Tool-assisted speedrun of elma

Look for replays and levels and ask people's times.

Moderator: Moporators

User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Tool-assisted speedrun of elma

Post by bene »

I started project to finish all levels with SL. The end goal is to make a video similar to EDQ but without the limitations of live playing.
Some times in table are obviously WIP like int09,int42 and some are superbad and ez improvable like int14 osv.
Go for it if you think you can, just sending a dat where you play nice from start and get a good min distance early can be usable.
Currently zanpe, finman, anpe and b0ne have bene driving to make the times.

Be creative, anything is allowed. Even apple bugs!
Want to improve an existing run? Don't be afraid to ask for the dat-file!

How to submit? Send pm to my or post here.

Rules
You have to submit a valid mila sl .dat or hourglass .wtf file that finishes the level.
All recs have to be theoretically drivable = they must be result of some time events and key presses on non-xiit elma.
End video will include credits for entire project as it is impossible to give single level credit to mans. Like int08 and int29 had 3 mans doing separate parts.
Be fast!

Current table of times

Code: Select all

1.  Warm Up              0:07,97
2.  Flat Track           0:08,09
3.  Twin Peaks           0:07,30
4.  Over and Under       0:10,08
5.  Uphill Battle        0:11,04
6.  Long Haul            0:12,57
7.  Hi Flyer             0:16,51
8.  Tag                  0:06,76
9.  Tunnel Terror        0:14,60
10. The Steppes          0:07,09
11. Gravity Ride         0:08,46
12. Islands in the Sky   0:08,25
13. Hill Legend          0:07,89
14. Loop-de-Loop         0:08,69
15. Serpents Tale        0:11,05
16. New Wave             0:22,10
17. Labyrinth            0:15,41
18. Spiral               0:11,21
19. Turnaround           0:10,44
20. Upside Down          0:18,11
21. Hangman              0:07,48
22. Slalom               0:05,82
23. Quick Round          0:07,07
24. Ramp Frenzy          0:09,53
25. Precarious           0:08,82
26. Circuitous           0:07,98
27. Shelf Life           0:08,74
28. Bounce Back          0:07,29
29. Headbanger           0:04,20 blaze it
30. Pipe                 0:09,73
31. Animal Farm          0:18,86
32. Steep Corner         0:10,10
33. Zig-Zag              0:18,81
34. Bumpy Journey        0:16,96
35. Labyrinth Pro        0:23,26
36. Fruit in the Den     0:14,66
37. Jaws                 0:07,66
38. Curvaceous           0:09,11
39. Haircut              0:23,28
40. Double Trouble       0:04,93
41. Framework            0:05,27
42. Enduro               0:12,41
43. He He                0:13,36
44. Freefall             0:06,36
45. Sink                 0:12,81
46. Bowling              0:26,75
47. Enigma               0:07,64
48. Downhill             0:13,22
49. What the Heck        0:09,07
50. Expert System        0:13,00
51. Tricks Abound        0:11,18
52. Hang Tight           0:07,86
53. Hooked               0:05,89
54. Apple Harvest        0:21,34

Total                   10:13,97
Last edited by bene on 14 Nov 2015, 00:32, edited 21 times in total.
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

bene wrote: 29. Headbanger 0:04,20 blaze it
what?
TT: 36:59:53 || Avg TT: 38:09:65
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: Tool-assisted speedrun of elma

Post by bene »

Super ez if you don't bang your head!
http://kopasite.net/up/3y697nm7p86a1oq/29be420.rec
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
User avatar
kuchitsu
Kuski
Posts: 1423
Joined: 13 Aug 2010, 20:31

Re: Tool-assisted speedrun of elma

Post by kuchitsu »

Not that anyone asked me, but to someone with zero experience in milatool this seems totally uninteresting since it feels like any kind of ridiculous lightspeed wheel travel through ground is possible, at which point there are so few limitations that nothing is impressive anymore. But I guess these actually require a lot of effort, intricate tweaking, etc and it's just that most people don't have enough knowledge to appreciate it?
iltsu
37mins club
Posts: 248
Joined: 25 Aug 2011, 19:15
Team: EF

