DS speedrun leaderboards

Discuss policy guidelines for the community and whether something needs to be changed or not.
Vulajin
Schoolkid
Posts: 46
Joined: Thu Apr 24, 2014 4:47 am

DS speedrun leaderboards

Post by Vulajin » Tue May 20, 2014 11:26 pm

For the leaderboards of DS Pokémon games, I would like to discuss the categorization of emulator vs. console runs. Currently, the precedent is set by the White 2 leaderboard, which lists Garfield's 3DS time in the same table as everyone else's emulator times.

Personally, my inclination is to have two entirely separate tables, with emulator runs listed in one table, and all console runs (DS/3DS alike) listed in the other table. It is not my goal to segregate runners on different platforms as though they were lepers or something. However, having the runs listed in the same table, sorted by time, implies a comparison between console and emulator times where one does not exist.

Before I make such a change on the two gen 5 leaderboards, I'd like to solicit a few opinions here. Does anyone feel strongly one way or the other? Does anyone have any solid arguments favoring a merged table?

User avatar
heero_fred
Youngster
Posts: 70
Joined: Thu Apr 24, 2014 2:37 am
Location: Mexico
Contact:

Re: DS speedrun leaderboards

Post by heero_fred » Wed May 21, 2014 12:20 am

I don't know if it's possible with the current leaderboards setup, but both console and emulator could be on the same table as long as there is an indicator of what method was used, and being able to sort it by it, similar to the Zeldaspeedruns leaderboards

anirocks99
Schoolkid
Posts: 30
Joined: Thu Apr 24, 2014 12:27 am
Location: California, United States

Re: DS speedrun leaderboards

Post by anirocks99 » Wed May 21, 2014 2:31 am

Vulajin wrote:For the leaderboards of DS Pokémon games, I would like to discuss the categorization of emulator vs. console runs. Currently, the precedent is set by the White 2 leaderboard, which lists Garfield's 3DS time in the same table as everyone else's emulator times.

Personally, my inclination is to have two entirely separate tables, with emulator runs listed in one table, and all console runs (DS/3DS alike) listed in the other table. It is not my goal to segregate runners on different platforms as though they were lepers or something. However, having the runs listed in the same table, sorted by time, implies a comparison between console and emulator times where one does not exist.

Before I make such a change on the two gen 5 leaderboards, I'd like to solicit a few opinions here. Does anyone feel strongly one way or the other? Does anyone have any solid arguments favoring a merged table?
You make a valid point Vulajin, where people mainly do a specific run on a specific Console/Emu such as in the case of Pokemon Black where Dhr, you, and Anem all use console, but I would use emulator. The thing is though, it just wouldn't be nice if one person did one specific platform unlike the other and have a seperate table. As well as that, despite Emu being generally faster, in many cases, WR wasn't on Emu and vice versa, so the need for two tables would be irrelevant, because it is comparable due to the fact that Garfield can definitely get overall wr, but the idea of writing what platform you use is great for comparisons, and should probably be put into the leaderboards like the Galaxy or ZeldaSpeedRuns are OneHand

User avatar
GarfieldTheLightning
Youngster
Posts: 54
Joined: Thu Apr 24, 2014 5:04 pm
Location: Scotland

Re: DS speedrun leaderboards

Post by GarfieldTheLightning » Wed May 21, 2014 3:18 am

Like Anirocks says, I don't think there's enough people running DS games on both platforms for split leaderboards to work well. The only possible exception would be HeartGold, because a decent number of runners have times on both platforms, but for everything else, only I have done a glitchless run on Pearl on console (and the fastest times for glitched are both on console), I think only Mike has done a Platinum run on console (and I know he wasn't happy with the time), I don't know the situation in Black, and only I have done a single segment run of White 2 on console. And even for HeartGold, the current world record is on console anyway.

I can't speak for everyone else, but my goal when speedrunning White 2 is to get a time that I'm happy with. Whether or not that turns out to be faster than the emulator record of 3:27 isn't really important - the way it is now, I can still look at the leaderboard and have at least some yardstick to compare my run to. Vulajin says that emulator and console times are incomparable, and while we may not know the time difference between emulator and console for DS Pokémon games, in cases like White 2 where only I have a valid console time, the current setup seems a lot more favourable than a leaderboard that doesn't compare my time to anything.

I do think that for all DS games, it should be made clear on which platform the time was set, though.
I speedrun Pokémon Pearl, White 2 and X!

