Tony Hawk's Pro Skater 3 is the third game of the Tony Hawk series. This game introduced the revert and double tap tricks, which allowed for much longer and higher scoring combos. There are 8 main levels in the game, five of which are with goals scattered across the level, and three competitions. Once the main career is finished, a secret level (Cruise Ship) unlocks which also has it's own goals.

Objectives

  • Emulator used: DolphinHawk (syncs on Dolphin 5.0-16426)
  • Optimizes for in game time first, then real time
  • Completes all goals
  • Gets a gold medal in every competition

Comments

Options

It's modern Tony Hawk baby! Autokick off is the way to go. Autokick off makes it so that to move forward you need to hold your kick button. In the thps2 engine, you don't really need to hold ollie much at all to keep max speed, but in 6th gen games you need to hold it a good bit more. This means that I can't get a million one frame ollies like 5th gen thps3, but I will move faster overall and we make up for the extra ollie button holding with stat changes.
The other benefit of turning autokick off is that you gain control of your skater earlier in the run and can skip the starting animations.

Skater Choice

Thps3 does something really interesting in that different groups of skaters have different goal layouts. SKATE letters will be in different positions and the "trick on x gap" goal will be different. Even after all these years it's still not worth it to switch from Street to Vert, so our options are Jamie Thomas, Bam Margera, and Andrew Reynolds (along with Officer Dick but he's an unlockable skater). Jamie is the fastest to select, and he also gets a really easy to adjust trickset.

Trick Swapping

This happens pretty fast because of how smooth the menus are and how close together what we need is (how did they make these menus so responsive they're so perfect oh my GOD). In order we change:
  • Nosegrab to One Foot Japan; Boneless 180 One Foot Japan is our bread and butter move to get special.
  • FS Shove-It to 360 Varial, it helps with score a lot and is very valuable in Tokyo
  • Big scroll here, but we change Crook Bigspin Crook to an FS540 for two reasons. First we do it for the style, and second is again for Tokyo (it takes a lot of prep)
  • 360 Hardflip to 540 Tailwhip; Tailwhip gives more score and is a pretty short special
One Wheel Nose Manual and Layback Sparks aren't altered because keeping a grind and manual special helps with meeting score thresholds, ESPECIALLY in competitions, but also in goal levels.

Stats

These stats are essentially the minimum you could hope to efficiently complete the run in. Losing even one ollie would ruin this route quickly. The stats menu is a lot simpler than the tricks menu, and goes as follows:
  • Lower hangtime, manual balance, and lip balance to 0 because they're entirely unnecessary
  • Lower rail balance to 4. Firstly because it gives us the exact amount of stats we need to fill the rest, and second because rail balance is pretty much a necessity for some of the movements done in the TAS. I'm sure there are slightly slower movements you can do in real time (like spacing out grind hops for example) that would allow you to put that rail balance back into ollie to make things easier overall, but it's a major help).
  • Max Switch (gives equal speed in switch stance when maxed), Spin (helps a ton with scoring), and Speed (it's a speedrun).

Competitions and RNG Seeds

I only found the seeds for first, second, and third place for each heat in the competition because competition scores are decided from the frame you skip the final intro comp cutscene and get into your first run. Final scores are calculated by adding your two best heats together, higher score is better. I'll discuss it more later, but the minimum scores with perfect rng are 100k for Rio and Skater Island, and 160k for Tokyo. The rng refreshes if you wait a long time in cutscenes (working theory is due to particle effects or something similar moving it forward) or if you retry the competition. A retry or two will still come out a decent bit faster than having to do an extra jump and trick in each heat to make up for having unfavorable rng. These addresses are used for all the competition levels.

RNG Seeds (All 2 Byte Signed)

Positionheat 1heat 2 heat 3
First place00B4659E00B465A200B465A6
Second place00B465C200B465C600B465CA
Third place00B465E600B465EA00B465EE

Ending Run vs Retrying Run

On Gamecube it’s generally less efficient to retry the level and then hit end run. The reason it’s so valuable on PC is because the game takes almost no time to load, while on Gamecube a retry loses over half a second on its own.
The second reason (and more important one) is the ability to complete objectives within the few frames after ending the run. The main examples of this would be the end of Suburbia, where ending run into the cutscene saves almost a second and a half vs the rta route where they watch it in the middle, and Canada, where real time runners would wait well over a second at the end of the level for a goal to complete that instantly finishes with end run.

