Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
I'm not against rejecting submissions. I have one of those myself.
I'm against a list of published runs listed showing the "worts" of the site. It's the principle of a 'wall of shame', not whether or not the author is or is not insulted.
A wall of shame is made to make fun of and insult people that are on it.
By creating such a list, we are publicly saying "here's a list of TASes that the audience rates as sucks."
I'd believe the overall aims of the site to not stoop to such childish displays of "let's display, point and laugh at these bad movies!"
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
"worst rated movies" page showing who has the statistically poorest TASes to me sounds like a wall of shame. The site had more than enough content and statistics already. The only thing destructive here is downplaying TASes that may be sloppy and not widely liked by the audience, but which the authors did put time and effort into.
And this is a poll. I'm guessing my opinion won't change it (although technically I could undo these efforts since it's site related, but I find it nonconstructive and intrusive to do so) but that's my opinion, that this addition is insulting to the efforts of the TASers who have their work displayed as "Failure" for having low technical/entertainment quality.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
AKA wrote:
would't spin dashing be better? I know the unassisted videos don't use it becuase much of it would require insane accuracy and luck.
spindashing for sure is better, however you have to take into account the game's 3d overhead view, compared to other 2d sonic games. There is a lot of turning in this game, where spindashing fails. if you spindash, the best way to maintain the speed is not hold any buttons and just let it take it's course in a straight, and bounce off walls. However, with all the twists and turns I have to make, that's no good. In other sonic 2d games, spindashing is far superior because it can be charged up if you hit a flat surface even if you were going at high speeds, and you're always going either left or right, so the straight directions of spindashes are very good.
In 3d blast, the controls are very slippery; in order to start a spin dash you have to be at a complete stop, which is near impossible to do when trying to slow down from running. the 3d means that you have to turn left, right, up, or down, or go on diagonals.
spindashing requires you to wait for a Flickie to settle down and land on you. sometimes it is impossible to stop, and make a spindash charge up in the direction and have the Flickie land right on top of you. Sometimes it's just faster to jump up and start running, especially if the next destination is short and has lots of turns involved.
My last gripe about spindashing is that it makes hitting the next target harder. when rolling you'll hit the robot, but coming to a complete stop when curled is near impossible, since again, slippery physics and sonic has this personal gripe about coming to a complete stop quickly.
But these are my initial observations. In redoing GG1, armed with the speed memory address I should be able to save a lot of time, with optimizations that I found. Expect tighter play and more/better spindashes.
MAJOR EDIT: I found the mem address for speed, and I made comparisons as to what is fastest in each situation. So I'm going to redo the entire GG1, based on these findings:
Speed: 00FFC204 1 un/signed (I think these both give the same value.)
Max speeds oscillate. Letting go of directions does not give the same Super Mario World effect, so no 'hop glitch", unfortunately.
All "slope" effects increase/decrease based on the steepness of the slope. Going down steeper slopes gives higher speeds, going uphill negates speed. Commonly you can go down hills so fast that you will be running in mid air.
-Max running speed takes 36-7 frames to attain from a complete stop.
-Max running speed: 76-79 (varies).
-Speed when jumping at max running speed: 77-79
-Acceleration gained from jumping is less than acceleration gained form running.
*Jumping before max speed is reached is slower than running to reach max speed.
-Rolling down slopes speed: 79-82
-Jumping on a down slope, when rolling/running at max speed: 100+ slight boost
*This is a slight 'boost' in speed; it brings the speed to 100 instantly, and then in ~10 frames the speed drops back down to normal max jumping/running speed. It is always worth it to jump when running down hills, not matter how small the slope may be.
-Running uphills: 75-79 (based on steepness. Jumping and rolling up hills is to be avoided entirely. Only run or spindash up hills.)
-"optimal" spindash: 119, slowly decreases by 1 over time (moreso when going up hills)
(note: A and C are the same button in this game, alternating A and C to charge up faster does not work.)
-from a standstill, an optimal spindash takes ~16 frames.
-spindashing down slope jump speed: 127 boost
-You can bounce into walls, and this will rebound you, keeping the same max speed with which you hit the wall.
-When at max speed, if you bounce off of an item or enemy, you do not lose or gain any speed.
-Turning causes a centripetal affect, in that the faster you turn, the more you slide, and your desired speed decreases. It is best to never switch immediately from one direction to another without first using the diagonals to 'ease' your way to your desired direction. This lessens the decrease in speed the most.
