FranceAlayan22 days ago

The removal of objects in low geometry levels is not consistently working in 1.4, and probably in several previous versions. The easiest test is to launch Black Forest in the lowest geometry detail. If trees are there, the game is bugged. It does sometimes work, and I'm not sure why.

This option provides a massive FPS boost, and since it can technically be used in other 1.x versions if the game cooperate, it will also be present in 1.5, but fixed to consistently work.

It is impossible to collide with track objects that get removed, which affects gameplay on some tracks. Although I'm not aware of tracks where it helps to get the record, it at least makes it easier to get a better time on some tracks.

FranceAlayan2 years ago

I have been refraining from trying to do times on some track because of stutter issues. As anyone here knows, stutters make it much harder to do a great time and much less fun to try too.

So without a good PC, advanced pipeline off is a good help.

But if that's still not enough, force enabling the legacy renderer in config.xml can reduce stutters significantly compared to advanced lighting off.

FranceAlayan2 years ago

I've collected the updated data and run it through my spreadsheet.

I used a snapshot of PBs on 05/05/20, 05/11/20, 05/05/21 to have an evolution of points. The October 2019 update was done on the 29th.

I didn't bother to include some players who only have one mediocre time. Players with multiple times or one really good time are ranked.

  • #1 Theodore (816)
  • #2 Andet (727)
  • #3 Alayan (186)
  • #4 Wax (99)
  • #1 Theodore (825)
  • #2 Andet (632)
  • #3 Wax (310)
  • #4 Alayan (162)
  • #5 FabianF (27)
  • #6 Mimiz (5)
  • #1 Theodore (816)
  • #2 Andet (611)
  • #3 Wax (357)
  • #4 Alayan (154)
  • #5 Fouks (40)
  • #6 FabianF (24)
  • #7 RareBeeph (22)
  • #8 Mimiz (11)
  • #1 Theodore (806)
  • #2 Andet (537)
  • #3 Wax (458)
  • #4 Alayan (139)
  • #5 SerpentS (89)
  • #6 Haenschen (40)
  • #7 Mimiz (40)
  • #8 Fouks (33)
  • #9 FabianF (21)
  • #10 RareBeeph (18)
  • #11 Bauju (4)
  • #1 Theodore (825)
  • ~ Theodore's October 2019 PBs (551)
  • #2 Andet (502)
  • #3 Wax (450)
  • #4 Alayan (209)
  • #5 SerpentS (77)
  • #6 TheZilag (70)
  • #7 Haenschen (59)
  • #8 Mimiz (39)
  • #9 Bubblee (37)
  • #10 Fouks (21)
  • #11 FabianF (16)
  • #12 RareBeeph (13)
  • #13 Bauju (4)

My points go back up in the last update because I have counted some new PBs not yet submitted for verification. I'm going to submit those and update the November 2021 snapshot with any new record done before the 6th to be consistent with the previous snapshots.

The fight between Wax and Theodore is evident. The improved WRs have steadily made Andet's points go down. Some new runners popping this year but with only submitted times on a few tracks.

Of note, even if Theodore hadn't improved a single PB in 2 years, he would still have the best overall record.

FranceAlayan5 years ago

It's nice to compete across the various tracks to try and get as many records as possible, and when it isn't in reach, at least get close to the best time.

But who has the best overall times across the ILs ?

Total time isn't that great because people who haven't posted a time for every single IL (even with the restriction to official tracks from a version) can't be compared at all to the others. A single poor time can also make a big difference.

Hence this alternative : points awarded by IL. It also gives some more incentive to post PBs which aren't WRs :)

Update : The previous formula dropped from 20 points for WR to 13 points for a time 1% slower to 10 points for a time 2% slower. It appears to me now this is too harsh. The fight between theodore and Andet has shown well that several % of improvements can happen, and while we don't have a lot of players submitting times right now, the points would be meaningless except at the very top.

New formula :

  • WR gets 40 points.
  • A time <0.1% longer than the WR gets 39 points.
  • = 0.1% and < 0.2% gets 38 points.

  • = 0.2% and < 0.3% gets 37 points.

  • = 0.3% and < 0.4% gets 36 points.

  • = 0.4% and < 0.5% gets 35 points.

  • = 0.5% and < 0.6% gets 34 points.

  • = 0.6% and < 0.7% gets 33 points.

  • = 0.7% and < 0.8% gets 32 points

  • = 0.8% and < 0.9% gets 31 points.