General Comments

RTA runners adopted the strats from the old TAS and then some, so I felt it was time to make a truly superhuman TAS and I think it came out super well. Compared to the previous TAS and RTA a lot of levels aren't exceptionally rerouted, but there are significant optimizations everywhere, mostly from limiting backtracking.

Stage by stage comments

In Game Timer Comparisons

LevelFog TASRTA WRNew TASDifference (RTA)
Foundry1:311:331:354(2)
Canada1:231:201:296(9)
Rio heat 15557572(0)
Rio Heat 25556583(2)
Suburbia1:241:201:273(7)
Airport5454595(5)
Skater Island heat 15655571(2)
Skater Island Heat 25656571(1)
Los Angeles1:201:201:266(6)
Tokyo Heat 15555561(1)
Tokyo heat 25555561(1)
Cruise Ship1:111:071:176(10)
In game time difference: 39 seconds (46 seconds)

Foundry

The Foundry route hasn't really evolved too much from the Fog route except for where we get the A and the foreman. This run opens the same way, cutting the corner super tight for the K then grinding to the valve.
Then comes the first major roadblock I ran into, bonelessing the S. This is normally pretty easy in real time because runners frequently use max ollie, but because I have it set to 7 I have to be a bit more creative with how we grab it. The main problem is that the letters bob up and down and we absolutely need a low part of the cycle to even have a chance to grab it. You may be able to gather exactly what went wrong here, an optimal start simply can't reach it. The way I lose time here is by clipping the wall slightly to slow down just enough to reach it.
(A little sidenote, but another issue here is that sometimes your ollie can be super low and I have absolutely zero idea why this is the case. I can only assume it is something to do with landing from a late ollie or something similar. I asked in the community and didn't get an answer so if it gets figured out I'd be very curious what is found.)
The next bit is pretty simple to get the E then wallride up to the top area, and now we've reached the main route difference between this and Fogs TAS. A few years ago it was found that you can fall onto the A after grabbing the tape, and this opened a pretty big reroute in rta. Runners nowadays will normally turn around at the fourth valve, grind back towards the way they came, then follow the path to hit the foreman grind with speed and go to the fifth valve from there. However, it was found that you can go from the A to the foreman with a well placed boneless, which eliminated the need for some backtracking. The TAS optimization of the segment comes from the fact that IL runners go towards the end of the scaffolding with the tape and use that for a consistent setup for the strategy, and I found that you can do it by just dropping off right when you get the tape and don't need to bonk the wall. This allows me to get 1 igt better than the ILs (using the IL timing method). The route after the foreman is pretty basic and standard. A SICK 50-50, a Boneless up to the bucket, a few side hops, then grinds with some Ollies/bonelesses to avoid upward slopes or get slope boosts. After the last valve I delay ending the run (and also choose end run over retry) to skip a cutscene.