SO, I am going to redo, knowing full well that with this knowledge and these guidelines, I can save time, most likely more than 1 second, especially since 10 seconds into the course I already have ~20 frames saved, and that was in an already tight spot with mostly uphills and flats, no downhills.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
I'm pretty sure those are vs. only tricks. I remember reading an article where the battle and vs. modes are glitchier than GP, and such things like this occur in those modes, but not gp.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
so, would a TAS of the first level look something like this? / AVI / youtube
Green Grove Zone 1, in game timer: 0:47, which is 4 time units faster than the WR that I know of. (Zer0, I couldn't view your .gmvs... what was your in-game time for gg1?)
First time playing any Sonic game, so I'm pretty sure that there's a few mechanics that I'm not familiar with. This time probably is beatable, but I made this as halfway between test run and fully optimized, in order to get a feel for it. Those controls can get very slippery (I particularly hate the winding narrow uphill path).
(Edit: I found this stuff out all on my own. aren't memory addresses amazing little buggers?)
So, yeah. Comments/thoughts?
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
"Congratulations! But you missed 6 bonus levels!"
oh well.
Nicely played, well executed. I thought I spotted frame savers here and then, but regardless, awesome shortcuts, glitches and such, you really reduced the projected finish length.
"Yippee!" "Abracadabra!" "Shazam!" "Alakazaam!"
How about "Yes vote!"
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Zurreco, you took exactly what I said and made it seem like the opposite. I never said it was an official classification. I CLEARLY said it was not about being exclusive at all. I never said people binding themselves to one system (which by the way some people do) is a bad thing.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
a N64 TASer is someone who happens to be TASing/has TAS'd a N64 game, does not have to be exclusively. They have a pretty good knowledge about mupen and 3D movement to be able to make a N64 TAS. Sometimes it can be exclusive, but only in some cases where the only games they have looked at are few and far between, and only in the N64 system. and never in a negative way.
hope that clears that up.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
I think I'll handle this one.
They are technically harder to do. In theory 3D movement makes optimizing harder. N64 games are generally longer. People can debate about whether they are "harder" to make, but IMO, someone who specializes in N64 TASing (like myself) woill not produce as many speedruns as someone who specialized in NES TASing.
N64 TASing is more recent, started in 2005ish, compared to others, so it's fairly new, i.e. less time for TASers to make them.
N64 games are harder to get running in mupen. Many N64 games either have many graphical problems, or simply do not run at all. So the number of games that can be run is lessened.
The EXP Pak can make N64 TASes desynch very easily, thus ruining many potentially good games because desynchs are very hard to work with.
There aren't as many N64 TASers as other systems, but the number is steadily growing.
I'm delighted to hear that you're interested in trying it out. go see the N64 forum, and give me a PM, I can certainly get you on the right track.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Yes. During the route planning we did, we timed how long it took to superslide to different areas, and experiemtned with well early, and got farter than anyone else. Acryte tested getting to that chest with the bombchus in the area.
He was able to:
-use enemies to clip into the walls
-swim underneath the level
-clip into the wall when the water level was low
-Almost reached where he needed to go when the water level was low
After an hour or so, he concluded that it could not be done, the water level was just not the right height for him to swim underneath walls and get to where he needed to be. He got very close though. I'm fully willing to not let the idea be tossed, if anyone wants to give a stab at it, I'm sure Acryte could share how he did it, and someone could try and make it work.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Spider-Waffle wrote:
This time around we should do the micro planning before the actual run is started. This is exactly where, when, and how many of each item should be picked up. Most noteably, bombs, for supersliding. If a super slide will save fraction of second it needs to be considered into the route. This is because you can pick up 5 bombs in less than 5 seconds. So figure out how many bombs you can get along the way and how many superslides will save time. Then we might have to weigh the cost of getting some bombs to the time saved from super sliding.
I'm all for this. MM has an any% route that has every micro detail of what to do at what exact point (use 3 bombs to superslide to here, collect 10 arrows, manipulate 5 rupees, etc. etc.) of which you speak, I see no reason why that same can't be done here.
There are some spots in the MM run where it seems like Mukki is supersliding more than he needs to, wasting boms and time, wheras other times he has a perfect ratio. I agree more bombs should be manipulated, more SS's done, and everything planned out so that at the end Link has zero bombs, and each one was used effectively. with one published run, and another near finished run, I'm sure this could be done.
mwl wrote:
You guys sure got it done fast. I can't find any possible improvements to be made to the current route.
