<- 12
--
--
List results:
Search options:
Use \ before commas in usernames
F*ckin' sanity effects...
Great effort!  Looking forward tremendously to the finished product.
Thanks for the feedback Uilnslcoap. I saw your post earlier, but was too lazy to reply/wanted to have something to report on first. I had a streamed attempt today that ended near the end of 8-1 (available on my jtv-channel; it's the last run if anyone has a fetisch for failures), which is actually my best attempt so far. It felt like it was good enough to have been submitted if I would have reached the end. That being said, I hope that in the final product I can pull off something even better. Let's see when (and if) I can actually get a deathless run.
New pb, 46:25. There was one death in 6-3 and one or two not so hot-looking sections on top of that. The way it feels now, I'll likely settle for the first run without unintentional hits.
The #1 player on nico has posted a new run (39:39):
http://www.nicovideo.jp/watch/1320498647?nicorepomail (part 1)
http://www.nicovideo.jp/watch/1320500330?nicorepomail (part 2)

It looks like he beat me to getting a deathless run. Don't worry though, I have it under control. My pb (with one death) would still be 45 seconds faster if converted to ntsc (for whatever that comparison is worth).

Other than that, nothing to report from my side. No good attempts since my pb.
I'm just a friendly TASer visiting
I've been PMing back and forth to ktwo on the TASVideos forum, discussing the game's physics. When on the skateboard in this game, you can have 3 different top speeds - 80, 81 and 82 (unit is 1/32 pixel/frame). Anyway, moving at speed 82 compared to 80 through out the game can make a difference of 15-20 seconds in a speedrun.

I found a safe way to get speed 82, and here's how: make a jump without holding down B, but still holding down right - you will converge to speed 73, no matter your original speed (because you speed changes by 1 per frame in mid-air). Once you land, continue holding right without B, making you continue running with speed 73. Now, hold down B, increasing your speed by 3 per frame which after 3 frames becomes 73+3*3=82.
Waiting hurts my soul...
Randil, as ever, your TASing insight is very helpful. Thanks for sharing.
I can verify that this speed manipulation works well and will be useful in a console speedrun. Other than the pure speed-up aspect, there are also a couple of jumps, that I previously thought too crazy to include in a speedrun, that the extra speed-boost will render much more reliable. I will take some time testing out level-by-level how to profit from this as efficiently as possible. So, no more serious attempts for a while.

More information about how the acceleration works in AI 1 can be found here:
http://tasvideos.org/forum/viewtopic.php?t=2875&highlight=
(scroll down to Randil's post with a diagram)

Finally, I agree with Zenic, it's awesome and much appreciated when people from the tas-community use their knowledge to help us console speedrunners out.
How do you do? I'm Macco, Japanese time-attacker with AdventureIsland (Takahashi Meijin no Boukenjima).
I'm broadcasting in "NicoNicoNamahousou".  I uploaded movie 39:39 in 2011-11-06. (Mr.ktwo said)
2011-12-24, I cleared it in 39:16 in Nico!! No misses, but I lost skateboard sometimes.

http://live.nicovideo.jp/watch/lv75235849
This is my broadcasting log in Nico, 39:16. I speak only Japanese.
But If you are not NicoNico PremiumMember, you can't watch this.
Being NicoNico PremiumMember, you need 525 Japanese yen per month.
I'm just a friendly TASer visiting
What's the status on this project?
Edit history:
Icthus: 2012-01-19 06:36:36 pm
Ktwo, are you working on the PAL version?  I think I have an NTSC cart that I may try putting a speedrun of.

Either that or I may try Clash at Demonhead.


*Plays game*

*Watches speedrun*

Okay forget that, you enjoy ktwo.
I've been away for a while and the momentum has been lost. Other things have now come up so the status is a bit unclear. Before I left, I did plan out the route including the speed manipulation above. I'm not entirely sure, but I think it saves 15 seconds or so in the end. A pal-run without major mistakes would be around 44:20 (on emu), which with the same play quality would be around 37:10 on ntsc.

Congrats Macco (even if you probably aren't reading this). Nice to hear from you again. I wish I had your play skill to apply on my route.

Icthus, I'm sure people would be much more interested in your ntsc-run than my pal-run, so don't give up. Either way, I wanna finish this sometime just for the personal satisfaction of having conquered the challenge.
Hello! I came to broadcast AdventureIsland in Ustream, at last!!
Please see below URL.
http://www.ustream.tv/channel/adventureisland
I broadcast at about 22-24 night on every Saturday.
If you live in Europe GreatBritain, you can watch at about 13-15 afternoon.
I'm looking forward to watch my broadcast!  If you can't watch it at this time,  please watch recorded movie.
I speak only Japanese, but a little of English.
Remember your mantra.
Bump for interest.