User avatar
Cooltrainermichael
Site Admin
Posts: 163
Joined: Fri Apr 11, 2014 3:38 am
Location: USA
Contact:

Re: DS speedrun leaderboards

Post by Cooltrainermichael » Wed May 21, 2014 5:19 am

One thing I would like to note is I am thankful to Vulajin, that you decided to get a discussion started before implementing a change. It's a good sign for the forums and community if potentially major changes can be discussed as a community.
GarfieldTheLightning wrote:Like Anirocks says, I don't think there's enough people running DS games on both platforms for split leaderboards to work well.
GarfieldTheLightning wrote:I do think that for all DS games, it should be made clear on which platform the time was set, though.
Currently, with such few runners in most of the Gen 4/5 categories as is, I don't think it would be effective to separate leaderboards, at least this early on. Certainly the discussion can be brought up again once there are more runners consistently running these categories, but as Garfield also suggested, the leaderboard times' purpose is mainly to have some benchmark to compare your own personal run. Whether you want to compare with Emu or Console runs, both times are available on the leaderboards, and as long as the run is listed with what system that was used, then the leaderboard served it's purpose. So at this point, I believe the leaderboards should stay as is, but any runs that do not currently have the console/emu listed, should be updated.

That said, I think a somewhat more separate discussion would be what is considered "World Record." I guess this topic came up since Vulajin noticed that a Black/White runner got a 3:27 on emu, and since the community knows emu is faster, it would seem to be that the console run would be the faster run. However, my current issue is that there are several times considered "WR" already that were done on emu, such as Platinum and Black2/White2. Even the Pokemon Gold/Silver wrs were done on VBA, which calls to question what would the community decide is a world record.

That said, it seems consensus that times such as MHM's plat time, Werster/5upa's Gold/Silver times, and Vulajin/Sinstar's B2W2's times are considered the fastest times of the category, even on emu. None of the current console times for the respective categories are clearly better than those runs, and in order for consistency, I suggest we stick by this format. The reason is that I cannot conceivably see a way to be consistent without outright banning emu runs across all generations except maybe Gen 1, which with such a small but growing community, seems counterproductive to outright ban emus that may discourage people to run the games. I can attest to that personally when I first started out when I was told emus were outright banned and could never be competitive, but I ignored it and helped contribute to what the Stadium community is now.

Overall, I suggest keeping the format as is for now, update leaderboards that show what console is used, and I guess have some sort of discussion of what people's interpretation are for "world records." Not entirely sure of the best solution regarding world records, but would love discussion.
El.Psy.Congroo.
Image Image Image Image Image

Chuckolator
Preschooler
Posts: 8
Joined: Fri Apr 25, 2014 4:16 am
Location: Chucklehuck Woods

Re: DS speedrun leaderboards

Post by Chuckolator » Wed May 21, 2014 6:27 am

If emulators have an inherent time advantage, I don't see why they should be on the same leaderboards. Side by side of course, but to have them on the same leaderboard would be very misleading. If emulator saves 2 minutes in White 2 (I don't know the exact numbers, just an example) and someone got a 3:28 on console, it is factually a better run, so why should it be listed lower than an emulator 3:27 that is really a console 3:29? Listing them separately fixes this. I also don't see why a lack of times or runners should affect this issue at all, because the timing differences are the same whether there's 1 runner on console and 1 on emu, or 100 runners on console and emu.

Someone brought up the idea that because people consider certain emulator times world records, then that means they should be considered equal. I've always thought that was very sketchy. Yes, the play quality for those runs were (and in some cases, still are) far better than anything people have done on console, so I always assumed that was the only reason they were treated as WR, but that doesn't mean that they should be considered the lone, indisputable WR when similar console times start showing up. Again, what if someone gets a console 3:28 in W2? What should be considered the WR, the "fake" lower number or the better run?

If the leaderboards were split into console times and emulator times, then neither platform would lose out. 3:27 would still be the emulator WR (because it's still a great time and deserves recognition) but the 3:28 would get the recognition it deserves as well, and there wouldn't need to be silly disputes over which is really the better run, or what the emu time would be on console, or any of that. And in the case where there's a great emu time but no equal console time yet (as is really the case with White 2, as well as Platinum) then the times would speak for themselves and it would be clear that the emu time is still the better run regardless. As I said before, low amounts of times to put on the leaderboards changes nothing with this, and it's silly to put arbitrary requirements on number of runners before action is taken on an issue that is going to still exist regardless of how plentiful runners happen to be.

edit: actually, I thought of something else. Assuming they were using the same timing standards and stuff, console vs. emu is like Japanese vs. English. What if the saved text time was equal to the saved loading time on emu, would it be okay if someone picked up console JP White 2, got a 3:27, and put on the same leaderboard with the note "Japanese"? From what I've seen, there is universal opposition to putting JP times on the same leaderboards as English runs because of the inherent advantages to those versions (though I am fighting for JP recognition, obviously it would have to be in its own section with its own rules and timing and stuff, and that's a debate for somewhere else) so I don't see why emu is any different for comparison purposes.