Sure did. It was mostly in figuring out that 2a vs. 2b path. But even that was easy to figure out, took us less than a day's worth.
Spider-Waffle wrote:
Another thing to investigate would be the chuko gathering to get the bottle. People were saying you could hit them to make them run to where you want them or make them stay put or something. It seemed like there was at least one that ran way far away. Might want to do some super slides for this too.
Yes, but it would be very tricky in managing where 8+ cuccoos end up after you attack them. Don't know about superslides in this one, although it's worth considering.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
hm, well that portrait thingie only accurately shows my choice of cap:
but finding my right eye color was tricky (still not satisfied with it), and my facial hair interestingly happens to be every natural hair color when that scruffy- black, brown, red, copper, blonde, and white, which the site is unable to replicate.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
I actually think hex editing will work wonderfully in this game. The reason is that the luck is all based on the TIMER that shows when you land on a square. So if you shorten the length of one course, the next one will still have the kart landing on its boxes at the same timer, not the same frame #.
However, I don't know what will happen with the other kart CPU's, and if they will do exactly what they did before, and be in the exact same places.
If you're thinking of redoing an earlier course, one idea might be taking the very first cup and using Toad/Koopa for the 5 courses, because I am 95% certain that either of those two drivers will save a significant amount of time compared to Bowser, because of better turning (DP1) and much faster, and longer long boosts. If that seems like too much work that may not be successfully hexed in, I'm sure an obsoletion to this movie should be sure to use either Toad or Koopa from the beginning.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
The Quad run set off a whole blaze of ideas for any other sets of games that, when played with the same controller, would give similar, and cool looking results. The many so far suggested bigame ideas are scattered throughout the forums and on IRC, so here will be the comprehensive wishlist for people to suggest what games they want to see completed at the same time.
However: not all games played together would make a good dual/quad/tri/quasi/whatever run.
I think it'd actually be best if the bigame runners: DeHackEd, Baxter, and Angerfist, made their way here and gave the list of what makes a good bigame run. They picked the best candidates, and they looked closely at the games, and tested immensely to see what would happen, and showed that their choice of game was flawless.
For those that don't know the definition, bigame run is having one movie file that completes more than one game at the same time. The methods for how to make such a run, and some in progress, can be found elsewhere.
Start making wishlist ideas. Here some criteria/things to think about when making suggestions:
A bigame run has games that, with the same input, flow together smoothly.
Be careful in what games you suggest; you must list a good/logical/technical reason for why they would work well together.
Of note, the two Megaman big game runs (Quad and dual X/X2) are absolutely PERFECT candidates. watch those runs/wips to get a sense of why the games work so well when played at the same time.
Games that have ignoring input, where one game can do something, but the other can't, is good. MegaMan games are good; when you're sliding in a Rockman game, you can't jump or go left and such. these kind of movement restrictions are good. transition scenes are good, where the character does not move.
The games should have about the same ending time. a 1 minute and a 5 minute game do no go well together, since only the first minute is hard. make sure they end pretty close to each other.
Think about what crazy input will do to the game. If all of the games go right at the same time, that is probably a good thing. If the games have to go in different directions, you may have running against walls a lot of the time.
The games might be better if they share at least the same genre, and preferably gameplay. an RPG and a shooter next to each other might not work so well. Games in a series that have almost identical gameplay, like MegaMan, are good candidates.
So 'dems the suggestions/(rough) rules. Go!
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
so instead of copy pasting it anytime someone asks about it, just look here. I'll be updating it every time a route change is made. If you forget the link, look in my homepage profile, it's listed under my "contributions" section near the top.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Criminey! I had completely forgotten you had showed me this. The biggest question is can it successfully be used to good measure in the new any%? Heck, can it be used for any of the tricks that you postulated in the first place?
I'm sure the skipping cutscenes potential is what this trick has the most to add for any%... as would passing through damaging barriers, like, say, the ganon time trials barrier? ^^
I myself am a very bad individual to be given the basics of a trick, and figure out a massively useful application on my own. (I suck like that.) but if anyone can get it to do one of the tricks theorized, getting faore's wind (or any other spell; does it work with any other spell?) might be very beneficial.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Tragedy, are you still working on this? that strat nfq suggested does not go against the 100% restrictions to not receive damage that the other gyrocopter technique uses.
