Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Backwards isn't the right term. I did a powerslide to face the wall head on, released the shell, and it shot off to my left and went backwards on the course.
So, proceeding as intended. Land in the tunnel and turn around, manipulate a mushroom, so a perfect mylestyle (easier said than done) and go for the blue shell.
edit: hey antd, have you worked any on Time Trials?
Wow, comincalflop. I just watched your first LR TAS, and was absolutely amazed. And now I hear you are making it BETTER? I am completely in awe of you, sir. Good luck with all runs you are working on. :)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
antd wrote:
I've have not had the time to work on any TAS yet. This should change soon.
Good luck on GP.
I await for when you start. You're going to be very hard pressed to optimize every single frame to squeeze out every ounce of speed you can muster; those WR guys are incredibly competitive and scour the game for every time saving trick known to man.
Isn't it weird that you and I always look at or run the same games? (and I know you're not fond of submitting stuff, because you get bored of them, but this is one game where you should systematically beat/match all 16 WR's and have it as one run, ending input on the very last course.)
TheFrigz wrote:
Wow, comincalflop. I just watched your first LR TAS, and was absolutely amazed. And now I hear you are making it BETTER? I am completely in awe of you, sir. Good luck with all runs you are working on. :)
Definetly what would've held me back would be all of your reactions/feelings/praise. It is SSSOOOOO heartening and motivating for me to see all of your responses over the past day or so. I'm very glad you've all appreciated the first WIP; the revised Luigi Raceway should hopefully stun you even more, and with luck the next 15 courses will do the same.
For those player boosted shortcuts, why don't you just make each level seperate, and end it when player 1 finishes. And for levels like that penquin island one use more than 2 players if it's faster. Then just put all the levels into one video. If bisqwit doesn't like it who cares. You can get way more views on youtube.
"Great spirits have always found violent opposition from mediocrities. The latter cannot understand it when a man does not thoughtlessly submit to hereditary prejudices but honestly and courageously uses his intelligence."
For those player boosted shortcuts, why don't you just make each level seperate, and end it when player 1 finishes. And for levels like that penquin island one use more than 2 players if it's faster. Then just put all the levels into one video. If bisqwit doesn't like it who cares. You can get way more views on youtube.
I think that is the only way it would really be feasible since it may not be faster for one player to wait a lap so that they can do the same trick off the other player. Yeah, I think I may work on this while Comicalflop works on his 1-player GP run. I will see what happens this weekend, I've been meaning to get this started for sometime now and the good reception of Comicalflop's work (even pre-optimization) may be enough to motivate me.
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Kitsune wrote:
In essence, this will be a 100% TAS of Mario Kart 64. I'm looking forward to it's completion. :3
I wouldn't call it 100% actually. maybe "4 gold cup completion?" or more likely "150cc Full GP". anyway, whatever it will be called, I don't think it falls under the 100% category since it's not collecting 100% items or doing everything that is possible in the game, like different cc's, the TT or Battle mode options, etc.
Spider-Waffle wrote:
For those player boosted shortcuts, why don't you just make each level seperate, and end it when player 1 finishes. And for levels like that penquin island one use more than 2 players if it's faster. Then just put all the levels into one video. If bisqwit doesn't like it who cares. You can get way more views on youtube.
That to me does not sound like a solid run, but certainly something for the concept demos section or to post on youtube etc. And may be not needed if I can pull off some CPU boosting that I'm thinking about in future courses.
Weatherton wrote:
I think that is the only way it would really be feasible since it may not be faster for one player to wait a lap so that they can do the same trick off the other player. Yeah, I think I may work on this while Comicalflop works on his 1-player GP run. I will see what happens this weekend, I've been meaning to get this started for sometime now and the good reception of Comicalflop's work (even pre-optimization) may be enough to motivate me.
You're correct, my run would have a shorter time than a 2P because of getting Bowser to finish. I would love to see you work on a 2 player demo run of some sort because I know you've always wanted to do it. Just be warned, it was incredibly difficult to just do the unoptimized version of LR; it is going to be very tough to do it 2 player. managing driving two different karts on two separate screens in frame advance is tougher than you might think. But best of luck!
2 players? How would that work?
On an unrelated (but fairly important) sidenote: Dammit Spider-Waffle, when is that AVI for Half-Hour Half Life going to be released? :(
On an unrelated (but fairly important) sidenote: Dammit Spider-Waffle, when is that AVI for Half-Hour Half Life going to be released? :(
Uh? there are a few different flavors of that run on sda, and a horrible horrible youtube encode.
edit: or so I thought. All I can find are the youtube and slightly better googlevideo streams. no nice avi :(
On an unrelated (but fairly important) sidenote: Dammit Spider-Waffle, when is that AVI for Half-Hour Half Life going to be released? :(
Uh? there are a few different flavors of that run on sda, and a horrible horrible youtube encode.
