Posts for JoshButro


JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Nice to See the X*Bert levels TASed. It was on my "to-do" list years ago, but I stepped away from all this in 2016. Glad to see you are awesome at TASing this game. My best X*Bert run was 21:XX, something, I forget. Sorry I deleted my Twitch channel, or you could have used that run as the comparision. That 35 minute run on display, I keep yelling "Hurry Up" hoping the would finish faster. Congratz again, this game isn't the easiest to optimize, even at 1 frame per whatever. --joshbutro Funny how my signature still reads this...
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Great TAS! I am impressed. I didn't think 54 minutes was possible. My version 2 TAS was hoping for sub hour. I would be interested in seeing a screen by screen comparision of frames. I know some screens had lots of frames to save, and some had little if any. I really would like to see this accepted as a good improvement on the current TAS. Maybe someone could memorize this route and RTA it under an hour lol. Great job nymx. --josh
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
Post subject: Passing the torch
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Spoke with Nymx. He may TAS this game. Any part of my TAS can be used in the new version if needed. I know this should be stated as confusion of using the same old TAS isn't well received. (See: Super Metroid Ceres escape). This game is essentially 80 separate puzzles. I'm not saying my solution is the perfect optimal solution, but reusing my old TAS is permitted if needed. I'll change here periodically if a new TAS is started. Good luck!
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Patashu wrote:
Except maybe by having an 'arbitrary code execution demonstration' style category, that doesn't have to be the fastest any%? Don't think there's precedent for it but this would be a good place to start.
But now doesn't this run fit the same reason that the Super Metroid 100% Map Completion TAS was rejected: Too many categories for a game. This seems like a new category name is being created to publish this, instead of dissecting the existing movies and seeing if there is a faster way. It is an awesome video and looks like a lot of work went into it, so that makes this decision difficult on what to do.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Status Update: Level 2-1 is done, better than I thought, but I have a question: First the fanfare of saving 43 frames total on one screen... by wasting 23 frames. Is that a standard practice of TASing? I saved 1 jump (35 frames - not the focus of this question) but delayed the final jump by 23 frames, in order to cancel out counting 1 object for 32 frames. Delaying (+)23 frames, took out (-)32 frames for a net savings of 9 frames. I guess my question is: Is it ok to delay and take no action, to have an overall savings in frames? After 2-1 Fade To Black: 79 frames total so far.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
I feel slightly silly. Been awhile since I uploaded a WIP and had to find it... but I did. lol. Anyways... http://tasvideos.org/userfiles/info/31620324977922482 is the WIP for Level 1. Notes to look for on Level 1: -Improved Game Start and Level 1 Intro sequences [call it I learned to TAS better lol] 1-1 - Screen reroute based on it's faster to take the disc and less have less enemies on the screen (I shall call it the "Disc for less enemies" strat lol) and it is 1 jump shorter. [9frames saved] 1-2 - same "Disc for less enemies" strat as 1-1. This strat will be primarily used on screens that have too many enemies when the board is finished. Also 1 jump shorter [8frames saved]. 1-3 - same "Disc for less enemies" strat. 1 jump shorter. [8 or 9 frames saved, i don't have my notes in front of me.] 1-4 - I could not find a faster route than the v1 TAS. I believe this level is optimized. I ran it differently and saved 1 frame... overall i remember at this point through the fade to black on 1-4. I have saved 36frames. I do believe the first 4 screens are completely optimized unless someone can see a completely brand new route, but the number of jumps are minimized for those screens. Level 2 is where optimization will start to be fun lol. I say that with all the sarcasm in the world. I am starting to look at mathematically what is the minimum number of jumps needed and how to route to complete the board minimally or just over. Enemies ruin optimal movement, so it is a tradeoff, but there may be a right and wrong answer.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
