Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Discuss policy guidelines for the community and whether something needs to be changed or not.
User avatar
iMAX1UP
Bug Catcher
Posts: 12
Joined: Mon Aug 18, 2014 7:36 pm

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by iMAX1UP » Sat Jul 16, 2016 8:00 am

astralslide wrote:As far as Gen 6 goes, in the case RTA is put in place I would advocate the timer start to be upon pressing the Start Game and end with the fadeout after the Congratulations page showing your team after beating the E4. That would save 10 minutes of credits and a pretty much superfluous fight. The Japanese timing rules end the timer when the THE END label appears after beating Brendan after the credits, but as it is gradual there's no precise way to tell when exactly to split. Neither of these two are the IGT cutoff time, which is after a soft reset when the game autosaves after the Brendan cutscene. I don't really mind having to sit through the credits and using the JP timings, but there have been complaints before regarding the matter.
I dont seem to have a problem on my timing in RTA for Gen 6 X/Y & ORAS. I have the start/end timer exactly so it matches to be precise with the IGT (that also includes the seconds). How do I know this? Out of so many runs I've done I've payed attention when I was off by seconds to getting a PB and based from that I started correcting it & splitting for precisement. There's a reason why Japanese end timers ends at THE END on ORAS, have you thought about that it could be that is precise there? As far as where to start the timer, simple, when it says BEGIN GAME ^_^

As far as changing to RTA overall, while is a good idea to change it for some generations of Pokemon, it doesnt mean is a good idea to change them for all, they dont all work in same terms. Not going to cover points to why is not a good idea since major runners in the pokemon community doesnt give a fuk of what i say. It would be nice to see RTA display on another row along with the IGT just for the purpose of knowing the differences of RTA along runners but not to make it the official time that matters of whos run is better/faster/WR etc for many reasons, people have covered some of those reasons as well. Overall, we are humans and we have errors, by changing to RTA do you really think a run is going to be more accurate/precise when we humans dont even know when to split correctly? LOL
♥ CHECK ME OUT ♥
Image Image Image Image Image

User avatar
RXFADEZ
Pokémon Trainer
Posts: 121
Joined: Sun Nov 09, 2014 6:32 pm
Contact:

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by RXFADEZ » Sat Jul 16, 2016 8:28 am

entrpntr wrote:The Gen 1 thru Gen 3 Pokémon games have in-game timers that are very accurate in WHAT they measure (I have limited knowledge on Gen 4+ so can't speak to them). Yellow and Gen 3 have some notable distortions with respect to real-time, but these are minor distortions and have no significant impact on the way the games are optimally run. Not many non-Pokémon games can claim this, and most people are missing what a luxury this is.
But WHAT they measure isn't the fastest run. The measure the best manipulation of In-Game Time.

entrpntr wrote:I've brought up this reddit thread several times, as it brings several additional considerations to the table (including a LiveSplit developer suggesting console games use an in-game timer if one's available). There are a handful of known cases where LiveSplit has been inaccurate up to 10 seconds per hour due to various PC clocks being inaccurate. Even with the atomic clock functionality introduced in 1.6, LiveSplit needs to "learn" from the PC clock before it shows an accurate time (source). I'm aware of at least one unresolved issue, even with the atomic clock functionality activated. I don't think this issue has been fixed yet, but someone correct me if I'm wrong.
The LiveSplit inaccuracy is basically a moot point. Mods are there to check it's accurate and there is nothing stopping a run from being retimed if the user accidentally left a PC game running in the background or something else to alter LiveSplit's accuracy.

User avatar
Chromatrope
Preschooler
Posts: 5
Joined: Sat Jul 16, 2016 12:14 pm

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Chromatrope » Sat Jul 16, 2016 12:59 pm

I am personally heavily in favor of RTA. It offers higher granularity and avoids ties that way, invokes more motivation to run because you don't need to be up to an entire minute faster to PB at all, (presumably) allows for S+Q, which is a tremendous tool for beginners and in certain situations like the Kalos E4 even intermediate/advanced players, opens up the door for more manip strats and removes the burden of 'invalidating' your run by using a safety save. Not to mention everyone uses splits already and compares their own runs RTA, so we'd really just be getting rid of a weird double standard.

