zwataketa, you told me in PM that you cannot use my input because I abuse zips. May you please show where I abused zip? I think going offscreen should not be considered as a zip…
Also I am quite sure that you need some practice with this game, no offence, but your runs are still way less optimized than mine, and instead of starting a serious project, I would suggest you to learn a little… For example, try to get 0:17 in EHZ1. Or 0:25 in GHZ1 (Amy Rose in Sonic 1). Then try to optimize it further. My best results for the moment are 0:16.28 and 0:24.22…
I will write you a PM about Lua HUD in a few hours…
well, by zip I mean I don't abuse parts where I go to fast for the camera and go offscreen and/or teleport to a compltely different part of the level. so yeah.
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
No, since I want to make EH1 in 0:15 and EH2 in 0:33. My last result in EHZ1 is dangerously close to 0:15. I need to make 0:15, because I need to be sure that you'll not feel easy to improve me when you start a serious project :). So, please, try it all yourself, I promise to give you tips. By the way, did my screencast help you?
I also have a feeling that each time from the current published Sonic 2 run is (in theory) (beatable|reachable) with Amy, despite her inability to get so much slope acceleration as Sonic. I am simply bad at deep optimization and do not have enough spare time to brute-force things, but I already managed to prove me right in CPZ1, which I actually beat faster than Sonic.
It's not Amy@S3K, where it's way easier to beat Sonic's times. In Amy@S3K Amy is pure speed, and her abilities to spin dash through inspindashable things and abuse corner glitch make her extremely good for speedrunning that game.
Just for your interest: most of my runs, including this one, actually have a lot more rerecords than it’s shown in the movie, because I often cancel failed attempts by reverting the input file to a previous state (I use Subversion) and often merge separate input files.
(upd) over 9000 rerecords to make one single level… I hope you’ll enjoy improving me, zwataketa, like I enjoyed improving Crramalama in GHZ1 and marzojr in GHZ2 of Amy@S1 * :-p
Link to video
Only one little trick remains to finally get under 0:16, but that trick is painfully hard to pull off. I think if someday I make it, the number of rerecords will exceed at least 20000 to that day…
* both with no success, but thousands consumed rerecords, right ;)
S3&A [Amy amy%] improvement (with Evil_3D & kaan55) — currently in SPZ2
my TAS channel · If I ever come into your dream, I’ll be riding an eggship :)
No, since I want to make EH1 in 0:15 and EH2 in 0:33. My last result in EHZ1 is dangerously close to 0:15. I need to make 0:15, because I need to be sure that you'll not feel easy to improve me when you start a serious project :). So, please, try it all yourself, I promise to give you tips. By the way, did my screencast help you?
I also have a feeling that each time from the current published Sonic 2 run is (in theory) (beatable|reachable) with Amy, despite her inability to get so much slope acceleration as Sonic. I am simply bad at deep optimization and do not have enough spare time to brute-force things, but I already managed to prove me right in CPZ1, which I actually beat faster than Sonic.
It's not Amy@S3K, where it's way easier to beat Sonic's times. In Amy@S3K Amy is pure speed, and her abilities to spin dash through inspindashable things and abuse corner glitch make her extremely good for speedrunning that game.
Just for your interest: most of my runs, including this one, actually have a lot more rerecords than it’s shown in the movie, because I often cancel failed attempts by reverting the input file to a previous state (I use Subversion) and often merge separate input files.
(upd) over 9000 rerecords to make one single level… I hope you’ll enjoy improving me, zwataketa, like I enjoyed improving Crramalama in GHZ1 and marzojr in GHZ2 of Amy@S1 * :-p
Link to video
Only one little trick remains to finally get under 0:16, but that trick is painfully hard to pull off. I think if someday I make it, the number of rerecords will exceed at least 20000 to that day…
* both with no success, but thousands consumed rerecords, right ;)
No way ill be able to replicate that. And dont forget, I'm trying to do it WITHOUT going offscreen and left+right, so I guess I can just optimize my run as much as possible but not go WAY fast like you. After all, theres no submitted runs of Amy in Sonic 2, so I don't HAVE to go too fast yet. Don't forget, either, that I'm still new to TASing, so I can't do THAT much yet. |:/
Also, I haven't really watched the screencast yet 'cause my dad wanted me to try TASing something else since I was ... Obsessing a bit with amy in sonic 2. I've been doing Poptropica Adventures DS for the time being and I'm almost done with it, so when I'm done, I'll continue the Amy in Sonic 2 run.
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
Sorry for a bit trolling post above. Was in bad mood after way too much rerecords.
By the way, this was TASed without left+right ever pressed.
zwataketa wrote:
After all, theres no submitted runs of Amy in Sonic 2, so I don't HAVE to go too fast yet.