Canada

Canada's gotten a major facelift in this TAS. In the previous TAS (and real time runs), the level starts with going to get Chuck unstuck, then going out of bounds to get back to the start and go towards the tape. This adds a ludicrous amount of backtracking in the level that's a huge waste of time.
This iteration of the TAS instead goes for the tape immediately, using a snowbank and some wires to reach the tape in an unintended way. After jumping from the house to the wire, I do a quick shuffle to build more speed and get a way lower/forward jump (this is also one of the major reasons that I didn't bother removing the rail stats. Not only is it faster to leave them in, but if your balance gets too bad you can catch the grind and basically explode with no balance meter even appearing for you to even TRY and save it).
When getting near the tape, I side hop to the left, do a really fast grind hop, and instantly jump out left again. The reason for doing the extra shuffle is that it gives you extra speed and forward momentum. We need the most speed we can get to reach the upper platform and making it up there in one jump saves a second or two by itself.
One question may come up: why not jump straight forward? Good question! Remember that shuffle we did when we first got on the wire? That extra speed comes back to bite us when hitting the tape segment. I know almost nothing about physics (chemistry>>>>) but I'm assuming that it's because of how much faster I'm hitting the rail perpendicularly, I have less speed overall. This on top of having to jump at a different time because of the different speed means that there's no chance of making the jump straight forward. It 's still a net timesave to do the shuffle even without the forward jump.
After grinding around the horn and jumping to bury the bully (plus impressing a skater), we get the next big route adjustment, say hi to Chuck! He's a genius and got his tongue stuck to the pole. It's up to us to rescue the legend, which we do while cutting the corner tight and jumping over the rail to get to the S.
While grabbing the S, I do a 180 One Foot Japan to get enough score to impress the second skater. Then we're immediately at another 7 ollie problem. Every other run in history will side hop to the right for the K, but again I just can't get the height because of the cycle. A quirky Boneless into side hop lets me get through it clean without any issues, and is a pretty unique piece of movement in the run.
The next bit is just following skate letters while adding in a slope boost after the A to build a bunch of speed. You need to side hop to the right to get the E, and then I hit a sticky situation. I can't make the Fog strategy because of the ollie stat, and was struggling with the normal rta left turn because if you jump at the earliest possible time to grab the E you won't go far enough to get past the rail and be able to wallride the billboard. A later ollie made quick work of it. As far as I can tell, you can't turn left enough to immediately wallride without bonking the wall without losing a good amount of time.
After climbing up the wall, I jump right and get a big slope boost to zoom through the final two skaters. The IL uses an ollie to cut the corner at the fourth skater tighter but you can't make it on 7 ollie. The speed that I get into the grind after the hill makes up for most of the time that could be lost from not doing the ollie.
In rta runs (outside of Foundry) you'd normally see runners retrying before ending the run to skip the cutscenes at the end. The drawback to retrying here is that you have to wait for a long time before the Impress The Skaters goal completes, and ending run lets me end the instant I hit the ground.

Rio

I haven't really mentioned it yet, but rng is a real bastard for these comps. I need to move forward the rng by 1 instance, and I can either do that by waiting forever in the cutscene or retrying instantly, so I choose the latter. The retry is a small annoyance but I can’t find a way to push the rng ahead before Rio starts, so I just have to deal with it.
After performing the Rio Retry™ my rng is considerably more favorable (going from a 180.7 to a 176.8 for the gold medal). This allows me to skip the manual at the end of heat two and save one in game time second.

Suburbia

Starting off the level I’m buffering ollie and kick, which means I can’t Boneless until I complete another action. I do a 180 off the platform into the slope boost then do a quick Fakie Ollie to 180 One Foot Japan to get special before using the sloped driveway to do a big Boneless to get the E. The next bit is just a grind, side hop for the axe, then skating over to and clearing the pool.
This is where the route differentiates itself from Fogs TAS, instead of going to the right when getting the pumpkin and grinding around the outside ledge, we go left and climb up onto the roof instead. One big optimization is that I get virtually no air time when clearing the slanted roof with a very tight ollie.
I grind over a power line (pretty dangerous, very extreme), grab a magic floating letter K, bonk a wall to turn around faster, then grind another power line (with a twist!). I grab the first tree branch, then jump off the wire to get the pumpkin below the wires. Once it does a squish, I bonk the tree and wallride up the wall and start grinding to the final branch. Immediately after the cutscene I jump off to the S and follow the path around to the pumpkin.
After this Pumpkin I ended up meshing the two existing routes together. Kami in his IL cuts the corner and jumps over the closer trailer, but then does a little bunny hop off the trailer hop gap. Fog instead goes super wide at the start, but clears the trailer landing on the wire giving him a faster and better path to the A. I just do first half Kami IL, second half Fog zoomie grind.
The ending from the fourth pumpkin onward follows Fogs route because that route saves almost 2 seconds of rta by skipping the Axe Man cutscene. The only difference is my tape jump ends up a little awkward. Due to being on seven ollie, I can’t get the super forward jump that Fog does and land on the rail, so I instead choose to land short and drop onto the house steps to get a massive amount of grind speed towards the man.