I don't agree with having IGT shown as well on the boards, or as primary, because we'd either have IGT be screwed up compared to RTA with S+Q strats, or we'd have to senselessly ban S+Q in spite of its undeniable utility, in which case IGT would still do nothing except tell us less about the time.

My general idea for RTA timing is "Start where IGT starts, end at the moment IGT stops ticking/is shown."

This does not punish runs that use IGT strats like depositing and makes the transition easier. For Gen 4 timing would start on movement, while for Gen 6 timing would start at the start of the game. Timing for X&Y, for example, would end after the game saves right after the credits (since the credits count as IGT and resetting puts you back at the hall of fame).
iMAX1UP wrote:Overall, we are humans and we have errors, by changing to RTA do you really think a run is going to be more accurate/precise when we humans dont even know when to split correctly? LOL
If we define start and endpoints properly and check whether or not splitting is done correctly in verification, compensating as necessary, this is not an issue.
Perpetually lazy and uninspiring.

Exarion
Cooltrainer
Posts: 249
Joined: Thu Jul 17, 2014 10:07 am

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Exarion » Sat Jul 16, 2016 6:13 pm

I am in favor of RTA. We now have evidence that the in-game timer for Gens 1-3 does not count all frames of gameplay. This runs contrary to the definition of a speedrun: to beat a game as fast as possible. This would be mostly insignificant if the excluded frames came in the form of unavoidable lag; however, this is not the case. In Gen 3, for example, runners often must choose between two forms of spinner manipulation. One form, known as bag manipulation, creates many lag frames that don't count toward IGT. All other spinner manips have little to no lag. Thus, runners who abuse bag manipulation will see a greater difference between IGT and RTA. My own runs have seen a variance of 20 seconds in the IGT-RTA difference.

Being able to abuse IGT is a problem by itself, but to even reach that stage of optimization, runners would need to know which frames count and which don't. It is possible to acquire this knowledge using a tool, but this is an unnecessary hassle, especially for newer runners. Also, runners are currently unable to know their time during a run, which makes certain decisions unnecessarily difficult. Let's say a Yellow runner defeats Lance at 1:54:08 RTA. A Champion split with standard strats takes about 2:20. Some runs with a final RTA ending in :28 have had an IGT that's 2 minutes faster (in this case, a 1:54), while others have been 1 minute faster (1:55). Without knowing the exact IGT, the runner may be forced to use an extremely risky strat to ensure the 1:54.

These types of examples have played out in several actual runs on the Gen 1-3 leaderboards. In one of my runs, I played safely because I thought I had secured a certain IGT (3:26). I got a 3:25 instead, and now I'm left wishing I would have played differently. This would be avoided entirely with a switch to RTA.

RTA would also allow save & quit strats, which have many benefits and no downsides. It's also counterintuitive to disqualify a speedrun for resetting; you wouldn't say a game hasn't been beaten just because it included a reset.

The issue of inaccurate timers is one of convenience, not accuracy. There are many accurate timers available to the public, and if someone's LiveSplit timed a run inaccurately, it could be retimed by anyone with a functioning timer (usually a mod, or the runner with a different timing device). Other communities have had little to no problem with this.

As a final note, my knowledge is limited to Gens 1-3, so my opinion should be taken with a grain of salt for Gens 4+. Also, while certain games have particularly bad in-game timers (Yellow, Sapphire, Emerald, FireRed), others are close to being accurate. It might make sense to keep IGT for Gen 2 and switch to RTA for Gens 1 and 3. We could also keep IGT for Red and not Yellow, although it makes more sense to change both (since RTA makes sense for both, while IGT makes sense for only one).
Twitch channel: http://www.twitch.tv/exarionu
Pokemon Red speedrunning guide: http://pastebin.com/CkVA5yvJ

User avatar
Decon082
Jr. Trainer
Posts: 90
Joined: Thu Jul 31, 2014 3:47 am
Contact:

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Decon082 » Sat Jul 16, 2016 6:24 pm

here's an idea i haven't seen mentioned before:

assuming we make a switch from IGT to RTA in an individual category, we should keep IGT and RTA both visible on the leaderboard.

because you can now see the difference between the old IGT and the new RTA, you can easily identify any cases where someone has faster RTA but slower IGT, or vice versa, and learn more from seeing those outliers.

by doing this, you can possibly learn of inaccuracies in the in-game timer that you didn't know about, or have a reason to inspect those runs more thoroughly to see where they might have got messed up in timing.
I am the person who has ruined Red "Glitchless" forever and seconhandedly, Blue NSC Single Segment too
http://www.twitch.tv/iateyourpie/subscribe
Image

MKDasher
Schoolkid
Posts: 41
Joined: Fri May 02, 2014 5:31 pm

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by MKDasher » Sat Jul 16, 2016 7:30 pm

First of all, I find more important to analyse the cons and pros of switching to RTA for each gen specifically. I want to talk about the 2 gens I've worked on more which are gen 3 and gen 4. I also feel to include some information about RTA vs IGT timing on this gens before giving my opinion.

- Gen 3:

- IGT properties:

- IGT doesn't count lag frames, and lag frames happen really frequently on this game, as you can see on this video.

- IGT shows seconds, as opposed to other games that only shows hour+minutes.

If we keep IGT:

- IGT timing clearly times differently than RTA, so runners would need to be more aware of what saves time and what doesn't. I don't think being able to manipulate IGT is the correct term for this, since IGT just times stuff differently, but it's consistently different. A problem for this is that if a runner wants to compare 2 strats, RTA comparison won't be accurate.

- Final time won't be known until the runner checks IGT. The variance between IGT and RTA is big enough, specially on firered round 2.

If we switch to RTA:

- It should be important to know when we will start timing. I suppose it will be when we gain control of the character. The problem I see is, people who wants to RNG manip will need to set 2 timers, one for the RNG and other for the RTA timing. That can somehow be annoying for the runner, or actually for the moderator, since some runners may just use a timer for the RNG and the runs would have to be retimed later.

- There might be a possibility that abra manip is consistent with s&q, but I haven't looked into it a lot yet.

- Gen 4:

- IGT properties:

- IGT is accurate.

- IGT doesn't show seconds.

If we keep IGT:

- More ties can happen since IGT doesn't show up seconds.

If we switch to RTA:

- If we start timing from character control, we would maybe have the same issue than gen 3 (double timing). If we start from reset, then manipulating for late seeds would have a clear disadvantage.
- Ability to S&Q may allow different manip strats, I haven't looked into it a lot yet though.

As a whole, I'm in favour of switching to RTA on gen 3, but I feel slightly more comfortable with IGT on gen 4, but I wouldn't care about going to RTA there as well as long as RTA starts timing the run in a proper place. Also I'm assuming RTA would mean having save and quit, else just remove those arguments from the list.

Also I want to point out (even if it's obvious), people are saying about some runs turning worse / better with the RTA switch. Allowing S&Q will make changes irreversible, so I really hope a good decision is made.

Keizaron
Site Admin
Posts: 273
Joined: Wed Jul 16, 2014 6:13 pm
Location: Spokane, WA
Contact:

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Keizaron » Sat Jul 16, 2016 7:58 pm

Decon082 wrote:here's an idea i haven't seen mentioned before:

assuming we make a switch from IGT to RTA in an individual category, we should keep IGT and RTA both visible on the leaderboard.

because you can now see the difference between the old IGT and the new RTA, you can easily identify any cases where someone has faster RTA but slower IGT, or vice versa, and learn more from seeing those outliers.

by doing this, you can possibly learn of inaccuracies in the in-game timer that you didn't know about, or have a reason to inspect those runs more thoroughly to see where they might have got messed up in timing.
This was actually a proposal in the Discord, but to be fair that channel is filled with like 302948 messages now.
MKDasher wrote:- It should be important to know when we will start timing. I suppose it will be when we gain control of the character. The problem I see is, people who wants to RNG manip will need to set 2 timers, one for the RNG and other for the RTA timing. That can somehow be annoying for the runner, or actually for the moderator, since some runners may just use a timer for the RNG and the runs would have to be retimed later.
My idea when it came to this was people could start using EonTimer a little more frequently. While there is the annoyance of running another program, we could make the "true" timer more obvious this way.

