GV NG 3lap

[glass summit]

Marth has a ghost from 2021 that gets a very low trick since he was able to RFH for 1 frame, though i was not able to recreate it with an updated startslide (that saves about 5ms)
however i did found a very small improvement (about 0.3ms) after the trick, but of course it needs more testing
(maybe a right slide might have a chance? police already has a run with it, but marth's ghost is slightly faster #tas-focused message)

absolutely would appreciate some help since i definitely cant do this alone lol

marth's ghost

[dusty lantern]

Might be interested in tasing this

[dense owl]

any potential for spear to be faster for NG?

i feel like i have asked an already answered question but couldn't find a clear answer after searching

[dusty lantern]

Not spear, though sneakster is maybe possible

I have low faith in it though

[glass summit]

i do remember a sneakster conversation, but idk

[dense owl]

is the trade off between speed and mt worth it? + the height off of tricks would cost a lot of time

[raw harbor]

mt and trick height are not the problem

it's drift that makes it and spear completely out of the question imo

[dense owl]

oh right, tbh i didn’t think drift would be much of an issue on gv

only 1-2 turns i can think of that pose an issue

[raw harbor]

https://www.youtube.com/watch?v=Abu_N90iVBc this is what spear looks like on gv if youre curious

i cant believe sending that summoned ejay

[lunar cove]

time for my daily watch of this

[raw harbor]

i miss the days when i had a regular catalogue of tases i would cycle through improving

[patent wren]

If you watched this everyday for the past five years, you would of watched this 1,870 times.
Given the current view count of 27,659 views, you make up less than 7% of views.

[solid panther]

some small improvement carried past trick, this feels very unclean there should be more to save

[dusty lantern]

I still have faith in a right slide

[solid panther]

i will test it again, but they seemed equal to me. the low is strongly limited by forward position, so with a right slide i think you'd aim to hop at the exact same spot anyway

[dusty lantern]

If I have time tonight I’ll give some attempts at the ramp

[solid panther]

try different amount of sta too

i ruled out landing earlier being faster but i'm not sure about landing later. all the airtime qm at landing makes it difficult to conclusively test

[dusty lantern]

I think a right trick might be good to test as well

[solid panther]

oh maybe i forgot to try it lmao

[dusty lantern]

I remember it giving pretty good results when doing tas comp a while back (though I was in a time rush so take that with a grain of salt)

[warm cedar]

Police can you compare this to our task 20 2023 submission

[dusty lantern]

Look at the initial post in this thread

[warm cedar]

Ah ok

[dusty lantern]

I think we get a better ramp+landing but a bad slip chain

Or something like that

I do think if we get a very fast start we can beat the qm lock on the cave uphill

[crude grail]

i will try to work on it

[dusty lantern]

got this with a right trick

I land a frame or two later but it's just ahead 🤷‍♂️

the trick is also kinda awkward and weird, still probably a decent amount of improvement in the sta but landing qm is buns

[solid panther]

right slide is faster. going to send something by this morning

[lunar cove]

hooray

that's the final unknown NG startslide direction to be figured out

[solid panther]

right ends up being faster bc you can carry more of the slide timesave by being further right on the ramp, so it ends up being better to just begin moving right instead of left + zigzag

[solid panther]

right slide is also a better line, so it ended up saving a lot over left. ghost is police.
while i couldn't replicate the bounces into the slipchain, i got a different 2 f bounce right before the trick boost ends, which is about as fast

not sure if this is fully optimized yet but it feels much better than what i sent yesterday

[south badge]

Oh yeah that's much faster

[south oyster]

damn

[glass summit]

~0.4ms (went slightly more left on RFH)
i also tried left trick and got pretty close but right seems faster yea

[crude grail]

saved 2~3 ms over previous ghost

[glass summit]

very nice

[crude grail]

Thanks !

[solid panther]

nice job both of you

[crude grail]

Ty !

[glass summit]

i tried so hard but i was not able to get a stick, both before the re-wheelie and after lmao
but considering this only loses 0.5ms, if a stick is possible then there would 100% be timesave

[solid panther]

i can't get anything, i don't think i will try this trick more the qm sucks balls

[glass summit]

^ here i do some neutral gliding, but you can actually extend it to get more farther right but i still wasnt able to get anything with that

[glass summit]

ikr

[solid panther]

i got the slipchain i showed in my latest ghost but it's just too slow unfortunately bc of the speed drop

the other one is too annoying for me

[raw harbor]

Just a heads up, seeing as this project is relatively new, don’t forget to talk about it in #tas-focused so others don’t miss out on hearing about it!

[glass summit]

can you try editing these stickX inputs on this frames? just goes more right with neutral gliding (as i said in my latest ghost i only do partial neutral gliding for some reason), maybe that can change the QM somehow

i tried but still no stick

[glass summit]

@solid panther got anything?

if you tried

[lunar cove]

are you sure about 100% certain timesave?

looks behind before getting airtime

[glass summit]

well i get airtime before that too

i am still at 119 in the air but i would lose less if i'd sticked

[crude grail]

managed to get a bit ahead of my previous ghost by modifying shxd's ghost

[glass summit]

[glass summit]

i was not able to improve it, needs more testing though

[solid panther]

small improvement to trick + continuation. it looks like my ghost is behind alego, but the left rotation helps for the first turn.
i can't get good uphill qm, also i'm pretty sure the second low is qm locked. not sure if there's anything we can do if not lose a bunch of ms?