Re: Tool-assisted speedrun of elma

Post by iltsu »

I wont even watch that video because i hate such shit bug rides.
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

bene wrote:Super ez if you don't bang your head!
http://kopasite.net/up/3y697nm7p86a1oq/29be420.rec
ye... looks legit.
kuchitsu wrote:Not that anyone asked me, but to someone with zero experience in milatool this seems totally uninteresting since it feels like any kind of ridiculous lightspeed wheel travel through ground is possible, at which point there are so few limitations that nothing is impressive anymore. But I guess these actually require a lot of effort, intricate tweaking, etc and it's just that most people don't have enough knowledge to appreciate it?
i kinda have to agree. i assume those recs are super tricky to make. but watching one bug bounce after 3 seconds and then BOOM flower, it's very uninteresting.
TT: 36:59:53 || Avg TT: 38:09:65
User avatar
Schumi
39mins club
Posts: 866
Joined: 27 Jul 2003, 06:44
Location: Ikrény, Hungary

Re: Tool-assisted speedrun of elma

Post by Schumi »

agree with Lukazz, it has totally no sense... sorry... shitbug recs are a total ruin of elma feeling
Elasto Mania - 34:22.20 | #416 - 12. April 2021
Ancient Internals
User avatar
anpdad
38mins club
Posts: 933
Joined: 6 Nov 2003, 06:04
Location: Russia

Re: Tool-assisted speedrun of elma

Post by anpdad »

It doesn't have to look legit. That's the whole point - to abuse every single known bug including bugbounces, going through polygons, going out of bounds, hooked bugs and so on. Ofc some recs will end up looking boring and difficult to follow, but stylefinding still exists in this category. And there are still limits that won't be surpassed, and imo it's interesting and exciting to find what those limits are.

I haven't really played anything for this project yet (not sure if i will cuz laziness overcomes excitement atm, but who knows), so i can't really describe common obstacles and limitations to making such recs. But as far as i've seen, bugbounce can't magically happen at any speed with any wheelpress. Also, direction of a bugbounce depends on how you position your wheels, so you cant arbitrarily bug anywhere and fly towards any chosen direction at will, hencewhy there are two bounces in Headbanger for example, instead of one. Theoretically you could clip downwards through the ground on the very first bugbounce, but you'd be left with supersmall speed and stuff. Another main obstacle are crashes. It crashes all the time if you get too speedy bounce or fly away too far. Thanks Balazs!

So basically, it seems you need to come up with a plan on where to get hooked bugs, where to get bugbounces in the route and then spend a few hours fiddling with random fps values trying to make bug behave in a way that doesn't crash elma while also making it behave in a way that works for the chosen route. This last part doesn't seem to require much skill, but is the most time consuming one, probably. As an example, in int08 bene gets a bugbounce early at the top of lev, the wheel shoots downwards through the ground, swings like a pendulum, flies through the bottom right apple, takes bottom left apple first, then swings like a pendulum back, takes bottom right apple, and then finally goes back up a bit, picks up the top apple and couple frames later suddenly takes flower. That's some really funky stuff imo, and is funny to watch and analyze. Probably not as funny to drive, but who knows :D

EDIT: why so shit reclay b0ne?
Kortsu
Kuski
Posts: 412
Joined: 4 Oct 2009, 21:12

Re: Tool-assisted speedrun of elma

Post by Kortsu »

Wow, this is an interesting project. It'll be fun to see how broken the game really is, which I guess is one point of such undertaking. Will you submit this to TASVideos or are there some restrictions there to not make it acceptable?

Anyway, I wish you good luck and resilience.
User avatar
Zweq
34mins club
Posts: 4055
Joined: 28 Nov 2002, 15:54
Location: suo mesta

Re: Tool-assisted speedrun of elma

Post by Zweq »

Yeah there are limitations, you can't do whatever you want and magically get 5seconds in every lev just like anpdad said. Of course you are free to dislike a subcategory in elma like I dislike piping, but ironically this is something more than just a subcategory, this is the final step in finding answer to what are the best theoretical times, the fundamental point of speed running. If you call it shit or whatever then you're playing the wrong game. And I'm not even contradicting my own earlier comments about bugs because bugs in live playing are arbitrarily limited, which obviously works somewhat in legitplaying but applying such method in toolplaying would be mindbuglingly stupid, here we have no arbitrary limits, the limit is decided by the code in .exe!!