User avatar
RXFADEZ
Pokémon Trainer
Posts: 121
Joined: Sun Nov 09, 2014 6:32 pm
Contact:

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by RXFADEZ » Sat Jul 16, 2016 8:17 pm

Keizaron wrote:My idea when it came to this was people could start using EonTimer a little more frequently. While there is the annoyance of running another program, we could make the "true" timer more obvious this way.
I agree this would be better, but many people dislike EonTimer. Another suggestion you made was for non-EonTimer users to try and split when they gain character control to make mods/themselves have an easier job.

User avatar
iMAX1UP
Bug Catcher
Posts: 12
Joined: Mon Aug 18, 2014 7:36 pm

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by iMAX1UP » Sat Jul 16, 2016 8:29 pm

As stated on previous post, having IGT & RTA displayed helps out in different ways. Main reason is for the difference of accuracy between both times. In Gen 6 is really not a big deal of change to RTA as long as the time is being split si accurate start & end. The way it has been at least for X/Y to be accurate is such a way that you start your split when it says Begin Game and ended right after the 2nd cutscene of HOBO AZ. I didnt come up with the end split, other runners before me did, but i did change the start split for the reason to be as accurate/precise with the IGT. In other words if you split start/end precisely at those points you get the same time with the IGT including the seconds even though 3DS doesnt displayed. As far as ORAS, the start is the same but the end split is when it says THE END, again for purpose of accuracy with IGT.

I cannot speak about other Gens in terms of how they should be split start/end, those should be decided with the people that know about them. In case RTA change is decided then, we gotta decide when to split and for what good reason(not because i hate/like it here/there). I have stated Gen 6 start/end split and is not because I've decided that way but for the purpose of being accurate with the IGT and it helps knowing if a run is going to PB or not. I dont know if others Gen can do that, whether they can or not it helps knowing the difference of IGT & RTA as long as being displayed.
♥ CHECK ME OUT ♥
Image Image Image Image Image

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

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Shenanagans » Sat Jul 16, 2016 8:52 pm

If we have both RTA and IGT on the leaderboards runs with S+Q will have a large IGT advantage (not that that really matters if we judge based on RTA). It would be silly to time RTA and not allow S+Q imo, so showing both on the leaderboards may look silly.
Feel free to follow me for all your Pokemon Resetting Needs!

Image Image Image

Keizaron
Site Admin
Posts: 273
Joined: Wed Jul 16, 2014 6:13 pm
Location: Spokane, WA
Contact:

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Keizaron » Sat Jul 16, 2016 8:58 pm

iMAX1UP wrote:As stated on previous post, having IGT & RTA displayed helps out in different ways. Main reason is for the difference of accuracy between both times. In Gen 6 is really not a big deal of change to RTA as long as the time is being split si accurate start & end. The way it has been at least for X/Y to be accurate is such a way that you start your split when it says Begin Game and ended right after the 2nd cutscene of HOBO AZ. I didnt come up with the end split, other runners before me did, but i did change the start split for the reason to be as accurate/precise with the IGT. In other words if you split start/end precisely at those points you get the same time with the IGT including the seconds even though 3DS doesnt displayed. As far as ORAS, the start is the same but the end split is when it says THE END, again for purpose of accuracy with IGT.
I disagree with having both listed. The leaderboard would look very odd if one person had a 1:54 in a game with an RTA of 1:54:50, yet somebody with a 1:53 had an RTA of 1:56:00 because of S+Q. I'm of the camp that if there's a switch, we only display what the timing method is, and if the runner really wants to show the other timing method, they can put it in their run comments upon submitting.