User avatar
Luckless
Site Admin
Posts: 113
Joined: Fri Apr 11, 2014 3:28 am
Location: UK
Contact:

Re: DS speedrun leaderboards

Post by Luckless » Wed May 21, 2014 6:34 pm

If there were separate leaderboards for console and emulator then that would mean every single category on the World Records would have to have two times listed much like it is now for the Stadium records. That would be stupid!

Also If you're worried about the better run not getting the recognition it deserves because a worse run was done on a faster platform, you'll still have the same problem even after splitting up console and emulator because there is also slight differences between different consoles and different emulators as well. So does this mean, for example, that if someone gets a run in BW2 using a DS, we would have to have a leaderboard for Emulator, one for DS and one for 3DS? Does it mean that in Gold/Silver there would be separate leaderboards for VisualBoyAdvance and BGB and Gambatte? No. Again, that's stupid.

In my opinion the way it is now is fine. Even though it would be incredibly difficult to determine the exact difference between emulator and console, it's listed clearly enough which platform each time was completed on so if a time is more impressive because of the platform it was completed on, then it just is. That's all there is to it. It's also inevitable that people will compare their times against other platforms no matter how much you separate them up so having them in one list from lowest time to highest time seems the cleanest way to show it. Plus no-one feels segregated from the rest of the runners of the category just because they use a different platform that changes their final time by the smallest margin.

EDIT: After Vulajin has explained it further I am not thinking like this anymore, I was just asking questions to try to understand this better and now I do.
"I see now that the circumstances of one's birth are irrelevant. It is what you do with the gift of life that determines who you are" - Mewtwo

Image Image Image Image Image

Vulajin
Schoolkid
Posts: 46
Joined: Thu Apr 24, 2014 4:47 am

Re: DS speedrun leaderboards

Post by Vulajin » Thu May 22, 2014 8:38 pm

A lot of points have been raised in this thread, and I would like to clarify a bunch of them.
Cooltrainermichael wrote:That said, it seems consensus that times such as MHM's plat time, Werster/5upa's Gold/Silver times, and Vulajin/Sinstar's B2W2's times are considered the fastest times of the category, even on emu.
That is not the case, at least for me. I consider the BW2 emulator WR to be 3:27, and the BW2 console WR to be 3:34. The only reason they are listed as they are is that Sinstar's run and my run were completed long before Garfield began console attempts, and whenever Garfield (or whoever else) added his time, it was added to the same table. I did not do this, and frankly I would not have added it to the same table if I had been the one to do so.

Additionally, I do not believe the gen 2 times are relevant to this conversation. It is well-accepted that emulation for GB/GBA games is far more accurate than DS emulation, and most importantly, is not known to affect the in-game timer, which we use for comparisons. In DS speedruns, the emulator has faster load times, and these do factor into the in-game timer, so no matter what the emulator has an advantage. Only DS speedruns (gen 4-5) are in scope of this discussion.
Cooltrainermichael wrote:The reason is that I cannot conceivably see a way to be consistent without outright banning emu runs across all generations except maybe Gen 1, which with such a small but growing community, seems counterproductive to outright ban emus that may discourage people to run the games.
No one is talking about banning emulators. I am specifically considering strategies to avoid banning emulators. As Chuckolator pointed out, if emulator has a speed advantage, comparing them on the same table to consoles is inherently confusing. However, banning emulators would obviously be detrimental to the community for these games, which get little enough love as it is. Frankly, I expect far more people to play on emulator, so I have no problem with the few weirdos who play on console (including me, when I do so) getting a separate table, and I'm unsure why anyone else would.
Cooltrainermichael wrote:Not entirely sure of the best solution regarding world records, but would love discussion.
We had this discussion back when I found out DS emulation yielded faster in-game times than console. We determined that emulators and consoles would qualify for separate WRs. Console would include all DS models, including 3DS, because they are all legitimate methods of playing the game.
Luckless wrote:If there were separate leaderboards for console and emulator then that would mean every single category on the World Records would have to have two times listed much like it is now for the Stadium records. That would be stupid!
We are talking about DS games specifically, so that means four games only. I don't see how this is stupid if it provides all the information we can provide about who holds the WRs for a given game. As has previously been stated, we can't compare a 3:27 emulator time in BW2 to a 3:34 console time - we simply don't know which is the better run. So we shouldn't arbitrarily choose based on the lower game time.
Luckless wrote:So does this mean, for example, that if someone gets a run in BW2 using a DS, we would have to have a leaderboard for Emulator, one for DS and one for 3DS? Does it mean that in Gold/Silver there would be separate leaderboards for VisualBoyAdvance and BGB and Gambatte?
This is spurious. We are not discussing gen 2 because GB/GBA emulation does not have the same issues about in-game time as DS emulation. We would not separate DS from 3DS because they are both legitimate means of playing the game. Emulator is significantly less legitimate and the rules for emulator should be considered separately from console as a result (even if you end up with the same rules! the important thing is they have different context). For comparison: PS2 speedrunners do not generally allow PS2 emulation, but they do specifically seek out the fastest model of PS2 to get the best possible time.
Luckless wrote:Plus no-one feels segregated from the rest of the runners of the category just because they use a different platform that changes their final time by the smallest margin.
We are not talking about a small margin here. I proved this back when we first discovered the difference - my first emulator run of BW1 with completely shitty performance, including a death, was a 3:26, where my 3DS PB was 3:30.