The v1 TAS is good to compare to, however that TAS was done deathless, which is not the fastest possible time. There are 2 screens I immediately think of, that have drastic improvements by taking a death, and a few more that could have smaller saves. Life management (I think) will be important in this TAS. Death abuse can save a lot of time [especially on levels of 'Frogg-death]. The old movie will help, and some of the screens I believe are already optimzed and can not be anymore, but we will see. So far the first 4 screens have seen improvements I didn't know were possible.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
I'm not sure if anyone will care if I double post. I think I'm the only Q*Bert 3 TASer, lol. Re-TASed and left the music on, so it is a fair comparison to the v1 TAS. So far... I found out how bad of a TASer I was on the v1. 3f saved on the Start Menu, 4f saved on the Level 1 intro [call those learning how to TAS] The real improvements: Saved 9f on 1-1 [The math is above]. Saved 8f on 1-2, same strategy as 1-1. Having 4 less objects on screen, and 2 less jumps, and using 1 disk, saved the frames. This strategy works for now as it seems all disks (so far) take 189f to use. Jumping speed is 35f/jump, but when that decreases later, is disk speed the same, and will using a disk overcome the time spent vs jumping, Right now 35f jumps vs 1 disks equals 5 jumps, later when jumping is closer to 15-20f, that is a big difference.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
I definitely appreciate the feedback from both of you. The more I think about it, the cheesier it seems to save frame from basically a hardware thing, instead of gameplay. Only being (1/80) into the game, won't take long to reTAS and leave the music on. For entertainment value: the game will be between 60-62 minutes, so an hour of silence might take away from the game. I will reTAS the first screen, and post a WIP in the week of how it goes so far. After 1/80 I am 9f ahead, so already an improvement lol. Thanks again.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
I appreciate the reply. Turning the music off for saving lag frames isn't the main reason for this TAS. It seems silly it works that way to 'save frames' lol. It most screens (probably greater than half), I can find more optimal routes due to new RTA strats and new learned TAS strats. There is also time to be saved by this run not trying to be deathless. That of itself caused the last TAS to be not optimal. You do bring up an interesting point. Altering the v2 TAS by taking out the music is now comparing apples to oranges to the v1 TAS. So I guess I'm posting this question now, since I'm 1% into the game. Would it be better to TAS it with music ON to compare it to the v1TAS easier, or music OFF to 'save arbitrary lag frames' for the best possible time? There are frames to be saved from lag music and from gameplay.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Version 2 of the TAS has started. I'll work on it a couple nights a week, so no estimated completion date at this time. Just some notes I'll be adding to this, if anyone else has any comments or suggestions. As tested before, turning the music off saves over 60 frames, possibly as much as 90 frames per playable screen, so 80 screens x 60 to 90 frames = a whole ton of frames lol. The options menu loses 300 frames, but makes up for it, but the end of Level 1. Crossover - jumping on a block that is already painted the final color and does not change its color. A wasted necessary jump. 1-1 Q*Bert jumps at 35 frames per jump. just a fun fact. So far 9 frames were saved on Level 1-1 and I'll explain why: At the end of the level, it takes 32 frames to count every enemy/object on the screen. The more objects, the bigger the time loss. Through completely different routing from the RTA/TASv1, 9 frames were saved as follows. v1 completes the level using 0 discs, 4 crossovers, and 5 objects remaining on screen. v2 completes the level using 1 disc, 3 crossovers, and 1 object on screen. The difference: The extra crossover involves 2 extra jumps (35frames x2 = 70f), the disc takes 189f, and 4 objects (32f x 4 = 128f). So 2 less jumps, 4 less objects, and 1 disc => -70f -128f + 189f = -9f: 9 frames saved. Taking the disc at the end of the level can save frames, so this will be calculated into future screens. FYI 1-1: Each disc, (even though they are at 2 different heights - 5th row, and 7th row), they both take 189f, so neither one is faster. Surprised me. Next is 1-2, and the 'disc to finish' method might be faster.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
http://tasvideos.org/userfiles/info/25282349202682415 First Level, Please be nice, im still relatively new to TASing,
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