The biggest problem right now is now the ultra frustrating internal errors, really hindering progress. Thanks Balazs!
Image
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: Tool-assisted speedrun of elma

Post by bene »

There are restrictions in milasl so it will not be accepted to TASVideos. Their videos have to start from 0 and go to finish, timing is made from first input to last input or something whereas I have to use stats.txt totaltime because there is no menu switching involved.
And even if you would fix milasl to include all of this I don't know if they will just randomly accept milasl as a tasing tool for this game without testing it osv.
I am sure their community is filled with people that is interested to watch even though it cant be submitted though, so I will post over there later in the project.

anpe didn't mention that you are also not invincible while you do bugs, so in tag I didn't only have to höyl for a bug where the wheel heads to the right position, I also have to make sure that my head moves away from the ground and not into the roof (That cave in the start is really tight during a bug bounce) so you don't die later on. Sure you can clip with your head, but that requires you to move the entire distance through the wall in a single frame.
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
Cold
Kuski
Posts: 25
Joined: 10 Apr 2015, 16:27

Re: Tool-assisted speedrun of elma

Post by Cold »

Zweq wrote:Yeah there are limitations, you can't do whatever you want and magically get 5seconds in every lev just like anpdad said. Of course you are free to dislike a subcategory in elma like I dislike piping, but ironically this is something more than just a subcategory, this is the final step in finding answer to what are the best theoretical times, the fundamental point of speed running. If you call it shit or whatever then you're playing the wrong game. And I'm not even contradicting my own earlier comments about bugs because bugs in live playing are arbitrarily limited, which obviously works somewhat in legitplaying but applying such method in toolplaying would be mindbuglingly stupid, here we have no arbitrary limits, the limit is decided by the code in .exe!!

The biggest problem right now is now the ultra frustrating internal errors, really hindering progress. Thanks Balazs!
If you want, I could pm you my anti-crash program (with source code) which can either restart the level when crashed or put you back into the main menu or I could change it to do something else.
Canada
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

i have two questions:
1. how does the head-going-through-ground bug work? is it that at a certain speed the head just goes through the wall or is it something else?
2. someone said mila-sl isn't working with windows 8 or the ddraw.dll. is there any workaround available for this at the moment?
Zweq wrote:If you call it shit or whatever then you're playing the wrong game.
I think most of us aren't saying that the subcategory itself is shit, it's just way less interesting to watch these recs than the "doable" mila-sl recs with new styles, although they are probably as unrealistic as these bug bounce ones. at least i assume that the Tricks Abound mila-sl rec with the wheel stretch or the hi-flyer rec style are as unrealistic to be ever driven without any tools as these bug recs. i understand that it's very hard to create styles for this kind of playing and actually finishing them, but for those of us (i.e. 99% of the already very small scene) who aren't familiar with the possibilities of mila-sl and bug bounces, we just can't appreciate it, because we can't compare it to anything else. i can't say if bone did that 29rec in 5 minutes or hoyled 2 weeks for it. and i have no clue if it would be hard or easy to improve 0.5 in that lev or what a good time would be.
TT: 36:59:53 || Avg TT: 38:09:65
User avatar
Zweq
34mins club
Posts: 4055
Joined: 28 Nov 2002, 15:54
Location: suo mesta

Re: Tool-assisted speedrun of elma

Post by Zweq »

Lukazz wrote: 1. how does the head-going-through-ground bug work? is it that at a certain speed the head just goes through the wall or is it something else?
Head appears behind wall in someframe and after wall in nextframe due to long timestep(<50fps) and high speed(imo 600+ ?), so you skip the collision. Nab code or something? 8)

Lukazz wrote: i can't say if bone did that 29rec in 5 minutes or hoyled 2 weeks for it. and i have no clue if it would be hard or easy to improve 0.5 in that lev or what a good time would be.
I dont know how much exactly was used but I heard bene used 2 hours just for the end, and I also heard finman did some part of that rec. I did the going through walls part in about 10 mins. I also tried to finish but I only wasted 1 hour. Then I just gave .dat to bene hoping he can do it.
Image
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