User avatar
Luckless
Site Admin
Posts: 113
Joined: Fri Apr 11, 2014 3:28 am
Location: UK
Contact:

Re: DS speedrun leaderboards

Post by Luckless » Thu May 22, 2014 10:32 pm

So it seems you want this change for DS games but not for any other games then? The points me and CTM are bringing up are more broad across all games because with only one or two generations of games having this change, people will start questioning the consistency of the leaderboards as a whole. I feel like this argument will be brought up for Gen 2 and Gen 3 games too asking why this was only done for DS games and not those as well. How do we answer them when this happens? Do we set a certain criteria of how much difference between emulator and console there needs to be? If so, we'd have to be careful with what we set it as because we don't want to have to get too specific with defining the exact time difference if a better emulator comes along later that's closer to the criteria.

EDIT: After Vulajin has explained it further I am not thinking like this anymore, I was just asking questions to try to understand this better and now I do.
"I see now that the circumstances of one's birth are irrelevant. It is what you do with the gift of life that determines who you are" - Mewtwo

Image Image Image Image Image

Vulajin
Schoolkid
Posts: 46
Joined: Thu Apr 24, 2014 4:47 am

Re: DS speedrun leaderboards

Post by Vulajin » Thu May 22, 2014 10:39 pm

I feel like you are completing missing the point. In gen 1-3, the IGT is known (to my understanding) to count frames, not real time. As such, when you do a run, even if you're doing it on an emulator that runs faster or slower than console (e.g. 60 FPS vs. console's 60.0949348384whatever), the IGT is not different from what you would get if you did the same run on console.

In gen 4-5, the IGT is known to count real seconds, based on the DS clock. Emulators have less loading time than consoles, so runs on emulator are not only faster in RT, but also IGT. As such, the IGT does not allow us to fairly compare console and emulator runs for DS games specifically.

To turn this into a broader discussion about emulator legitimacy would be to miss the point. DS speedruns have a specific issue that we are discussing here. GB/GBA speedruns do not suffer from the same issues. If you'd like to have a general discussion about the legitimacy of emulators, that should be an entirely separate thread.

User avatar
heero_fred
Youngster
Posts: 70
Joined: Thu Apr 24, 2014 2:37 am
Location: Mexico
Contact:

Re: DS speedrun leaderboards

Post by heero_fred » Thu May 22, 2014 11:07 pm

Vulajin wrote: In gen 4-5, the IGT is known to count real seconds, based on the DS clock. Emulators have less loading time than consoles, so runs on emulator are not only faster in RT, but also IGT.
This may be a little bit of a tangent, but at the same time related to the issue: Does the 3DS and all DS models's clock run at the same speed? if not, pretty much you can have up to 7 different IGT measures (DS Phat, DS Lite, DSi, DSi XL, 3DS, 3DSXL and 2DS) with official consoles alone.

Even if there is a single frame difference between models in a run as long as the DS games, it will add up to 3 to 4 minutes of a difference.

So the question in this case would be? where would you draw the line? would you make a separate leaderboard for each model? or just make a single leaderboard with all times listed that can be sorted by platform used to run it, as well as a note indicating the differences?

Vulajin
Schoolkid
Posts: 46
Joined: Thu Apr 24, 2014 4:47 am

Re: DS speedrun leaderboards

Post by Vulajin » Thu May 22, 2014 11:11 pm

Back-to-back posting because I want to provide some background reading for those who may not be aware of everything being discussed here.

