Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
zwataketa wrote:
I could handle the ocean sections. And there should be other ways to make it faster, even without the slope glitch.
If you want to TAS this game (not just the wave cyclone sections), I'll help you by looking over your .dsm to see where it can be improved upon. That's as far as I'll go with TASing SRA. As I said before, I have no intention of TASing this game because it's not appealing compared to Sonic Rush or even an improvement of Colors.
Also, I never 100% completed SRA, so I have no idea.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
zwataketa wrote:
Hey, if you want to do this, will you do the full game (meaning you beat the egg wizard too)? Unlike Sonic Rush, the game is linear (both characters have the same story), so it should be easy. You won't be able to soft reset at the credits to save time (since the auto save only happens AFTER the credits), but you can hold A to speed it up to non annoying time, so it should still be entertaining if you decide to do that. I'd like to see that! :D
Also, just like on the Sonic Rush Topic, here's a tip: you can get a free hit on both Johnny and Whisker if you spindash as SOON as the fight starts. Pretty easy to figure out yourself, but I figured I'd tell you just in case.
It's been so long that I think it's safe to say that Sonikkustar isn't working on his run anymore. I never had intentions of doing a TAS of Rush Adventure. There's a couple of reasons why, one being the Wave Cyclone sections (long and boring). More importantly is that neither aerial boosting nor the slope speed glitch are available in Rush Adventure, which makes it unappealing to TAS to me. Especially with Blaze, she would be terribly slow in a TAS without the slope speed glitch from Rush.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
I don't think that Amy should go for any of the special stages because she'll waste the most amount of time getting to them, especially if you decide to allow ultra spindashes (which you should either allow completely or not, a mixture is too arbitrary). I just don't see how Sonic or Tails couldn't reach any of them quicker than her.
I rewatched the Sonic w/ ultra spindashes TAS, I guess I'm against them if each character would look like a copy + paste of each (well, besides Amy). You should definitely allow them for Amy though hahaha.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
Yeah, I'm steadily making progress, but didn't feel like it warranted a post (but since you asked). I did Blaze NC1 twice more and have now moved on to NC2, final time in NC1 is now 54'37" (7 frames faster than last, 38 frames faster than what I uploaded on youtube). I'm in the middle of the WPB with Sonic, but I'm not stuck. The stages are more enjoyable to work on than bosses, so the Sonic run won't be worked on at least until I reach the NCB in the Blaze run.
The route for the first 10 seconds or so of NC2 is very interesting with Blaze. It is basically 3 frame hovering to the right and then slope glitching the first slope. Afterwards it's using/maintaining/storing the x-velocity over the course of 10 seconds to blast through the first 25% of the stage. I'm not sure if Sonic will be able to pull it off (but if he can, he'll do it much better since he'll be able to glitch the whole slope instead of half of it).
Now for some hovering stuff;
3 Frame Hovering is holding hover (R) for 2 frames, then releasing it for one. It gives the lowest average y-velocity out of all the different combinations over time which is 22.
I've also been experimenting with jumping from either a ceiling or certain slopes and then hovering. When jumping from a ceiling or slope, Blaze is "tilted" and hovering for one frame while tilted at certain angles does some weird things. For instance, jumping from the ceiling and then hovering can do two different things.
1) Flings Blaze upward at usually a medimum y-velocity (-1500 or so)
2) Reverses all x-velocity to the opposite integer (So if x-velocity was 14300, it would be come -14300), but this can be somewhat controlled by forcing the x-velocity cap.
Neither which are particularly useful so far (a certain amount of progress must be made in the loops before being able to exit them making this not applicable, plus this would miss out on slope glitches), but I imagine there will be a use for them later on down the road. I'll refer to this glitch as "Tilted Hovering" from now on.
Weird how many different mechanics the simple hover move has lol.
EDIT
Finished my first optimized try of NC2, starting it over again tomorrow. The final time for this attempt is 1:28:55. I've got a few ideas for how to improve the time, so it should be at least sub 1:28 in the next attempt.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
I enjoyed that test run a lot, I hope you decide to do some more. It seems like there's a lot of potential for a very entertaining TAS with those links that you shared in provided in the opening post.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
Is there a simple way to convert .mkv and .avi files into .mp4 format without A) losing a ton of quality or B) Having a huuuuge file afterward (Tried converting a 11GB 1080p mkv in VLC, came out 79GB lmao).
Any help or suggestions are appreciated.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
FUN FACT
Finding Dory, the sequel to Finding Nemo, was announced the day after this was submitted.
A COINCIDENCE? PIXAR PLOT TO CASH IN ON THIS RUN'S POPULARITY? Odongdong has got some explaining to do... or should I just call you John Lasseter?