Zweq wrote:
Lukazz wrote: 1. how does the head-going-through-ground bug work? is it that at a certain speed the head just goes through the wall or is it something else?
Head appears behind wall in someframe and after wall in nextframe due to long timestep(<50fps) and high speed(imo 600+ ?), so you skip the collision. Nab code or something? 8)
you gotta love the elma code. such a simple explanation actually. ;D

it's kind of funny actually. the very idea of elma is "speedrunning" the levs, yet bug-exploitation has never been a part of that (up until now), although it's very common in the speedrunning community.
TT: 36:59:53 || Avg TT: 38:09:65
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: Tool-assisted speedrun of elma

Post by bene »

Lukazz wrote:1. how does the head-going-through-ground bug work? is it that at a certain speed the head just goes through the wall or is it something else?
The game works like this:
1. Move bike depending on speed, direction and frame time, frame time is 1/fps.
2. Check collision (Head touches ground = die, Wheel touches ground or inside ground = Move wheel to be on top of ground osv.)
3. Show frame to player.

This works at normal speeds and high fps because you move small distances and constantly check collision and adjust as needed.
At high speeds and low fps the bike moves too far and code stops working.
So there is no check if the head touched the ground between the frames.

Wheel through ground is super easy and common because the wheel adjusting code is bad and you also don't die from the wheel hitting the ground. You can do this live in int44 in many spots by just going 30 fps and volting like a madman while falling. http://www.recsource.tv/r/qpxdjvguye

Head through ground is hard because you have to move the full distance in exactly one frame.
Check this frame by frame image in 29.rec
http://kopasite.net/up/46r6i2eao1z55x8/Frames.png

I'm just guessing about everything I'm saying here I have not checked the game code for an exact explane. But basically high speed + low fps = clip, as zweq mentioned.
Lukazz wrote:i can't say if bone did that 29rec in 5 minutes or hoyled 2 weeks for it. and i have no clue if it would be hard or easy to improve 0.5 in that lev or what a good time would be.
Well finman drove the first 2 bounces, dunno how much he spent for that, the 2nd one was bad and impossible to work with so after some experimenting (2 hours over 3 days or so) i managed to go from 0.0002 min distance to 0.0001, this is when zweq started playing, now bounce was in almost right direction but too low speed (around 400) for clipping. So after spending some more time I lucked 0.0000 min distance, which is extremely hard to get. zweq drove to flower and then for the last 0.20 or so I used 2-3 hours and zweq played maybe an hour. And that is to just finish, höyling the end would maybe give a better time, the flower gets grabbed on a really late frame. But with these runs just finishing makes you happy enough.
Lets estimate that everyone together put in 10 hours total to make this one finished rec. And this is just an easy 4 second rec. Our current lab pro rec is 10 seconds long and several hours in the making taking the first 2 apples so far, but the 2nd grab is not apple bug and must be fixed somehow if possible because apple bugs are very valuable.
Some recs are super ez like hooked took maybe 30 mins, drive normally hooked bug and bugbounce on that apple to the flower. It's easy cause its 99% normal driving with a really short and normal bug bounce. Maybe hooked is improvable with some crazy new style or route that I didn't think about who knows.
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

thanks for the detailed explanation! and thanks for the frame by frame pic.
what can you actually do to control where the bike is going after the bug bounce? does it depend just on the "angle of entry" or are there other factors too? also is it always 'closer wheel distance = stronger bounce', or does it have some randomness or other factors too? and what are you actually doing after the bounce? because when you watch these recs, for me it looks like you've been trying 50 different bounce angles/places and after one of these the bike magically touched the flower somehow, but now after your explanation i understand that the harderst part is to hoyl the part after the bounce. but that seems totally random to me, wheels just go crazy, i can't imagine how you would actually control the bike in this state.
TT: 36:59:53 || Avg TT: 38:09:65
User avatar
Lousku
Kuski
Posts: 2925
Joined: 5 Feb 2010, 00:25
Team: BAP
Location: expensive land of dads

