Just thought i'd mention that i've at least temporarily taken up the 00A improvement project that i and RT started a year ago or so. I finished Dam by myself then and since i couldn't find any obvious ways to improve it now i decided to keep that and continue with Facility. So hopefully in a few days i'll be finished with it and WIP will be available.
This sounds awesome. Three questions:
Is RT doing this improvement with you, or is it just yourself?
How much did you beat the old Dam by?
So your PD project is now dead?
1: For the moment by myself, haven't spoken to RT yet but if he's interrested in joining then that'd would be awesome.
2: 2 seconds, it's now 1:47.
3: Kinda lost motivation, not the funniest levels coming up, someday i'll pick it up, unless someone else wants to finish it(only 5 levels left after all, and i can only think of 2 things that should be difficult to TAS, Getting Fast Elvis on Deep Sea and timing everything on Attack ship(elevators and stuff)).
A little progress report here.
First i tried taking advantage of the 2.3 glitch luring a guard to open first few doors for me, a little faster in mission time but it was a significant loss in real time so i decided to skip that. Now for the last few days i've been trying to manipulate a good Doak again which was harder than usual but today i finally nailed it. So hard part done, only rest of the level to go.
That was sexy.
One thing I wondered about RT's run, and you did it too, is in Facility, why do you get the keycard from the guard if the other guards open the door anyway? Wouldn't it be faster to just wait outside and go in? Or is the keycard actually needed elsewhere and I'm overlooking something?
Also, how much do you think you'll be able to improve RT's run by overall? And are you including Aztec and Egyptian (I think you should. You could just press B after Cradle to get back to the level select screen then go to them)?
Problem with pressing B is that i will never reach the credits unless i play cradle again, so button codes would be preferred, but yeah, i'll try to include them if i ever get that far(I tend to loose motivation quickly, though this time i have experience of every level in the game and previous TAS'es to help me so it might be easier).
Anyway here's some estimations:
Runway - Prolly 35 again, seems good.
Surface - Very hard to improve as well, maybe a second faster.
Bunker - Think Simon already cut a second here, 2.3 will allow open doors in beginning saving even more.
Silo - A second or two max, not sure how much is possible, RT's run was good.
Frigate - I'll try really hard to manipulate this perfectly, so 10'ish seconds.
Surface2 - 3rd camera can gain a few seconds by shooting through woods, simon already did this.
Bunker2 - A second or two has already been done so that's what i'll aim for.
Statue - Think i already did a second faster here, maybe another sec is possible.
Archives - My 52 TAS is probably the best TAS ever done in this game, gonna be hard to do again but i'll try.
Streets - 1:54 is great, depends on decimal if 1:53 can be done.
Depot - 2.3 is a major advantage here, 3-4 seconds possibly.
Train - Not sure here, a second or two maybe.
Jungle - A second?
Control - If i remember correctly this is improvable by a few secs as well(2.3 allows for open elevator door too).
Caverns - Beginning can be faster, so 2 secs perhaps. Also 2.3 advantage.
Cradle - 33.
Aztec - Already did 1:29, just gonna do it again.
Egypt - Also did this already, 44.
I see. Good luck on finishing this.
As for not seeing the credits, I seriously doubt anyone would care. What you could do is just let the final cutscene at the end of Egyptian play, and end the movie after the last shot at Baron Samedi.
Also thought it was kickass, especially Facility.
I made an encode of Facility for Comicalflop last night because it desynched for him; maybe it can be helpful to others.
And I strongly agree with SL that Aztec and Egypt should be included.
Yeah that's another option SL, though isn't there a nice estetical touch to ending the movie at the credits? Also i think it'd be pretty cool seeing button codes entered within a second :P
Guess further discussion on ending will be done as that moment approaches. For now i'll just continue as planned. Gonna startup Runway now, My knowledge of this level is extremely limited so it could take a while to master.
I put up an encode of the wip from the beginning, and I'll probably do future levels as well: http://bkdj.net/index.php?dir=tas/ge
EDIT: oh, there's already an encode of the latest part. oh well.
I put up an encode of the wip from the beginning, and I'll probably do future levels as well: http://72.233.35.202/index.php?dir=tas/ge
EDIT: oh, there's already an encode of the latest part. oh well.
Yeah you're correct Wouter, Simon told me yesterday that he had successfully done the door warp on egypt to get 42 on agent. Condering 00A requires a worse shot on last baron, end time should be 43 or possibly 42.