[raw harbor]

yeah second ramp is a pretty tight lock

[solid panther]

wait i thought the uphill was supposed to be a full stick but the bkt gets a slip

the only bounces i get are late, is that usable? i have no clue

[dusty lantern]

Yeah the ramp is qm locked

[dusty lantern]

Yeah you want very late bounces

[solid panther]

last frame of airtime happens here

bkt slips starts here

wish there was a ghost but this is the best we can do

it looks like the uphill qm i got is workable then

oh it's not 11.093 the 093 is the speed lmfao, okay it looks like bkt also gets a slip at about 11.8

[raw harbor]

I love watching new tasers get confused by old tas community things

[lunar cove]

wish this was faster

https://www.youtube.com/watch?v=GMNUmc2gQVQ&t=50s

[raw harbor]

iirc i tried that when working on nu lap 1 in 2020

[solid panther]

are we 100% sure this is slower? or is it close like the ryf mt

[raw harbor]

it's close like ryf

problem is the straight isnt long enough for mt to catch up

[solid panther]

but if the ramp is a qm lock, unless a new strat is found for that like some magic slip, it's not going to be useful

[glass summit]

are we closing the first trick?

as in, its finished

[solid panther]

two people have confirmed that the second trick is qm locked (i have to lose ~2 ms already) so i think it's fine to move on

i guess maybe test that mt strat although i don't think there's a high chance it'll save significant time

[crude grail]

got better bounces ( i think ? ) and ok cave first mts

[solid panther]

small improvements after 1st turn cave give bad qm. i was trying bigger improvements not sure if this is viable

[raw harbor]

Those mts are pretty wide, i may take a look if I find the time

[solid panther]

this gets ahead but i don't think it's actually viable

this turn is fucking crazy i really don't understand how to optimize it

[crude grail]

yeah

will re take a look into it tomorrow

[crude grail]

didn't got anything better for now

[solid panther]

@raw harbor can you take a look at this now?

[raw harbor]

I’m busy for most of the next 2 days, I can take a look on Monday earliest

Sorry for not already doing it

[glass summit]

who is down to work on this again

i might be able tomorrow

[astral rivet]

i will try to work on this

[raw harbor]

Ah yeah I was meant to work on this

I'll work on it after the task is over

Please ping me if I forget again

[crude grail]

got that for the cave mts, behind but with way better angle for the next turns

so end up ahead afterwards

[solid panther]

i don't think i will be able to take a look until monday, but i like this strat for 1st cave turn. this feels like it will be the way to go

bc the slip into the 2nd turn is qm locked and getting about a frame ahead to get the slip again leaves you with a horrible angle. although i am not sure if the angle from the ghost in the clip can be improved considerably

[crude grail]

improved a bit cave mts, and got good downhill qm

[astral rivet]

have we tested justin's cave mt strats?

ig i could test it

[crude grail]

here is justin mt strat

if it's possible to stick at the downhill, it could be faster

[raw harbor]

hey maybe i wasnt crazy when i got this to be faster in task 10

[raw harbor]

granted i didnt get the god stick here like you did, but still

[crude grail]

got this ( vs this one #1384507571825541270 message )

[distant blade]

i was gonna say

[distant blade]

this looks better than Justin's line

[distant blade]

tested this for a bit and it felt really clunky

not sure how much it could be optimised, but the current line seems to have more potential

[distant blade]

saved 1ms with input changes around 14.7s IGT, causing a tighter cave overall. i also saved on the cave exit turn's wheelie tilting without sacrificing angle

[astral rivet]

looks like right trick is faster

if done properly

faster one + attempt at boost ramp

changes inputs on the downwards slope

[raw harbor]

arent we still on the first cave turn? 😭

[astral rivet]

really?

lmao

i actually tried improving them

but wasn't able

[dusty lantern]

second

[lunar cove]

yeah this

[solid panther]

aren't there two equal strats for one turn as well why the fuck are you continuing past that lmfao

[astral rivet]

cause i tested stuff

did not work

and saw mitch talk about the rfh trick

[raw harbor]

this is just a case of people not communicating where theyre up to and it having a domino effect

alego happened to do an attempt at the rfh trick at the end of the improvement to the cave lines, then mitch took that to mean that we were up to the rfh trick and improved that by a really small amount, and you took that to mean we were almost done with rfh trick so you improved it and did a test for the next turn, when the whole time we were never even working on the rfh trick

[dense owl]

so then, where are y'all

[raw harbor]

the double mt entering the cave still

[dense owl]

thought so, ty

[distant blade]

did you read my improvement

i improved the beginning of the cave which carried to the RFH ramp, and only continued the ghost because Alego's TAS code allowed me to get an accurate screenshot to demonstrate an improvement.

i never tried to convey that the cave was finalised because it obviously wasn't, nor had been even reviewed by a different TASer

i should've communicated it better that my improvement was only minor, and not a completed cave. so i apologise if it gave off the impression that i was indicating we were up to the RFH ramp

[lunar cove]

it was Cesar's message that said boost ramp attempt

[solid panther]

it's fine, it happens. not a big deal because nobody rushed too many sections.

in the future, everyone should keep in mind that it's best not to move on from a section, until someone (or better, several people) finds a qm lock. we only moved on from optimizing the beginning because the second low is qm locked, for example