Airport

I seriously have no idea how Fog didn’t find this route considering how exceptional and far ahead of his time his TAS was, but it was found by Adelyn that it’s actually faster to grab the tickets at the start of Airport, grab the S, then get both of the pickpockets on the way back towards the helipad and baggage claim.
The TAS saves .7 over the original version of the strategy because I’m able to cut out an awkward side hop with good angles on the final turns. In the proof of concept video she jumps to the right to trigger the first pickpocket and move him to an easier position to hit on the way back and allows for an easy setup for the 180 turn to get to the helipad, but that’s no problem for TAS.
After we grind through the helipad and Crooked the baggage claim, we go to grab two more pickpockets in the bathroom. This was actually really hard to deal with because of how fast you move with autokick off. You have to start holding down super early and fitting between the pedestrians is super awkward.
Immediately after leaving the bathroom the movement deviates slightly from the original TAS. Instead of doing a big wallride and empty air to get the K, I now jump left over the moving walkways, then go up a kiosk and get to the upper ledge/K that way instead. Overall it only saves about .1-.2 over the Fog route, but it’s something.
After grinding down the escalators and grabbing the A there’s another pretty big route difference. It’s faster to turn sharply left and do the flags backwards (ending with giving the tickets to your friends). Once that’s done we immediately turn left and start climbing up the red rails. Jumping off of the second tier we wallride and wallie to reach the T while quickly falling.
The route is pretty much identical after that. Final pickpocket, E, flags 6-10, then the tape. Only difference is very slight movement optimizations. There was a theoretical strategy to grab the 7th flag, wallie up to the tape, then fall back down onto the rail to skip having to do the extra loop at the end. From my testing you can get the tape but the gap between the wall and the flag rail is too wide for you to catch the grind again.

Skater Island

There’s a pattern with these competitions and I’ll just spoil it now. The rng is absolutely dreadful and just does not get better. In Skater Island and Tokyo the rng I got was about 5-6 points higher than the IL or RTA world records would get. There’s nothing I can do to manipulate it in any capacity apart from retries or waiting a long time, so I just need to deal with it.
The strategy itself starts with the new Any% strategy. A special manual to pivot is used when landing from the Tailwhip, along with a kiss the rail and a crazy tight 360 Varial to get enough score (along with buffering the Boneless after ending the run). Such a brutal competition.

Los Angeles

This level is just the standard All Goals and Golds route but incredibly optimized from end to end. I found that doing an ollie off of the first tremor rail right before the rail kinks down allows you to get the gap while landing on the right side of the rail. Landing on the right is important because it gives us the ability to cut the corner super tight and start moving towards the next rail.
A similar technique is done at the second and third rails. The main goal is to jump super short so that you can get the gap and start heading into your desired direction faster.
After finishing the rails and jaywalking through a police chase, I get a slope boost off of the fountain with the K and do a varial kickflip over the stair gap. Right after landing in that grind I do a short ollie to clear the rail and angle myself left to boneless up to the top of the car wash.
After yeeting a guy off a bridge, doing an awesome FS540 (I hope that guy liked it), and freeing the balls, I wallride up the purple wall and finish the SKATE letters before getting the tape in an unintended way. Instead of using the bridge, I use the ramp under the tape instead. I boneless and air out of the purple quarter pipe, then spin into a wallride. I’m not sure why, but it’s so much harder to catch the wallride if you don’t spin. After catching the wallride, you can immediately ollie to catch the rail, then instantly ollie off the rail to get up to the secret tape. I’m really glad I did this tape strategy in Any% first because otherwise I never would’ve realized that spinning is the key factor and it would’ve taken forever to solve.