That technique actually reminds me a little of DTD's in F-Zero X... you gain amazing speeds in the air, but it's the subsequent, higher than average speed that you hit when you reach the ground that is sought after.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
goron bracelets?
you need to go in goron city, do some fire torch stuff, get into durania's room, play Saria's song, and watch through a very long cutscene of Durania dancing. I think it's long enough to make it not worth it, but I am not sure. Getting bombchu's allows access to dodongo's very fast, and then lets bomb hovering get to the bomb bag fast.
Editor, Experienced Forum User, Published Author, Experienced player
(730)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Song warps are faster than you might think. dying requires to plan carefully ahead to take damage enough to be low health, than the death animation, choose to not continue, select the file again, then load inside ToT. Playing the prelude is taking out Ocarina, play the song, and 2 short screen scenes, easy cheesy. Of course, it's 6 seconds slower if you don't already have the ocarina out.
Getting into dodongo's early was exactly a flaw I pointed out. (massive time could be saved had we thought of that earlier with one route suggestion.) we've concluded that going to spirit temple early as normal, and doing a deathwarp is still fastest. we also considered getting 10 skulltulas to buy an adult wallet and then buy some bomchus, and also doing well early and getting the bomchus in the chest. The first was way slower, the second was sooooo close to working but not quite, because Acryte wall clipped in many areas, but was centimeters shy from having the water level be at the right height to swim underneath the walls and get to the chest..
Yes you need a sword to do bomb hovers, so no BA/reverse BA stuff is done before bomb hovers in adult section are needed.
In 3d blast, the controls are very slippery; in order to start a spin dash you have to be at a complete stop, which is near impossible to do when trying to slow down from running. the 3d means that you have to turn left, right, up, or down, or go on diagonals.
spindashing requires you to wait for a Flickie to settle down and land on you. sometimes it is impossible to stop, and make a spindash charge up in the direction and have the Flickie land right on top of you. Sometimes it's just faster to jump up and start running, especially if the next destination is short and has lots of turns involved.
My last gripe about spindashing is that it makes hitting the next target harder. when rolling you'll hit the robot, but coming to a complete stop when curled is near impossible, since again, slippery physics and sonic has this personal gripe about coming to a complete stop quickly.
But these are my initial observations. In redoing GG1, armed with the speed memory address I should be able to save a lot of time, with optimizations that I found. Expect tighter play and more/better spindashes.
MAJOR EDIT: I found the mem address for speed, and I made comparisons as to what is fastest in each situation. So I'm going to redo the entire GG1, based on these findings:
Speed: 00FFC204 1 un/signed (I think these both give the same value.)
Max speeds oscillate. Letting go of directions does not give the same Super Mario World effect, so no 'hop glitch", unfortunately.
All "slope" effects increase/decrease based on the steepness of the slope. Going down steeper slopes gives higher speeds, going uphill negates speed. Commonly you can go down hills so fast that you will be running in mid air.
-Max running speed takes 36-7 frames to attain from a complete stop.
-Max running speed: 76-79 (varies).
-Speed when jumping at max running speed: 77-79
-Acceleration gained from jumping is less than acceleration gained form running.
*Jumping before max speed is reached is slower than running to reach max speed.
-Rolling down slopes speed: 79-82
-Jumping on a down slope, when rolling/running at max speed: 100+ slight boost
*This is a slight 'boost' in speed; it brings the speed to 100 instantly, and then in ~10 frames the speed drops back down to normal max jumping/running speed. It is always worth it to jump when running down hills, not matter how small the slope may be.
-Running uphills: 75-79 (based on steepness. Jumping and rolling up hills is to be avoided entirely. Only run or spindash up hills.)
-"optimal" spindash: 119, slowly decreases by 1 over time (moreso when going up hills)
(note: A and C are the same button in this game, alternating A and C to charge up faster does not work.)
-from a standstill, an optimal spindash takes ~16 frames.
-spindashing down slope jump speed: 127 boost
-You can bounce into walls, and this will rebound you, keeping the same max speed with which you hit the wall.
-When at max speed, if you bounce off of an item or enemy, you do not lose or gain any speed.
-Turning causes a centripetal affect, in that the faster you turn, the more you slide, and your desired speed decreases. It is best to never switch immediately from one direction to another without first using the diagonals to 'ease' your way to your desired direction. This lessens the decrease in speed the most.
SO, I am going to redo, knowing full well that with this knowledge and these guidelines, I can save time, most likely more than 1 second, especially since 10 seconds into the course I already have ~20 frames saved, and that was in an already tight spot with mostly uphills and flats, no downhills.