Probably not, even if you could start with every upgrade. Just the fact that one player is always going to be ahead at the start means that another is going to be behind. And even though your position placement is dependent on how well you did, the only way you could exploit that would be to wait for CPUs to pass you up. You can't bump CPUs to move ahead of you, as getting bumped by CPUs tanks your maximum top speed, which in turn ruins any nitro routing you're attempting to do. The only way 2-player could be just as fast as 1-player is if both players cross the finish line when the game graphically updates positions, which is dependent on how much lag there is when it happens. As far as I know from TASing the game, it's like frame rules in SMB, but more volatile.
Spikestuff wrote:
Again I'm not a person who does racing games, so I'm not going to go over Auckland and do all 5 laps to see if more can be skinned off but it is something worth noting that the TAS isn't using racing lines efficiently.
feos wrote:
What I think about turning in this game is that it may be generally losing time (notice the sound), so one might need to travel in straight lines that are chosen carefully. But this hasn't been tested.
From my experience, racing lines aren't a thing in this game. Any noticable difference between turning, I think, might be due in part by how much lag you might have received when the game visually updates. This games feels just like Outrun or Slipstream, in that it's 2D made to look 3D.
feos wrote:
I tried improving the last level and couldn't. Sure there are some frames lost on starting up: when you initially hit X, A, when you shift gears. But the bigger problem with this game is nitro management on slopes, and it's really hard to get right. And it loses tons of time too. I saved 50 frames on the first lap and lost almost 200 at the finish. I even admit that improving the first level wasn't exactly easy, for the same reason.
...I couldn't have said it better myself. Nitro management in this game is tricky business, especially when you're trying to figure out how to stay above that mark where you barely decelerate for as long as you can.
Washed up Top Gear 2 player, here. Message me on Discord if you have any questions related to speedrunning or TASing the game.
Discord: Rally#0001.
From the first SNES Top Gear game I know that managing the gears is extremely important: there you can accelerate way faster while being at the 4th gear. Also, the frame you switch the gear at greatly influences the distance and speed you reach, because it would accelerate you differently depending on some tricky internal mechanics. Does gear switching have the same effects here? If yes, was it as carefully managed?
While I don't speedrun the first Top Gear game, the second game has a very big emphasis on staying in lower gears to gain most of your acceleration (namely, 1st gear). You don't accelerate as quickly in higher gears to get to your top speed as you would in the lower gears.
feos wrote:
From the first game I also noticed that if you reach a sharp turn with too high speed, for instance from having used nitro, you either get thrown off-road, or you have to brake not too lose too much speed to this. Isn't spending nitro only on (relatively) straight track segments faster?
In this game, unless you're on a snow stage that incorporates some sort of hydroplaning mechanic, turning while using a nitro doesn't really hinder your handling, especially if you've accounted for it beforehand.
feos wrote:
Was all the menuing optimized to death? I see tons of mashing there, and also that sometimes the game would just not react at certain buttons. If this run is to be redone and made more optimal, leaving any sloppiness in the menus would be disappointing.
It wasn't.
feos wrote:
How much time would wall glitch save overall?
A lot. I mention the entertainment value of this glitch in my last post, along with what might be the most entertaining way to perform it - while we don't know the true limitations of this glitch (e.g. where we can perform it, how fast or slow we need to go to perform it, how to clear an entire track with only one wall glitch, etc.), we know that it can be done, and that it can finish a track very quickly.
feos wrote:
Do all the cars have identical parameters that don't affect time?
I'm not sure if this is a car paint question, or if it is a AI question. Regarding the former, paint doesn't matter, and it's changed as a meme. Regarding the latter, AI is a very peculiar thing; from what makes sense in my mind, each AI will initially go as fast as a permitted, programmed constant top speed will allow it to go. If it is bumped from behind, and the AI increases it speed as a cause of it, it will stay at this new constant speed, without dropping any speed (unless it is stopped by getting bumped in front from either another AI or the player). I'd like to make a note here that this is speculation from the 15 years I've been casually playing this game, and not actual information found from reading code from the game itself. As much as I'd love to say I currently know the most about how this game operates, I have no knowledge of exploring the game's code.
feos wrote:
Is it correct that once you've entered the track using the password glitch, you get money just as a reward for quitting?
