Splitting of console and emu
6 years ago
Ohio, USA

Sorry, my apologies. I actually had forgotten who did the run, but that's really no excuse. My point was that the run was nowhere near as technically good as the console runs, yet was over a minute faster. Those two things don't add up. I should have said it in a nicer way, though.

Asre likes this
Texas, USA

No problem just glad I have found a good answer to the split, just wanted to know is all thank you for the information on the split it makes sense now to keep them split until there is a fully accurate emulator :)

Netherlands

So I watched both runs of @Asre and @starbiits at the same time at both screens. Both made mistakes, selecting the wrong ham-chats. And both bumping in walls. To be exact you should time every mistake made by both runners. But this isn't the reason why I am replying ^^

I did notice something different. At the very beginning, when Boss is talking, it seems at Asre's run the textboxes are faster, also when Asre is selecting a ham-chat it seems a bit faster (not alot just some miliseconds) I am not sure this is the reason why Asre's run is faster. It could also be Starbiits is clicking slower lol or the Gamecube Player is acting slow. But it seems the Emulator doesn't have as much delay for clicking through textboxes and selecting.

Should be checked to be 100% sure! It's only what I noticed.

Edited by the author 6 years ago
Asre likes this
Florida, USA

Yeah. I've ended up being busy this week and haven't had time to try it like I wanted to. May look into it further over the weekend.

Asre likes this
Netherlands

Because I want to help I made this video today. I did a run on the Bizhawk emulator and other consoles I was able to record.

As you can see, holding B button while spamming A has a huge impact! But it doesn't seem that the Bizhawk latest version is faster at all. But comparing Starbiits and Asre's video, Asre is quicker through the tutorial. I think it has due to do with spamming the A button.

@Asre, if I may ask, which Emu did you use?

Edited by the author 6 years ago
MelonSlice and Asre like this
Texas, USA

I used VBA no modified, so I should run it on Bizhawk and see what I get. I do hold B and spam A through all dialogue. @Seydie

Scotland

there really is no difference pretty much have you tried GambatteSpeedrun it seems very accurate as it is what the pokemon community use, if you are worried? :) also no point putting down the wr speed run seen as they still ran the game and tried there best ^_^ and if the emu are this accurate maybe they made up time in another area.

Edited by the author 6 years ago
Asre and Seydie like this
Ohio, USA

If you guys think that a certain emulator is exactly the same as console then I have no objections at all, however, we would still have to require all emulators to show what emulator they are recording on through the entire run in order to verify the proper emulator. This would still reject 100% of the current emulator runs and delete them from the leaderboards which is what I was trying to avoid. But if you'd prefer allowing specific emulators under one category then I am 100% in support of that.

josef733 and Seydie like this
Scotland

did you watch the comparison video really really I think having 1 emu is a good thing even tho they all run the same but deleting runs they should be allowed to remain and then all future runs should have to follow them rules it makes more sense it would be rude to the people who already have runs and Asre's amazing run idk why they have to be removed

Asre likes this
Ohio, USA

That's not how it works. Either we make a rule where you can use a specific emulator and it all goes under one category, requiring all the old runs to be deleted, or you just keep them split and all the old runs can stay.

You can't pass a rule requiring all runs to show the emulator they use and then just give all the old runs a pass from the rule

josef733 likes this
Ontario, Canada

Its a well-known fact that Gambatte is a very accurate emulator, damn near identical in performance to real hardware. Only days ago did the Pokemon community approve speedruns using that specific emulator.

What should be done is requiring everyone to use Gambatte if they will do an emulator run. What should be completely off-limits is something like VBA because the emulation for GB/GBC is very inaccurate. I'm surprised this isnt mentioned whatsoever in the rules.

http://www.gbrunners.com/game-boy-speedrunning-explained Guys, it doesn't hurt to do a little research before creating your rules and leaderboards.

