Hey Spezzafer, are you planning to continue your 100% run? In case you needed the ability to search for RAM addresses in the embedded chips, Snes9x 1.51 offers that functionality, though it means you'd have to start over or something. :\
it's not very likely at this point, it's way too demanding
although if I do there's that separate memory watcher app that I could use with snes9x 1.43 that can display them (I've tested it and it works fine)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
With mentions of improvements available from a restart, it would probably be a pertinent time to point out that there's corner boosts and going through walls present in this game... now only to poke Arne enough to get him to post about them in here....
*poke poke*.
there are so many variables that affect randomness in this game that hex editing is completely out of the question, the game might not desync for the rest of the level and maybe even the next but it is guaranteed that it will desync soon enough
for reference here is a doc with a whole bunch of ram addresses i put in that could be used with a tas (note the speed, x/y position, and boss damage/health addresses):
http://spezzafer.home.att.net/other/yi_data.txt
a list of tips: http://spezzafer.home.att.net/other/yi_tips.txt
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Well, would this be a cool start to an improved any% run?
1-1 / youtube / AVI
23 frames saved.
I lost maybe 2-3 frames from increasing entertainment, i.e. picking up and throwing more eggs; this is because my speed decreases slightly and I lose a pixel or two each time I do it. If it were to go for pure 100% speed, you'd get pixel perfection but very easily lose out on entertainment. Pixels lost to gain a lot of entertainment seems a good investment to me.
Frames were saved from menu optimizations at the beginning (3 frames) using the acceleration trick 6 times instead of 3 (2 frames) shooting the flower with the egg as I'm about to float off a shy guy (6ish frames, let me keep maximum speed) using acceleration when floating (10+ frames) and landing at the very bottom of the circle at the end makes Baby Mario touch the next Yoshi sooner (5 frames).
So yeah, anyone that believes 100% will be superior to any% entertainment wise, will hopefully see how much can be added even in a short level like 1-1. Having eggs and coins to collect simply increases the number of eggs you can use for zany stuff, but any% is not short on that. All you have to do is collect more eggs and bounce them off tons of stuff, like the 100% does and you should be good. I got a gazillion other ideas for yoshi craziness, both shared by 100% and any%.
So yeah, hope this looks cool.
That looks great comicalflop.
A note on Yoshi's speed: looking at the speed using s9xwatch you'll notice that even while running on a level ground at full speed, the top speed doesn't remain constant. In fact you can actually acheive a slightly higher top speed range depending on the number of times you do the acceleration trick (sometimes using it 7 times, maybe waiting 1 or more frames after the 6th will give you a higher top speed) and you might save several frames depending on how long you can keep it.
Why not do a 100% run though? You don't seem averse to major time investments, and spezzafer's incomplete 100% TAS has room for improvement too. I don't know if anyone else feels this way but I would be disappointed if the next YI run isn't a 100% run, after seeing how good it looks and much effort has already gone into it.
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Well, I tried 100% a long time ago. It was too much work for spezzafer, it most certainly will be too much work for me, and likely will be too much work for other people. Plan what eggs to make, throw, and keep, how to get the flowers/coins, routes, etc. and it's just an insanely massive task. Any% is a much older run, and is in more need of increased precision, new time-saving tricks, and a lot of increased entertainment. I think that, in the end, any% is easier to do and can achieve 95% the same results, especially if the levels are played out like what I did and better. I myself would love to see a 100% done, but eventually; I can more easily see an any% improvement come to pass.
Joined: 4/30/2006
Posts: 480
Location: the secret cow level
I really hope someone eventually does a 100% run, the one at SDA was so good you would swear it was tool-assisted, so it would be really damn cool to see how much room for improvement there would be.
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
To be honest (and to be fair), I've seen the 100% SDA individual levels, and it does not look even close to tool assisted. Impressive for human play to be sure, but nowhere close to TAS quality as spezz's WIP beats the pants off it time-wise and entertainment wise.
And as I'm trying to demonstrate, 100% is a shitload of work; any% is easier to accomplish and can have the same results.
Sure, it's insanely massive if you do it all in one go. But that's like saying "We don't know the best route yet for a Super Mario 64 run, let's not make it yet". If we all had that attitude, we would have a completion time of 16 minutes for that run. Improvements can come later.
