhancements active. This should be a nice quality of life update for people wanting to run the game.
Personally, I'd vouch for this too. Although I'm thankfully not having to run on emulator any more, ePSXe is very outdated, 1.9.0 especially. The fact we still require this right now is a relic of the past and is partly just because of lazy moderation (including myself).
It's time for a change, to make speedrunning more appealing by not requiring only one option of emulation, which is currently 10+ year old software.
As for accuracy, I haven't done frame checks, just barebone timing, but it times the exact same - if not MORE accurate. I don't see an issue in that regard whatsoever.
Heya and welcome! I unfortunately had to reject your run, refer to the reasoning of my rejection so you can be safe in the future.
Also, check out the rules before submitting. They usually specify everything you need to know and prevent runs from being rejected. In this case, it's not specifically worded when to start recording, but I do require seeing you in the hub and selecting the level. Hope it's not too big of a hassle for you to try again! :)
So I just tried it out and I have no idea what you mean by "UNLOAD YOUR ENTIRE JAR OF JELLYFISH INTO HIS BUTT". He has invincibility frames in which you can't keep attacking him, which we already take in account for and try to minimize the time we spend NOT attacking him, so we chain shots from a distance with a delay in between, he dies and we're already right new to the item. Have you watched the way the speedrun does it? Can you post a video of your method? Because it doesn't sound faster.
I believe school already answered in the discord, but just for clarification: 1.9.0 is the most accurate in terms of loading and lag since it replicates the original PSX pretty well. There are other emulators which are generally accepted on PSX leaderboards, but I don't think they've been tested good enough with World 3 so I can't say for sure. Better just stick to 1.9.0. It's not only the loading times as I mentioned that play a role here, lag matters a lot in this game. The amount of time for the battle menu to pop up can take several seconds on emulator. Though, if you use a PS2 with fast disc speed for example, you'll have them come up much much quicker and more consistently.
Can't help with the audio glitches unfortunately, I don't think I've had that happen, then again it's been a while since I used ePSXe.
Not gonna lie, I really dig the 2nd batch of icons. Looks super unique, though it does look more like a Kula World Leaderboard, but eh :D
I'm quite a fan of icons which represent something from inside the game (and have only encountered ones which are exactly that), but since there's no icon right now at all, why not.
Okay, that's interesting then. Banning emulator for having faster cycles is an obvious decision because it's not legit, but if an actual official console runs it faster, then that is just the fastest way to run the game. I don't see any issue accepting PS3 runs, that's for my part at least.
I don't think PS3 has been considered yet, so testing would be required here yeah. I have no way of testing that, so if you could do that that'd great tbh. The cycles are described in-depth in the other forum thread, so you know what to look for. Recording it would be great too and then posting it here afterwards.
I didn't dive very deeply into it, but the decision of banning emulators was pretty straight forward given the huge inconsistencies within different emulators. Would have been fine if they didn't provide such a great advantage with the cycles now that it's loadless. It would be possible to allow just one specific emulator to be run on for submission, but it would still be a different leaderboard and I'm not very sure if that would hold up in terms of runs. I mean, I guess you'd submit, and that's about it.
Though if you're coming forward to demand that, I'll have no issue with that. If you actually run on PSX and discover that cycles actually load later (getting them earlier), then PSX is the way to go with loadless. Until then, this is a dead speedgame anyway : (
If you don't know the game too well, I'd recommend playing it casually, it's plenty fun. You'll learn the map layouts, get used to menuing and just get to know the game. When you're comfortable with the game, get to know the notes, read through them, see if you understand what they want from you. Everything from that point onwards should resolve itself from jumping into the cold water and running the game! :D
You didn't effect the community in a negative way, if anything, you helped me discover this major difference!