For six dollars, New Adventure Island (TG16) on WiiVC is really, really pretty, and runs on what seems to be the fastest engine of any in the series. That being said, it beats the piss out of me, and like the originals, an SS run would be insanely difficult.

Anyway - if Icthus, or anyone else, still have an interest in submitting a run of the original, I sympothisize with your struggles, and wish you luck.
Quick post to say that I'm down to 37:22 (ntsc, sda-timing). It's really not a great run though. However, I mainly picked this game up again to get myself a time listed for the ntsc-version and not with the intention to make an optimized run. And as far as I know, it's currently the fastest reported run. I will have to watch the video again to make sure the recording looks ok to me (I'm not sure I can 100% trust my capture card). If it is, I'll send it here first. If it's not good enough for sda, I can always put it on youtube or something.
twitch.tv/ccaakk
Congrats ktwo...I'm really really looking forward to this one.
Wow ktwo, good job. I'll admit I didn't expect you to pull that off.
What sort of improvements did you make to takenoko's run, and what kind of time do you think you can reach yourself?
Smörgåsbord
<3
Hey guys, thanks for the support.
I have watched the recording now and my fear of some kind of crap was not completely unfounded. Parts of the video are a bit tainted by inserted frames (at least it looks there are inserted frames and not just lost, so the timing should be ok as far as I can tell). I described the issues to Nate and he already replied that it didn't sound like a deal-breaker though. We'll see what happens when I send over the files. Either way, this is a bit annoying as this means I don't want to use this recording setup anymore. Shortly prior to this run, I recorded Heroes of the Lance, which was basically just to get some experience with the setup and to be able to catch stuff like this. Unfortunately it didn't help.

sledge, this run was a fairly quick and dirty job. I hadn't played the game in almost a year and didn't spend much time looking into new strats. I basically just relearned the same strats as in my pal-run + a few new minor strats that I came up with on the fly (I did also watch takenoko's run and picked up two things that I sadly hadn't noticed before). The major difference between me and takenoko is that he doesn't manipulate the speed, while I do it as much as possible. Each instance doesn't save much time, but it adds up. Without having done any real comparisons, I think this run is maybe 10-15 seconds worse in play quality than my pal-run. It might not sound like a lot over 37 minutes, but in this game I think it actually is. So I think around 37:10 (sda-timing) would be pretty good already, but I wouldn't be surprised if a dedicated person could bring the time down to a low 37. I should also mention that there is a tas with a time of 35:59 (tas-timing), but I didn't find much useful for a console run in it. It's also on an unspecified emulator, so that leaves of course some question marks. There is lag in that tas, but it's really difficult to say what kind of lag a different play style should result in.
Edit history:
ktwo: 2017-09-08 03:48:43 am
I thought I was done with this game after last time, but the terrible recording quality and mediocre run of the NTSC-version have been bothering me for a while. A few weeks ago, I decided that it was time to start working on this game again and try to obsolete my previous run so I can have peace of mind again. The last run was pretty much a copy-paste of the PAL-route I was used to with a few improvements here and there. A major flaw, which I also knew at the time, was not taking into account the additional lag in the NTSC-version, so the routes weren't well adapted. I've tried to correct that this time around and have also implemented a lot of tweaks in many places.

Right now I just got a time of 37:29, which included a death that cost 21 seconds (my current pb is 37:21). That's the first time since I started playing the game again that I completed an attempt on a correct pace without multiple deaths. There were a couple of other mistakes as well, but overall I'm not gonna complain too much other than that death. With the routes and pace I have at the moment, 37:05 would be a great time, while anything over 37:10 would probably include some ugly mistakes. Sub-37 is theoretically possible, but without any additional time savers, it looks like it's too close to the limit for me to aim for.

Edit: In case anyone gets interested in looking into this game, I've collected resources, routes and a description of the game mechanics in the knowledge base: https://kb.speeddemosarchive.com/Hudson%27s_adventure_island
I got my first deathless attempt since I picked this game up again, www.youtube.com/watch?v=o3mELZuq0xk. The time was 37:08, so 13 seconds faster than my previous best. The quality of the gameplay wasn't good enough though. Major mistakes in 4-2 (1.5 sec) and 6-1 (2.5 sec from losing the skateboard) and then a couple of other minor mistakes too many scattered throughout. The boss fights were about average of what can be expected. #4 and #7 were poor, but clean boss fights all the way through seems quite unlikely.

Since my last post, I've found several small improvements that add up to a couple seconds. I hope I can shave off another 5 seconds or so from my new time. I've had a couple of runs on a sub-37 pace reaching the second half of the game, but not far enough to consider it my goal.