Tokyo

This strategy starts identically to the Any% run. Due to 7 Ollie stat you can’t fit a 360 Varial and another trick in the first air. After doing a special manual to pivot, I do a fast grind hop to get a bunch of air and speed.
Now Tokyo evolves into a whole different beast. Instead of doing a 540 Tailwhip, I do an FS540 then land in a special grind. I just out right when the grind triggers to do a 360 One Foot Japan (the 360 is very important) to another special manual to pivot.
Ending the competition on the second heat is delayed because I need enough time to pass to get a higher score. One frame earlier and I lose by a few tenths but I just squeak out a victory (and no igt was lost).
I really wanted two 57 heats for Tokyo, but after fully finishing it I don’t think that’s ever going to be feasible without using unlockable skaters and tricks like the IL does. If not because of the RNG, then definitely from missing the first trick due to 7 Ollie stat. Still happy with a 56 since that’s an improvement over RTA and the previous TAS.
(Future me here) Since I finished this whole segment I rerouted the combo a little bit. It’s currently useless because the rng is way too bad (and can’t be manipulated to be good enough), but this could save around 10 frames per heat if it was possible (https://youtu.be/O0m1Fjn1jxE)

Cruise Ship

The route for this level remains remarkably unchanged in the many years since the previous TAS was created. It starts off similarly by turning around and jumping down a deck. One big optimization over every other run of this level is how I utilize slight angle adjustments to save a large amount of time. I land entirely sideways from the deck drop, then do a short hop while slightly rotating left to instantly end up next to the wires to destroy the museum. A time save over Fog's TAS is that I skip the cutscene way sooner. To be honest I think that's mostly just because I can just manually insert one frame at a time while holding frame advance to find the spot where it skips. Couldn't imagine doing half of this run on actual Dolphin.
One solid optimization is instead of grinding around the outside of the room after skipping the cutscene, I ollie, get a slope boost, then grind hop off of the inside rail. It makes the exit angle a lot better and I carry more speed by landing on flat than on a slope. When leaving the museum I do a trick to impress a Neversoft girl, then do two tight jumps from the pool to slide and from the slide to the flags.
After I go up the flags and grab the tape, we get to the big "oh dear lord I need points" segment. After grabbing the S I do a seemingly awkward turn to jump down towards the next Neversoft girl. The reason for the turn is that if I went sharper left I would float on the underside of the ramp and lose a bunch of time. the route is pretty standard through the Nosebluntslide and then we get to the biggest change in the level.
After the Nosebluntslide, instead of going for the E, we go right to lowering the boat. This is because the E is an almost 3 second detour that we could partially avoid by rearranging goal order. After lowering the boat the route looks similar to Fog, but instead of ending at the T we grind the rail and head back to get the E.
After using the freshly lowered boat as a ramp to reach the E, we can finally end run and enjoy our trip on the Cruise Ship.

Other Comments

I did not expect this run to be so fast. I knew there were a decent amount of improvements over the years, but the time saves just kept rolling in. This run (provided I did the math correctly) would be a 5:52 with the normal RTA timing convention and PC loads. (Minor note: I forgot that PC players remove the movies so they don't get the cutscenes after completing Tokyo. I'm not sure exactly how much time keeping them loses but it could very well be a few extra seconds.)
Shoutouts/Thanks to:
  • Sawwl, Adelyn, and Slime for their various routing endeavors/movement optimizations in Canada, Airport, and Rio respectively.
  • Frankie not only for having the wr that I was comparing to, but also being so supportive of my work on the project.
  • Kami for the various strategy ideas and corrections he pointed out, and also just listening to me ramble on and on in voice calls while I worked.
  • Fog for the temp encodes and, more importantly, being my motivation to get into the TAS community. You pushed these categories to new heights and I can only hope to achieve the same with this project.

Samsara: Claiming for judging.
Samsara: Yes, hell yes, absolutely hell yeah. He'll yeah. He will yeah. Everything I said in the any% submission holds true here as well. Fantastic work, both in reaching the high bar set by the currently published TAS and the RTA, but also in knocking said bar up by a good hundred notches or so.
fsvgm777: Processing. Holding off (see below).
Samsara: Setting to Delayed at the request of the author, pending an incoming improvement.
Samsara: File replaced with a 158 frame improvement, and re-accepting!
fsvgm777: Processing.


TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 15576
Location: 127.0.0.1
Noxxa
They/Them
Moderator, Expert player (4124)
Joined: 8/14/2009
Posts: 4090
Location: The Netherlands
Now this is a treat. Always love seeing new TASes of this game, and it's great to see some of the new routes and optimizations (the Los Angeles secret tape route blew my mind). Very fun to watch, big yes vote.
http://www.youtube.com/Noxxa <dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects. <Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits <adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
CoolHandMike
He/Him
Editor, Judge, Experienced player (895)
Joined: 3/9/2019
Posts: 695
That was an easy yes vote for sure. This was my favorite game of the series and had some of my favorite levels like the airport and the cruise ship. You destroyed the game.
discord: CoolHandMike#0352
Dimon12321
He/Him
Editor, Reviewer, Experienced player (596)
Joined: 4/5/2014
Posts: 1222
Location: Romania
I've only seen my friends playing THPS series and never tried it myself. This TAS looks very dynamic. Yes vote BTW, the credits video doesn't sync with audio.
TASing is like making a film: only the best takes are shown in the final movie.
Rocker
Any
Player (108)
Joined: 6/30/2022
Posts: 10
User movie #638282478162665538 Updated due to a new Skater Island route. Here's a list of changes: Time saves -Swapped FS540 for The 900 (saves the scroll) -Cut out retries on Rio and Skater Island -Saved 1 igt per heat in Skater Island -Faster Tokyo heats Time losses -Bad judge rng in Rio heat 2 requires an extra trick (couldn't find a way to get 3/2 heats like in the original submission) -Added two retries to Tokyo because the rng was dreadful -Not a time loss, but not having fs540 definitely dinks the entertainment for me. The things we do for speed, Yippie Sub 8!
Banned User
Joined: 1/6/2023
Posts: 263
I really enjoyed these TASes.
Published TASes: #1, #2, #3, #4, #5, #6, #7, #8, #9, #10, #11, #12 Please consider voting for me as Rookie TASer Of 2023 - Voting is in December 2023 My rule is quality TASes over quantity TASes... unless I'm bored.
Post subject: Movie published
TASVideoAgent
They/Them
Moderator
Joined: 8/3/2004
Posts: 15576
Location: 127.0.0.1
This movie has been published. The posts before this message apply to the submission, and posts after this message apply to the published movie. ---- [5514] GC Tony Hawk's Pro Skater 3 "All Goals and Golds" by Rocker in 07:58.70
Noxxa
They/Them
Moderator, Expert player (4124)
Joined: 8/14/2009
Posts: 4090
Location: The Netherlands
Are there by any chance plans for a run that gets all stat points and decks with one character? Still holding out hope to see a run like that, lol - as great as these runs are at zooming through everything, there's still a lot of hidden content/secret areas in the game that go unvisited in all goals and golds (especially in the competition stages), so I think there still would be merit to such a run.
http://www.youtube.com/Noxxa <dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects. <Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits <adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
Rocker
Any
Player (108)
Joined: 6/30/2022
Posts: 10
Noxxa wrote:
Are there by any chance plans for a run that gets all stat points and decks with one character? Still holding out hope to see a run like that, lol - as great as these runs are at zooming through everything, there's still a lot of hidden content/secret areas in the game that go unvisited in all goals and golds (especially in the competition stages), so I think there still would be merit to such a run.
There aren't plans for that currently, though I'm personally not opposed to creating something like that. I agree that the competition levels are pretty much entirely unused and that it'd be an interesting avenue to explore. The main issue is it ends up in a sort of middle ground between All Goals and Golds and 100% which is a gray area. I reckon something like "100% one character" could work as a category, but I don't know how well that'd be received/accepted. Could maybe be considered more of an IL and then get obsoleted by a full 100% category (if that ever exists) but I'm unsure. I think that if that category/idea was ever going to come to fruition, it'd be during a full 100% (all characters) run. Only issue with that is that there isn't a gap list in game (or a definitive list anywhere) to define 100% so the category wasn't developed. I know that RTA runners ran a category called Neversoft Eyeball% which was essentially just AG&G for every skater, could just add the collectables to that and make it a pseudo 100% (due to the aforementioned gap tracking issues). I'm currently working on THUG1 100% and THPS4 Any%/100%,, but they're still so early on that I'm super down for more THPS3 if I had a reason to work on it.
Noxxa
They/Them