[dense owl]

speaking of qm locks, does the ramp after the cave have a qm lock

[solid panther]

very likely but each qm lock is different

the second low has a pretty strict forward position requirement to allow for rfh. almost an entire frame would need to be saved to carry more timesave which is not happening. the next ramp could be more lenient, or even more strict

if i didn't have an exam literally tomorrow i would've looked into this, sorry for delay you'll have to be a bit patient

[raw harbor]

the stick on the downhill exiting the cave and the ramp immediately after both have very different qm locks and what alego does here is the only time they coincide

the whole sequence is incredibly complex because there are so many different slopes you can stick on and get rfh from, and each of them work slightly differently

[solid panther]

yeah this cave has a very high concentration of bullshit. we'll need to spend several days to understand all the nuances

[raw harbor]

the double mt alone is stupid because the sequence starts from the uphill before it and ends at the slip drift after it

[solid panther]

only thing i can tell you is that the slip into the 2nd cave turn is qm locked, and with this alt strat for 1st turn i think a bit more can be squeezed out before we begin to stick

[raw harbor]

change anything on either end and the whole thing gets thrown off

[solid panther]

been trying to improve the latest ghost here. i can only make small improvements to cave before i get bad qm for the rfh, not sure if others can get around this. i improved the section after the cave to save ~1 f at the flip trick, still very unoptimized

idk if i want to commit to this track myself. hope activity picks up from other people

[solid panther]

i have not tested this mt strat at all

[lunar cove]

should be able to take this turn very tight like so

[solid panther]

oh cool you guys can figure it out lmao

[lunar cove]

https://youtu.be/K0hdlYZnMGg?feature=shared&t=30

the much lower ramp trick is possible to get from this line

https://cdn.discordapp.com/attachments/735589843513638940/1061156924827107348/image.png?ex=68d77866&is=68d626e6&hm=bab6b9b9140407abf599071b97e18161a31e0ee0ff4c1f620ea1e67f257fabe7&

[lunar cove]

ghost for this

if anyone wants to match the line onto this 3lap

obv you'll have to change everything after the miniturbo boost runs out, since we're tricking left

[crude grail]

i'm gonna work on this today

[crude grail]

improved plateform section + better slip ( as fast as ejay's ghost one )

[grim ibex]

Is there ever a way to know when a turn is 100% perfect

[dusty lantern]

Not really, there were ideas of trying to calculate turns at some point but it was deemed too complicated and case specific

[solid panther]

ejay have you tried with that other mt strat in the cave at all?

[lunar cove]

no

[dusty lantern]

Has anyone compared it to the old flame runner cave?

[lunar cove]

GV NG 3lap - Improvement on platforms carried to boost ramp. Improvement allowed for very good boost ramp QM (you can get a trick with <50f airtime)

not sure whether delaying the trick is faster or not

from my brief testing, instant trick gets more airtime, but has increased XZ speed in the air

[dusty lantern]

Instant trick would result in better vta

[solid panther]

trying this out rn

[solid panther]

small improvement to platforms + better turn after low. i couldn't get tricking early to be faster but don't trust my conclusion

[lunar cove]

did you get the same ramp QM?

[solid panther]

yes

[lunar cove]

I wonder how big the window is

[solid panther]

it's at least 1 ms but not sure when it ends

[lunar cove]

good work

[solid panther]

counted frames against bkt, i think it's 2 f better

[jade crypt]

Improvement to cave + platforms

[haughty night]

waitttttttt

[distant blade]

a little bit from the airtime before the slip, and a little bit on the MT realign before the ramp

[glass summit]

i should work on this again

literally only did first ramp

if i have time i will

[crude grail]

resynced and improved thomas old l1 ending, ( 35.789 l1 ) ghost is thomas' ghost with 5f of delay

[lunar cove]

There's a faster ghost than Thomas

#rksys-archive message

[warm cedar]

What happened to the other shroomspot

On the turn

[lunar cove]

what other shroomspot

the TAS shroom is faster than everything else

[warm cedar]

Oh wait was that just for flap

[lunar cove]

yes

[crude grail]

oh yeah i didn't saw it

.763 l1 with monster's ghost

[lunar cove]

i think kierio gets a great last ramp trick in his flap https://youtu.be/K0hdlYZnMGg?si=NbrGh_sTscSkftxx&t=42

[crude grail]

yeah

.704 l1

[solid panther]

lmfao three different authors just from resyncing

[crude grail]

i didn't used monster's input in this ghost, so only kierio and thomas

[lunar cove]

@raw harbor good work

I'll try and save time anywhere on lap 1 post-platforms tomorrow when I wake up

[haughty night]

he definitely does

I wonder how strict the qm is for that

[thick scaffold]

I got something good

took some tighter turns and avoided the slip chain before the shroom

[dense owl]

any update on this?

i might have a go at this track

[thick scaffold]

yeah one sec

didnt get the trick, comparison vs the flap here

[dense owl]

trick might be pacelocked, nice improvement

[thick scaffold]

ty, it could also just be the alignment at this pace. not sure

[dense owl]

is tricking on the 7th frame of airtime on the 2nd to last ramp optimal or something

[haughty night]

the offroad bounce is pretty specific so I assume you just want to trick on whatever frame allows for that airtime/landing combo

[lunar cove]

^

[haughty night]

iirc if you land too early you don't turn as far right in the air

[lunar cove]

you could get varying height from the end of the rfh

so there's no exact frame of tricking which gives you the desired landing spot

[dense owl]

.699 l1, 1f more of airtime compared to alego's trick, not sure if avoidable

another 6ms should be saveable from the ending trick

[coarse hearth]

I got something good

[dense owl]

better shroom or better trick?

or something else entirely

[coarse hearth]

Something else

[dense owl]

👀

[dense owl]

is that 2nd bounce after landing unavoidable?