It’s not a good motivation, because if such a run gets accepted, any TASer (me for example :-p) will be able to simply take your run, shave off a couple frames from every single level and submit an «improvement» without any efforts. Also there is a rule that your movie will not be accepted if there are known better movies (not only within this website — on youtube, for example. Do you know that some TASers are even not registered here and so do not post any submissions, but their movies on youtube are quite well known)?
S3&A [Amy amy%] improvement (with Evil_3D & kaan55) — currently in SPZ2
my TAS channel · If I ever come into your dream, I’ll be riding an eggship :)
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
I have to remake some parts of my Amy@S3K run again. Found a zip in HCZ1 which is too significant to miss it. I am now discussing this trick with marzojr, maybe he knows how to take even more advantage from it (it’s possible to cause horizontal underflow, but seems to be hard or even impossible to take a suitable Y position after that).
S3&A [Amy amy%] improvement (with Evil_3D & kaan55) — currently in SPZ2
my TAS channel · If I ever come into your dream, I’ll be riding an eggship :)
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
Ok, here is marzo’s response.
marzojr wrote:
After going around a bit, I have to say that I don't think it is possible; if there were a way to do an horizontal underflow into higher up position (there isn't, at least not with this zip -- the highest you can go is terrain that makes you fall through), it would be possible. Relevant addresses:
camera X: RAM word at $FFEE78;
camera Y: RAM word at $FFEE7C.
The boss itself loads when camera Y between $380 and $780 and camera X is between $3500 and $3900, and unloads outside this range. However, he will wait doing nothing until camera X is between $3500 and $3700 and camera Y is between $300 and $400. If you can't get the camera high enough when this X range happens, you will not teleport to the boss' arena.
The highest you get the camera Y is $4F6 at 0:19::58 -- you would need to get around $100 pixels higher than that at least 35 frames sooner to be able to pull it off.
But horizontal underflow is still usable if someone wants to make a glitchfest.
The zip itself will be shown in the next preview of my TAS WIP, or maybe a bit sooner, if I decide to record a short glitchfest run.
LTRP wrote:
I bet that DMTM will ask for credit from that zip as well. Anyway, I'll wait patiently to see the zip itself.
His name is already listed in the title of my run, so I think it’s more than enough, even if he really was the first to discover that zip, I abused many other zips in my run, I do not see why I need to write about the discoverer of every single glitch, especially considering that they are all based on the same common knowledge.
(upd) made a short and not very optimized run of HCZ1, and it improves my optimized run by 7 seconds. The improvement comes not only from the zip itself — I also found a way to get to the higher level quicker, by abusing a red spring. This level definitely needed some deeper lurking, and I am happy about all the stuff I found, and maybe I will find something more.
(upd2) the improvement is now 8 seconds due to route change… With proper optimization it may become 9–10 seconds.
(upd3) waduh… Really wanted to make some progress in my run, but feel too tired, and probably will not have spare time next some days. So, I made this short and not optimized TAS. This TAS is 3 in 1: testing 2 possible routes + a little glitchfest (nothing new, just some walking through walls as usually). I made it only to see what times I can get if I use the found zip. Hope you’ll enjoy this. And sorry if I disappointed.
Link to video(upd4) optimized it to 0:32, feeling almost happy with that result :)
(upd5)zwataketa, I remember what you told me in PM, but when will you get back to TASing anyway? I was so happy about your interest in Amy@S2, and where are you now? :(
S3&A [Amy amy%] improvement (with Evil_3D & kaan55) — currently in SPZ2
my TAS channel · If I ever come into your dream, I’ll be riding an eggship :)
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
LTRP made a score attack TAS of Star Light zone act 1 (Sonic 1), so I let myself post this. It’s like a score attack too, but it’s improveable (it’s easy to see that you can earn points even after the timer stopps). I made it just for fun — actually, because a few days before the day when I made it, I saw Amy in my dream, and she told me that she wanted to be a millionaire… And I realized that I could help her :-p
Link to video
Sorry for this TAS being so long and boring, I placed a shortcut allowing to skip the boring part. Posted in hope that someone will enjoy it.
S3&A [Amy amy%] improvement (with Evil_3D & kaan55) — currently in SPZ2
my TAS channel · If I ever come into your dream, I’ll be riding an eggship :)
Sorry for a bit trolling post above. Was in bad mood after way too much rerecords.
By the way, this was TASed without left+right ever pressed.
zwataketa wrote:
After all, theres no submitted runs of Amy in Sonic 2, so I don't HAVE to go too fast yet.
It’s not a good motivation, because if such a run gets accepted, any TASer (me for example :-p) will be able to simply take your run, shave off a couple frames from every single level and submit an «improvement» without any efforts. Also there is a rule that your movie will not be accepted if there are known better movies (not only within this website — on youtube, for example. Do you know that some TASers are even not registered here and so do not post any submissions, but their movies on youtube are quite well known)?