Moderator, Expert player (4124)
Joined: 8/14/2009
Posts: 4090
Location: The Netherlands
Rocker wrote:
Noxxa wrote:
Are there by any chance plans for a run that gets all stat points and decks with one character? Still holding out hope to see a run like that, lol - as great as these runs are at zooming through everything, there's still a lot of hidden content/secret areas in the game that go unvisited in all goals and golds (especially in the competition stages), so I think there still would be merit to such a run.
There aren't plans for that currently, though I'm personally not opposed to creating something like that. I agree that the competition levels are pretty much entirely unused and that it'd be an interesting avenue to explore. The main issue is it ends up in a sort of middle ground between All Goals and Golds and 100% which is a gray area. I reckon something like "100% one character" could work as a category, but I don't know how well that'd be received/accepted. Could maybe be considered more of an IL and then get obsoleted by a full 100% category (if that ever exists) but I'm unsure. I think that if that category/idea was ever going to come to fruition, it'd be during a full 100% (all characters) run. Only issue with that is that there isn't a gap list in game (or a definitive list anywhere) to define 100% so the category wasn't developed. I know that RTA runners ran a category called Neversoft Eyeball% which was essentially just AG&G for every skater, could just add the collectables to that and make it a pseudo 100% (due to the aforementioned gap tracking issues). I'm currently working on THUG1 100% and THPS4 Any%/100%,, but they're still so early on that I'm super down for more THPS3 if I had a reason to work on it.
I talked with Fog about 100% several years ago, although it was quickly dismissed as it basically requires beating the game AG&G 22 times, which is pretty excessive (and probably gets repetitive at some point). Therefore I think 100%ing one character would be a fair compromise. Six extra collectables per stage, including competition stages, should provide enough extra content and runtime to be interesting alongside any% and AG&G. I don't think gaps are tracked anywhere, so I don't think they'd be required for 100%. That sounds like hell to route anyway lol, if I remember correctly some levels have more than 100 of those.
http://www.youtube.com/Noxxa <dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects. <Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits <adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
Rocker
Any
Player (108)
Joined: 6/30/2022
Posts: 10
Noxxa wrote:
Rocker wrote:
Noxxa wrote:
Are there by any chance plans for a run that gets all stat points and decks with one character? Still holding out hope to see a run like that, lol - as great as these runs are at zooming through everything, there's still a lot of hidden content/secret areas in the game that go unvisited in all goals and golds (especially in the competition stages), so I think there still would be merit to such a run.
There aren't plans for that currently, though I'm personally not opposed to creating something like that. I agree that the competition levels are pretty much entirely unused and that it'd be an interesting avenue to explore. The main issue is it ends up in a sort of middle ground between All Goals and Golds and 100% which is a gray area. I reckon something like "100% one character" could work as a category, but I don't know how well that'd be received/accepted. Could maybe be considered more of an IL and then get obsoleted by a full 100% category (if that ever exists) but I'm unsure. I think that if that category/idea was ever going to come to fruition, it'd be during a full 100% (all characters) run. Only issue with that is that there isn't a gap list in game (or a definitive list anywhere) to define 100% so the category wasn't developed. I know that RTA runners ran a category called Neversoft Eyeball% which was essentially just AG&G for every skater, could just add the collectables to that and make it a pseudo 100% (due to the aforementioned gap tracking issues). I'm currently working on THUG1 100% and THPS4 Any%/100%,, but they're still so early on that I'm super down for more THPS3 if I had a reason to work on it.
I talked with Fog about 100% several years ago, although it was quickly dismissed as it basically requires beating the game AG&G 22 times, which is pretty excessive (and probably gets repetitive at some point). Therefore I think 100%ing one character would be a fair compromise. Six extra collectables per stage, including competition stages, should provide enough extra content and runtime to be interesting alongside any% and AG&G. I don't think gaps are tracked anywhere, so I don't think they'd be required for 100%. That sounds like hell to route anyway lol, if I remember correctly some levels have more than 100 of those.