  • = 0.9% and < 1.0% gets 30 points

  • = 1.0% and < 1.15% gets 29 points

  • = 1.15% and < 1.3% gets 28 points

  • = 1.3% and < 1.45% gets 27 points

  • = 1.45% and < 1.6% gets 26 points

  • = 1.6% and < 1.8% gets 25 points

  • = 1.8% and < 2.0% gets 24 points

  • = 2.0% and < 2.2% gets 23 points

  • = 2.2% and < 2.4% gets 22 points

  • = 2.4% and < 2.6% gets 21 points

  • = 2.6% and < 2.8% gets 20 points

  • = 2.8% and < 3.0% gets 19 points

  • = 3.0% and < 3.3% gets 18 points

  • = 3.3% and < 3.6% gets 17 points

  • = 3.6% and < 3.9% gets 16 points

  • = 3.9% and < 4.2% gets 15 points

  • = 4.2% and < 4.5% gets 14 points

  • = 4.5% and < 4.9% gets 13 points

  • = 4.9% and < 5.3% gets 12 points

  • = 5.3% and < 5.7% gets 11 points

  • = 5.7% and < 6.1% gets 10 points

  • = 6.1% and < 6.5% gets 9 points

  • = 6.5% and < 7.0% gets 8 points

  • = 7.0% and < 7.5% gets 7 points

  • = 7.5% and < 8.1% gets 6 points

  • = 8.1% and < 8.8% gets 5 points

  • = 8.8% and < 9.6% gets 4 points

  • = 9.6% and < 10.5% gets 3 points

  • = 10.5% and < 12.0% gets 2 points

  • = 12.0% and < 15.0% gets 1 point

We'll see how it holds with time.

theodorepringle likes this
FranceAlayan5 years ago

The speedrun timer is not included in the main release so we'll need separate modded binaries with it.

I have ported back my code to a recent version but there are a few nasty bugs to fix.

It may take a few days to get both linux and windows binaries after I publish the code.

I suggest to avoid doing full-game runs without the timer mod in the meantime, so all the runs can be timed consistently.

Alistair_Findlay likes this
FranceAlayan5 years ago

We recently asked @coolperson to submit replays for his times in order to confirm their legitimacy. He submitted the files, we checked, and we could confirm that things were in order.

However, if someone wants to check one of the times done by me, @Fouks and others, there is still only the video (which proves things, but it can be harder to detect something amiss).

Similarly, if someone wants to train having those replays on hand could be useful to better understand where time is lost.

However, having each player host his replay files is impractical. Sure, the same is true of videos, but this adds a layer of complication and increases the risk that some will be lost.

Is there some way we could make replay file sharing easier and more reliable ?

I'd like to share some of my own, but I don't fancy having to setup some "cloud" storage thingy to get it done;

FranceAlayan5 years ago

SR had its own issue and seems to have rolled a backup.

The runs validated yesterday have been deleted and have to be resubmitted.

FranceAlayan5 years ago

I made a compilation showing the 7 runs I had which were on pace to get the Any% WR and which got destroyed by the dumb randomness of FTL.

Kitoko likes this
FranceAlayan5 years ago

In a game I run, me and another guy are trying to get a time below 30 minutes, while we have PBs (and WR/former WR) in the 30m20s.

Now, analysis on our best segments show that this is very doable (an ideal run with no mistake and perfect luck would be at around 29m), but it's hard to avoid any significant blunder, and there will always be some amount of unluck (there are small random events across the run, most won't take out more than 2-3s, though in some specific spots it can result in a bigger run-killing loss).

And so this got me thinking : when is the best time to reset a run ?

I should clarify what I mean by resetting "early" and "late".

Resetting "late" means only resetting when there is no significant hope of besting the WR, i.e. enough time has been lost that it would require practically no mistake/bad luck until the end to barely succeed.

Resetting "early" means trying to get a significant advantage in the beginning of the run, resetting if on par/slower than WR pace at some point (or even faster, but not by much), even if there is still room for some bad luck and mistakes in the final part to still snatch a best time.

For some games and categories (especially those with no luck at all involved), if the WR/PB is optimized enough, there is no real difference between resetting early or late, as there is no room for error.

But in others, like mine currently, it's possible to lose 10s in a segment but still have a good shot for the record, so this question can really come up.

I can see some pro and cons to both strategies :

  • Resetting late means more practice on the later parts of the run. The sooner you reset, the more often you'll do again and again the same parts at the beginning while doing only rarely those in the middle and in the end.
  • Resetting late gives more opportunity to get a record with a strong finish even if the beginning was not that great.
  • Resetting early means that less time is wasted in runs which are unlikely to succeed.
  • Resetting early increases the chance to have a really good time in the middle of the run, and so leaves room for error, and increases the chance that when a run is finally done to its completion, the resulting time is lower.

So, what are you personal preferences ? How does it change whether you're 5% in, 25% in, 50% in or 90% in the run ? What do you think is the most efficient strategy to reduce grinding ?

FranceAlayan5 years ago

I expected Fouks to post an any% run this month, but there is no sign of it happening.

Rather than waiting a few months and having to relearn 0.9.3 tricks, I prefer trying to get "definitive" records while my technique is here.