About DS emulation speed: http://pastebin.com/BAMadVRg

About DS Pokémon speedruns: http://pastebin.com/YjKpH8dU

Vulajin
Schoolkid
Posts: 46
Joined: Thu Apr 24, 2014 4:47 am

Re: DS speedrun leaderboards

Post by Vulajin » Thu May 22, 2014 11:13 pm

heero_fred wrote:So the question in this case would be? where would you draw the line? would you make a separate leaderboard for each model? or just make a single leaderboard with all times listed that can be sorted by platform used to run it, as well as a note indicating the differences?
It's like people literally don't read my posts. We don't separate console models because they are all equally legitimate ways of playing the game.

(edit for clarity) We know that DS Phat runs gen 5 games faster than DSi/3DS. We're not entirely sure how DS Lite compares to DS Phat, but we know that it is slower than emulator and faster than DSi. No one has tested 2DS or 3DS XL to my knowledge.

User avatar
heero_fred
Youngster
Posts: 70
Joined: Thu Apr 24, 2014 2:37 am
Location: Mexico
Contact:

Re: DS speedrun leaderboards

Post by heero_fred » Thu May 22, 2014 11:24 pm

Vulajin wrote: It's like people literally don't read my posts. We don't separate console models because they are all equally legitimate ways of playing the game.
I understand that part and why would you not separate them, however

You also say that one of the reasons is because the quality of the run cannot be comparable thanks to the IGT not being consitant between Console and Emu thanks to how it works, but at the same time the way the IGT works can make 2 different runs not comparable if their clocks run at a different speed, there can be the same or worse difference.

So even if all console models are legitimate ways of playing the game, they can be different, would you say a run with a better console time in a model with faster loading times/slower clock is the same as one with slow loading/fast clock?

Vulajin
Schoolkid
Posts: 46
Joined: Thu Apr 24, 2014 4:47 am

Re: DS speedrun leaderboards

Post by Vulajin » Thu May 22, 2014 11:29 pm

I actually do not even understand the point you are raising about clock speed. Here are my three reactions to this:

1) I'm not aware of any differences in the rate the clock advances between DS models.
2) I'm not sure how one would even test for that.
3) I'm not aware of any evidence that leads anyone to believe this is a thing.

The issues we are aware of are that different models have different load times. I have never heard of "slow loads/fast clock" being a thing, and have no reason to believe that it is.

User avatar
heero_fred
Youngster
Posts: 70
Joined: Thu Apr 24, 2014 2:37 am
Location: Mexico
Contact:

Re: DS speedrun leaderboards

Post by heero_fred » Thu May 22, 2014 11:47 pm

Fair enough, I just used a clock as an example because you said the IGT counts real seconds, so if there is (note the IF) a difference it would make a huge impact in a 3 hours run, but it can be replaced with loading times with the same reason, hell that is why you want to separate console and emu in the first place, even a frame of a difference every loading zone make a huge of a difference in a long run such as these.

Granted, emu has an advantage over all consoles, but at the same time :
Vulajin wrote: We know that DS Phat runs gen 5 games faster than DSi/3DS
One model has an advantage over the others, Would you also want DS Phat runs to be separated from 3DS runs? DS Phat has a loading advantage, so a run on it can haveworse excecution than a run on 3DS and still have a better time. [Main reason version/console difference is one of (if not) the biggest BS in speedrunning]

That is why I suggesting a list that can be sorted out by both time and platform used, if you preffer it can be set as default for console or a specific console version, similar to the OoT leaderboards in ZSR

User avatar
Luckless
Site Admin
Posts: 113
Joined: Fri Apr 11, 2014 3:28 am
Location: UK
Contact:

Re: DS speedrun leaderboards

Post by Luckless » Fri May 23, 2014 12:16 am

heero_fred wrote:One model has an advantage over the others, Would you also want DS Phat runs to be separated from 3DS runs? DS Phat has a loading advantage, so a run on it can haveworse excecution than a run on 3DS and still have a better time. [Main reason version/console difference is one of (if not) the biggest BS in speedrunning]
I think what Vulajin is saying is both DS Phat and 3DS (as well as the other types of DS) are all legitimate so if someone wants to get really picky about that stuff then they will have to get the better version of DS if they want pinpoint optimization of their time. It doesn't mean the consoles have to be separated or a certain type of DS has to be disallowed or anything, it's just something for the really picky people. (I'm trying not to use "autistic" to describe them)

Emulators vs Consoles is different because there is a big enough gap between the times. The versions of consoles, although they are as you say technically different, are not so much of an issue because there isn't a considerable enough difference and I think when you say it'd make a "huge impact" you are overestimating it.