[coarse hearth]

Idk I tried for a little bit and that's the best I could get

[dense owl]

i'll take a quick look before going to bed

[coarse hearth]

Gl

[dense owl]

couldnt avoid 2nd bounce. implemented a slip for a slightly better alignment (closer to syncing with previous ghosts) but loses around 0.3ms

tried to go tigher to save more time but couldnt get the bounce to go over the offroad

[lunar cove]

nice jack

the offroad miss should be qm dependent

[crude grail]

nice job !

[thick scaffold]

its the wheelie timing mostly i think

yeah i got it in 1 min

[lunar cove]

How does wheelie timing affect it

[thick scaffold]

oh lol I completely misread that

I thought you were talking about the bounce after the ramp

carried up to here but I'm getting a bounce now

[thick scaffold]

but yeah this is qm/alignment dependent, if you dont get a bounce here ^^^ the alignment is automatically way better

[thick scaffold]

I think I'm gonna resync from the flap

[coarse hearth]

What... I thought I tried that lol nice job hopefully we can avoid that other bounce

[coarse hearth]

I avoided the airtime, trying to carry it

[coarse hearth]

vs paul, need help on the sc

[lunar cove]

I might mess around with the hop after the boost ramp landing

This boost distribution could be improved by getting the mt to begin charging earlier

[dense owl]

i think you're too far ahead for a bounce over the sc, could be wrong though

[coarse hearth]

When you say "bounce over the sc" what part are you referring to

[dense owl]

just before the shroom

sorry i worded that poorly originally. i noticed that when you're too far forward you cant get that bounce. not entirely sure if its required but i assume it is for the rfh

[lunar cove]

well then we should try and save more time beforehand to get the next QM cycle

[coarse hearth]

Ejay did you get anything

[dense owl]

yeah, idk how much time is required but better distribtuion might be enough

[coarse hearth]

Ghost is 35.699

[dense owl]

did you use ejay's suggestion

[coarse hearth]

No I worked off my own ghost

[lunar cove]

I haven't booted up Dolphin yet

[coarse hearth]

I think it would be a good idea to keep looking for improvements after the ramp at 0:24

[coarse hearth]

But if someone can carry this that would be good too

[crude grail]

.693

vs.699

[coarse hearth]

Nj alego

[coarse hearth]

I think a few ms can be saved on this part

[thick scaffold]

is the last ramp qm locked? at that point we're pretty much tied again

[dense owl]

charged the mt 1f earlier but got bad qm for the next ramp

i also ran into the same issue when i was ~1.3ms ahead

[coarse hearth]

Nj foss

[thick scaffold]

hacked this together really quickly

the question is still if we would even get a good last ramp with these sub frame improvements

the .693 loses almost .4f on the last ramp compared to the flap

[lunar cove]

one thing to try is intentionally releasing the miniturbo (before ramp) late

this will keep you even with the ghost for the line into the ramp

and then the excess miniturbo boost could pull the player ahead after the low trick

[thick scaffold]

comparison 693 vs the flap. we're already ahead by a bit so the rfh trick is lower so it desyncs quite a bit on the bounce. i think the yaw diff is the biggest reason why the current trick is worse/higher

[dense owl]

i didn’t think about this. i might try it later if i feel like playing this qm bullshit track

[thick scaffold]

i already got the stick and ramp in this ghost

[dense owl]

wait im slow did you use my ghost and carry it to the end

[thick scaffold]

yes but only up to the landing from the ramp

[crude grail]

nj guys !

will work on it soon

[crude grail]

improved the ramp + ending, 1f ahead of .693 at last ramp, but couldn't get a good last trick

[raw harbor]

very.

[thick scaffold]

small improvement on the sc of .06f, this should now be ever so slightly faster that it's tied with the flap's pace. I just cant get a good landing atm. needs to touch offroad for 2 frames but I get either 3 (in the pic) or 1 which sends me too wide.

ghost is flap

[thick scaffold]

finally got a drift trick and a .682

still a lot worse than the flap, by .8f

the qm is so fucking precise

[crude grail]

nice job paul !

[lunar cove]

maybe when comparing timesave we need to group the off-road bounce and the ramp as one thing?

since they both heavily affect each other

[thick scaffold]

yeah I have less roll/am more upright so I'm automatically higher when hopping

A better exit from the sc would require more timesave. Flap is about 1/6th of a frame ahead going into the shroom (although I dont think that much is needed)

[lunar cove]

is the current shroom faster than the flap?

[thick scaffold]

yes by this amount

[lunar cove]

ah i get it now

[thick scaffold]

got some good timesave here. would appreciate some help with the landing

[solid panther]

i'm going to try, am i just supposed to try to replicate the previous ghost?