So far, I did two promising runs which managed to get past the main obstacle in the run, the triple-skip in Hacienda (I decided the skip in Northern Resort to be way too unreliable). In both, I was more than 20s ahead of the current WR after Candela City.

Both ended in a tragic fail, the first killed by the FTL.

The second was killed by a mix of errors and by the game trolling me and throwing me off-balance.

Using my old graph comparing WRs, here is how the new runs (purple and light blue) went :

In that second run :

  • I somehow failed the last skip in black hill mansion despite my cautiousness.
  • The game decided that I want to play in intermediate difficulty in the gran paradiso challenge. As, usually, you select as fast as possible, I couldn't change it before being in the challenge. So I exited, went back, selected too fast once again, exited, and finally started the challenge in expert
  • Because the Old Mine challenge has only 3 AI opponents even in expert, I mistakenly thought I was again in intermediate (thanks to that Gran Paradiso issue...), and lost a few seconds exiting and going back in.
  • I somehow missed the gate skip TWICE, each time flying over the wall.
  • I lost 3-4 seconds with an accident going towards the final challenge, after the gate skip

Considering everything that went wrong, it's a small miracle that it was still a sub-32 time.

At the same time, I'm certain that with some more tries going at sub 9m50s after Hacienda, one of them will be free of most of these errors... Making sub-31 very possible.

FranceAlayan5 years ago

With Fouks recently giving me some serious competition, I had to improve many times which were until then "good enough" to be on top.

With that, what looked like a pipe-dream only a few weeks ago seems really doable now :

I only need to take off 2s somewhere...

FranceAlayan5 years ago

It is possible to start in a saved GP in story mode.

As this requires having done the beginning of the GP on another profile, and allows to have an arbitrary number of AI karts, this is obviously banned from full-game runs.

Beware though, as if the conditions for it to trigger are met, you can't avoid the bug and it can ruin your run.

Kitoko likes this
FranceAlayan5 years ago

How is that route balanced ? The goal is to get the hardest/luckiest part behind you as fast as possible, and reducing kart switching to a minimum. I kept it similar to the any% route at the beginning.

Considering the improvements made here and there, here is what I think is (nearly) the optimal route : 1)Start with a light kart in front of Scotland 2)Scotland 3)Oliver's Math Class 4)Switch to a medium kart 5)Cornfield Crossing 6)Cocoa Temple 7)Shifting Sands 8)Hacienda (with skip if executed reliably enough (>50%)) 9)Northern Resort (with skip if reliable enough) 10)Zen Garden 11)Snow Peak 12)85-points GP

  • a)Skip Cocoa Temple
  • b)Do Northern Resort (with skip if reliable enough)
  • c)Either skip Hacienda, or use the skip (confusing I know !)
  • d)If you did Hacienda, skip. If not, do Zen Garden
  • e)Do Snow Peak 13)Candela City 14)Switch to light kart 15)Do 40-points GP
  • a)Do Shifting Sands (use the skip)
  • b)Skip Cornfield Crossing
  • c)Do Oliver's Math Class
  • d)Skip Abysses
  • e)Do Scotland 16)Do the FTL 17)Do Blackhill Mansion 18)Do Green Valley 19)Do the 125 point GP
  • a)Skip Gran Paradiso Island
  • b)Do Green Valley (with skip)
  • c)Do Blackhill Mansion (with skip)
  • d)Do STK Enterprise (with skip)
  • e)Skip Volcano Island 19)Do Minigolf 20)Switch back to medium kart 21)Do the 165 point GP - this is one of the hardest part as you need to pull-off a double-skip in a 4-races GP
  • a)You don't have unlocked Fort Magma, it's not among the track you have to do
  • b)Do Minigolf
  • c)Skip XR591
  • d)Skip The Old Mine
  • e)Do Around the Lighthouse. (At this point, all major difficulties are behind you) 22)Volcano Island 23)Abysses 24)Gran Paradiso Island 25)Around the lighthouse 26)Old Mine 27)XR591 28)Fort Magma
FranceAlayan5 years ago

I did a single try of full-game run tonight.

I tried to use some new techniques, and it didn't work out very well (I tried to use the northern resort skip and failed on each of the 3 laps...).

However, by some miracle, despite some usual stupid errors, it was overall rather clean, and I didn't get trolled by basket ball or parachutes.

The new time is of 32m08s599.

So I'm now torn about submitting it or not.

Pros :

  • It IS the current WR
  • Accurate leaderboard history.
  • Allow the people interested here to see how it paned out and maybe get ideas.

Con :

  • It takes only 2,5s from the record ; while it is clearly possible to get 15-20s less without having to do a perfect run.
  • With some luck, I may be able to beat it again more convincingly rather soon.
  • My internet upload speed is poor, and it would take several hours to be uploaded to youtube.