User avatar
RXFADEZ
Pokémon Trainer
Posts: 121
Joined: Sun Nov 09, 2014 6:32 pm
Contact:

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by RXFADEZ » Sat Jul 16, 2016 10:30 pm

Keizaron wrote: if the runner really wants to show the other timing method, they can put it in their run comments upon submitting.
^this is a very good point and I think it should be highlighted more.

User avatar
astralslide
Preschooler
Posts: 8
Joined: Mon Sep 15, 2014 3:58 pm

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by astralslide » Sun Jul 17, 2016 12:18 am

I stand by my opinion that regardless of the technical hurdles and potential for abuse (even though many other games have always been using RTA with no issues whatsoever), the lowered barrier of entry to speedrunning a Pokémon game granted by allowing S+Q should outweight any con to switching to RTA. It's brutally punishing to die to Red in Gen 2 and HGSS for example, as it's pure RNG and if you did a perfect run and died there under IGT to submit it you would have to go all the way back from the Pokémon center losing so much time the entire run is wasted. I simply don't think that's fair to the player or healthy to the community in the long term.

Any person going for WR should be striving to get the best time possible and I don't think doing safety saves would come into the equation, at least not often, so I don't see this as a granularity issue for the top runs, but rather as a way to make speedrunning more fair and fun for newer runners.

I don't think anyone would question the integrity of the runs already timed with RTA that are in the leaderboards for glitched categories for example, evne those done on emulator. I don't see why would it be a real issue to apply it to glitchless. The Western PSR community was founded on the premise that runs should be timed on IGT, and because of them being timed on IGT, it was implied that runs should always be single segment, but nobody questioned why it should be IGT in the first place, especially for a game that has an IGT not accurate to the second. This isn't Sonic 2, where the IGT total of all levels to get the final time makes sense, and it has a timer down to the second. A Pokémon run is continuous from start to finish, you are in direct control of the action during the entire run.

I agree with iMAX, for Gen 6 whether using RTA or IGT doesn't matter at all since the 3DS uses the system timer instead of a frame counter so it's more or less accurate and I don't really mind using the Japanese RTA rules, or keep using IGT in case IGT is kept. But for other gens I understand that's not always the case, so I don't see a reason to really keep forcing IGT on them.
Last edited by astralslide on Sun Jul 17, 2016 12:31 am, edited 2 times in total.

User avatar
Decon082
Jr. Trainer
Posts: 90
Joined: Thu Jul 31, 2014 3:47 am
Contact:

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Decon082 » Sun Jul 17, 2016 12:25 am

Keizaron wrote:I disagree with having both listed. The leaderboard would look very odd if one person had a 1:54 in a game with an RTA of 1:54:50, yet somebody with a 1:53 had an RTA of 1:56:00 because of S+Q. I'm of the camp that if there's a switch, we only display what the timing method is, and if the runner really wants to show the other timing method, they can put it in their run comments upon submitting.
true, i completely agree here. but i think in our current state when the category initially switches over, there won't be any S+Q runs on the leaderboards yet. since all of the runs already there will have IGT and will be single segment, there shouldn't be any disparities like the one you mentioned.

obviously as more runs get submitted, this is likely to change though, but until then i think having both visible would be beneficial for finding possible timing errors like i said. this is why i am in favor of keeping both assuming a switch to RTA, at least for the beginning. it wouldn't be hard to go in later on and remove the IGT column once more S+Q runs get onto the leaderboards.
I am the person who has ruined Red "Glitchless" forever and seconhandedly, Blue NSC Single Segment too
http://www.twitch.tv/iateyourpie/subscribe
Image

Keizaron
Site Admin
Posts: 273
Joined: Wed Jul 16, 2014 6:13 pm
Location: Spokane, WA
Contact:

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Keizaron » Sun Jul 17, 2016 12:55 am

Decon082 wrote: true, i completely agree here. but i think in our current state when the category initially switches over, there won't be any S+Q runs on the leaderboards yet. since all of the runs already there will have IGT and will be single segment, there shouldn't be any disparities like the one you mentioned.

obviously as more runs get submitted, this is likely to change though, but until then i think having both visible would be beneficial for finding possible timing errors like i said. this is why i am in favor of keeping both assuming a switch to RTA, at least for the beginning. it wouldn't be hard to go in later on and remove the IGT column once more S+Q runs get onto the leaderboards.
I mean, the mods could just put the IGT in the comments anyway. I rather not clutter the boards with a pointless column that might be removed in the future when we can just put it in the comments.

