Tool-Assisted Highscores/Speedruns

Are you a rad N player or want to put some rad N players to the test? Talk about your skills and challenge each other at N in this forum.

Moderators: TheRealOne, EddyMataGallos

User avatar
If you ain't first, you're last!
Posts: 2250
Joined: 2009.02.07 (16:07)
NUMA Profile: http://nmaps.net/user/EddyMataGallos
Location: Spain

Postby EddyMataGallos » 2016.07.19 (23:37)

That's very handy jp thanks for sharing.
ImageImage
Impressive flag by ska -|- N2High - My software for N2.0 -|- Awesome sig by Hendor
About me & Stats (Spoiler)

Hi, im Eddy, a Spanish highscorer and current Grandmaster of N. I've been playing N since 2006, I beat it on the 8th September 2008, and NReality later too. Started intermittently highscoring Metanet in 2010, although serious highscoring started in late 2011.

Current Stats
• Completed N on 8th Sept. 2008
• Completed NReality on 24th Dec. 2010
0th Place on Metanet 0th Rankings with 443 0ths.
0th Place on Metanet Top20 Rankings with 599 Scores.
0th Place on Metanet Top10 Rankings with 599 Scores.
1st Place on Metanet Top5 Rankings with 595 Scores.
0th Place on Metanet Level 0th Rankings with 361 0ths.
0th Place on Metanet Episode 0th Rankings with 82 0ths.
0th Place on Metanet Total Level Score: 72142.900 (20:02:22.900)
0th Place on Metanet Total Episode Score: 35669.900 (09:54:29.900)
0th Place on Metanet Total Point Rankings with 11669 Points.
0th Place in NReality 0th Rankings with 235 0ths.
0th Place for a grand total of 676 Total 0ths.
• Winner of some contests (BUST, 100, 9000, Bounty Hunters, SOAST) and of some rounds of some contests (ROPP, NErratic, Red Ball, NReality Team Arena)

{Info as of 01/Feb/2015} The info may be outdated.

Achievements
• 6th highscorer to reach 100 Metanet 0ths
• 4th highscorer to reach 200 Metanet 0ths
• 1st and only highscorer to reach 300 Metanet 0ths
• 1st and only highscorer to reach 400 Metanet 0ths
• 8th ever Grandmaster
• 2nd* highscorer to reach 599 top-20 Scores
• 1st highscorer to reach 599 top-10 Scores
• 10th highscorer to reach 70k Level Score
• 5th highscorer to reach 71k Level Score
• 2nd highscorer to reach 71.5k Level Score
• 1st highscorer to reach 72k Level Score

Relevant Dates
• First 0th: 31-3 on 04/Oct/10
• Reached 100 Metanet 0ths on 25/Feb/12
• Reached 200 Metanet 0ths on 30/Sep/12
• Reached 300 Metanet 0ths on 16/Mar/13
• Reached 400 Metanet 0ths on 04/Sept/14
• Reached highest amount of 0ths on Jan/2013
• Became the Grandmaster on 12/Apr/2012
• Reached 70k Level Score on 25/Dec/11
• Reached 71k Level Score on 13/Jan/11
• Reached 71.5k Level Score on 01/Aug/12
• Reached 72k Level Score on 04/Jul/14
• Completed N 1.4 on 08/Sept/08
• Completed NReality on 24/Dec/10
• Reached 100 NReality 0ths on Oct/2011
• Reached 200 NReality 0ths on Nov/2011
• Reached 100 Top-20 Scores on Aug/10
• Reached 200 Top-20 Scores on Nov/11
• Reached 300 Top-20 Scores on Nov/11
• Reached 400 Top-20 Scores on Dec/11
• Reached 500 Top-20 Scores on Jan/12
• Reached 599 Top-20 Scores on 04/Apr/12
• Reached 599 Top-10 Scores on 18/Nov/12

Deds & Images (Spoiler)

Thanks to ska for dedicating this map to me
Thanks to BluePretzel for dedicating this map to me
Thanks to da_guru for dedicating this map to me
Thanks to zoasBE for dedicating this full episode to me!: 00-0 00-100-2 00-3 00-4
Thanks to zoasBE for dedicating this mappack to me
Thanks to zoasBE for partially dedicating this map to me
Thanks to VODKALOVER for dedicating this map to me
Thanks to runningninja for dedicating this map to me
Thanks to zoasBE for dedicating this map to me
Thanks to zoasBE for partially dedicating this map to me
Thanks to trance for dedicating this map to me
Thanks to zoasBE for dedicating this map to me
Thanks to zoasBE for dedicating this map to me
Thanks to llabesab for dedicating this map to me
Thanks to the23 for dedicating this map to me
Thanks to Traveleravi for dedicating this map to me
Thanks to 123leonidas321 for dedicating this map to me
Thanks to the23 for dedicating this map to me
Thanks to Uniden for dedicating this map to me
Thanks to Uniden for dedicating this map to me
Thanks to Ors_II for dedicating this map to me
Thanks to Sunset for dedicating this map to me
Thanks to blue_rocks for dedicating this map to me
Thanks to Leonhard for dedicating this map to me
Thanks to blue_rocks for dedicating this map to me
Thanks to Daggafork for dedicating this map to me
Thanks to VODKALOVER for dedicating this map to me
Thanks to blue_rocks for dedicating this map to me
Thanks to blue_rocks for dedicating this map to me
Thanks to Armaghan_Ali for dedicating this map to me
Thanks to Armaghan_Ali for dedicating this map to me
Thanks to Armaghan_Ali for dedicating this map to me
Thanks to RandomDigits/SpartaX18 for dedicating this map to me
Sorry if Im missing any ded.

