Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
My guess would be that the last frame of input was lost/desynced. The last frame of the movie is an A key press. This causes Sonic to use his shield and it saves 8 frames.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
It would be very difficult to perform in real time. One would have to perform the 11-12 revs within a window of 22-24 frames, otherwise the glitch fails.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
Zeupar wrote:
petrie911 wrote:
Awesome run. Is there a place where the glitches used in it are explained?
[17:35] <Mukki> Zeupar: I'll add information about the tricks nitsuja used, the new tricks that I found, a level by level analysis etc. It might take a little while .
I've have since updated the submission text. It seems that not everyone religiously follows the recent changes! You should find all the information that you need there.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
Zeupar wrote:
This is an excellent TAS. I didn't see this coming. Good job, Mukki!
I would like to pass comment on some acts:
- Neo Green Hill Zone Act 2: Did you try using a Spin Attack to deliver the last hit to the boss to see if it was faster?
- Secret Base Zone Act 1: At 0:15:4X, you could have released the rope as you grabbed it by pressing "A" and "down", saving some time.
- Casino Paradise Zone Act 2: Sonic doesn't seem to touch the boss in the final hit. Could you explain that?
- Angel Island Zone Act 2: It feels like you could have manipulated Mecha Knuckles better to make him do more Spin Dashes, defeating him faster as a result.
- Egg Rocket Zone Act 1: I liked the graphic glitch that could be seen at the end of this stage in nitsuja's TAS. Do you know how to make it happen? If you do, why didn't you include it?
- X-Zone: I like what you did after defeating the second boss to prevent the explosions from being heard. ;)
These are some great questions and I'll answer as best I can.
NGH2: Before you hit this boss a final time he does a front flip thing. While he does this he has an invincibility period and I hit him as soon as I can (as far as I am aware of course).
SB1: Ugh...you're right. There's a couple of frames for the next version. I tried hexing, but this game is far too random. Oh well...
CP2: My guess is that, because he is moving away from me, the hit is registered on one frame, but he is too far away on the next frame for the bounce to register.
AI2: Hitting him with tails in the air is the most difficult action to optimise by far so, believe me, I tried my best to manipulate him to use different attacks but I couldn't. I used lua to spew random input into the game to figure out what actions made knuckles 'tick'. For example, inserting skids against the last boss causes him to use different attacks, but I was unable to find something similar here.
ER1: I also loved the graphical glitch in nitsuja's run, but I have no idea what caused it. I initiate the glitch that ends the level from a much better position. I was able to get far up the ramp on the frame that it loads rather than going on top of the transporter. The positioning required to glitch through the top half of the level must be very precise because you have to hit a tile that will send you vertically to the goal. When this happens the game 'locks' sonic so that he can't move and so this pretty much eliminates any chance I had of manipulating a graphical glitch in some way.
X-Zone: :)
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
Response to this seems borderline so I'm going to give a few more days before making a decision. I'm attributing thin feedback to there being no encode. Now that we have an encode (thanks Mister Epic!) perhaps a few more of you will get the chance to watch and comment. Thanks.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
That was pretty awesome. It's great to watch an up to date 120 star run. I hope these guys intend to submit their work here as I'm really curious, with all the new discoveries, just how much time can be saved compared to the old run.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
No we're not. Even if we did go straight there you cannot catch anything without the Silph Scope and to get that we'd have to complete the game corner storyline. It is much faster to just trainer-fly the tower pokemon.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
Very nice improvements yet again marzojr and still the most entertaining run on the site etc. etc. etc.
The new strat at the start of Hydrocity 2 was interesting. Launch Base 2 was very odd. I'm surprised that that Mushroom Hill 1 glitch wasn't discovered earlier as it seems so obvious now. Entertainment-wise I think that Sandopolis 2 benefitted most from the new glitches.
...and I second the need for a glitchless run. Yes vote.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
After spending a few days playing through this game, watching the run and reading various sources I've decided to accept this submission.
Some of you disliked this run so I'll explain why I am accepting it. I will also try and explain some of the important aspects of this run (as best I can, in spite of being unable to read Japanese).
It is generally understood that TASes of RPGs are given some leniency in regards to their entertainment requirement. I agree with this as RPG runs tend not to have inherent qualities that make them entertaining to the lay viewer (Chrono Trigger being a notable exception). They can have many qualities that make them incredibly entertaining to those familiar with them, but some are still uninteresting tool-assisted. Getting the balance right here is important. RPGs are not completely exempt from entertainment and we must strive to publish material that maintains our high standards. With this in mind I am generally looking for an RPG run to represent three qualities to some degree:
1) Most importantly, an RPG run should demonstrate proficiency with skills common to good RPG runs i.e. the author should have to deal with luck manipulation and the author should have put considerable effort into battle strategy and route planning. This should be fulfilled in a comparatively surprising/entertaining way.