EDIT: I do really like the suggestion of doing the leaderboards in a similar way to ZSR if that's possible, their system allows people to view both the full unseparated leaderboard along with the separate leaderboards at the same time. However, I would imagine it would be quite difficult to do and with the way the PSR leaderboards are set up right now with it pulling things from the wiki I don't see how it could be done without people having to edit 2 (or 3 if it's WR) leaderboards just for 1 run.
"I see now that the circumstances of one's birth are irrelevant. It is what you do with the gift of life that determines who you are" - Mewtwo

Image Image Image Image Image

Vulajin
Schoolkid
Posts: 46
Joined: Thu Apr 24, 2014 4:47 am

Re: DS speedrun leaderboards

Post by Vulajin » Fri May 23, 2014 12:32 am

In an ideal world, all legitimate means of playing the games would work identically. Sadly, there are differences between the consoles, but I don't want to be in the business of disallowing legitimate means of playing the game to force a completely level playing field. Maybe someone would go and play on a faster console only for the purpose of getting a faster time, but frankly I don't think there are any speedrunners like that in our community and who cares if they do anyway?

The main distinction to me is effectively illegitimate vs. legitimate. If streaming DS consoles were cheaper and less single-purpose, I'd probably be all for restricting emulators from WR consideration and not including them on the leaderboard. Unfortunately, as a practical matter, emulators are necessary for most people to be able to feasibly stream DS speedruns. As such, I'd like to give emulator runners a reasonable blessing (separate WR consideration, a leaderboard of their own), but at the same time not implicitly or explicitly compare their times against console times.

luckytyphlosion
Pokémon Trainer
Posts: 182
Joined: Thu Apr 24, 2014 2:26 am

Re: DS speedrun leaderboards

Post by luckytyphlosion » Fri May 23, 2014 12:53 am

heero_fred wrote:One model has an advantage over the others, Would you also want DS Phat runs to be separated from 3DS runs? DS Phat has a loading advantage, so a run on it can haveworse excecution than a run on 3DS and still have a better time. [Main reason version/console difference is one of (if not) the biggest BS in speedrunning]
What do you mean by "a loading advantage"? Do you mean when you power on the 3DS, you have to wait a bit before you can load it, or is it a "frame" difference? Does this only apply to DS Phat or all models of the DS (Lite, DSI, XL)?

What really bugs me about console differences is in Explorers of Sky, for some reason you can't soft reset (L+R+Start+Select). This means that (even though the rules state "Single Segment"), you would have to turn off the DS to reset the Game. Since turning off the game and turning it back on with a regular DS (and Lite) is faster than going to the Home Menu and Re-Loading the game, the DS would obviously be faster. While this may not apply to Any% rules, if "Any% Postgame" (Beating Darkai) would be a category, this would be much more prevalent since there are a lot of extremely hard dungeons in the Late Game, with Pokemon that can instantly destroy you. (Any% Postgame is equivalent to Rumble Advanced% or redbaring Squirtle% (I guess)) Would this mean that DS Phat would have an advantage over resets, and should be a separate category?

Note that this is just an example on comparing Console differences, and isn't supposed to be focused on only one game.

You know, you can also do TriHard record your DS Screen with a Webcam strats
Jack-Flys are OpieOP.

"GO ASK SANQUI" - luckytyphlosion whenever someone asks about how something in Pokémon Red/Blue works. (this means you exarion Kappa)

anirocks99
Schoolkid
Posts: 30
Joined: Thu Apr 24, 2014 12:27 am
Location: California, United States

Re: DS speedrun leaderboards

Post by anirocks99 » Fri May 23, 2014 1:20 am

Vulajin wrote:In an ideal world, all legitimate means of playing the games would work identically. Sadly, there are differences between the consoles, but I don't want to be in the business of disallowing legitimate means of playing the game to force a completely level playing field. Maybe someone would go and play on a faster console only for the purpose of getting a faster time, but frankly I don't think there are any speedrunners like that in our community and who cares if they do anyway?

The main distinction to me is effectively illegitimate vs. legitimate. If streaming DS consoles were cheaper and less single-purpose, I'd probably be all for restricting emulators from WR consideration and not including them on the leaderboard. Unfortunately, as a practical matter, emulators are necessary for most people to be able to feasibly stream DS speedruns. As such, I'd like to give emulator runners a reasonable blessing (separate WR consideration, a leaderboard of their own), but at the same time not implicitly or explicitly compare their times against console times.
Vulajin I am seriously starting to get a little annoyed by what you are saying, you act as if whoever gets wr on an Emulator doesn't necessarily need to try hard to get one, and that its "legitimacy" doesn't exist. Getting a record is much harder than that. Even though there may be a time disadvantage for Console to Emu, That doesn't mean Console people can't take a record back or that Emulator People will always be better than Console. For example, on the Oot Any% Leaderboards, Cosmo aka First Place used the IQue. The IQue in the sense is a mock controller with a flash card loaded with the game, aka just like an emulator. There is a slight advantage with no lag, but a drawback with a quick reset. Could anyone who doesn't play IQue take it back, it would be hard but definitely. Does ZeldaSpeedruns however put all different consoles in different leaderboards, Of Course Not. And Neither Should We. If WR is such a bother, just have a comment section added like the Galaxy Leaderboards such as "Console/Emu WR." If there isn't any other drawbacks and would like to end this discussion pronto.