Re: Tool-assisted speedrun of elma

Post by Lousku »

Very offtropic: that looks like an figurehead.
then again i don't know anything
maybe easier not to think abouut alöl things thought than not things thought ... or something..=?
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: Tool-assisted speedrun of elma

Post by bene »

Lukazz wrote:1)what can you actually do to control where the bike is going after the bug bounce?
2)does it depend just on the "angle of entry" or are there other factors too?
3)also is it always 'closer wheel distance = stronger bounce', or does it have some randomness or other factors too?
4)what are you actually doing after the bounce?
5)for me it looks like you've been trying 50 different bounce angles/places and after one of these the bike magically touched the flower somehow
1) Well the bug happens in a direction. I don't really understand how it is determined mila could probably give best explaination. But basically it's determined by wheel positions relative to the polygon you are on. If you compare int47 32tt project to int02 zweq bug you can see that they have the same wheel positions on the same polygon and the bounce is exactly the same. 32tt project rec just uses higher fps. The direction of the bug is inverted if you gas on the min distance frame instead of braking is all I know for sure. The direction does seem to follow some logic of normal bounces however. With some magic fps change after the bug you can do all kinds of funny stuff like changing rotation completely, more on that in answer 4.
2)dunno
3)Well lower distance is usually stronger bounce yes but changing distance also changes bug direction so lower is not always better. With 0.0001 or 0.0000 you can make sick bounces even at 999 fps.
4)Playing from frame of the min distance you try various fps values several times and hope wheel or bike goes where you want. Then you can keep trying and saving when you get the result you want. The physics is extremely funky during big stretches the bike will behave normally during 999 fps but if you change to low fps for a few frames and go back to 999 you can get all kinds of funny results. I use this trick in int12 for a spinning end. Compare this 13.34 http://www.recsource.tv/r/yeqiudhznt finish rec to another one where i just constantly 999 fps http://www.recsource.tv/r/rywbvuclmn after 2nd bounce at like 7s. Also I am aware that this rec is pretty bad theoretical-time wise but it looks super funny! You can also do a really tedious walk with the wheel like int int03 http://www.recsource.tv/r/uaieyhsgto end where the wheel magically shoots in the direction you want. Here you basically live on the edge of internal error increasing fps gradually to not crash but low enough to keep bug going and save when the wheel moves in the direction you want.
5)No it's usually one bounce that works somewhat and then you alter it to your favor.

Edit) You don't always get a bug bounce just because you have low min distance. Sometimes nothing happens dunno why
Edit2) Can mention that I am still just guessing about everything. All is new and unexplored for me.
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
User avatar
Mats
39mins club
Posts: 1183
Joined: 5 Apr 2007, 12:30
Location: Norway, Sandnes

Re: Tool-assisted speedrun of elma

Post by Mats »

Hmm, is there normal volt delays on this?
TT:39.59.86|| AvgTT:41.49.24 || Multi TT:27:43:82 || Team [TR]
User avatar
Hosp
38mins club
Posts: 2105
Joined: 30 Aug 2009, 20:55
Team: MiE
Location: up yours
Contact:

Re: Tool-assisted speedrun of elma

Post by Hosp »

Yass
Image
ToMaT
User avatar
FinMan
36mins club
Posts: 2038
Joined: 13 Feb 2007, 11:14
Team: dat
Location: Jyväskylä, Finland

Re: Tool-assisted speedrun of elma

Post by FinMan »

Mats: "All recs have to be theoretically drivable = they must be result of some time events and key presses on non-xiit elma." :|

On a brighter note I am really, really excited about this. I even feel really bad that I have so much to do in elma, like HALF pack and non-tas speedrun, not leaving as much to this project as I'd like to. Whatever you say, dislike bugs or not, this stuff takes a lot of effort and a lot of knowledge. After all no one is claiming these recs "legit" or whateverthefuck, but non-legitness is almost the goal of this thing.
Image
User avatar
Fruitós
Kuski
Posts: 21
Joined: 30 Sep 2012, 20:22

Re: Tool-assisted speedrun of elma

Post by Fruitós »

Pretty excited on this :D

I used to TAS some games a few years ago, and I know the bunch of work it requires.