2) The game itself should not be manifestly unsuitable for TASing i.e. the visuals should not be ugly, the music should not be irritating, and the game must not be of unreasonable length.
3) Finally, there should be enthusiasm about the run and there should be many viewers that can appreciate the submission. Otherwise, there would be little point in granting the leniency to the run in the first place.
This run scores big points on 1). The game seems tailored to make speedrunning difficult. The author runs away from many battles, which confused some of you. When you run from a battle in this game you keep the money that you would have won from the battle, but don't get the experience points. Furthermore, the enemies level up depending on how many battles you have fought, which means that if you run from a battle the enemies are levelling up faster than the player. The author runs a lot but is able to get the balance right and the game is still beatable. This was interesting to watch. Also, when you run outside of battles the game hides the enemies from you in an attempt to cause more encounters and ultimately slow you down. This was another interesting quirk that the author dealt with well. There was also some impressive glitches towards the end. The game also has a completely non-linear storyline and you can beat the seven heroes in any order that you wish, which provided interesting route possibilities.
On point 2) this was pretty standard. The music was good, visuals acceptable and the run's length was quite short comparative to other SNES RPGs.
On point 3) this run is relatively unknown to Western gamers, but it is my understanding that this game is big in Japan and our large, but quiet, Japanese contingent seem to be quite enthusiastic about this one.
For these reasons, I believe that this submission makes the grade.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
Voting Yes. The improvements were very surprising and I didn't think there'd be so much time left to save in this. I look forward to the 101% even more now. I'm sure it will be truly epic. Nothing more to say to this that I already haven't said to a ton of submissions this weekend.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
I didn't see this one coming. Very nice improvements. The individual improvements weren't as large as they have been in previous submissions...perhaps this game is nearing its limit? Anyway, this remains one of my favourite runs on the site. Superb work guys. Yes vote.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
This one has been sitting for quite some time so I would like to see it judged soon.
Feedback on this has been scarce. One of the few comments in the thread suggests that this is one of the most boring runs ever created. Not a glowing review, yet 91% of you have voted yes! If you are one of the people that voted yes can you please make a post to explain why you voted this way, or if you do not want to post publicly you can PM me instead.
Also, knbnitkr, if you are reading this can you please finish your submission text? Especially information about the glitches used. Whatever information you can provide would be great.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
While this run has recieved a decent number of yes votes the yes voters have remained somewhat quiet in the thread. The rationale of those who did comment on this run positively I did not find compelling and I find myself sharing the sentiments of the more vocal meh and no voters. When Baxter asked for more opinions there was no enthusiasm for publication and instead there were further negative comments about the run.
The good first and that is that this is a well played run by a great TASer. I can't fault this as far as technique is concerned. The run appears to be well researched and the use of luck manipulation is impressive as is the wall glitch in Skynet.
Unfortunately, the choice of game really lets this one down. I have played through this game and, as many of you noted, it is notoriously difficult to play. This game presents the player with two main challenges; overly difficult enemies and overcoming the horrendous physics. The first of these challenges is mostly removed by luck manipulating the enemies away, an impressive feat in itself, but the result is that the viewer is left with empty, dull, linear levels. The latter of these challenges, the game physics, is trivial to a tool-assisted player and so does not add to the entertainment value in any meaningful way. Furthermore, the bad physics and enemy hitboxes give the run a clunky and laboured appearance, especially when descending down the fire escape. What little entertainment some of you experienced while watching this seems to have been derived from having watched AVGN's review, but this, of course, should not affect the entertainment value of this movie as a tool-assisted speedrun. A TAS should be published on its own merits and it is not an objective of tasvideos to supplement the content of cinemassacre.com.
Rejecting this run due to bad game choice and lukewarm viewer response.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
Based on the strength of the previous WIP I could tell that this was shaping up to be a sensational piece of work. This WIP furthered that point. The work gone into optimising every pixel is admirable (I am aware of the speed/entertainment tradeoffs, and fully agree with their ojectives, and so won't hold them against the authors). The entertainment value here is truly without equal. It is the hallmark of a great TAS to manoevre, with panache, large non-linear levels, but to do so whilst simultaneously dealing with the, very difficult, 100% criteria, as almost an afterthought, is incredible. Yoshi takes on an entire personality of an awesome pompous swagger that makes this fantastic to watch.