ExtraTricky
Bug Catcher
Posts: 21
Joined: Thu Apr 24, 2014 12:39 am

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by ExtraTricky » Sun Jul 17, 2016 2:21 am

MKDasher wrote:Also I want to point out (even if it's obvious), people are saying about some runs turning worse / better with the RTA switch. Allowing S&Q will make changes irreversible, so I really hope a good decision is made.
I think we should try to avoid the mindset that allowing a new technique is irreversible. If active runners have tried both categories and prefer no S+Q, it makes sense to have a leaderboard for no S+Q. But the mere possibility of a significant route change shouldn't be an argument against allowing S+Q. Managing people's expectations should help. People might (reasonably) expect that their run for "Any% Glitchless" stays valid for the "Any% Glitchless" category, but they shouldn't expect that their "Any% Glitchless" run is valid for "Any% Glitchless No S+Q" if they used S+Q.

gifvex
Bug Catcher
Posts: 10
Joined: Wed Nov 04, 2015 3:55 am

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by gifvex » Sun Jul 17, 2016 4:13 pm

I think RTA is cool. I have enjoyed that timing method in gen 1's glitched categories, as every little time-saving discovery is worth learning about. Some of those discoveries we found were viable for glitchless runs, but since that category uses IGT, 2 seconds mattered less when trying to break a minute barrier (though top runners did appreciate the seconds).

I would like folks not to sacrifice too much when deciding start/stop points in an attempt to perfectly preserve old IGT runs or create consistency across gens. Gen 1 has menuing techniques during the intro for naming, which have been optimization points in glitched categories. In the hall of fame, there are pokemon cries and varying amounts of text based on pokedex completion, both of which can be controlled by the runner. X/Y has a final fight that received a timesave with the strategic play of: "just die lol". I have no experience with other gens, but these would be nice to keep within the timing boundaries for those games.

Wadiwadum
Preschooler
Posts: 6
Joined: Sat Feb 20, 2016 12:38 am

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Wadiwadum » Sun Jul 17, 2016 5:50 pm

I find IGT pretty nice when it comes to gen 1 because we have a clean cut-off and we make a lot of decisions based on the time we go for. RTA would remove this and have a feeling of "only WR matters" when we have a very nice way to see our progress over time. You could argue -and you'd be right- that beating your PB is the way to go when it comes to take decisions but I simply liked IGT quite a bit. As far as other generations go I think RTA is the way to go especially when there is a good ammount of lag not counting through IGT involved.

Keizaron
Site Admin
Posts: 273
Joined: Wed Jul 16, 2014 6:13 pm
Location: Spokane, WA
Contact:

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Keizaron » Sun Jul 17, 2016 7:29 pm

ExtraTricky wrote:If active runners have tried both categories and prefer no S+Q, it makes sense to have a leaderboard for no S+Q.
My opinion is there shouldn't be separate categories on whether somebody S+Qs or not. We don't need to expand the leaderboards to encompass multiple styles of play. If it comes to a point where we want that displayed, we can always pull a PMD and add a "style of play" column that indicates whether it was S+Q or single segment, but a separate leaderboard is kind of pointless IMO.
gifvex wrote:I think RTA is cool. I have enjoyed that timing method in gen 1's glitched categories, as every little time-saving discovery is worth learning about. Some of those discoveries we found were viable for glitchless runs, but since that category uses IGT, 2 seconds mattered less when trying to break a minute barrier (though top runners did appreciate the seconds).
This is a good reason why I'm for RTA. One or two minor time saves don't feel as valuable with the current setup for any gens' glitchless categories besides gen 3, but this rewards routing further. It can change the mindset of some runners from "it's just a second or two, who cares" to "it's a second or two of visible time save".
Wadiwadum wrote:I find IGT pretty nice when it comes to gen 1 because we have a clean cut-off and we make a lot of decisions based on the time we go for.
I disagree with this. I find it counter-productive that instead of going for a "PB", people are playing needlessly risky to go for the cutoff because there isn't a visible reward for completing a faster run.