[thick scaffold]

yes up until the shroom. from there it'd be a good idea to resync this with the flap since the goal is to get the same ending trick

[solid panther]

is this shroom better than flap? could try to lose time before this low to resync otherwise

[crude grail]

yea i think so

[thick scaffold]

yes because it needs to catch up to the flaps pace. but that makes us end up with less roll / more height as I mentioned earlier in this thread

[thick scaffold]

oh and heres the flap data. delay is 3803

[solid panther]

i'm not sure if it's possible to carry the entire thing, but i got a stick landing which is ~3 ms ahead of the 682. can send early if you want i need a bit of time to learn the end

[thick scaffold]

yeah sure. I can try as well

[solid panther]

best attempt at continuation, idt i'm doing this right. i can try more after dinner

[solid panther]

i think this shroom is better but it's still nowhere near the flap

[thick scaffold]

I'm almost tied but the boost runs out a frame too early so the wheelie isn't full speed

I'm not able to get more than 30f of airtime going into the cut, flap gets 32f so thats the reason we're short on boost.
I intentionally lost some time before the cut to match the flap's pace but I ended up doing a different version of the sc anyway. The way to go is probably to avoid that timeloss and try to get 31f (like the .682 does, unfortunately the faster we go the worse the lower the airtime gets) of airtime exactly so we can shroom a frame later. As for the cut, only the exit matters (you wanna get some kind of slip to the left there)

[thick scaffold]

Need a bit more time but I've almost got it. 675 already

[thick scaffold]

got it

[dense owl]

hell yeah

[crude grail]

incredible job!

[dense owl]

is it time for lap 2?

[thick scaffold]

Maybe we could for a similar trick on the edge of this ramp as well since we're already quite far to the right. The small bump helps with alignment too

[thick scaffold]

cant get anything useful. alignment for the first turn would probably be an issue as well

[raw harbor]

That’s an interesting idea

Really nice work on the last trick of lap 1 btw

I knew it sucked to perform but I thought that was just me

[lunar cove]

Fantastic job Paul

[thick scaffold]

ty ty

[thick scaffold]

yeah it's stupidly precise. there's virtually no room for modification before it desyncs.

[solid panther]

messed around with start of l2, got a 40 f low which seems decent. couldn't get anything with rfh like on l1 but idt it's much slower if at all

[thick scaffold]

Yep that looks good judging by the Z distance. Seems like we have to lose a bit of time anyway if we wanna resync it from here.

[solid panther]

2nd low is qm locked, but maybe try to improve this before resyncing. i get airtime at ~118 u/f when i land, and i'm not sure what inputs are best during the trick either

[solid panther]

couldn't improve trick, i just continued a little. there are so many different cycles on l2 and l3 compared to l1 that resyncing will be high effort. for example this first platform is moving in the opposite direction and has lower height

not interested in continuing on my own, gl improving the trick / continuing the resync

[crude grail]

i'll try soon i think

[crude grail]

here what i got for the cave

[crude grail]

improved cave

[crude grail]

nice job !

[crude grail]

carried jack's improvement, + tried the ramp line, ghost is l1

[coarse hearth]

Ty and nj carrying it

[lunar cove]

i haven't been keeping up to date with this - is the l2 cave much better than l1? potentially usable on lap 1?

[crude grail]

no l1 is as fast as the l2 one

[crude grail]

this is vs the l1, the video below is an improvement over it so it's pretty much tied before the trick

[lunar cove]

cool

[crude grail]

better trick

[dense owl]

do we sync this part with l1?

[crude grail]

beside the trick ( since there is the fireball in the way ) i think we can think up from the mt release to the end of l2

[thick scaffold]

improved the trick by another 1/6th of a frame but to catch up to the next frame of l1 we'd need half a frame more. Doesn't seem feasable so I'll try to resync it from this frame now

[solid panther]

was trying this earlier, i agree that gaining another frame doesn't seem realistic

[thick scaffold]

It'd have to be as fast as l1 despite the fireball in the way

resyncing the last trick rn

[thick scaffold]

resynced the end on l2. idk why I couldn't get a similar grounded hop as l1 so I'm losing a bit by going over the small gap

[crude grail]

nice job

i think we'll have to chage the first trick of l3 too

bc of the rock that fell

[thick scaffold]

ty, I'll have a go later at the first ramp

the platforms after the cave seem really good l3, otherwise the cycles are similar to l2 (fireball in the way again)

[thick scaffold]

didnt get anything, i'll work on another track for now

[solid panther]

best l3 trick strat i could come up with. different angle makes the realignment more awkward. should be improvable

[mortal turtle]

I managed to improve it by about 0.25f, but is this usable?

[solid panther]

hopefully it is. i have a feeling it's possible to get to the ramp 1 f earlier than my ghost with good optimization

[solid panther]

more timesave, got good trick qm 1 f ahead. note this is ~1 ms behind l2 qm

[raw harbor]

[lunar cove]

nice job

[crude grail]

improved a bit the trick + resynced the cave

[thick scaffold]

improved the cave exit and tried up to the ramp where I'm 1-2ms short. seems like the fireball is gonna make us lose a frame as well this lap :/ (ghost is l1)

[solid panther]

nice track

[thick scaffold]

shit's just fallin apart

nj on carrying this I was still stuck on the flip ramp lol

[solid panther]

ty i'll try to crudely resync by losing a frame and send lmao, but we should look for potentially better strats for the shroom since there's one less rock in the way

[solid panther]

okay just going to send. i lose 4 f at last trick, may be possible to save a frame at shroom too

[solid panther]

1 f better shroom + good last trick. nearly done finally. last trick is a couple ms worse compared to previous laps

[raw harbor]

1:46 W

[thick scaffold]

757

I think it's done 👀

[solid panther]

nice you got the 22 f low too, i can give it one more session tomorrow morning but i also think it's likely done

[lunar cove]

I can take a look

I doubt I'll be able to do anything though

[dense owl]

gonna be hard considering this is a long ms due to rounding, 1.8ms to be exact

[raw harbor]

Excited to see another pre-2019 NG 3lap gone!

