The reason this level is taking so long, is that for the first camera I have to be a certain distance to it and no further away, or the shot goes right through the camera. Finding this radius is taking awhile.
I found the most ideal place to shoot it, but it's outside of that radius and the bullet goes right through where the camera should be.
Frigate is done.Youtube link for the lazy.
There is a long-ish wait time at the end of the level, some very key manipulation takes place here. While setting the hostages initial path is easy enough, at this point in the level I'm near the escape points of three of them, and if they see me while escaping they don't. This is why I hang out in the Bridge for some time. I opened two doors here so hostages wouldn't have to, this saved a little time. I had to wait to open the second door, otherwise the first hostage to escape sees me, and doesn't.
The last pause is slow to get a guard to do his rolling animation, so the hostage I just freed on the bridge doesn't get stuck.
I had originally wanted to shoot a picture or message into the side of the ship during the final wait period, but this was not possible as I had to look at a very specific angle to get the final hostage to escape when he was supposed to.
For the record, the speed of this run before the waiting/manipulation is good enough for a 0:56 fail.
Feedback is welcome, but unless I missed something really obvious I don't think I'll try to improve this.
Any ideas for the next level before I get to work on it?
I've figured out how the hostages work.
It's basically as Henrik says, at some point the game decides the path the hostage will take, and at certain points along the path he can escape. This is decided by the frame on which I kill the guard holding the hostage... well, hostage. However, my actions in the level can still influence guard behaviour and cause them to get in hostages way/accidentally explode hostages/or I'm seen by the hostage at their escape point and they don't escape.
This is still gonna take some time, likely, but at least I'm not running around in the dark anymore.
awesome silo, but you could have had the second door open if you had waited the intros.
Ah, but if you remember correctly in my 1:18 vid I got there too soon and had to open the door myself anyway, and lose time on the warp from killing the guard in my way behind the door. I did try cutting the cinema at various times to try and get him to be opening the door from the other side, but cutting to late just warped him in the room, and cutting to soon had him too far away. These two events occur a frame apart from each other, so I'm pretty confident what I did for that door is optimal.
Simply incredible! You are totally destroying this game RT :)
Ok so Frigate next, that's gonna be pretty damn hard to manipulate. I'm not sure if you read my post over at The-Elite but i found out how the hostages work.
When they are released they randomly choose between a number of predetermined coordinates and run to these. When they arrive they check if bond is in firing range(Means if the hostages can see bond, This is true even if bond/hostage is facing opposite direction, However, if bond is behind a wall or a fence and looks at the hostage, he'll still escape.
So for luck manipulation, you gotta make sure that the hostage runs to the closest preset(Predetermined Coordinate). And also that you don't see him after that which shouldn't really be a problem. I'm not sure exactly when the decision of preset is being made but i think it's right before Hostage Released! message comes up.
This is going to help. A lot.
*sacrifices virgins*
Check out this thread for a WIP I made of the NES Mario themed Golf game. It plays two courses and gets a perfect score on them both. Since the amount of shots seems predictable per hole (1 for short ones, 2 for longer ones), I tried to mix things up with some trick shots, near-misses and lots of unusual ball movement. Maybe, it can give some ideas to anyone who wants to try a TAS of this game.
I voted yes, but I don't like to see autofire in the input area. Some of the jumps pre-double jump item _look_ unoptimized, but I don't remember how well jumping controls in this game so it may just be me. This can certainly be improved, but it was a fun watch nonetheless.
I beat the turbo mode time... XD
Also, I don't think I told you guys why I haven't been working on this as much. It's because I'm moving at the start of June and have a piano audition at the end of June for University. So, I'm kind of really busy these days, and only found time yesterday because my piano lesson was cancelled. I ended up TASing almost all day and now I'm behind on my packing... whoops!
So, updates will be very sporadic until the start of July. I imagine I'll get most, if not all of Silo done by then if I can squeeze fifteen or so minutes of TASing into everyday. Who knows,,, we'll see.
Holding left right and up while a fairy zips you to the top of the screen fairly quickly, but you still need some frames of only left and right near the top to get over the barrier and glitch fall to the room below. This might save a little time if the fairy route becomes feasable.
Here's my test movie for the fairy-only route. I'm not quite ready to give up on this route yet, but I'm out of ideas for now. Palace 7 in this wip is really unoptomized, so don't let the overall time fool you. If we can find a way through palace 7, then this would be the ideal roure, as far as I can tell. I would appreciate it if someone wanted to test some stuff too ;) (todo: don't get first jar I get: I got it to test another idea that didn't work out. Use the north or south exit for the jar that I need to get for the small timesaver)
here's the route through palace 2 with the healer glitch I think. I forgot to watch it before I put it on my thumb drive to make sure :(
I can confirm that is the palace two fcm. It's definately faster with the healer glitch active.
As for the other one, if you tidy it up it should come out to a final time near the current glitch run considering how un-optimzed palace 7 is in this. Also, since you only need five magic containers to cast fairy, would you really need to get the one south of Zelda's palace? (just get the one next to the hammer cave, you need to warp there anyway it seems). Or if you do need it, activate fairy for the second one to forgo the "Hey, I got a thingy" animation of Link's.