Your take on this, guys ?

FranceAlayan5 years ago

Currently, the rules for the GP ILs are IGT. While there is good in it (it discards loading time and doesn't force button mashing in menu to matter), it also means that racing for a full minute before doing a skip counts exactly the same as managing to skip in a few seconds. Doing the latter without screwing up the GP is harder and more skillful, so it's not good that it is not rewarded.

Do you think the rules should be changed ?

The possibilities are :

  • Keeping the final GP IGT
  • Doing it full RTA
  • Some mixed solution : adding the game timer at the moment the player pauses to skip to the final GP IGT ?

Second : what should be the rules for the last GP ? Should it be 4 races (without fort magma), like it is in the story mode when done early, or 5 races, like it is otherwise defined in normal racing ? On one side, the double-skip in a 4 races GP is harder and it matches more full run experience, on the other, 5 races match the intended game design and one race more to do is one occasion more of gaining some time.

Fouks and Alistair_Findlay like this
FranceAlayan6 years ago

For training, I prefer using a version with a translation.

I guess that the rules forbid using something else than the Japanese version because the length of the text affects the time the run takes ? It remains forbidden even if the language patch increases the duration of the run ?

If I got a time good enough to be competitive, I'd certainly run with the original version to gain some more time, but if I'm at +50% of the WR I wouldn't worry much about losing three minutes more to this.

FranceAlayan6 years ago

I think we need a FTL hate-thread.

@Alistair_Findlay already has a video up about this.

Here is what happened to me tonight :

I was roughly on par with the Any% WR time, even maybe a few seconds faster (I finished Candela City 9 seconds earlier, then did Gran Paradiso rather than Green Valley so it's hard to compare), then FTL happened. I didn't race carelessly : I applied the "try to get a lot of nitro at the beginning to use when you're caked" technique.

But... Just watch it yourself. There are two races. Each make me hate the FTL mode. Having to follow-the-leader means that if you're skilled enough to go faster than the leader (and skilled STK players go A LOT faster) you're harmstrung. You can't use your skill to gain time, and when all karts are packed close, you can't avoid any bad item.

FranceAlayan6 years ago


@Dugglez is not active at all here, I don't see the point of having him as super-moderator for STK. Verifier would be enough, if he wished to come back later, I think.

@UbuntuJackson is very active verifying runs and is trustworthy, I'd put him as super-moderator for an easier management of mods in the future.

I'd also like to myself get moderator, would be nice to be able to verify runs, sticky forum posts, etc.

FranceAlayan6 years ago

Maybe we could add ILs for track in reverse ? The strategies and gameplay change slightly between the two, as "long straigt, curve, short straight" is not symmetrical, for example.

However, adding it for all difficulties would create an enormous amount of ILs.

And tbh, I don't like having categories nobody run, wants to run, which don't add anything, and seems to be just here waiting for some stupid kid to claim a bunch of "WR".

FranceAlayan6 years ago

I'm not a big fan in the gate skip, at least in its 0.9.3 iteration. It just makes the run shorter, but with the proper technique learned, it makes the overall run easier (expert% requires to pull off some track specific-tricks like the mansion skip twice ; and require to pull-off a double-skip in a 4-tracks GP ; both of which are very skill-selecting).

I like more GP skips, as messing up can be catastrophic, but it's still very much in control of the player.

Part of what make a game "speedrunning-friendly" is making the time of a run less luck-based and more skill-dependent. Even with a normal run through all levels, random events (items) often average out enough that the better time is the more skillful-one ; and at worst the better player will get better luck over a few tries. Still, the time difference between "excellent skill" and "decent skill" is not that big. It's shaving off these last percents which is the hardest, and if say in 0.9.4 a SuperTux% run takes 1h30m with extraordinary skill but 1h35m with good skill, we have just too small of a gap, there is no "wow factor".

So what is needed are things which are harder to pull-off, not random, and can make a significant difference.

Keep in mind that we need something which could be accepted into the base game.

I see two ways of doing this :

  1. Make some challenges punishing : doable, but with little margin for error. Realistically, this can only be done for itemless SuperTux challenges.
  2. Create "shortcuts" : while the casual players would complete the challenges and get the unlock normally, managing to do some extra-skillfull action could be rewarded with an extra making the overall completion shorter.

Any good idea of "shortcuts" which the main STK developers may accept ?

About Alayan
6 years ago
13 days ago
Games run
Last run 4 months ago
Cave Story
Cave Story
Last run 6 years ago
Games followed
Last visit 2 years ago
Cave Story
Cave Story
Last visit 3 years ago
Rayman 3: Hoodlum Havoc
Rayman 3: Hoodlum Havoc
Last visit 4 years ago
SuperTuxKart Add-Ons
SuperTuxKart Add-Ons
Last visit 2 years ago
Games moderated
Last action 22 days ago