Seriously though, I enjoyed it. Yes vote, I liked it even though it is a bit on the long side.
Just because he was sitting on this run isn't a reason to hold it against him. Sure there was a cut-off point, but the quality of the run is impressive. There has to be some sort of compromise that can be reached in this case.
As far as the run itself, I loved it. This was a great game that I stumbled on years back. This run does a good job showing off what the game has to offer and does so in a very entertaining fashion. The auto-scrolling levels are especially impressive because of their difficulty when playing in real-time, but this run makes them look like a joke just like any good TAS. The same can be said about some of the later levels too, this run makes them look deceptively easy.
"Did you find this movie entertaining?"
It gets a big yes vote from me. I hope the depreciation issue can be overcome with a comprise. I know I'm biased from having played the game, but I think it'd be a shame to let such an entertaining run go to waste.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
ALAKTORN wrote:
nevermind me
I caught your reply before the edit, I just didn't get a chance to reply to it. I used "slow" was because using the diagonal trick would've taken 20+ frames since I would have had to fallen low enough to not go over the loop. I've never encountered a delay on either of Blaze's tricks besides the one that occurs naturally when traveling too far with the horizontal trick (but I know that's not what you were talking about). My guess would be that doing B tricks before doing the R tricks cancels the delay on the Up/Diagonal tricks, which would explain why I never came across it.
Paused wrote:
Was not expecting that.
Going to be very interesting comparing the Sonic/Blaze times and routes.
Yeah, I think so too. Blaze has a lot of potential to get big speed with slope glitching in places where Sonic might not be able to. In most situations though, Sonic is going to be able to get more out of most slopes compared to Blaze provided there's enough room before the slope to get up to 4300 x-velocity. A good example is the first slope in NC1.
Cruizer wrote:
The last levels you've uploaded here are amazing, it's great to see such progress :P
Thanks, I appreciate it. Your maps are a huge help, as always. Sorry I hadn't been making good use of them sooner haha. If you want, I could PM you how to do the slope glitch so you could take some of those top times for yourself over on TSC haha.
zwataketa wrote:
wut?
I appreciate the tips and the video you provided. I most likely won't use your trick, but not because it's bad or anything like that. The route I've loosely thought about for AL1 will be abusing the initial rail for speed (since iirc it's possible to keep landing on it without touching the ground first) and then the right curve on the top of the loop to get some big x-velocity and speed. I've actually got a really awesome route for AL1 I've roughly thought out, it'll be interesting to see if it's possible when I get to that point in the run (If it is, easily sub 1 minute). Keep the tricks coming though.
As for being blocked, I don't know what's the issue <_<. My blocked users list is blank lol. Also, in the future if you want to tell me about something that that doesn't relate to Rush, just send me a PM rather than posting here.
---
I've got my route finished up for Sonic WP2. Hopefully no boss trouble *crosses fingers* once I get to the Water Palace Boss lol.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
Can someone explain what the benefit to glitching out the graphics in level 5 are? I'm not familiar with any of the Zelda 1 runs, so I haven't seen it done before. Looking at the submission text didn't help either ._.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
SoulCal wrote:
Also, anyone know if a Wii homebrew app to play GC or Wii games could achieve a similar loading effect? since you're loading from a HDD or SD card, and not a spinning disc.
Yeah, cfg loader mod sped up all the Gamecube games I've played by a noticeable margin (I haven't played anything loading intensive, mostly Melee, Wind Waker and the Mario Partys, but even so). The Wii games that I've played both on cfg and in Dolphin (Brawl, Kirby's RTD) had very similar load times.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
3/11
Link to video
This ended up a lot faster than I initially thought it was going to (Almost 20 seconds ahead of the fastest Blaze time on TSC). There's plenty of slopes and springs that made planning the route take awhile. For instance, the section at 0:15 - 0:17, I optimized about eight different strategies in the test run until I finally squeezed out the last one that skips the switch entirely with the R + Right trick (5300*16 x-velocity) rather than the slow Up+Right R trick (2300*16 x-velocity).
The hardest part in the route planning was picking the route starting at 0:47, since it's possible to jump down to the bottom route and do the slope glitch in the loop and slope afterward, horizontal R-trick onto the last platform and get a much bigger slope glitch on the final slope before the ring. While it isn't as flashy or high speed as the bottom route, the middle route ended up being 26 frames faster since it avoid using the pressure spring (The object at 0:21/0:38. The longer down is held, the higher Sonic/Blaze will be sent upward, up to 4300*16 y-velocity). The bottom route could be faster for Sonic since he's not able to hover over the swings like Blaze in the middle route. Boosting right before releasing the pressure spring will easily allow Sonic to skip them and the floating platforms.