These are all the NG 3laps from before the purple bkt era

[solid panther]

nice we saved all the most annoying tracks in the game for last + rds

[raw harbor]

• rDS lmao

Really MT should be included on that list

The kart bkt is an alt Strats tas that happened to be faster

[solid panther]

mt is the worst track in ng but the new tas is tentatively done at least (thank god)

[lunar cove]

need to resync everything

but the bulk is done

[solid panther]

mh prob won't be too bad but it'll probably have to be redone like 4 times

[lunar cove]

where's that list of NG tracks hang on

LC 🟩
MMM 🟩
MG 🟩
TF 🟨 1-0-2 & fast walltrick lap 2?

MC 🟩
CM 🟥
DKSC 🟥
WGM 🟥

DC 🟨 Test Daisy + Mach Bike
KC 🟥
MT 🟥
GV 🟩

DDR 🟩
MH 🟥
BC 🟨 First walltrick and halfpipe rfh lap 1?
RR 🟥

rPB 🟩
rYF 🟩
rGV2 🟩
rMR 🟩

rSL 🟩
rSGB 🟩
rDS 🟥
rWS 🟨 2f improvement on lap 1 (need someone to AGCync it)

rDH 🟨 Last turns?
rBC3 🟩
rDKJP 🟩
rMC 🟩

rMC3 🟩
rPG 🟩
rDKM 🟨 A bit unoptimised?
rBC 🟩

Getting there

[solid panther]

rdh i want to try fully optimizing last turn l1, first turn l2, add a-tech, and see if i can squeeze out a frame on l2 and l3 that way. (also i'm still on the double trick copium)

dkm i also want to get around to testing shooting star. i messed with it for a few mins and didn't seem any worse at slope boosting than wild wing

[dusty lantern]

odb isn't strictly worse at slope boosting, but it has a hard time maneuvering with the EV afterwards without superhopping

since you need to turn towards the direction you're facing

[lunar cove]

I'm honestly super excited at the idea of not a single timesave being known about for NG 3laps

[dusty lantern]

no spindrift=lose ev
spindrift=banned

[icy nova]

How thin is the line between slope boosting and superhopping on an odb

[solid panther]

is it even superhopping if you do a spindrift on downhill? they're like spaced out

[lunar cove]

I think I suggested a while ago that if it's proven useful on ODB rDKM then we do another vote

because I don't think anyone really knows the line

It depends on what the driving looks like I guess

[dusty lantern]

I just feel like if people already complained about the 1-2 EV gotten from the rfh low trick, there's no way people will be happy about spindrifting and gaining like 14

[solid panther]

i mean is odm not allowed in ng unless it's on slow ramps? i would be okay with just saying "no glitch does not exist for odbs"

cause like could you not do something like the mt strat and just get a bunch of bounces to maneuver down the mountain

[icy nova]

I agree. ODB’s must be a glitch cause there’s no way they were intentionally programmed to be that bad

[lunar cove]

ODM is currently only banned on MG

imo banning ODM for individual problematic tracks makes more sense than just banning an entire class of vehicles in a category

anyway we should probably move channels

[solid panther]

police i would love if you cooked up some basic wip so at least we know what we're going off of

i don't think i'd be able to find the best strats on my own

[lunar cove]

Latest rkg

Who's uploading this btw

Or who would like to

[dusty lantern]

I don't have much experience with dkm and there's a lot of possibilities, maybe could but probably not soon

[solid panther]

i would be happy to upload. if someone else wants it more that's fine it's not a big deal

[lunar cove]

dude GV is perfect for alestorm

or your other go to artists

is there a song about lava lmao

[solid panther]

i have had a song choice in my head for months, honestly i don't care who uploads i care about the song more lmfao

[lunar cove]

send it

[solid panther]

https://www.youtube.com/watch?v=EOU2HjLCgJA

[lunar cove]

hard

[icy nova]

Slipknot mentioned

[lunar cove]

I wonder if that'll get blocked by yt

[chilly anchor]

for alestorm the closest thing is probably "mead from hell"

[lunar cove]

it's not under the UMG label so there's a chance not...

[solid panther]

oh is there like a gay version i can use

[lunar cove]

Sulfur (kidz bop)

YouTube used to have an official page where you could look up a song and see the copyright outcome if you used it. Genuinely worst day ever when they removed it for no reason

[jade crypt]

fast walltrick l2?

[lunar cove]

I can't remember where that's from tbh

Maybe Corvid mentioned it

But from what I can tell, l2 looks optimised and the same as lap 3

[jade crypt]

the wall trick is the same on all 3 laps

[lunar cove]

it came from this message, I just assumed he knew something I didn't

[coarse hearth]

We should try to save another frame. The turns after the ramp at 1:35 can be improved by multiple ms. The shroom also loses 10 ms

[solid panther]

the trick exiting the red room is 1 f faster on l3 than other laps

[solid panther]

i lost 10 ms because the final trick is very qm locked. i'm pretty sure the previous ramp is too. i can try it ig

[lunar cove]

thanks, i tried to find the info in dms but couldn't

[solid panther]

on tf i wanted to spend some time trying to make the reject road faster on every lap and not just l3. idk if it's possible that section is so annoying. if it works out i think 0-1-2 would avoid the stompers handily

[grave pike]

maybe this can lead to something but the bounce would need to happen later

[coarse hearth]

I don't think the previous ramp is qm locked. I'll try later

[crude grail]

vs the l2, with marth ghost managed to lose no time over l2, but not fully synced at the end, i think it can be improved

[solid panther]

742, should be possible to get 740 with the same trick as before but i'm done trying for now

[thick scaffold]

hows this trick different than before? looks the same to me w/ 22f airtime

[solid panther]

had to drift 1 f more to get 22 so i think i take a worse line to finish

[dense owl]

what's better than a 740?

a 739

[solid panther]

got a slightly better 739. personally i think this is finished but feel free to try improving this more

[dense owl]

how close to 738?