I have more ideas?
When you go back to the Hammer cave to get the magic container and warp, I think it's faster to just walk through the graveyard and go over the long bridge, like you did at the start of the run. Or, is there no way to exit Mido town without warping to Bagu's Hut (or wherever),,, I remember reading somewhere that you only need to wait for the graphics to scroll over to learn sword techniques, not magic spells, so if this is right that's more seconds off the test run,,,
Also, if it's possible to warp to Bagu's Hut from Rauru town, it may be faster to do this instead of walking when you get the Water of Life for the fairy spell. I'm not 100% on how the healer glitch works, so I'm not sure.
If I get more ideas later Inz, I'll let you know, I need to go learn pianos for awhile now.
I think he meant to say that he doesn't use any of the ninpo, not that he doesn't collect power-ups...
I was able to watch the whole movie without being bored, and a few parts were interesting where some sneaky ninja moves had to be done to avoid getting hit. A few of the boss fights where you couldn't use that edge of the ledge quick strike trick dragged on irritatingly though. Also, why use the moonwalk the whole run? That started to annoy me fairly quickly.
Meh vote from me. It's well done and all, but some things about it just really bugged me.
*watches the latest (I think) WIP*
You do know, that the cannon outside Potos can take you to Gaia's Navel after the Water Palace event. Yeah the animation takes some time, but walking through Pandora, the area outside Pandora and most of Gaia's Navel seems to take a fair bit longer. The funds needed for the flight could come from the Chest in the Elder of Potos's house.
Is levelling the weapons really worth it in this run? As soon as you get offensive magic, that's probably the fastest way to kill bosses until Moon Energy (and maybe beyond considering your low level or unforged weapons). Yeah, it probably saves a bit of time against Tropicallo, Spiky Tiger and that miniboss you fight to get the magic... and it didn't lose you too much time to level the sword to 1 in the Dwarf Cave.
Also, while the Tropicallo fight was pretty fast, I think it could be faster still with some really sneaky manipulation. There is a small window where you can hit him and he wont dive under. Of course, it's unlikely to manipulate critical damage on such frames without ridiculous study of what affects the game's RNG.
Hmm,,, how well does that charge up glitch work? I haven't seen it done myself so I don't know how fast it charges to Lv. 8 or what the damage is like with weaker weapons. Does it have to be switching from an at least level 2 weapon to a weaker one? Couldn't switching a level 1 weapon for a level 0 weapon work as well?
Also, if the charge glitch does work well,,, it could be a good idea to charge one such attack up during a chain of spells and unleash it once the spell chain's damage is resolved, that way you basically get two extreme hits in the time of one,,,
I'm still interested in messing around with this game, but I'm supposed to be doing Goldeneye. Oops!
The file desyncs for me after the first two courses. Also, the time at the end of each race in the movie is different from what you listed in the submission. Can anyone tell me if it gets more interesting to watch after the first two courses? Otherwise I'll be inclined to vote no from lack of entertainment and beating the game half-assedly.
Here's my crappy test run of the Bunker.
It went fairly well for about half the level and I was able to get Boris to co-operate. However, tripping the alarm and having all those guards in the way really slowed things down- the second message during Boris's hacking seemed to be on screen much longer than usual. When I redo this, I will get rid of the guard at the beginning who trips the alarm and see if this makes a difference. The "Boris Shot" doesn't save much time even when done perfectly, ten frames was the best I could save with it.
The ending flat out sucks, but I think it's still a good watch.
Comments or feedback before I redo would be appreciated.
EDIT: Youtube link for the lazy.
Agent route is roughly 42 frames faster to the control room (about 2/3's a second). I started the level over using this route and tripping the alarm for fun. A rapid pause in front of Boris gets him to start moving ten frames faster. I've just finished clearing out the guards in the control room, and got two boosts up to Boris. I'm already over a second ahead of my previous run.
Now, I just need to manipulate Boris's running to get him into the MF room, and pull off a good Boris shot. This and a few good boosts during the escape sequence will hopefully put the mission time close to the 1:00 barrier.
Well,,, back to work.
CF: The weapon switch wasn't done because it isn't needed to zip up the ladder. While it does save a few frames to do so, I'd lose much more from having to wait for the weapon switch to finish to fire the sniper rifle at the locks.
Also, avis of each level of the current run plus the silo from the old can be found at http://youtube.com/profile?user=R151NGT3MP357, so there's no real need to encode them unless you can improve the quality or something. Do note that each of these videos show the mission result screen for a few seconds so the viewer can see the final time, headshot score etc. I do want each level available on youtube in this format for a casual viewer to watch, in addition to the full run on this site when it is finished.
Also, if the Agent route is faster (for Bunker), I will definately use it- having infinite guards to bother me would make the level more frantic & interesting to watch. Maybe more boosts in the final run to the exit too.