Edited by the author 5 years ago
Asre and MelonSlice like this
Texas, USA

Did some stuff with gambatte, got a 6:35 on my 1st split which was 4.8 seconds slower mainly from minor mistakes running past things. We dont have to put emu and regular back together I honestly dont care at this point. But if we do I say make emu on gambatte only. But again we can keep them in their own categories.

josef733 likes this
Canada

I'm not sure I like where this discussion ended. So for people wanting to run on emulator now, should they purposely not use Gambatte since it's accurate and therefore slower than other emulators? Where would I put a run I did on the Gambatte emulator? If your answer is in the Emulators category, think about it a bit more. In the current state of that section, that would be much less fair than putting it with consoles.

EDIT: In my opinion, regardless of whether the community wants to merge emulator and console runs, all emulator runs should have to use the same emulator (or if not, then one of the emulators generally agreed-upon to be accurate). This would, of course, disqualify VBA (unless we agree to all use VBA for emulator runs, which would be ridiculous given its inaccuracy), but I'm sure the runners with VBA runs on the boards currently would be able to redo them on an accurate emulator, or maybe even improve them. This would certainly heighten the quality of the leaderboards.

Edited by the author 5 years ago
Netherlands

I completely agree, but is it needed to wipe every run till now who didn't use Gambatte?

I didn't do a run so I can't really argue. But maybe wiping the runs who are in the top 3 and didn't use Gambatte? Or no wipes..

Atleast adding a rule we all use the same emulator is a good one.

Canada

It doesn't have to be Gambatte - it can be Bizhawk, or even BGB, going by the site linked earlier by josef. But if we leave a couple runs up that used VBA it would set a bad precedent.

A way around this could be like "Any time faster than X time cannot be submitted with VBA." This introduces something arbitrary so it's not great IMO but it could work.

Seydie likes this
Canada

My AV capture card came in today so I was finally able to stream console runs. I am almost 100% confident that the Gambatte emulator is accurate, considering I was able to gold multiple splits on console despite having pretty good golds from emulator. For reference, since neither run has been verified yet and so they aren't on the leaderboards, here is my 44:53 run on the Gambatte Emulator: And here is my 45:15 run on the Gameboy Player using the Gameboy Interface as a replacement for the official disc (the disc/official software costs $$$$):

My conclusion: We should merge the emulator and console leaderboards, and only allow Gambatte/Bizhawk/BGB on there. This seems like the most fair thing to do, and will hopefully save future runners money (from having to order a gameboy player, old capture card, gamecube, etc). @Totalled , @ambaharmony , @Asre , @starbiits

Edited by the author 5 years ago
MelonSlice and Seydie like this
Ohio, USA

Thanks for doing those tests, guys, it definitely helps. Personally, my opinion is still the same. That being, it depends on whether you want to wipe the current emu runs. If you guys are okay with wiping the current emu runs, then I'd be perfectly fine with merging the categories, only allowing certain emulators that we know are accurate, and requiring that all emulators capture the program as well, to show proof that they're running on the program they say they are. The reason we would have to wipe all the current emu runs is BECAUSE we'd have to add the rule about capturing the program. I'm not personally a fan of "they can use inaccurate programs unless they get a good time."

That's just my opinion, of course I'd love to hear from the other mods too. You guys seem passionate about merging the categories, so I'd like to do whatever makes you guys happy. It's just the age old question of to wipe or not to wipe a ton of standing runs. Maybe we could hold a community vote?

Canada

I am all for wiping the emu runs with no proof of which emulator they're on. That being said I'm not everyone, so I agree with the community vote idea. Hopefully we can get a vote going relatively quickly.

As for the requirements for emulator runs, I think the proof method I used in my video is fine - the emulator looks like this: https://i.imgur.com/lwbvNWy.png so there's nothing really to show apart from the name at the top (and doing so would hugely restrict possible overlays). Being able to move the window around seamlessly, along with the colour scheme not changing at the end of the run, is likely enough proof.

Seydie likes this
Ohio, USA

For other games that require proof of emulator, it is usually customary to record the entire window. You can easily avoid the black bars by changing the size of the window.

As for the vote, how would we go about that? Maybe a strawpoll posted in the discord?

sonicyellow likes this
Canada

Sounds good to me.