User avatar
Shenanagans
Site Admin
Posts: 116
Joined: Fri Apr 11, 2014 3:59 am
Location: Ohio, USA
Contact:

Re: DS speedrun leaderboards

Post by Shenanagans » Fri May 23, 2014 3:41 am

anirocks99 wrote:
Vulajin wrote:In an ideal world, all legitimate means of playing the games would work identically. Sadly, there are differences between the consoles, but I don't want to be in the business of disallowing legitimate means of playing the game to force a completely level playing field. Maybe someone would go and play on a faster console only for the purpose of getting a faster time, but frankly I don't think there are any speedrunners like that in our community and who cares if they do anyway?

The main distinction to me is effectively illegitimate vs. legitimate. If streaming DS consoles were cheaper and less single-purpose, I'd probably be all for restricting emulators from WR consideration and not including them on the leaderboard. Unfortunately, as a practical matter, emulators are necessary for most people to be able to feasibly stream DS speedruns. As such, I'd like to give emulator runners a reasonable blessing (separate WR consideration, a leaderboard of their own), but at the same time not implicitly or explicitly compare their times against console times.
Vulajin I am seriously starting to get a little annoyed by what you are saying, you act as if whoever gets wr on an Emulator doesn't necessarily need to try hard to get one, and that its "legitimacy" doesn't exist. Getting a record is much harder than that. Even though there may be a time disadvantage for Console to Emu, That doesn't mean Console people can't take a record back or that Emulator People will always be better than Console. For example, on the Oot Any% Leaderboards, Cosmo aka First Place used the IQue. The IQue in the sense is a mock controller with a flash card loaded with the game, aka just like an emulator. There is a slight advantage with no lag, but a drawback with a quick reset. Could anyone who doesn't play IQue take it back, it would be hard but definitely. Does ZeldaSpeedruns however put all different consoles in different leaderboards, Of Course Not. And Neither Should We. If WR is such a bother, just have a comment section added like the Galaxy Leaderboards such as "Console/Emu WR." If there isn't any other drawbacks and would like to end this discussion pronto.

First of all the iQue is a Nintendo official console, and therefore is not comparable to emulators. Emulators also pose a much larger advantage than the iQue (iQue saves like 10-20 seconds Emu saves MULTIPLE minutes)

When I look at this issue I think about the sm64 community. Virtual console (while nintendo official) is FASTER than the original 64 games, Yet almost every top runner runs on original console. They simply split the times into separate categories (see here: http://mariospeedruns.blogspot.com/p/su ... io-64.html ).

This is why I like Vulajin's suggestion. If we simply list the times as different categories it would allow runners who want to play gen 5 and compete for times to have a way to achieve this. The biggest problem I see with this is about which run is the "World Record" Would it be fair to call the fastest console time the "WR" even if an emu time is clearly better. Would we just have 2 categories on the leaderboards, one for Emu and one for Console? Or would we find a way to acknowledge the best run as the sole Record? This would be a question that needs addressed if we take this route.

Overall I am greatly in favor of at least adding a emulator section to the leaderboards, because it will allow new runners to have a larger variety of categories they can pick up.
Feel free to follow me for all your Pokemon Resetting Needs!

Image Image Image

Vulajin
Schoolkid
Posts: 46
Joined: Thu Apr 24, 2014 4:47 am

Re: DS speedrun leaderboards

Post by Vulajin » Fri May 23, 2014 8:19 am

anirocks99 wrote:Vulajin I am seriously starting to get a little annoyed by what you are saying, you act as if whoever gets wr on an Emulator doesn't necessarily need to try hard to get one, and that its "legitimacy" doesn't exist. Getting a record is much harder than that. Even though there may be a time disadvantage for Console to Emu, That doesn't mean Console people can't take a record back or that Emulator People will always be better than Console.
If that's what you're talking away from my posts, you're misunderstanding me. Let me set forth a few basics of my position:

1) The vast majority of people who want to speedrun DS games do own a DS and a copy of the game in question, and therefore they are at least technically capable of running such games on a legitimate console.
2) The vast majority of people who want to speedrun DS games do not own a capture-capable DS or a capture-capable 3DS, and buying one simply to stream one or maybe two games is not feasible (without, for example, donations).
3) Emulators are less legitimate than consoles.
4) Official consoles are all equally legitimate.
5) Speedrunners these days want to stream, or at least record their runs to upload to the internet.
6) I want more people to speedrun DS games.
7) Therefore I think we must permit emulators in some capacity.
8) But emulators have a major time advantage, such that for a player of the same skill level, their best possible emulator time is several minutes faster than their best possible console time.