ExtraTricky
Bug Catcher
Posts: 21
Joined: Thu Apr 24, 2014 12:39 am

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by ExtraTricky » Sun Jul 17, 2016 8:04 pm

Keizaron wrote:
ExtraTricky wrote:If active runners have tried both categories and prefer no S+Q, it makes sense to have a leaderboard for no S+Q.
My opinion is there shouldn't be separate categories on whether somebody S+Qs or not. We don't need to expand the leaderboards to encompass multiple styles of play. If it comes to a point where we want that displayed, we can always pull a PMD and add a "style of play" column that indicates whether it was S+Q or single segment, but a separate leaderboard is kind of pointless IMO.
When I said that I was assuming that the category's route significantly changes between allowing or disallowing S+Q, either mid-run RNG manipulation, or save warps (which no pokemon games have yet), or something else. If S+Q is just safety saves then I completely agree with you.

Keizaron
Site Admin
Posts: 273
Joined: Wed Jul 16, 2014 6:13 pm
Location: Spokane, WA
Contact:

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Keizaron » Sun Jul 17, 2016 8:30 pm

ExtraTricky wrote:When I said that I was assuming that the category's route significantly changes between allowing or disallowing S+Q, either mid-run RNG manipulation, or save warps (which no pokemon games have yet), or something else. If S+Q is just safety saves then I completely agree with you.
I still think if it comes down to mid-run RNG manip, I rather just have a column and not an entire leaderboard. Just my opinion, though.

Ryziken
Preschooler
Posts: 9
Joined: Tue Sep 23, 2014 5:09 pm

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Ryziken » Mon Jul 18, 2016 1:53 am

I am in favor of the switch to RTA, mainly for the reasons of there currently being excessive ties on a lot of Gen 1 - 3 games and the whole "playing for the next minute" mindset that others have already detailed.

I do feel, however, that for runs that currently utilize IGT, that both times should be visible in some way on the boards. Keiz has suggested putting IGT in the comments, and if that's the route we take, my suggestion is that it goes BEFORE any other text that runners have put in themselves (since for some reason speedrun.com has started clipping longer comments when viewed on the boards), so that it will always be visible just by hovering over the run.

Shiru666
Preschooler
Posts: 3
Joined: Sat Jun 27, 2015 9:25 pm

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by Shiru666 » Mon Jul 18, 2016 8:35 pm

I have been thinking about this switch since I joined the community (more than a year ago), and I'm still uncertain.
I see many issues with both RTA and IGT, which actually ends up making me unhappy with both (there's obviously no other way to measure, but that fact makes me hesitant on which one to choose). Many people are willing to switch to RTA due to inaccuracies in the final time, but what I wonder is if RTA will actually fix that. An answer that I've found is that it will, but it will bring another issue: "my computer is really old and that makes LiveSplit pretty innacurate; the timer shows 2:00:01, but it's actually a 1:59:55". You can obviously just retime the run and this problem is fixed, but the former problem is not: timing is still inaccurate.

Entr made a very good point with a living proof: Caran. This runner has had many, many issues with LiveSplit and its accuracy issues. This still happens, as one of his very recent FF X (I believe?) runs was .5 seconds off. You may think this is nothing, but there goes your so wanted sub 2 in Fire Red any%.

But what about IGT? MK made a very good video showing the frames missed in the final time in the third generation. This is plain dumb, and it's not speedrunning friendly, since it makes the final time a complete mess for this generation. This is not only for generation 3, obviously, but for games such as Yellow: Pikachu's cry (and many more other factors) makes it so that the final time will also be imprecise, which is not fair.

I conclude saying that IGT, although inaccurate, is way more comfortable to deal with than RTA. With RTA, however, you have the chance to fight that inaccuracy and make it more fair for the runner and for the run itself, but will get rid of many cool strats that you can choose to use in order to have a 1:49 instead of a 1:50 in games like Pokémon Red.