I'm taking a break for a few days and then I might work on Sonic some more, but probably Blaze. Blaze is a lot more fun to work on since the moveset is a lot more flexible for creative routes, so I might end up finishing NC2 first instead.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
I must've been unlucky enough to extract while it was saving, which lead to the corrupt import or was just really tired and doing something wrong when I posted this morning. I tried it again today following your foolproof steps and it worked hahaha.
Thanks Zanoab.
EDIT
I figured out the reason the imported sram would become corrupted. If I manually set the size of the memory to 64k, the import works perfectly fine. Leaving the size as automatic results in corrupted sram.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
I'm using 0.9.8 (without the lag frame fix).
So I can't seem to get a usable SRAM after playing back a verification .dsm I made. I've made sure there has been no SRAM there when booting up the ROM or letting one be created before loading the .dsm. No matter what methods I use (loading/not loading savestates from various points) the SRAM created during the start of the .dsm is always "blank" as if the game was never played. I've tried loading a multiple savestates that were created during different parts of the .dsm (even though that SRAM technically wouldn't be acceptable) while having no movie playback and none of them will save to the SRAM either (even when doing something that should clearly write to it).
I can get the SRAM to save progress/be written to so long as there is never a .dsm loaded. If one does get loaded, SRAM stops being written to until there is a hard reset of the ROM. Does someone have an idea as to what my issue here is? I'm guessing it's some kind of setting that I'm overlooking.
Edit
I also can't use the Import/Export Backup Memory feature as a workaround. Whenever it is exported from some point during the .dsm, it is corrupted upon importing it back.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
RachelB wrote:
It's a conservative estimate based on almost nothing. 2 hours would not really surprise me too much. Keep in mind though that loading times alone are probably going to save over 30 minutes compared to a console run.
Also, new video of goats, because i know everyone loves watching them:
Link to video
I'm guilty, I loved watching it lol. I'm surprised how the improvements are by a pretty big margin (for TAS standards) in each goat video.
Experienced Forum User, Published Author, Active player
(458)
Joined: 12/24/2010
Posts: 297
Location: CT, USA
zwataketa wrote:
Another trick :)
In water palace, whenever you need enough speed to go to the upper path instead of the lower path at those sideways waterwheel things, you'll either immediately get out when you reach the top or you'll touch the top on the left side and have to waste time waiting to be swept to the right and exit. Try to manipulate it so you always come put on the right side. Also, the faster you go into the waterwheel, the faster up you go, but that may affect the exit point manipulation, so be careful about that. :)
Yeah, you're right about manipulating coming out on the right side. It's actually possible to come out so low that you'll touch the ground and be able to boost for a shortways before triggering the "swept away" animation (which caps out at 1600*16 x-velocity), which is a bit faster. However, the speed you hit the wheel doesn't effect how fast you go up (although it'd make sense in theory). Y velocity is always -384*16 as long as the wheel was hit with enough x-velocity to take the upward path.
Thinking about the whale skip in Colors, I think that a similar skip is possible in Mirage Road 2. It's possible to briefly touch the platform (specifically, for one frame) to remove the invisible wall on the right, then take a leap of faith off to the right. There's two problems, one is that there's no solid ground and the slopes leading down to the platform are too steep to do the slope glitch on, so Sonic doing the skip just isn't possible. However, I think that with careful manipulation with hover and full tension gauge, Blaze should be able to hover all the way to the exit ring. The other problem is that there isn't solid ground, so Blaze is going to die after touching the ring. Not sure which would take priority, touching the goal ring or dieing.
Anyways, the route planning and practice runs are done for Water Palace 1. Almost sub 38 in the last attempt. Should have the level done in a couple of days. I'm also going to finish the verification movie I started awhile back for a Blaze run so that I can work on both runs at once.
3/5
There's a "pause" that makes blaze stay in place for a few frames when using hover with a high positive y-velocity. It also resets any stored Speed and becomes the current x-velocity value unless backwards is held. I'm going to have to figure out what y-velocity value triggers the pause and avoid it.
3/6
It turns out that regular hovering also resets speed to x-velocity's value when R is released unless the opposite direction is held for two frames. It can potentially be useful, I'll have to wrap my head around it some more. Hover also extends the hurtbox that interacts with objects (Springs, Tubes, etc) but not enemies, which is pretty useful. Still experimenting with the route for Night Carnival 1 for Blaze, will probably get it done by Monday.
Water Palace 1 is done, viola.
Link to video