Point 8 should clarify what you seem to be upset about. It is trivially easy for a particular individual to get a lower time on an emulator, because of the huge advantage. Getting to the point where I could accomplish a 3:30 on 3DS in BW1 took me a lot of work. After I got to that point, from the same baseline, I did a low-effort emulator run with much worse performance and got a 3:26. However, a lot of overall work still went into accomplishing that emulator time.

My implication is not that someone who can get an emulator on 3:26 did not work to achieve that time. My implication is that the taint of the emulator advantage means that that run should not be directly compared against console times. Even if someone comes along and gets 3:20 or lower on emulator, which I think would almost certainly be a superior run to mine, it should still be considered separately because we just can't be certain.

Unofficial emulators absolutely should not be allowed a competitive advantage over consoles. Every other speedrunning community that I'm aware of bans unofficial emulators [and, in some cases, official emulators] that have speed advantages. In the interests of practicality, we should allow emulator runs to be considered independently, but we should not ever be in the position where a DS emulator run is given primacy over a console run, regardless of what the times are.

(Again, please note that the preceding paragraph implies nothing about GB/GBA emulation, since to my knowledge we have established that such emulators are accurate for game time and therefore comparing against real console game time is valid.)

anirocks99
Schoolkid
Posts: 30
Joined: Thu Apr 24, 2014 12:27 am
Location: California, United States

Re: DS speedrun leaderboards

Post by anirocks99 » Sat May 24, 2014 3:31 am

Vulajin, I apologize for my lack of background towards my statement. I do agree with you about time advantage and its supremacy over Console, but at the same time, It varies in many cases such as slow computers = slow runs and Mega Speed = Mega Play. Seperating it would be a good idea if it were not for the fact that the world isn't perfect, not everyone is in the financial stability to get ds capture like you, and the fact that PSR just isn't popular enough to warrant different leaderboards. If it were a perfect world, I would support the banning of Emus with time advantage, but for now due to lower popularity, having 2 different boards just wouldn't make sense. And BTW Vulajin, what exactly did you not do great in your "effortless" run cause 3 2 6 on even Emu is a damn good time TriHard

Sanqui
Jr. Trainer
Posts: 93
Joined: Thu Apr 24, 2014 10:05 am
Contact:

Re: DS speedrun leaderboards

Post by Sanqui » Sat May 24, 2014 9:36 pm

I agree with Vulajin

Also, this is an interesting and relevant blog post: http://adshomebrewersdiary.blogspot.cz/ ... nswer.html
We can see that DMA takes near-zero time on emulator while it takes pretty long on DS Lite, presumably that's what Pokémon is using.
I am the person who has ruined Gold Glitched forever and seconhandedly, Red NSC too
Ask me to make a dumb romhack for you!
Image

Jul, a board.

User avatar
Dabomstew
Site Admin
Posts: 162
Joined: Fri Apr 11, 2014 3:26 am

Re: DS speedrun leaderboards

Post by Dabomstew » Mon May 26, 2014 5:38 am

Alright well I thought I would make some mockups of what the proposed changes to the current leaderboard pages we have now could be, along with some fundamental assumptions I made here.

WRs page (relevant section):

Image

The main fundamental assumption I made here is that an emulator time should only be shown if the "emulator WR" is actually faster than the console time. I think there are two things we can say fairly objectively about this situation - emulator is faster than console, and the difference in times is at least a minute in recognised categories. Therefore emulator times that are higher than, or even tie, the console WR are objectively worse than the console WR and should not really be considered a WR by anyone.

There is a small problem here in that the only console Plat time I could find is MHM's, and that doesn't have a video anymore - but seeing that MHM and potentially Viskiv too are going to start console runs really soon, this shouldn't be a problem by the time these changes are implemented.

Times page:

Image

This is one of two proposals, with the other being that we have a combined table but put all the console times at the top. Similar to the WR rule, there probably isn't much point in a single runner having both a console and an emulator time listed if their console time is equal or better, but that's really up to the individual runner.

Thoughts on where we should go from here?

Locked

Return to “Policy Changes”

Who is online

Users browsing this forum: No registered users and 5 guests