I hope im only an amateur TASer (lol), but ive seen a few people play this game, and may try to see if i can make something resembling a TAS. If anyone else is working on it, let me know and i'll see if i can help. I may not be able to make it optimal on my own but would be willing to contribute something.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Just wanted to congratulate you Nymx on finishing this TAS. You have spent a lot of time on it, and you should be proud of that. [Before i start, i want to mention, i have not read all the comments/posts, thats a lot for me to read, and frankly will take me too long, so my opinion is my opinion.] Nymx, I have enjoyed watching your TAS on Sunday afternoons on Twitch, i my not have been active some weeks, but i was lurking. I liked the community approach, as some people have ideas but do not put the effort into making a TAS, so sharing ideas is constructive, and i plan on doing a TAS live like you did. I was entertained by the run, i enjoyed seeing shinesparks where i did not think they could be performed. I have occasionally watched SM runs and TASes, but it was interesting to see this 100% map completion, because parts of the map are never seen in other runs. Having said there are ~90seconds of improvements, should be a starting point for another TAS by you or someone else. Along the way improvements can be found, and a first TAS completed, and these improvements are motivation for another run. if you plan on 100% map v2 TAS, you may find even more. I enjoyed the run, and watched how much effort you put into it. Don't let anything discourage you from making a v2 or another TAS. You put the effort in and completed this. Be proud of that. Completing a TAS is an accomplishment.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Well done again Hetfield. I enjoyed watching the making of this, and really enjoyed the whole run when finished. Definitely a YES vote. Keep it up. What game is next? lol
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Awesome TAS, i loved it. Very good route to minimize extra visits. I didn't know Culex could be defeated that early in the run, so that surprised me. Great TAS. I vote definitely yes.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Thank you for the feedback and I guess this answers a lot of questions about this game on this forum. If many people saw this run as boring, I apologize. I wished I would have been told sooner. There are no game breaking or sequence breaking glitches found yet. No way to skip levels or credits warp. There was no existing TAS or any speedrunning information before i started. The interaction during the WIPs and submission was minimal at best and zero help provided on gameplay. If there are any suggestions on how to make this run "not boring", please let me know. The first TAS was intended to be a fast deathless run, to show off the game and show it can be completed deathless, because honestly no one will even attempt this game because of its difficulty. I've seen three people play this game on stream, and each of their death counts was easily in the hundreds. So I thought a deathless run would be impressive. The second version of the TAS I'm working on with have a few strategic deaths in it that save time. There are times that taking deaths is faster, but some of them look sloppy though. They look careless to immediately (or some are very delayed) jumps directly into an enemy or its intended path. I didn't include those the first time, because some look very noobish. But the next version won't have "time wasting for entertainment". To maintain the first TAS deathless, there had to be some stages (6-2 around the Derby and 16-4 around Coily and the Frogg) had to involve extra jumps for entertainment to maintain block color but also dodge multiple enemies and time/manipulate enemies too. This run may be seen as boring, and I tried to explain what all went in to this run, that someone may not specifically see, regarding enemy spawning, enemy manipulation, and polarity, which can add many minutes over the course of a run, if someone doesn't understand these concepts. In closing, this game is a difficult puzzle game about hopping onto colored blocks in an optimized or minimal path, and i tried to demonstrate that. If the run was boring because it didn't contain game breaking glitches, none exist to date. If the run was boring because it is an hour to watch it, 80 stages isn't short to beat (even the RTA WR is 1:08:35). If the run was boring because important concepts aren't understood, it will just look like hopping. If the run was boring because it's Q*Bert, the community for this game and series is only a few people. If the run was boring for any other reason, I received zero help on this TAS, no one showed any interest whatsoever in helping with gameplay aspects. To make this run "not boring" for the TAS community, I'm open to suggestions. And to be honest, I'm working on next improvement version of this TAS, but I'm very very hesitant to even submit WIPs or the completed TAS in the future based on the little feedback I've received, criticism on non-gameplay mechanics, negative votes having no feedback, and a general lack of interest. This run may have been boring to most, but that's Q*Bert 3.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