Good luck with it!
User avatar
milagros
Cheatless
Posts: 4560
Joined: 19 May 2002, 17:05

Re: Tool-assisted speedrun of elma

Post by milagros »

you can add my 55 time (even though it got way better time with extreme megabugs)
[carebox]
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: Tool-assisted speedrun of elma

Post by bene »

Update:
Loop-de-Loop 0:08,69

Can ez add a 55 time, what was time with extreme megabugs? where is .dat?

Edit:
Update:
Turnaround 0:10,44
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: Tool-assisted speedrun of elma

Post by bene »

Using a frame perfect turn+volt you can clip with head at relatively low speeds cause head moves instantly to a new position when you turn.
http://www.recsource.tv/r/dahkegsocw

This is made with 214.86 speed at frame where turn is done.

Edit to explane some more i guess)
This is possible at any wall. Increasing speed will increase window for input. You can also remove volt and/or turn with higher speed. The lower fps you have the lower speed you need. This is with as low speed as i was able to do. Some 204 couldnt get to work.
Theoretically ofc.

Ofc this has been known and discussed since 1997 but in all cases i know it was just animation bading and no real clip.
Last edited by bene on 2 Sep 2015, 09:20, edited 2 times in total.
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

omg, elma ;D
TT: 36:59:53 || Avg TT: 38:09:65
HatCat
Kuski
Posts: 39
Joined: 22 Apr 2015, 18:58

Re: Tool-assisted speedrun of elma

Post by HatCat »

I'm feeling the same way as some others: a tool-assisted speedrun of Elma might be more interesting if all the moves performed were legit, that is, as would be accepted in a normal run by WR-table standards.
User avatar
FinMan
36mins club
Posts: 2038
Joined: 13 Feb 2007, 11:14
Team: dat
Location: Jyväskylä, Finland

Re: Tool-assisted speedrun of elma

Post by FinMan »

HatCat:

1. All the wrs on the current wr table are not legit (bugs are included). Enigma bug, in general standards, shouldn't be accepted to the wr table, also stuff like Pipe sc and for example Animal Farm start sc are arguable.
2. Wheel pops? Where do you draw the line between a legit wheelpop and a jailbait pop?
3. We already have stuff quite close to it, also people may not want to share their secret styles.
4. Fps changing mid-run, what's allowed? It really is cheating and physics manipulation.
5. Fuck the line-drawing.

but what do i know anyway, i'm only interested in tasing because no real skils in elma osv and i cant even beat the real wrs without buging by pressing secret bug buton in mila sl or such
Image
User avatar
Zweq
34mins club
Posts: 4055
Joined: 28 Nov 2002, 15:54
Location: suo mesta

Re: Tool-assisted speedrun of elma

Post by Zweq »

We are only working with the input we give to the simulation. Simulation decides whats legit whats not. Simulation is our god. As long as milagros code is bugfree we are still in the bounds of legitness.

Changing between 999 and 30 fps is probably the most debatable thing though. Can such circumstances occurr on any PC? I believe someone should prove it to be possible.

What would not be considered legit: edit gamestate in memory, bypassing the simulation. You dont fuck with simulation.
Image
User avatar
milagros
Cheatless
Posts: 4560
Joined: 19 May 2002, 17:05

Re: Tool-assisted speedrun of elma

Post by milagros »

changing fps during run is also arguable - John did it by turning on/off his harddrive:)
[carebox]
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

milagros wrote:changing fps during run is also arguable - John did it by turning on/off his harddrive:)
when i hoyled hi-flyer-wr-style i turned the timer off mid run, because first bounce was easier with a little lower fps and the brutal was easier with high fps. so ye, possible.
TT: 36:59:53 || Avg TT: 38:09:65
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: Tool-assisted speedrun of elma

Post by bene »

Update
37. Jaws 0:07,66

Edit:
21. Hangman 0:07,48

Edit Edit:
13. Hill Legend 0:07,89
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: Tool-assisted speedrun of elma

Post by bene »

Found fun when playing int16 with bug.
http://kopasite.net/up/of1c8wl5hb9p32z/ ... elogiC.rec
http://kopasite.net/up/p4648vjtb212zfb/ ... elogiC.dat

You can try this yourself, just play after 2nd bounce and press brake. Works anytime until wheel hits ground.