Yeah there are a lot of gaps it would be a nightmare. I'd love to do it if there was a list (I like routing and problem solving more than actually doing the input for the most part), but it would be a mess. https://pastebin.com/tFppyCyp here's a list of the different goal sets for each skater. There's definitely variety in each skater selection but at the end of the day there's not that many options. I think a restriction for either 100% category (one character or all) that could be interesting is putting a restriction on trick and/or stat editing. I think the former would create a much more interesting dynamic. Like "oh no Tony Hawk doesn't really have a special grind/manual how can we figure out points?" or something along those lines. If you'd just want one skater, do you have a favorite?
Noxxa
They/Them
Moderator, Expert player (4124)
Joined: 8/14/2009
Posts: 4090
Location: The Netherlands
Restriction on trick editing could be interesting for an all character run - I imagine it would result in some extra variety for each character in move selection, alongside the SKATE/stat point/deck differences. Restriction on stat points feels unnecessary, if you're going to collect stat points anyway might as well go all out in statting up the skater as optimal/strong as possible (not that that's very hard, as you could already min-max stats pretty efficiently even without extra points). Personally for a one character 100% run I don't really care about which character it would be. Could be Tony Hawk, or Jamie Thomas or whoever would be fastest. I don't think anyone has mapped out what the fastest combination for SKATE/stat points/decks is, so I guess it's still up in the air. Regarding gaps, I found a gap list on IGN/GameFAQs, but it's impossible to tell if it's complete. Still, I don't think it's really worth going for, because they aren't tracked, and it's pretty much impossible to know if you might have missed any.
http://www.youtube.com/Noxxa <dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects. <Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits <adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
Rocker
Any
Player (108)
Joined: 6/30/2022
Posts: 10
I'm currently leaning towards doing 100% all characters w/default tricks instead of doing single character. Partially just because like you said the fastest selection isn't mapped out and optimized. There might've been some data for it 10 years ago, but nothing that would be relevant to the current standard of RTA or TAS. Who's to say I try the big players (Jamie, Bam, Andrew) then all the sudden Geoff or Rodney is better? I wouldn't want all the work/research to end up wasted. The main reason though is I feel that all character 100% would be a nice category to end it on. Essentially I just don't wanna leave anything left on the table with thps3. Any% and AG&G are maxed or exceptionally close to maxed, there won't ever be another character or goal reroute for those unless a major game breaking glitch is found. I feel like I'd finish one character and question about the others if I didn't do them. That said, it'd be no small feat to do the full thing. I'm still up in the air on whether or not I'd choose one character or do the full thing, but those are my general thoughts right now. Could also look into doing a few characters with the most unique layouts to meet somewhere in the middle.
Noxxa
They/Them
Moderator, Expert player (4124)
Joined: 8/14/2009
Posts: 4090
Location: The Netherlands
Whatever you decide to do in the end, looking forward to it :)
http://www.youtube.com/Noxxa <dwangoAC> This is a TAS (...). Not suitable for all audiences. May cause undesirable side-effects. May contain emulator abuse. Emulator may be abusive. This product contains glitches known to the state of California to cause egg defects. <Masterjun> I'm just a guy arranging bits in a sequence which could potentially amuse other people looking at these bits <adelikat> In Oregon Trail, I sacrificed my own family to save time. In Star trek, I killed helpless comrades in escape pods to save time. Here, I kill my allies to save time. I think I need help.
Rocker
Any
Player (108)
Joined: 6/30/2022
Posts: 10
Link to video Of course there's something new found right after everything was published :) In Any% this loses 30-40 frames adding 900, loses 1 igt/heat in Rio (worse 3rd place rng), saves 1 igt/heat in Tokyo. Might be frames faster rta depending on what a new Rio would look like (personally don't feel like resyncing everything and resubmitting again right away for a few frames rta and no igt benefit) In AG&G this would save 2 igt/heat in Tokyo and loses no extra time because we already have the 900 for S.I. It's essentially the same score but faster (retries are still required for better rng). Will try to update this soon enough but I don't wanna throw the same thps3 category back in submissions again instantly :p