Im not THAT unserious about it. But I'm still new and don't know everything about it. Well, let me think ... Say, what's you're fastest run of Emerald Hill zone act 1 where you DON'T go off screen, and that includes outrunning the camera
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
I don’t really know, but I think I can make it in 0:17 without going offscreen. Anyway, I simply suggested you to join my TAS projects instead of starting your own ones. But mentioned that I would help you if you prefer your own projects.
S3&A [Amy amy%] improvement (with Evil_3D & kaan55) — currently in SPZ2
my TAS channel · If I ever come into your dream, I’ll be riding an eggship :)
I don’t really know, but I think I can make it in 0:17 without going offscreen. Anyway, I simply suggested you to join my TAS projects instead of starting your own ones. But mentioned that I would help you if you prefer your own projects.
I think you could help me, possibly. I really think we should use the amy and tails setting to make some parts faster and boss fights a bit easier to tas. Also, dont forget the funny scenes i put at the end of my wips XD
Btw, did you see my oil ocean 2 test run? I wanted to show off the famous psuedo speed glitch without zips (since it is a zipless run of the game), and i didn't want to swim in the oil like everyone else (but i will in act 1). Also, if i had swam in the oil, it would require a zip to get to the boss fight since the oil stops at a certain point of the stage, possibly due to boundaries, so it would violate the rules of the run as well. I think the path i chose was a good one for a zipless run, dont ya think? Needs optimization, i know, but still a good no zip route, right?
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
Did you mean this?
Link to video
Well, then I have some questions to you.
1. You said that you were going to make a zipless run. Isn’t that a zip right after applying the «atomic» glitch? Okay, I understand that different people have different understanding of what zips are, but if you consider going offscreen as a zip, why not to consider this as a zip as well?
2. How about applying the «atomic» glitch 2 times in a row? Three times?
3. In many places it is possible to spin dash through spikes and other obstacles. And it will be faster than to jump over them. Should it be permitted or not?
And the main question, of course: can you make a better optimized run of this? I will be able to give more advices (and maybe even contribute my input, if you need). Your project looks interesting for me, nothing to hide, but until something is really made, it will be only a „project”.
S3&A [Amy amy%] improvement (with Evil_3D & kaan55) — currently in SPZ2
my TAS channel · If I ever come into your dream, I’ll be riding an eggship :)
Did you mean this?
Link to video
Well, then I have some questions to you.
1. You said that you were going to make a zipless run. Isn’t that a zip right after applying the «atomic» glitch? Okay, I understand that different people have different understanding of what zips are, but if you consider going offscreen as a zip, why not to consider this as a zip as well?
2. How about applying the «atomic» glitch 2 times in a row? Three times?
3. In many places it is possible to spin dash through spikes and other obstacles. And it will be faster than to jump over them. Should it be permitted or not?
And the main question, of course: can you make a better optimized run of this? I will be able to give more advices (and maybe even contribute my input, if you need). Your project looks interesting for me, nothing to hide, but until something is really made, it will be only a „project”.
1. I actually consider zips to be zips if they go offscreen to another part of the level WAY FAR from where you started the zip. so that doesn't really count. don't worry, I took it into consideration. although maybe I should change the tag to "no offscreen" ...
2. I don't know of any place AFTER the part where i stop the glitch where i can activate it again. if i just go into a checkerball again, it will send me in a set direction (although the checker ball i entered was necessary seeing as how theres no other way to pass it without zips. believe me, i tried.) and deactivate the glitch. if you mean stacking the effect of the glitch by performing it multiple times, then that would be impossible without having to go in the same spot over and over (depleting of entertainment and loss of time) or dragging an enemy to a checker ball and figuring out how to get hit and land in the checker ball (Losing lots of time from making sure the enemy stays on screen and positioning to fall in the checker ball when hit and waiting to get hit by the enemy).
3. I suppose i can do that, as long as it doesn't shoot me off the screen (for reasons you already know). but I won't be able to do that in oil ocean 2, because spindashing also deactivates the glitch, making me slower and not able to jump as high, thus making me inable to go through the route.
for your last question, the better optimized version of the stage will come once i actually get there in the run. I used level select to practice it :P
Now, lately, my homework has been keeping me till late time, so ill probably have to wait till the weekend or when i have little homework that i can get done easily and quickly before we can work collaboratively. sorry, school's helpful, but annoying sometimes -.-
Hey, you think we can ask Aglar how he perfects the tails control?
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
zwataketa wrote:
Hey, you think we can ask Aglar how he perfects the tails control?