I for one would love to see a 100% run. I've played through the game several times, and have gotten all 100%'s only in the first few worlds. It'd be cool to see some of the later levels done.
I'm pretty sure, though, that it's necessary to grab some (forgive me here, my memory's not quite clear here) star items to boost your points in many of the levels. Not quite sure how a TAS would elegantly work around that.
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Let's assess this...
We do know the best route for Super Mario 64. We know how to get 16 stars. We're very close to seeing a sub 15 minute submission.
SM64 16 star runs are easier to make, time wise, compared to Yoshi's Island. So SM64 was a very bad example to compare with.
YI is a massive undertaking of time. It is very tricky, you are expected to make it flashy and entertaining, and also hard pressed to keep speed at a high when navigating levels.
Also, as Spezzafer has pointed out, the egg management is very tricky for 100%. Figuring out which eggs you are going to make/use, where you are going to get your 30 stars, as well as which stars/flowers to hit with eggs, which to hit with enemies, which to grab manually, etc.
Here's how I see it:
100% has a 50% complete, 99% optimal WIP. It is insanely hard to do, to the point where our conventional YI runner fizzled. It is in theory the most entertaining, on the basis that it has the most opportunities to use dazzling egg throwing in the collection of stars, coins, and flowers.
Any% has a 100% complete, 75% optimal run. Many new tricks/precision techniques that have been found: skipping areas in 1-2 (which may work elsewhere) using different acceleration to gain a faster speed, using acceleration when flying, removal of mistakes (like passing through too many cloaked walls), using the spinning ball better for vertical ascension, and most of all: increased standards in entertainment because of 100%, which any% can attain. 100% only has more opportunities to showcase entertaining Yoshi play, not better.
Furthermore:
100%, despite having SDA runs to compare to, still has a lot of exploring, planning, and route making for the remaining levels, from 4-3 onwards. The 'hard stuff" of 100% is not just doing the motions, but in planning everything out. Any% already has everything already mapped out: improve the precision, improve routes when needed with new shortcuts, increase entertainment; the hard works is already done for the entire run.
Do I want to do a run of YI? Yes, I love this game. I probably can't at the moment, but I'd love to complete any%, and maybe eventually 100%. We shall see. I did 1-1 any% to show that the current run can be improved with precision and does not have to be "lesser" than 100%.
Oh, and the extra/special levels and such confuses me a little. I don't know too much on what their place is in the run, like when they're done and such.
The "special" levels are simply unlocked. The first for each level after killing Bowser. The second secret levels are unlocked upon scoring 100% in all 9 other levels (heh heh).
Comicalflop, before you progress further, I URGE you to use Snes9x 1.51 for this run. Not only you gain the access to the FX addresses, but also much greater graphics.
<common knowledge mode>
I don't know about the SMW:YI remake for GBA, but the Special levels in the SNES version are one additional level added to each world, as well as the one of the available bonus games unlocked for direct access. (As opposed to appearing (randomly?) after obtaining a flower at the finish roulette.)
Beating Bowser in itself has nothing to do with unlocking them. It's all about achieving a 100 pts. score for all the 8 already available levels in that world.
After achieving 100 pts. for a special world, you are awarded a star at the start-up screen. There are no other game-related benefits to accomplishing it.
</common knowledge mode>
Andreas
Yes, obviously.
I am interested in doing a 100% run of this game, but I wouldn't start this before all of the movies I'm currently working on are completed. (And I also think Comicalflop should finish the 15 projects he is currently working on before adding another one.)
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Spezzafer already found the addresses for YI, as shown in a .txt. I tested them myself and they work perfectly, so there is no need for that. I also dislike 1.51; I do not notice any improvement in graphics nor sound, the occasional lowered frame rate my snes9x 1.43 experiences is not fixed by 1.51, and 1.51 has experienced desynchs. Honestly, there is no reason for me to use it. Don't fix what isn't broken.
I wouldn't start until I have finished or have less projects as well. (and it's 4, not 15....)
I had a thought for the any%. When fighting a boss, you often reach a point where the boss can't be hurt, but you can be, right?
So, during one of these points, take damage with timing such that Baby Mario has 0 stars left when the boss dies. That way, the score screen goes faster. The entertainment comes not from the extra speed, but from knowing attention was paid to such a fine detail.