Image
ImageThanks to leo.
Image


Plus (Size) Member
Posts: 50
Joined: 2008.12.22 (12:31)

Postby zapkt » 2016.07.21 (10:17)

Hmm this is interesting. We should see with Unreality, who has a lot of experience with modifying the game, if it would be possible to add a function in NReality (or a modified version of N dedicated to producing MBD runs) were you could automatically compute some stuff. For instance, having a button which would automatically produce the different possible runs for the next frame, and test which one lead to death, would be an interesting first step.

User avatar
Smoothest Taint in the West
Posts: 3131
Joined: 2008.09.29 (14:22)
NUMA Profile: http://nmaps.net/user/ska
Steam: www.steamcommunity.com/id/
Location: Australia
Contact:

Postby ska » 2016.07.23 (16:17)

zapkt wrote:Hmm this is interesting. We should see with Unreality, who has a lot of experience with modifying the game, if it would be possible to add a function in NReality (or a modified version of N dedicated to producing MBD runs) were you could automatically compute some stuff. For instance, having a button which would automatically produce the different possible runs for the next frame, and test which one lead to death, would be an interesting first step.

It might sound simple, but I wouldn't be surprised if programming such a function was much harder than it sounds.

User avatar
Damn You're Fine
Posts: 389
Joined: 2008.10.13 (10:10)

Postby Unreality » 2017.03.13 (03:01)

um, interesting idea :p an interesting idea to demotivate all human on earth to be creative to find the best route in a map when we finally got the AI doing the job.

and then for every new N scoring contest ppl will run the AI once to figure the best route.

then the map creator will start to adapt to it and specifically make a human-almost-impossible route for the shortest path to nullify the best route calculated.

Plus (Size) Member
Posts: 50
Joined: 2008.12.22 (12:31)

Postby zapkt » 2017.03.13 (08:28)

Finding the best route is incredibly complicated and computation-expensive. It seems to me like the last things AI will be able to do.
If I happen to do something, I guess it would first find micro-optimizations at best.

User avatar
Smoothest Taint in the West
Posts: 3131
Joined: 2008.09.29 (14:22)
NUMA Profile: http://nmaps.net/user/ska
Steam: www.steamcommunity.com/id/
Location: Australia
Contact:

Postby ska » 2017.03.14 (05:31)

Well, to conclude my point, if someone found a way to automatically run an MBD program that got the best possible time in all level scenarios, you wouldn't just be looking at an incredible breakthrough in this community—you'd be looking at an incredible breakthrough in AI itself.

Plus (Size) Member
Posts: 50
Joined: 2008.12.22 (12:31)

Postby zapkt » 2017.08.12 (21:42)

Well, this year has seen quite a lot of breakthroughs in game-beating AIs (with go and poker), so I wouldn't be that surprised :D

User avatar
Smoothest Taint in the West
Posts: 3131
Joined: 2008.09.29 (14:22)
NUMA Profile: http://nmaps.net/user/ska
Steam: www.steamcommunity.com/id/
Location: Australia
Contact:

Postby ska » 2017.08.26 (19:06)

Man... it would be fascinating to see a self-learning AI run N. I'm pretty sure you could get it to beat all levels in the game (given enough time, and assuming it was programmed optimally); but, like you said, the inherent non-linearity of N makes this much harder to achieve.

If I had to list a starting point (just for completion's sake), I would make each attempt that guides the ninja closest to the exit be the one that you branch from (which is essentially a form of virtual natural selection/evolution). However, you'd need to have a more in-depth approach than this. Ideally, you'd prune attempts where the ninja doesn't have a direct line of sight with the exit, otherwise you might get situations where the AI hopelessly gets stuck in a loop that results in constant dead ends. Also, maybe some kind of subroutine might be a good idea; for example, have a series of waypoints once a computer has determined that a certain section of the route has been optimised; this could also save on time/computation power.


Return to “Highscores & Contests (Nv1.4)”

Who is online

Users browsing this forum: No registered users and 3 guests