What happens is that the 2nd bounce causes the wheel to spin very fast because wheel bounces in line with surface and takes megarotation. Braking with high wheel spins causes fun to happen.
This is a likely explanation to why brakvolting(braking before volting with gas) causes a stronger volt if it happens at any speed. Same thing but less wheel rotation.

http://kopasite.net/up/u9e934pfd2h4bbq/34b0ne.rec here xiited rec with 100k spin on wheel and press brake
http://kopasite.net/up/uk9lo5lgt2ja21o/06b0ne1.rec here xiited rec with 100k spin on wheel and land on polygon
http://kopasite.net/up/f6uzodmb9h72920/06b0ne2.rec here xiited rec with 100k spin on wheel and press brake
http://kopasite.net/up/8i2o824lc388d7h/34b0ne1k.rec 1k spin on wheel and press brake

These xiited recs are not driven with milasl.

Thanks to finman for figuring out it was wheelspin that caused bug and thanks to domi for creating recs.
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
Thundr-
37mins club
Posts: 398
Joined: 10 Jun 2010, 17:02

Re: Tool-assisted speedrun of elma

Post by Thundr- »

milagros wrote:changing fps during run is also arguable - John did it by turning on/off his harddrive:)
I fiddled with resolution and zoom settings along with some computer settings (all legal and normal shit osv) to get max fps with timer off, and 200-300 with on. Timer on fps worked differently if I capped to 999 in config or just set to unlimited with all sorts of variations. Could pretty much get the on-fps to anything between 200-999 while the timer-off fps always stayed at max.
User avatar
Bludek
38mins club
Posts: 1725
Joined: 23 Sep 2009, 10:56
Team: CART
Location: Some pub in Prague

Re: Tool-assisted speedrun of elma

Post by Bludek »

bene wrote:Update
37. Jaws 0:07,66

Edit:
21. Hangman 0:07,48

Edit Edit:
13. Hill Legend 0:07,89
Just seeing those times impresses me quite a bit. And a rec here and there would be highly appreciated :) I started to love this project after u explained a bunch of stuff, b0ne.

That high speed wheel looks seeeeriously weird, but I quite get why it is like that.
Image
EOL top 10 kuski of 2014 and 2015.
21:03:48 <umiz> i like 99% of bludek levels
User avatar
milagros
Cheatless
Posts: 4560
Joined: 19 May 2002, 17:05

Re: Tool-assisted speedrun of elma

Post by milagros »

i'm really impressed by bene's times (and everybody else who played for a few milliseconds of the 2nd second)
maybe i should give it a try some day
[carebox]
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: Tool-assisted speedrun of elma

Post by bene »

Update:
35. Labyrinth Pro 0:28,29
11. Gravity Ride 0:08,46
16. New Wave 0:31,43
52. Hang Tight 0:09,69

That lab pro is easily the most impressive bugrec so far, I spent up to 20 hours in it doing stuff and zweq also played a couple of hours.
It has slow end for now because can't finish. Some 22-24s should be poss.

Here recs for bludek:

Volts at high stretches are superstrong. At 7.70 a single alovolt changes everything.
http://kopasite.net/up/563m7n34y27n3ca/21be1183.rec
Norm finish that fm did:
http://kopasite.net/up/ry4x6c1rn4kvf2x/21fm748.rec
Here slowfinish 9,12 and result after getting norm end:
http://kopasite.net/up/9z9ewgci1h2zmc1/13be912.rec
http://kopasite.net/up/s7wh7ti72sflto5/13be789.rec
Last edited by bene on 14 Sep 2015, 22:15, edited 1 time in total.
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

sick stuff... 8O
TT: 36:59:53 || Avg TT: 38:09:65
User avatar
Igge
38mins club
Posts: 6393
Joined: 7 Apr 2007, 12:15
Location: Stockholm, Sweden