Depending on what do you mean by that. Breaking the capsule with Tails should not be that hard. The more hard thing may be to manipulate animals forcing them to disappear faster (which may involve writing a script). But it actually makes sense only when the overall quality of your run is enough to want to save a few non in-game frames. In what I see in the OOZ2 run above, equal amount of frames is lost during 1–2 seconds of the game process. Shave off those first, and then we’ll talk about perfecting minor parts.
zwataketa wrote:
Hey, you think we can ask Aglar how he perfects the tails control?
Depending on what do you mean by that. Breaking the capsule with Tails should not be that hard. The more hard thing may be to manipulate animals forcing them to disappear faster (which may involve writing a script). But it actually makes sense only when the overall quality of your run is enough to want to save a few non in-game frames. In what I see in the OOZ2 run above, equal amount of frames is lost during 1–2 seconds of the game process. Shave off those first, and then we’ll talk about perfecting minor parts.
zwataketa wrote:
Well, i mean how to make tails appear in time to hit the capsule and etc. and also all those other moments. I know where i need to optimize, and ill get to it in the real run. I just gotta go up faster at the first rising elevator without falling back down, or jump up ALL the way and not have to worry about the cliff next to the checker ball. And then entering the checker ball to progress i gotta enter it without missing, right.
Hey, you think we can ask Aglar how he perfects the tails control?
Depending on what do you mean by that. Breaking the capsule with Tails should not be that hard. The more hard thing may be to manipulate animals forcing them to disappear faster (which may involve writing a script). But it actually makes sense only when the overall quality of your run is enough to want to save a few non in-game frames. In what I see in the OOZ2 run above, equal amount of frames is lost during 1–2 seconds of the game process. Shave off those first, and then we’ll talk about perfecting minor parts.
zwataketa wrote:
Well, i mean how to make tails appear in time to hit the capsule and etc. and also all those other moments. I know where i need to optimize, and ill get to it in the real run. I just gotta go up faster at the first rising elevator without falling back down, or jump up ALL the way and not have to worry about the cliff next to the checker ball. And then entering the checker ball to progress i gotta enter it without missing, right?
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
The worst thing about TASing Sonic for me is that everytime some TASer becomes really highly skilled, he decides to stop TASing. Last years, a LOT of people advanced so much with Sonic - mmarks, Upthorn, nitsuja, Sonikkustar, Crramalama, - but Sonic tasing "industry" keeps missing skilled players. Why? Because they all left! It’s not a reply on previous message, but a reply on the last video from LTRP, who seems to be going to join those guys >_<.
S3&A [Amy amy%] improvement (with Evil_3D & kaan55) — currently in SPZ2
my TAS channel · If I ever come into your dream, I’ll be riding an eggship :)
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
First I got 18 seconds. Long time ago, in the galaxy, far far away… It was my second tool-assisted run ever made.
From the second attempt I managed to beat it in 17 seconds
It made me go really mad, and I made it in 16 seconds.
In my attempts to reach 0:15 I wrote my first Lua script for positioning me by brute-forcing. However, 0:15 was not reached, but I got painfully close to 15 seconds.
I started from scratch, and… Guess what… Right, I am ahead more than enough to beat the level in 15 seconds!
And the fact that 2 days ago Aglar improved one of my runs by 60 frames keeps me sure that even 14 seconds are reachable.
This TAS begins to be really crazy and kinda reminds me of the battle between LTRP and NaturelLorenzo on TASing Sonic 2 HAZ.
(upd) 15 seconds reached, huh. What’s next? 14? :)
S3&A [Amy amy%] improvement (with Evil_3D & kaan55) — currently in SPZ2
my TAS channel · If I ever come into your dream, I’ll be riding an eggship :)
This TAS begins to be really crazy and kinda reminds me of the battle between LTRP and NaturelLorenzo on TASing Sonic 2 HAZ.
I remembered that scenario. It was funny because when I had improved his time, I said to myself "there was no way that he can beat this". Boy was I surprised what a battle he put up. In the end, he wanted my GMV so he could improve my run by 4 frames.
Joined: 10/6/2011
Posts: 1690
Location: RU · ID · AM
Finally beaten Sonic’s time in Emerald Hill 1. Last result is 0:15.31. And, beleive or not, the movie includes running backwards ^^ it’s probably avoidable, but funny that a faster movie may include going opposite direction.
Enjoy this for now. So my new goal is 14 seconds.
Link to video
Please tell your opinion about it :)
S3&A [Amy amy%] improvement (with Evil_3D & kaan55) — currently in SPZ2
my TAS channel · If I ever come into your dream, I’ll be riding an eggship :)
My first language is not English, so please excuse myself if I write something wrong. I'll do my best do write as cleary as I can, so cope with me here =)
(ﾉಥ益ಥ）ﾉ
Link to video
