By the way, if your TAS doesn't break all existing records (both RTA and TAS) in its category, don't submit it. Just upload to userfiles the input file (necessary for other TASers to easily critique your work) (make a video too if you want it to be easier for people to watch, though!) and post it in the respective game thread on the forum if you're looking for feedback.
EDIT: Added... advice from Spikestuff.
Great TAS! Loved the storyboarding and Spongebob sequence.
Agree with the comparison to Demoscene, I think as long as a TASer has the drive to push technical limitations further than before, or to create a new visual audio experience, there will be new and greater ACE TASes.
Can we agree on 'item glitch, gate clip' being the category name? A little clunky but describes all the major skips in use, and means that the branchless category foregoes item glitch and gate clip (e.g. it would be your 'no major skips' TAS).
Also, the submission text indicates that an extra ring can be collected in the main stages when playing normal Sonic the Hedgehog 2. (It would also be greatly beneficial in the special stages due to split ring paths.) I'm not sure how if would affect the entertainment value, but it terms of pure rings count, picking the Knuckles version seems sub-optimal and arbitrary to me.
Yes, Knuckles misses 1 ring, but he more than makes up for it with other rings only he can get:
'Casino Night 2
This is the second zone where appear the second group of unreachable rings, 14 in sky at the begining 2 groups of 7, only Knuckles can grab these rings. '
So in the end it's a wash - each character's Ring Attack collects a different subset of all rings in the stages.
I remember that, for one game, going for the shortest input file lead to a really boring end of movie where it takes ages to reach credits. So the audience asked for a quicker ingame time instead.
Was bored and came up with a math challenge for myself:
To roll a y sided die is to randomly select a value from 1 to y inclusive. As a probability distribution, has an expected value of (0.5+y/2), a variance of (y^2 - 1)/12 and a skewness of 0.
Normally dice have a positive, integer number of sides, but I would like to extend this function to all real values of y >= 1, and the expected value for non-integer values of y should still be (0.5+y/2). For example, if I roll a 2.5 sided die, I should be able to roll either 1, 2 or 3 with non-equal weightings, and the expected value should be (0.5+2.5/2). Additionally, this function should be indistinguishable from normal die rolling whenever y is an integer.
1) What is the algorithm for rolling a real sided die that satisfies these criteria? With what probability does it roll each side?
1.5) Extend also to negative integer y, then to negative real y.
2) Does this algorithm still have a variance of (y^2 - 1)/12 for all values of y? If not, can you come up with a second algorithm that does NOT have expected value of (0.5+y/2) all the time but DOES have variance of (y^2 - 1)/12 all the time? How about skew?
EDIT: Fixed variance formula. Also, here's my answer for 1:
xDy, where x is integer and y real:
Values from 1 to floor(y) have 1.0 weighting, value floor(y)+1 has floor(y)*frac(y)/(floor(y)-frac(y)+1) weighting. select randomly via weighting for each roll. Examples:
1.5 (should have EV 1.25) is a 1.0 weighting for 1, 1/3 weighting for 2. e.g. 3/4 chance to roll 1 1/4 chance to roll 2. EV is 1.25.
1.25 (should have EV 1.125) is a 1.0 weighting for 1, 1/7 weighting for 2, e.g. 7/8 chance to roll 1 1/8 chance to roll 2. EV is 1.125.
2.5 (should have EV 1.75) is a 1.0 weighting for 1, 1.0 weighting for 2, 2/5 weighting for 3, e.g. 5/12 chance to roll 1 5/12 chance to roll 2 2/12 chance to roll 3. EV is 1.75.
2.25 (should have EV 1.625) is a 1.0 weighting for 1, 1.0 weighting for 2, 2/11 weighting for 3 e.g. 11/24 chance to roll 1 11/24 chance to roll 2 2/24 chance to roll 3. EV is 1.625.
3.5 (should have EV 2.25) is a 1.0 weighting for 1/2/3, 3/7 weighting for 4 e.g. 7/24 chance to roll 1/2/3 3/24 chance to roll 4. EV is 2.25.
The convention for TAS timing is that it starts when the console powers on and ends when you stop giving input to the game (and the game must be completed with no additional input necessary, of course). RTA timing conventions are different - they go from when you gain control to when you lose control, because RTA speedruns have no input file to determine 'when the player stopped entering inputs'.
AlphaGO is super cool and a massive achievement, don't get me wrong. But it is still categorically different from an AI that can learn in a generic sense (rather than being tailor made for a certain game or simulation), especially one with emotions or biases.
Something I dont understand. Why at marks close of 30 seconds are optimized to less than 30 seconds? It is faster to add a score of 10,000 than one of 50,000.
Mysctic Cave Zone 2 is a case.
It is considered less interesting to optimize for real time than for in-game time, because once you are optimizing for real time, there is no point in trying to make stages faster than 30 seconds but slower than 23 seconds faster for in-game time (as it is slower real-time than just waiting until 30 seconds is hit then taking the sign). As a result, all speedruns and TASes of Genesis Sonic games are timed by in-game time, not real time.
The submission notes go into further detail, ctrl+f for 'Real Time Savings'.
Typo: The submission is a 3 frame improvement over the published run (5.9 seconds in-game time). -> 354 frames
The submission text was very nice to read! I'll have to watch this TAS later.
If it's essentially random whether VBlank starts on or off, and other such things related to startup state, should a NESHawk TAS be able to specify the starting states of such things (similar to how you can specify the starting state of RAM)?
and since I want to make sure I don't end up violating any rules this time, I'd like to know if it suffices just to create a verification movie that gives the desired SRAM?
Yes, and you have to make sure your submitted TAS uses that SRAM, that the verification movie doesn't desync, etc.
The verification movie doesn't have to be fast, though. It can be as unoptimized as you like.
http://tasvideos.org/Movies/GruefoodDelight.html
I think only editors can put TASes here. It would go under 'Technical Foul' for being completed on an unaccepted emulator. (Unrelatedly, I think this section is getting big and can be split. Maybe into 'emulation problems' vs 'does something cheaty'.)
I have an idea. Is there a save state, SRAM or simple cheat code that could be used to see all 13 worlds contiguously in one FDS warpless TAS?
It would be an arbitrary category, but it would be entertaining enough for Moons. There's basis for such TASes being acceptable, see [2744] Genesis Sonic 3 & Knuckles "newgame+" by marzojr & Aglar in 27:31.29.
I had the same idea, and I've already found the responsable SRAM address to enter World A, but it's still too arbitrary goal for a warpless run. An FDS warpless run till World D is probably not a good idea because Mario has to game over at 9-1, return to the title screen, and use 8 stars from a save state or whatever to enter A-1.
I have an idea and I'm working on it, but I'm leave you in suspense till I'm done. :)
I think as long as every level is completed contiguously besides the game over, it would be good enough (basically, it would be the least arbitrary way to achieve the goal of a warpless all levels category).
But, I await to see what your idea is. :D
Those are decent TASes but won't be approved because people perfer the SNES version that shows World A-D.
In my opinion, an FDS warpless run (till 9-4) can show several special endings (skipping castles) and some interesting routes (going through upside-down pipes).
I have an idea. Is there a save state, SRAM or simple cheat code that could be used to see all 13 worlds contiguously in one FDS warpless TAS?
It would be an arbitrary category, but it would be entertaining enough for Moons. There's basis for such TASes being acceptable, see [2744] Genesis Sonic 3 & Knuckles "newgame+" by marzojr & Aglar in 27:31.29.