[solid panther]

can someone with more patience than me confirm that the author list is accurate? alego, aoiro, cesar, corvid, ego, ejay, foss, jack, kierio, marth, mitch, paul, shadow, thomas

[astral rivet]

massive collab wow

[solid panther]

idk what the fuck even happened to get to this many people

[astral rivet]

yeah crazy

wanted to help a bit more tbh but was unable

iirc i worked from the end of the cave till boost ramp (?)

[lunar cove]

I worked on the cave exit ramp lap 1 to the ramp before the shroom

My inputs might have been reused on later laps too

[astral rivet]

^

[dense owl]

is this the biggest ng collab?

[frosty swan]

rsgb ng had 15 i think

[thick scaffold]

bold to assume that there'll be an author display when Corvid is uploading

[glass summit]

if i still have inputs they are like on the First second of the run

i tried to improve mid-run before but yall Just too good

[solid panther]

this guy gets it...

[solid panther]

i finished rendering the 739, in theory i could upload today. should i wait 2-3 days for tf flap?

[dense owl]

2 uploads on the same day sorta undermines one of the runs no? imo atleast

[solid panther]

which is why i'm asking if it's fine to upload today before ejay, and we trust the run is finished, or wait

[lunar cove]

I vote for wait personally

[coarse hearth]

Do you think there's any chance another frame can be saved between qm locks somewhere

[astral rivet]

i can try check tomorrow

[solid panther]

l1 no. l2/3 not sure, i didn't work for most of them. i doubt there's enough timesave at first trick on either lap to save a full frame. flip trick loses a frame on both laps due to the fire, i haven't tried to save it at all

[coarse hearth]

I tried to save the frame on the fire turn and didn't really get close. I also tried getting a clean landing on first trick l2 and couldn't get anything. The only other thing I can think of would be the cave turns because they are all different every lap. Might be worth looking there again every lap even if it's unlikely to save a frame

[solid panther]

oh iwasn't aware of the cave turns being different, i'll have a look later

[dense owl]

I saved like 9ms on the l2 cave but qm didn’t let me get much further

pavelocked by the stick exiting the ramp, i imagine it’s the same for l3

[raw harbor]

that was fake 15, all the contributors were on the last turn

[haughty night]

MT UR was 16

oh you asked about ng

[raw harbor]

Yup

I think biggest NG so far was bc3

[haughty night]

DDR NG ?

[raw harbor]

Lemme check I literally just made a list last night

All the biggest collabs have happened this year

2025 collabs

16 - MT UR 3lap
13 - DDR NG 3lap
12 - rBC3 NG 3lap
10 - rWS NG 3lap, MG NG Flap, BC UR Flap
9 - rGV2 NG 3lap, MC NG Flap
8 - MG NG 3lap, MC NG 3lap, rPB NG 3lap, rGV2 NU 3lap, rBC UR 3lap
7 - LC NG 3lap, MMM NG 3lap, rBC NG 3lap (Daisy), WGM UR Flap, rGV2 NG Flap, rSL UR Flap (5.0)
6 - rSGB NG 3lap, rMC NG 3lap, rMC3 NG 3lap, rSL UR 3lap (5.9), rDKJP UR Flap
5 - MG NU 3lap, BC NU 3lap, rPG NG 3lap, rBC NG 3lap (Funky), LC NU Flap, GV UR Flap, DDR NG Flap, rPB NG Flap, rSGB NG Flap

I cut it off at 5 because I wanted to

[dense owl]

was rsgb contributors actually just carried by the last turn

funny asf if so

but anyway wow biggest ng collab yet

[solid panther]

i am not able to save a frame at l2/l3 cave either. i think that's a dead end

[coarse hearth]

Sounds good, I can try tomorrow, I'd like to try out marth's strat

[coarse hearth]

Tried for a bit and couldn't improve it

[astral rivet]

i saved half a frame on cave exit

okay it's actually not half a frame but 0.4f

if someone wants to try

imma keep trying as well

stuck here

i got a great drift that i could softdrift entirely and get the mt earlier

[thick scaffold]

wow that low is fucking insane, nice job. I'll have a go

[solid panther]

i can't get more than 10 ms ahead. idk if this is viable

[thick scaffold]

I got a frame ahead already looking to resync now

[solid panther]

okay damn! corvid is washed in 2026 alright

actually before you resync this; have you tried the same thing in previous laps? l1 should work for sure, l2 has different cycles so off the top of my head i'm not sure

[thick scaffold]

l2 the platform is way too far left and l1 saving half a frame doesnt do anything (i think)

[solid panther]

wait how did you save a frame then? half a frame before flip trick half a frame after?