so this is interesting. if the table is hard to read, i'll explain. I have ran the game with Backgrounds ON and Music ON (TASv1) Backgrounds OFF and Music OFF (TASv2) Backgrounds OFF and Music ON (Test1) Backgrounds ON and Music OFF (Test2) I found an interesting result. Using the options menu takes ~360 frames. The Intro for the Level number always takes the same number of frames (bkgd/music doesnt matter) [3 frames different is optimization] The 'Loading Time' for the Level is 97frames for Bkgd OFF/Music OFF and Bkgd ON/Music OFF, and 182 Frames for Bkgd ON/Music ON and Bkgd OFF/Music ON, -so music OFF is a faster loading time, backgrounds have no effect. Level 1-1 is played at ~1122 +/- 2 frames regardless of bkgd/music The 1-2 Loading time is 577 Music ON vs 502 Music OFF, again bkgds make no difference. Level 1-2 playtime is ~913 +/- 1 frame, regardless of bkgd/music. The 1-3 Loading time is 580 Music ON vs 512 Music OFF, again bkgds make no difference. --In conclusion, the backgrounds have no effect on the time difference of the game, the Music does. Has anyone seen this before, because this doesn't seem to make sense, but that's what the numbers show.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
I am currently testing that, I have been busy the last couple days to fully test it, but i should have an answer today or tomorrow. My guess is music shouldn't affect it, but i didn't think backgrounds had that much of an affect.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
Post subject: Breakthrough
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
A discovery, a breakthrough, a... disappointment. To read: First column is frame count for TAS v1 Snes9x Second column is TAS v2 Bizhawk 1.9.1 Third Column is TAS v1 done on Bizhawk 1.9.1 [as a control to compare Snes9xv1.53 to Bizhawk 1.9.1] Results: Turning backgrounds and music off DOES NOT AFFECT Gameplay, however it affects the load times between levels. The intermediate number show frame difference. Frame difference for gameplay is negligible. Frames saved are loading time between levels.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
Now, I am working on a more detailed, version 2.0 TAS for this game. I am using Bizhawk 1.9.1 and I love this TASeditor it has, so expect alot, as I ask for feedback and such. The first movie was accepted even though it got very very little feedback. I hope this one receives more feedback. The first big difference in v2 is turning off music and backgrounds, they lag the game, about 1-2 seconds a level. So 6 seconds to turn those off is made up by ~80 seconds over the run. http://tasvideos.org/userfiles/info/20428769902681613 level 1
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
As far as I've seen no one had anything negative to say, and Spike could only save a few frames on Non-gameplay areas. So I haven't seen anyone say that gameplay could be faster.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
I don't know if it needs to be said, but i thought there would be more interest just based on the feedback I received in the SNES forums while putting this run together. If the run is too long, it's because no game breaking/sequence breaking glitches have been discovered, so this run is all levels. If it's because it is Q*Bert, it is a game most won't even attempt because of the high difficulty level, but i hoped this TAS would attract people to the series. I'm not sure what happens now, since the vote count is lower than all o the other movies posted on similar dates, and the viewing rate for this movie has essentially gone to zero. I had played on TASing the X*Bert levels after this to really show off the game, but i may just for my use, it really depends on how this TAS goes, and it didn't get the reaction i thought it would. I guess i just needed to vent, since i put alot of my time into this TAS to have it go unnoticed. Vent over.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
I enjoyed watching this TAS being made, and it looks awesome Het. Very nicely done. You get my vote. Now you have to do an RTA run that matches the TAS, lol.
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels
JoshButro
He/Him
Experienced Forum User, Published Author, Former player
Joined: 1/8/2015
Posts: 43
game has been submitted for review and voting, any support is appreciated
current TAS: Q*Bert 3 any% v2TAS<br> next TAS: Q*Bert 3 X*Bert levels