News:

Herr Otto Partz says you're all nothing but pipsqueaks!

Main Menu

Unstable replays at ZakStunts

Started by Duplode, April 10, 2023, 02:48:47 AM

Previous topic - Next topic

dreadnaut

Quote from: Daniel3D on May 08, 2023, 04:46:54 PMYes, they all load not crashed if I'm not mistaken.
It would be good practice to rewind a bit to check if the finish line has been crossed.

Some replays show a cracked windshield when loaded, but play correctly from the start. Crossing the line is covered by the first sentence: "A replay must be a complete lap of the circuit."

Quote from: Duplode on May 08, 2023, 10:47:00 PMIs it worth it to mention playing with multiple cameras, or would that be excessive detail for that page?

Switching to different camera would be "custom instructions", and fall under #5. Option #4 was meant to cover only replays that play at normal speed with the default camera.

Duplode

#91
Quote from: dreadnaut on May 09, 2023, 01:03:55 AMSwitching to different camera would be "custom instructions", and fall under #5. Option #4 was meant to cover only replays that play at normal speed with the default camera.

Fair enough. While this would have ruled out Frieshanen's final ZCT251 replay, he wouldn't have needed to fix it in the first place, as the withdrawn laps are complete on the fast-forward timeline. The hardest part might well be not switching to F2 on instinct when validation time comes  :)

Friker

Quote from: Duplode on April 22, 2023, 10:34:40 AMFast playing an unstable replay from either the beginning or 0:00.05, depending on where the divergence point is, will lead to the fast-forward timeline.
@Duplode , please, confirm that this happens only with fast-forwarding.

Quote from: dreadnaut on May 08, 2023, 11:43:50 AMA replay must be a complete lap of the circuit. Please validate your replay before uploading it, starting from "Options → Load replay". A replay is valid if it shows a non-crash frame when loaded, or if it plays from start to finish at normal speed. Invalid replays will be disqualified.

I would still add something about crossing a finish line (extended to the whole block). But maybe it is obvious for others.. Then a question is - what is a real finish time? - because @Argammon showed very bizarre things. :)

Duplode

Quote from: Friker on May 09, 2023, 10:12:42 AM@Duplode , please, confirm that this happens only with fast-forwarding.

It's fast play (also known as double speed) rather than fast forward, but other than that, yes indeed. Fast play doesn't actually make a difference when it comes to validation: depending on the starting frame, you'll get either the fast-forward timeline or the "main" divergent one, which are the ones checked by #4 anyway. (In this context, fast play is mostly useful for providing a way to watch the successful timeline without pausing.)

Quote from: Friker on May 09, 2023, 10:12:42 AMThen a question is - what is a real finish time? - because @Argammon showed very bizarre things. :)

I belive the sensible thing to do is to keep using the file length-based time shown by the site, which that corresponds to what the pipsqueak got while driving. The only tricky case, which should be very rare, is if the lap completes successfully in the main timelines, but neither matches the file length time. Options in that case include sticking with the slowest time, and ruling the replay invalid.

Friker

Quote from: Duplode on May 09, 2023, 01:03:25 PM
Quote from: Friker on May 09, 2023, 10:12:42 AM@Duplode , please, confirm that this happens only with fast-forwarding.

It's fast play (also known as double speed) rather than fast forward, but other than that, yes indeed. Fast play doesn't actually make a difference when it comes to validation: depending on the starting frame, you'll get either the fast-forward timeline or the "main" divergent one, which are the ones checked by #4 anyway. (In this context, fast play is mostly useful for providing a way to watch the successful timeline without pausing.)
You are right, I had fast-playing in mind.

Quote from: Duplode on May 09, 2023, 01:03:25 PM
Quote from: Friker on May 09, 2023, 10:12:42 AMThen a question is - what is a real finish time? - because @Argammon showed very bizarre things. :)

I belive the sensible thing to do is to keep using the file length-based time shown by the site, which that corresponds to what the pipsqueak got while driving. The only tricky case, which should be very rare, is if the lap completes successfully in the main timelines, but neither matches the file length time. Options in that case include sticking with the slowest time, and ruling the replay invalid.
One can argue that he drove with fastest time in mind (even if that does not make sense because then a replay would finish sooner..) that's why I would like that clarified. :) I am also inclining to the file length/slowest time.

Daniel3D

Quote from: Duplode on May 09, 2023, 01:03:25 PMThe only tricky case, which should be very rare, is if the lap completes successfully in the main timelines, but neither matches the file length time.
In such cases you could rewind to just before the finish line and continue from there. (Right?)
Edison once said,
"I have not failed 10,000 times,
I've successfully found 10,000 ways that will not work."
---------
Currently running over 20 separate instances of Stunts
---------
Check out the STUNTS resources on my Mega (globe icon)

Duplode

Quote from: Daniel3D on May 09, 2023, 07:19:34 PMIn such cases you could rewind to just before the finish line and continue from there. (Right?)

In the unlikely situation I'm thinking of, it wouldn't. For instance, consider a replay which is completed by the pipsqueak on the divergent timeline, crashes on the fast-forward timeline, and diverges in a different way on dreadnaut's machine so that it clips the finish tile and ends prematurely with penalty. In that case, the only laptime dreadnaut will see isn't the file length one. Outside of perfect storm cases like that, though, the file length lap time will be good enough.

dreadnaut

Whops, I forgot about this thread :o

I have one question remaining: when should the updated rule come into play?

A) ZCT262 - This race, I'm hiding a really tricky powergear lap!
B) ZCT263 - Next race, this is urgent
C) Season 2024 / ZCT270 - No in-flight patching

I'd rather not (A), but let's see what you think!

Duplode

Quote from: dreadnaut on May 30, 2023, 12:02:21 PMI have one question remaining: when should the updated rule come into play?

I'd be happiest with (A), but (B) is an acceptable substitute if you'd rather stick to the general principle of not changing rules mid-race. There's no point in waiting until 2024 IMO.

Cas

Earth is my country. Science is my religion.

KyLiE

As I mentioned previously, I'm happy with the current rule, so I don't have any preference when the updated rule comes into play.

alanrotoi

Quote from: Duplode on May 30, 2023, 05:29:00 PM
Quote from: dreadnaut on May 30, 2023, 12:02:21 PMI have one question remaining: when should the updated rule come into play?

I'd be happiest with (A), but (B) is an acceptable substitute if you'd rather stick to the general principle of not changing rules mid-race. There's no point in waiting until 2024 IMO.

Same here!

Daniel3D

Quote from: alanrotoi on May 31, 2023, 04:52:59 AM
Quote from: Duplode on May 30, 2023, 05:29:00 PM
Quote from: dreadnaut on May 30, 2023, 12:02:21 PMI have one question remaining: when should the updated rule come into play?

I'd be happiest with (A), but (B) is an acceptable substitute if you'd rather stick to the general principle of not changing rules mid-race. There's no point in waiting until 2024 IMO.

Same here!
Same
Edison once said,
"I have not failed 10,000 times,
I've successfully found 10,000 ways that will not work."
---------
Currently running over 20 separate instances of Stunts
---------
Check out the STUNTS resources on my Mega (globe icon)

dreadnaut

Quote from: Duplode on May 30, 2023, 05:29:00 PMI'd be happiest with (A), but (B) is an acceptable substitute if you'd rather stick to the general principle of not changing rules mid-race. There's no point in waiting until 2024 IMO.

There is also a principle of not changing rules mid-season — even when the replay validation was introduced, it only came months later if I remember correctly — but (B) seems the best balance, I'll update the rules on Sunday 👍

Duplode