Um, where are these various flavors you speak of? All I've seen is the Youtube.
btw, Comicalflop, sorry I didn't get on AIM this weekend, my wife got me a Wii and that took priority for the weekend. I will hopefully be able to work on this with you though :)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
haha that's alright. I had a 3 day weekend and somehow managed not to make a single rerecord in mario kart 64 (maybe get some today in between homework? I generally have a lot of free time after classes, I should start getting into the habit of getting a minimum of 2 hours or more on a daily basis).
Progress: 2nd lap, finished redo of MT'ing around the curve.
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
It's very circular progress, not linear. First, there's optimizing every single movement, which makes a maximum of 3-4 seconds (mostly less) gained after 6 full hours of TASing work, which usually needs to get redone again until absolutely perfect. I was on lap 2, about to do a mylestyle when I realized I'd have a shitty time trying to get a mushroom on pure luck like I did last time. the only thing that can manipulate an item when you already have one in reserve (the blue shell held behind the kart) is kart movement, so I'd have to mylestyle over and over again until I ended up with a mushroom by chance. (a mylestyle takes at least 1,000 rerecords. the latest one I did took 1,700.)
However, a new trick was discovered and used to successfully skip a lap. This has great potential to either skip one or two (I'm unsure yet) laps at the beginning, so I'm going back (saving all prior progress in a separate .m64 if I fail to pull off this new stunt) to see if more time can be saved. It might lead to a completely new new route.
My hope is that no future course takes this much time to do. So far (yet) I know of no other trick as difficult as the mylestyle, however that could easily be proved wrong, especially as Weatherton, myself, and the mariokart64.com people devise exceedingly diabolical shortcuts.
So progress was slow, slowed again as I experiment with this new discovery, and will continue to be slow. (Luigi's Raceway is supposed to be the easiest of all the courses =D) The end result will, I guarantee, be even more fantastic than the 1st version.
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
so will DK jungle, Frappe Snowland, and Yoshi Valley. Although optimizing those will also be tough in their own little way, as there's precious little time for movement so everything has to be perfect, I have to test many times what's the fastest strat for getting flaps, and see if any item boxes are close by to grab mushrooms from. That, plus aim to beat the WR's despite not starting with 3 mushrooms.
Update:
The end of Luigi Raceway is so close! Almost there!
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Good news and Bad news.
The good news is I have a wip to show, getting a final time of 38"46, an 8 second improvement.
The bad news is that it didn't quite accomplish the spin around shortcut I wanted.
I will work on it more to try and get the shortcut, so that more seconds can be saved.
After 1 month I am sick to death of this course, but I hope you all enjoy the results of lots of hard work in redoing laps, and other segments over and over in order to achieve perfection.
DeHackEd: no need to encode this one, unless by tomorrow or Monday Weatherton and I deem the shortcut impossible.
So have fun watching this, and stay tuned for the (hopeful) even greater improvement!
Edit: .m64 file updated to be faster, Final time of 38"46 (.WMV).
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Yes to the first.
Yes to the second.
Anything less than perfection for the rest of the run is unacceptable. So the same quality shalt be maintained throughout the whole run.
OMG HAX Update: newer LR time, thanks to Michael Fried:
37"18 .m64
It's very circular progress, not linear. First, there's optimizing every single movement, which makes a maximum of 3-4 seconds (mostly less) gained after 6 full hours of TASing work, which usually needs to get redone again until absolutely perfect. I was on lap 2, about to do a mylestyle when I realized I'd have a shitty time trying to get a mushroom on pure luck like I did last time. the only thing that can manipulate an item when you already have one in reserve (the blue shell held behind the kart) is kart movement, so I'd have to mylestyle over and over again until I ended up with a mushroom by chance. (a mylestyle takes at least 1,000 rerecords. the latest one I did took 1,700.)
However, a new trick was discovered and used to successfully skip a lap. This has great potential to either skip one or two (I'm unsure yet) laps at the beginning, so I'm going back (saving all prior progress in a separate .m64 if I fail to pull off this new stunt) to see if more time can be saved. It might lead to a completely new new route.
My hope is that no future course takes this much time to do. So far (yet) I know of no other trick as difficult as the mylestyle, however that could easily be proved wrong, especially as Weatherton, myself, and the mariokart64.com people devise exceedingly diabolical shortcuts.
So progress was slow, slowed again as I experiment with this new discovery, and will continue to be slow. (Luigi's Raceway is supposed to be the easiest of all the courses =D) The end result will, I guarantee, be even more fantastic than the 1st version.