You get money from the Continue Country menu (at the sum of $10,000 per Country being skipped from Australasia). Using the password glitch is a method we use to force the game to give us a password, instead of having to put in a pre-made password with extra stuff. What's great about this is that we don't have to worry about people adding extra money to get better upgrades - we can just use arbitrary limitations set by the game itself to complete the game as fast as we can (and the limitations are good enough to where we can abuse some of the games flaws, more so than Dooty's TAS does).
feos wrote:
Can you guys upload TheRallyFTW's movie file to our userfiles so we could do proper comparison? If this run loses to some existing records, or is overall not optimal enough (for example, if some of us can easily and significantly improve it), it will have to be rejected.
...nah, you good fam.
Seriously, though, I really only made the TAS as a proof of concept for speedrunning this game - while I wanted the quality of it to be as good as something that could be submitted to TASvideos, I didn't want to submit it here, because I knew there was still some room of improvement after finishing it.
But more than this, the file is on a different computer, and on an outdated version of Bizhawk. If I were to recreate this TAS (which I barely have any interest in doing at the moment, never mind not having the time to do it at a reasonable pace), it'd already be a submission here. Shoot, maybe I've should - it'd be a waste not to, anyways.
If you still want it in its current form, I could happily provide it for you, but I wouldn't want it to be an actual submission. Let me know if you'd like it.
Washed up Top Gear 2 player, here. Message me on Discord if you have any questions related to speedrunning or TASing the game.
Discord: Rally#0001.
Hello, washed up Top Gear 2 speedrunner here. I have a lot of quoting to do, so let's see if I can't destroy the character limit in one post.
Challenger wrote:
I haven't posted before (on the topic of this game), but if you can soft-reset (starting from frame 33 - power-on), you can skip the copyright screen and entering title screen earlier.
This isn't wrong. On the topic of soft-resetting, you can also soft-reset after you've inputted the first password, and then just Continue Country to the United States once you're back into the main menu - instead of having to load the track and quit to the main menu from there.
Spikestuff wrote:
And then there's the issue of not using all your fuel.
From what I've been told, rimsky82 on GameHacking.org built a password generator for this game years ago. I asked him what he knew about this game, and he told me that he also found that Fuel Tank upgrades were checked in the password generator. His guess was that Fuel Tanks were originally coded to be in this game as something else to upgrade, but due to a constraint somewhere, the feature was turned off before release. While fuel is important to take note of in the first Top Gear, it doesn't matter in the second one. This explains some of the lengthier tracks in this game (London and Sheffield quickly come to mind based on the sheer brutality of those maps at the start of the game).
xxNKxx wrote:
I see sometimes you have run out of road, of course maybe it was speed tested. But at New York and San Francisco start race, I think I see have some space can help you run to midle road. I'm not sure, maybe you was tested it very carefully.
This point has already been discussed, but I'm going to use this quote to mention that the AI is very shifty. In New York and San Francisco, you can't pass the car in front of you (the 16th place car) on the right and expect to instantly nitro. With San Francisco (the last track), we can bypass this by going off the left side of the road. As far as speed differences go, they're negligible when using a nitro with low speeds. New York, on the other hand, doesn't have this luxury. Moreover, that car will sometimes not let you pass on the left either, causing a reset. If there's a way to consistently pass that car by tricking the AI at the start, I haven't found it.
Spikestuff wrote:
Now you have him beat on the later levels except the first one at Las Vegas. He pulls in a 1'51"06 while you come in behind with a 1'51"11.
Do we have an explanation for this?
My money is on his last nitro. He can get over that last hill with 228 MPH+ speed if the last nitro is placed properly. This should have been fixed in the TAS. I don't think this was actually discussed in the submission - there's a weird speed property that happens at 228 MPH/276 KmH where the rate of deceleration decreases dramatically at that speed. Nitro placement is kept with this in mind. For reference, check the 3:34 mark in the YouTube encode - after his nitro ends, he keeps above 228MPH for just over 10 seconds.
To my understanding of what makes an acceptable run on TASVideos (at least for a Vault acceptance), I don't think this TAS is optimized enough for it to be placed there. That's not to say a new Top Gear 2 TAS won't ever be approved - there are a lot of interesting mechanics that aren't explored in Dooty's TAS, simply because he doesn't reach a speed where they can be shown. There's also my TAS, which Technickle does improve on during the actual racing. There's a wall glitch that can occur in Los Angeles - it'd be hilarious to see a runner drive backwards for 20 seconds, just to see him clip the left wall with enough speed and instantly finish the stage. This would definitely take a lot of testing to get right, since this wall glitch is extremely inconsistent to get.
If you guys have any questions, feel free to ask - I'll be paying attention to this post over the coming hours.
Washed up Top Gear 2 player, here. Message me on Discord if you have any questions related to speedrunning or TASing the game.
Discord: Rally#0001.