[thick scaffold]

yeah the fire only lost half a frame

[thick scaffold]

used this

[solid panther]

oh interesting i thought the flip trick was qm locked

[thick scaffold]

it felt like it lmao

[solid panther]

like i always got big air before it i guess i saved too much time lmao

[thick scaffold]

i dont even know how i got it

[solid panther]

yea that's kinda crazy gg i'm going to try it out on l2 tomorrow just to make sure

[astral rivet]

🙏 ty

[thick scaffold]

722, thanks again cesar for that crazy low.

[lunar cove]

fantastic

[raw harbor]

Is any of this usable for l1/2

[lunar cove]

^

[raw harbor]

Oh thanks

I literally read all that

[solid panther]

the platform cycle is way off on both laps anyway

[raw harbor]

right

[solid panther]

has anyone tried to look into why this part is so different on l3? (ghost)

there's too much going on i don't get it

[haughty night]

presumably the platform is moving the other way... it also looks like the ghost lands from the first trick earlier so it has to land past the platform earlier to realign during decel (?)

[solid panther]

i got this but i have no clue how to carry it

[solid panther]

1 f improvement to l3 cave

i think this can be applied to l1 and l2 as well

[raw harbor]

looks much better :D

thanks for speaking up corvid

i actually had the same suspicion when i saw alego resyncing l3 cave but wasnt really thinking much about it

[solid panther]

it definitely isn't l3 only

[solid panther]

got the frame save on l2 as well

[lunar cove]

nice!

[solid panther]

i need help for l1 though. we need a pretty specific sequence of bounces to happen for this and i always get a stick instead for some reason. because the platform is moving in the opposite direction, there's a slipchain to align for the uphill, maybe that's causing the suspension to be different?

[solid panther]

gave l1 more tries but i just can't get anything. i'm throwing the towel and writing it off as cycle issue personally, but i would like you guys to also try it a bit before resyncing

[dense owl]

forgive me if i'm wrong but wasn't there plans for a sneakster run to be made

[solid panther]

no thomas was deluded that it's fast because there's one rehop but it isn't fast

not enough drift

[thick scaffold]

I got the first set of bounces by being more ahead but I'm short on yaw

touching offroad mid air, see if you can fix it cuz I'll work on something else

[thick scaffold]

vs ur ghost here

[solid panther]

gained 0.7 f with your ghost, not quite enough for a slip yet

i hate this turn so much

[thick scaffold]

what the fuck is this bullshit

how do you not touch offroad there

[solid panther]

if only i had abit more room it would've been so easy

[thick scaffold]

I was working on more room atm

I'm .01f behind and I'm no longer getting bounces

[solid panther]

yup...

[thick scaffold]

wait I got something in between tho which still gets bounces

try this it's 2 units more right and slightly more yaw

[thick scaffold]

wait I'm insanely stupid the offroad wasnt slowing me down I was just turning more after landing 💀 💀

[solid panther]

got the slip with that though still a bit behind should be close though

nice word vomit

saved the frame. fucking finally

[thick scaffold]

[solid panther]

hardest frame of my life man god damn

are you interested in resyncing paul?

[thick scaffold]

I am

lets get this shit done

[solid panther]

🔥

[haughty night]

lfg .6 now

[thick scaffold]

l1 done

[raw harbor]

omg i cleared these notifications on autopilot because i almost missed the fact that you guys got the l1 resync 😭

nice work

[solid panther]

i just remembered that this was suggested at one point. tried it again and it's definitely slower.
i also tried first cave turn without a slip. if it weren't for the chain drift qm being so stupid it would be faster

[thick scaffold]

im gonna need help on the flip ramp

the issue is most definitely that I'm .01f behind, maybe the platforms need to be reworked entirely

[thick scaffold]

oh I should mention the cave turns can be taken better. May cause issues with the qm lock at the cave exit but that should be the first thing to try imo

[thick scaffold]

small improvement which should be enough for the ramp but i cant get the quickhop qm that gives you 1 yaw for free

fuck this

[crude grail]

got it

[crude grail]

.678, last low can be improved ( 23f of airtime rn )

[solid panther]

672, should be finished for real now. ty guys for help

[thick scaffold]

ggs

l3 had hands god damn

[solid panther]

yea i was trying it too, maybe the platforms were on a slightly different cycle and it was messing up the suspension just enough to all be different

[thick scaffold]

cave took an hour too, I did it with way different bounces

[solid panther]

the only frame that has a remote chance of being saved is l2 from cave exit to shroom. i messed around with different lines for the platform, but saving enough time for that seems pretty bleak, i'm not sure it's possible. if someone wants to try that i'll give it a couple days of wait

going right early at the first platform is faster but then you're moving against the other platform so you lose a bunch there. i was also able to get a 21 f trick

[lunar cove]

🔥 🔥 🔥

[crude grail]

did u got like a 0.4 0.5 f improvement on the l2 plateform ?

[crude grail]

bc if so we can do this right ?

( not the low, the improved part after the platform )

[solid panther]

i didn't find enough timesave sadly. i couldn't get a lower trick without a bad platform line. i noticed the mt before the flip ramp is held for 1 f long so we only need to find about 0.3 f

idk if it's possible though

[crude grail]

i'll give it a shot today

[solid panther]

nice gl

[crude grail]

ty !

[thick scaffold]

it's not I just started the drift a frame later

looks like .4f is needed

[crude grail]

didn't got anything

[dense owl]

i think this run is done then since you need to save 1ms for a 671

[solid panther]

yea me neither. i will upload the video on the 26th if someone wants to keep trying until then

[dense owl]

tried for a bit at the ending but could only save ~0.25ms which still isn't close