I guess it won't ever solve the equation of what the fastest route is, but to even be able to predict where each missile is a what time would be useful + I think that's something that's going to be useful in lots of levels. So if someone has time, it wouldn't go to waste to find the missile speed and create a basis for that sort of calculation.
It takes 632 frames for a missile to go from left to right and back again, and 232 frames for the missiles in the upper areas.
So one complete cycle is 18328 frames (~4min).
New level 5!
Link to video
I'm pretty damn happy with this one, as it saves 2.6 s off the previous video. There's a lot less waiting time and very little backtracking. I'm gonna call this one done unless someone sees something significant I can do to improve this.
EDIT: Video is not working right now, taking way to long to process... FINALLY working now.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Looks pretty good.
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Edit:
We're all doing this build your own staircase technique in the lower right corner. Is that really possible?
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Edit:
We're all doing this build your own staircase technique in the lower right corner. Is that really possible?
Good route, even if it doesn't pan out I may incorporate the ending as an alternate to the end of my 1st route. The "staircase" technique does work, I was able to do it in real time.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Not that I like either of these, but here's non-build-your-own-stair-case routes:
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
linear_transform = function(x, a, b)
return (math.floor(a / 65536) * (x % 65536) * 65536 + (a % 65536) * x + b) % 4294967296;
end
rng_state_to_index = function(state)
local A = 0x2925141D;
local B = 0xD6DAEBE3;
local C = 1;
local D = 2;
local i;
local j = 0;
for i=0,31 do
if (state % D) ~= 0 then
state = linear_transform(state, A, B);
j = j + C;
end
B = linear_transform(A, B, B);
A = linear_transform(A, A, 0);
C = C * 2;
D = D * 2;
end
return j;
end
So Slamo and co, it occurred to me that isn't ladder climbing faster when jumping up the ladder? It only became relevant with level 05. Was that taken into account?
Further about moving on foot: Could a table showing what jumps are make-able and what aren't be useful for fast reference in planning routes in levels where you go without fuel? I started listing them in the tricks section, but my information is based on non-TAS trials so the list should be updated by using frame perfect jumps. I don't think things like moss or conveyors affect how far you can jump, but you seem to fall off ladders a bit sooner so those would be different unfortunately.
Added a todo about including a little more info on getting the game (and movie files) to run.
For those interested in .png-compression - for what I could tell, using ImageMagick's 8-bit convert didn't really save any more space in the end compared to a file that went through otherwise the same compression process (or maybe 2 bytes, don't remember). I used (as per instructions page on the site) optipng, PNGOUTwin and DeflOpt. Also, using ScreenshotCompress on top of those (with its numerous trials per image) seemed to be saving only a few extra bytes at most.
Finally, I have a list of image links (imgur) to ready-to-use pictures. Next up I'm going to see (within a couple of days) what kind of neat table can be set up on the RP.
So Slamo and co, it occurred to me that isn't ladder climbing faster when jumping up the ladder? It only became relevant with level 05. Was that taken into account?
Further about moving on foot: Could a table showing what jumps are make-able and what aren't be useful for fast reference in planning routes in levels where you go without fuel? I started listing them in the tricks section, but my information is based on non-TAS trials so the list should be updated by using frame perfect jumps. I don't think things like moss or conveyors affect how far you can jump, but you seem to fall off ladders a bit sooner so those would be different unfortunately.
Jumping up ladders is indeed faster, it's crucial for hitting certain cycles in level 5. For longer ladders I can just come up with a cycle of inputs to optimize the speed, I will definitely figure that out for level 6.
Interesting point on the jump possibilities, I will have to check that out. e.g. I just barely miss a 3-block jump in real time but it might be doable in a TAS.
Level 6 WIP:
Link to video
This is my route #1 with Nach's alternate ending (faster by ~10 frames) plus some minor optimizations. I'll try another couple routes before moving on.
It seems to me that a lot of sounds are missing. Shouldn't there be a sound whenever you collect an emerald?
Yes the sounds are glitched, I forgot to mention that. For a full video the sound will be normal, Ilari knows the technical explanation better but the sound is weird because I started dumping after the game started.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
Thinking about it, would it be faster to ignore the second gem on the way up, and instead nab it on the way down from the ladder directly to the right of it?
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Thinking about it, would it be faster to ignore the second gem on the way up, and instead nab it on the way down from the ladder directly to the right of it?
You're right, in fact that part turns out to be even faster since you're falling down a portion of a ladder instead of climbing it! I keep forgetting about this, have to minimize climbing when possible...
Maybe you'll figure out also how long a ladder has to be to make it worthwhile to jump before going down (i mean even if the jump is not stopped by a block above). Also remember, if there's room you can just fall off and keep pressing alt and I think that prevents you from sticking to ladders.
So I noticed the bigger png's and the editor preview png's that we have on the levels subpage have a different frame happening. Unfortunately, this messes up how the autoladders look. I recommend looking at the editor picture and seeing which end looks "closed". That's the direction the ladder is going. For conveyors look at the stripes, they're tilted the way it's moving.
So I'm off tomorrow for 2½ weeks with little editing possibilities. Will finish the tables after that, but here's a link to a .txt file containing the image links anyway if someone wants to speed it up.
Level 6, route #2 (glitched sound):
Link to video
It will never cease to amaze me how totally different routes can result in such close times. This is 22 frames faster than the first WIP I posted, and if I factor in Nach's idea about the 2nd gem this is still faster than the optimum route #1 by a whopping 3 frames.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
When going down the middle of the screen from the very top, it's really faster to take the ladder than busting the floor out from under you?
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
When going down the middle of the screen from the very top, it's really faster to take the ladder than busting the floor out from under you?
Yes, this is one of the areas where I tested a few things. Using the ladder was faster by a few frames (probably saved by the fact you can just fall off halfway.)
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
And with the leftmost ladder, going up it instead of busting the block as in my route images is also faster?
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
Joined: 3/9/2004
Posts: 4588
Location: In his lab studying psychology to find new ways to torture TASers and forumers
When going down to grab the lowest bottom right gem, is it faster to build a staircase as you did, or phase out the block on your way down, then return with a single jump instead of two?
Warning: Opinions expressed by Nach or others in this post do not necessarily reflect the views, opinions, or position of Nach himself on the matter(s) being discussed therein.
I just tried Nach's first route and it was slightly slower, and we've discussed a ton of optimizations so I think I'll move on. Level 7 idea:
I don't think there are too many possibilities with this one. The blue switch is required, the red one isn't but I think it's undoubtedly faster to use it.
Route 2:
I will try both of these for sure, I have no preference between either of them.
Edit:
We're all doing this build your own staircase technique in the lower right corner. Is that really possible?
I don't think there are too many possibilities with this one. The blue switch is required, the red one isn't but I think it's undoubtedly faster to use it.