Emulator runs, from this day, have and will be banned from submission. Refer to my other thread, I will not go into detail on why this is the best decision:
No, there will not be a seperate leaderboard for emulator runs, differences are still too major between them. The people whose run has been banned now and really feel unjustified, hit me up on Discord (Mentro#3169). If there's enough demand, we might work something out. But for now, the people who have emulator runs on the leaderboard are inactive and probably forgot that they even have a run for this game. This will do.
Also, runs will, from this day onward, be timed without the loading screens to guarantee a more competitive and just competition. I will retime all current runs, and all further submissions. Runs will be submitted WITH loads, I will edit the time WITHOUT loads. Leaderboard placement will, of course, be judged by your loadless time. I will time everything that the player still has control over, for example skipping cutscenes is timed, choosing stages also is timed, loading screens are not. I will start timing once the loading screen goes into the black screen with the stage's name on the screen, and when finishing stages or selecting them on the map I'll start timing once the entire screen is filled black.
About 100%: If there's enough demand for the category, I will create it. It's a legit run which will take longer, and show more of the actual game. Though, I have the feeling once creating it, it would stay empty for a long while, and I refrain from creating empty and dead leaderboards just for the sake of completion.
Auw wow, that tool is very interesting, good stuff! And yeah, about 100%. I'd also like to see that as a category, I already finished one like 5 days ago. Took me way longer than it should have though, I believe it was almost 8h long.
So as I recently found out, there's a huge issue with emulator runs. As of now, all of the emulator versions are allowed, which wouldn't be an issue if there wouldn't be cycles in certain stages that only emulator can get (I don't know why that's the case). Those cycles are not replicable on an actual PS2. Mangombage's 23:03 was utilizing those unreplicable cycles.
The following cycles are not replicable on an actual PS2:
Kelpazoic Jungle: https://gyazo.com/ec371a4a773f92aa3c234a243c142a14 (If you try to just walk, you will definitely get hit and pushed back to the left)
Canning Factory: https://gyazo.com/55cf698f1ffa0eb3dd28450c103e3b1f (This is not even close on PS2, you will always have to wait for like 5 seconds until they even come out)
Another issue that was already talked about were loading times, specifically Laure's 21:57 loads. We assumed that the faster loading times resulted from using the "Greatest Hits" version of the game, but that's probably faulty. Cyberdemon531 is using the same version of the game as Laure, but is still getting the same loads as me and Hdot12. The faster loading times are resulted from a different model. Me, Hdot12 and Cyberdemon531 are using a SCPH-90001, which is generally speaking the fastest model regarding loading times and lag reduction, but there are exceptions. This might be one of them. Laure is using a SCPH-79001, which is probably the reason for the big difference of loading times, that model seems to be the best for this specific game.
There's a lot of stuff going on, but there are solutions. While they might seem rough, they are definitely necessary to create a fair leaderboard.
My suggestion regarding emulators: Remove the current emulator runs and ban any further emulator submissions. Having cycles on emulator that are not replicable on console is enough of a reason for that action, at least for me. Another solution would be to split console and emulator runs into two different categories. Certain emulators would then have to be banned though, because they load the game too fast. ePSXe 1.9.0 would be a good limitation for emulator runners.
My suggestion regarding loading times for console runs: Loadless, as simple as that. Submit your run, mods will time all the loading screens of your run and then subtract that from your RTA time. Simple, efficient, fair (further definition for loading times have to be defined among mods).
The current runs would have to be retimed for loadless, and I would definitely do that to help the mods.
Update: Still crashed, copied the settings of someone whos emulator isn't crashing, who is also using OBS. Changed video plugins, also crashed.
I changed it to Window Capture instead of Game Capture, doesn't crash anymore, so I'm good. But it's still weird, I mean crashes happen every now and then, and often have a fix. But why did they appear every exact 15 minutes (checked with timer, crashed 3 times, always after 15 minutes). That seems so weird, there has to be something OBS does to the game capture every 15 minutes, can't explain what though.
Window capture for life!
First, audio stops and screen freezes for like 2-3 seconds, then the emulator just closes on itself.
I'll try re-downloading it, alright.