Re: Tool-assisted speedrun of elma

Post by Igge »

can someone recsource for lazy mans?)
John: lol hittade ett popcorn i naveln
(19:52:06) (@Madnezz) The Golden Apple Award goes to.....
(19:52:36) (@Madnezz) ib9814.lev by igge!!!
Zweq wrote:99.9999% of nabs haven't even opened the book yet and most of those that have are still on the first pages
User avatar
John
first 35tt
Posts: 4738
Joined: 28 Sep 2002, 19:42
Team: WNO
Location: Luleå, Sweden

Re: Tool-assisted speedrun of elma

Post by John »

milagros wrote:changing fps during run is also arguable - John did it by turning on/off his harddrive:)
Never during runs. But yes, the bike behaved differently with HDD on/off.
Image
User avatar
Zweq
34mins club
Posts: 4055
Joined: 28 Nov 2002, 15:54
Location: suo mesta

Re: Tool-assisted speedrun of elma

Post by Zweq »

this i did alone, after tricking bene to höyl wrong slower start(nat orka go for details but wasting bene's time is always fun)))) http://kopasite.net/up/83kiclplq9d827w/19zb1044.rec http://kopasite.net/up/862s4k3s1i4bx3l/19zb1044.dat
another keulakuva rek for lousk, it also the start of bene's 9s rek http://kopasite.net/up/2pg4649jiesrsux/14zb900s2.rec
Image
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

why don't you all use recsource? we are all lazy!
TT: 36:59:53 || Avg TT: 38:09:65
User avatar
umiz
Kuski
Posts: 381
Joined: 22 Oct 2005, 09:32

Re: Tool-assisted speedrun of elma

Post by umiz »

Yeah, recsource is fantastic! Please use it :)
[i cruise, i lev, i live]
User avatar
bene
Hot kuski
Posts: 906
Joined: 18 Aug 2003, 23:33
Team: dat
Location: Sweden
Contact:

Re: Tool-assisted speedrun of elma

Post by bene »

Update:
35. Labyrinth Pro 0:23,26
52. Hang Tight 0:07,86

wow managed lab pro end and int52 got better time.
Image
Image
Image
Signatür ruined by SveinR - smaller plz :*
User avatar
Zweq
34mins club
Posts: 4055
Joined: 28 Nov 2002, 15:54
Location: suo mesta

Re: Tool-assisted speedrun of elma

Post by Zweq »

I'm mega hapy that lab pro is finally done, at least in somekind of good-enough-as-basetime-for-now manner. It was very frustrating nightmare lev, and I only spent about 5h+ for it, I cant imagine how b0ne felt. I let him do al the imposible clip stuf while I spent time optimizing easy parts)). In the end we had to do 3 apple bugs and about 9000 drivings through wall, and we didnt have a plan so we also had to waste a bit of time for stuf that didn't quite work in the end. The rec looks a bit like I had imagined years ago when I jokingly joked to mila about theoretical lab pro ride, what I couldn't imagine was that it's possible to slow down and survive after riding through walls, not even 1 week ago I could imagine it (because of my experiences in turnaround), b0ne just did it somehow. It's like you make a positive bug and then later a negative bug to even out all the forces and just calmly continue ride as if nothing big ever happened.
Image
User avatar
milagros
Cheatless
Posts: 4560
Joined: 19 May 2002, 17:05

Re: Tool-assisted speedrun of elma

Post by milagros »

elma at its best
[carebox]
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

pls lab pro rec ;o
TT: 36:59:53 || Avg TT: 38:09:65
User avatar
Bludek
38mins club
Posts: 1725
Joined: 23 Sep 2009, 10:56
Team: CART
Location: Some pub in Prague

Re: Tool-assisted speedrun of elma

Post by Bludek »

this is like hardcore nerd porn for selected few.
Image
EOL top 10 kuski of 2014 and 2015.
21:03:48 <umiz> i like 99% of bludek levels
User avatar
Lukazz
36mins club
Posts: 5241
Joined: 4 Jul 2004, 12:10

Re: Tool-assisted speedrun of elma

Post by Lukazz »

Bludek wrote:this is like hardcore nerd porn for selected few.
ye, i can fap to this.
TT: 36:59:53 || Avg TT: 38:09:65
Post Reply