I realise that this project must have ben an ordeal so far, but I hope that you guys finish it. I greatly look forward to seeing the finished product (and future WIPs).
Voting yes.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
What's up guys? Over the last month or so p4wn3r and I have been working on the "Gotta Catch Em' All!" run together. Progress has been great and despite a few RNG related setbacks we are as far as the SS Anne.
http://dehacked.2y.net/microstorage.php/info/824232114/POKEBLUE.vbm
The goal of this run is to beat the Elite Four as fast as possible with a completed pokedex.
A few notes (this will be elaborated on further in the submission text):
- When you use a wild pokemon for its attack modifier, catching it causes the attack modifier to be reset. This is why we do not catch geodude or bellsprout at those opportunities.
- The rhydon that we catch is a real rhydon as far as the game's memory is concerned. Rhydon is hex no. 1 and so the game automatically changes certain missingnos into rhydons and you can see it in the status screen (which we obviously don't do). We think that you will be pleasantly surprised with the glitchiness here.
- On route 25 we catch caterpie and weedle. The caterpie generation went well, but there are not many A presses in this battle to shift the RNG D-Sum far enough to catch a weedle. We therefore had to insert a second or so of delay to wait for the RNG to move into the correct range as we had to encounter the bellsprout last.
- We ran many tests and are now absolutely convinced that the game does not allow you to encounter another pokemon within three tiles of an encounter. This is why we move erratically through the route 25 grass.
- The rocket that gives you TM28 does not reset your start button as he does not walk to you. Therefore the forced trainers outside Vermillion have to walk to us or the game will crash.
Enjoy the WIP and any feedback that you would like to give would be great!
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
Nach wrote:
Was there more than one arcade Bionic Commando?
Because I played one years ago, and aside from the same main character, none of this game seems familiar.
I'm not aware of any other arcade games. I get most of my Bionic Commando information from here. It is a useful resource for generic BC information and history, so you might find what you are looking for there.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
I never thought that I'd find a Bonk TAS to be particularly interesting, but the flora springer trick really saved this one. The fluid motion of bouncing off enemies when travelling at high speed was impressive and looked great. Voting yes.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
Is there a reason why you don't jump to grab the orb after the boss battles, as in the published run?
That aside, I am surprised by some of the overpoweringly negative criticism. When I play Castlevania I use a huge amount of lives and powerups, so I found this concept interesting. I strongly dislike, and always have, that whenever someone submits a run with goals that deviate from the standard any%, 100% etc. there is a deafening cry of "arbitrary!" "random!". I'd like to thank scrimpeh for trying to give Castlevania tasing a new lease of life. While I hope that more authors make creative completion runs of great games I agree with adelikat's sentiment that when you remove damage and subweapons there isn't much left. I found the Grim Reaper battle most entertaining and the bat battle least interesting. Without the damage boosts etc. the linearity of this game is quite outstanding and damage seems too easy to avoid to warrant a separate category. This balances out to a meh vote from me.
On the arbitrariness/random goal choice issue... I would consider such a goal to feel 'tacked on', as if the author created the goals to fit the movie rather than the movie in accordance with the goals. Arbitrariness in Castlevania would be "doesn't use holy water", "takes damage only in level 6" etc. Arbitrariness has nothing to do with entertainment. It describes goals that are inconsistent or goals that cannot logically be derived from the game's challenges. The goals here are quite clear. No damage at any point of the game. No subweapons at any point of the game. They are consistent and they notably affect the way the game is played and so can be rationalised by comparison to the game's challenges.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
I was wondering how long we'd have to wait for a run of this. With so much going on it is difficult to watch, but that's just a case for watching it again. I think that the combined damage was an important discovery. Some of these bosses, especially Mr Burns, take an age to go down when playing with one character. I also like how you managed to show off most of the game's cool easter eggs etc as I expected that it would require a devoted playaround run. Voting yes.
Experienced Forum User, Published Author, Skilled player
(1091)
Joined: 8/26/2006
Posts: 1139
Location: United Kingdom
I played Timesplitters 2 to death back in the day. I think that a TAS would be just as good as Goldeneye. The best way to do it would be with 2 players in Hard Mode. This could be really interesting as much of the level progression is done by completing various objectives and splitting this up between two characters could cool.
I'd also love to see a run that played through the challenges etc. as, with good luck manipulation, you can get some crazy fast times.