EDIT: editing to mention mention Best Time ties. Those are way more common with IGT, and I think they are healthy for the community. These are way more rare with RTA, however, but not everyone likes them.

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

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by GarfieldTheLightning » Tue Jul 19, 2016 1:02 am

At the start of this debate on the Discord channel, I was apathetic about using game time or real time. Regardless of the timing method used, I would not speedrun any differently. For the games I run, the only thing that concerns me is that runs that previously strived for the lowest possible IGT might be penalised for actions that occur outside the IGT measurement. Hopefully, if a switch to RTA were to occur, timing would align with IGT timing conventions (not saying it has to run to, say, the end of the credits in Pokémon X, since no input affects your time after the final AZ battle). I do think re-timing glitchless Johto runs would be finicky, but other people seem confident that this wouldn't be a problem so I'll trust them on that.

Anyway, when I saw the video detailing which frames are skipped when calculating IGT in gen 3 (found at https://www.youtube.com/watch?v=ymgdQ9oLd-Y if anyone hasn't seen it yet), my opinion changed (for these games, at least). IGT is not representative of how quickly you beat the game, so I would be in favour of a switch to RTA here for sure. Are there similar issues with other games? I don't know for sure, but all in all RTA seems like the metric that better represents how quickly you beat a Pokémon game.

Allowing save+quit would also be a nice benefit of a switch to RTA. It helps newer runners out, doesn't invalidate an entire run if you have a tiny bit of awful RNG (looking at you, Siebold), and may open the door to some cool new strats.
I speedrun Pokémon Pearl, White 2 and X!

User avatar
werster
Site Admin
Posts: 110
Joined: Thu Apr 24, 2014 4:38 am

Re: Revisiting Timing Methods for Pokemon Speedruns (IGT vs RTA)

Post by werster » Tue Jul 19, 2016 7:42 am

I really didn't want to post here after reading this thread and seeing a whole bunch of "I like this, so it's objectively better", but I thought I may as well anyway.

For reference, my stance is entirely neutral on the whole thing. I don't give a shit either way, and have always preferred IGT because of that simply because it's way fucking easier (especially since we didn't have a good gbc emulator back in the day, and still don't have a good one for RTA for gen 3+). Either way it just is what it is, saying "IGT isn't accurate" because it isn't accurate relative to Real Time is stupid, since...they are two different timing methods. If it is accurate to itself, it is accurate.

End of the day, there really is no objective better measure, and no matter how much anyone might disagree the best option is simply going to be what makes the most people happy. If having something that always times down to the second to break a whole bunch of ties (which is another ~opinion~) makes more people happy, then it is better, even if I disagree. If having an option that gets runners to play for a more universal and "easy to understand" timing method is seen as "better", then it is, even if I disagree.

The only thing I really came in here to comment on (other than arguing is dumb), is the "where to start/stop timing" if/when the switch to RTA is made. First off; obviously Gen 2 should not have credits (anyone who thinks otherwise is actually a complete moron), and that time should be taken off old runs that sat through them.
More important point: Y'all talking about starting timing at character control? Uhh.....nope. Every game that has any function that has any impact on the game must be timed from when that function starts. E.g, Gen 4/5 RNG starts when the game starts, games must be timed from reset. I'm not as familar with Gen 1/3/6, but Gen 2 has RNG running during opening, must be timed from reset. Even if Gen 1/3/6 aren't known to be the same, unless they are known to -not- be the same, they should also be timed from reset (in case something is found in the future that could be used from manipulation)

As far as I'm concerned that should simply be non negotiable, and simply common sense. Obviously past runs should be compensated for accordingly, but if you want to switch to RTA that simply has to be done. You can't say you want the "actual real time because the IGT doesn't count everything"....and then not count everything that impacts the game. That would be incredibly hypocritical and dumb as hell.

Post Reply

Return to “Policy Changes”

Who is online

Users browsing this forum: No registered users and 1 guest