For Runway you were correct as well, just finished it with a time of 34! :
http://www.yousendit.com/transfer.php?action=batch_download&batch_id=WnBUbUpUb0JOMUJMWEE9PQ
Easier than i thought it would be, didn't have to redo anything and managed a better strafe line for third drone.
Surface coming up next and my estimation appears to be pretty bad too :P simon already did 1:46 so at least a 2 sec cut will be done here.
Thanks
I just remembered that nfq i believe had the memory address to the timer and could check exact decimals, If someone could explain how to use this and provide a link to the progam(s?), that would be extremely helpful in fully optimizing the game and achieving best possible times.
Problem with pressing B is that i will never reach the credits unless i play cradle again, so button codes would be preferred, but yeah, i'll try to include them if i ever get that far(I tend to loose motivation quickly, though this time i have experience of every level in the game and previous TAS'es to help me so it might be easier).
Anyway here's some estimations:
Runway - Prolly 35 again, seems good.
Surface - Very hard to improve as well, maybe a second faster.
Bunker - Think Simon already cut a second here, 2.3 will allow open doors in beginning saving even more.
Silo - A second or two max, not sure how much is possible, RT's run was good.
Frigate - I'll try really hard to manipulate this perfectly, so 10'ish seconds.
Surface2 - 3rd camera can gain a few seconds by shooting through woods, simon already did this.
Bunker2 - A second or two has already been done so that's what i'll aim for.
Statue - Think i already did a second faster here, maybe another sec is possible.
Archives - My 52 TAS is probably the best TAS ever done in this game, gonna be hard to do again but i'll try.
Streets - 1:54 is great, depends on decimal if 1:53 can be done.
Depot - 2.3 is a major advantage here, 3-4 seconds possibly.
Train - Not sure here, a second or two maybe.
Jungle - A second?
Control - If i remember correctly this is improvable by a few secs as well(2.3 allows for open elevator door too).
Caverns - Beginning can be faster, so 2 secs perhaps. Also 2.3 advantage.
Cradle - 33.
Aztec - Already did 1:29, just gonna do it again.
Egypt - Also did this already, 44.
Oh boy! Improvement time!?
(Wall of text)
Well, my main reason for not jumping in right away was that I was utterly SICK of the first two levels. There were also some other nuances that have been taken care of at this point on the technical side of things (one of them was just that I wasn't aware of a particular feature), but then again I haven't seen the WIP yet (on the laptop for a couple days so I'll come back monday with the verdict). I'll watch what's done so far- I'd be glad to help out too!
But.... I had some preconditions I wanted before I would even try to do an improvement by myself or otherwise.
1- glide Napalm graphics plugin; as far as I've seen testing it myself (played through the game on emu) this plugin has no graphical issues whatsoever. this is a must for any improvement.
2- 2.4 control style- NOT 2.3. 2.4 has the forward movement and side strafing on the same joystick so I can still use my dated but preferred method of Nrage input. I refuse to use the TAS input plugin.
3- Aim for ingame time. I really have a lot of things I don't like about my first run and trying to please everyone is probably the worst. I think that if I were to do a run I'd rather just do it my way and leave it at that, by aiming for ingame time, and sacrifice real-time readily and often for that purpose.
4- Avoiding skipping of cinemas. First one on Dam's kind of required for the optimal strategy (so the dual gate is open). But otherwise, since some cutscenes would be watched to an extent to use the fire weapon during cutscene glitch I think just watching them all would give the run a more "complete" feel. Not skipping the end cinemas would also allow for some nice playaround techniques without ruining the sacrifice of ingame time which the run would aim for.
5- More blatant showing off. Watching my old run, I missed plenty of opportunity for unnessecary, yet super-sweet moves. Control was fun sure, but think of the fifteen odd levels before that where I focused on being quick and almost nothing else but.
I highly doubt henrik's WIP conforms to my lofty and possible unreasonable standards here, so I might not end up being too involved this time around, unless I start this on my own and just publish it on youtube or something. Don't know if the majority of users at this site want to see the run I have in mind, beat the game as quickly as possible seems to be more the standard I guess. I'd like to see it at least so for anyone who actually cares I'll get around to it... sooner rather than later if Dam and Facility magically apparate finished on my harddrive somewhere.
Oh, now on to the actual shit I quoted, above.
Runway- You already did this looks like, but I was going to suggest 34 should be doable. I had an idea about shooting grenades though I don't remember the specifics offhand. 2.3 can get a boost to start the level by luring that guard near your starting position but I don't think this is to helpful on 00 where health is at a premium later (I remember my health bar was completely empty on the last run)
Surface- if you can get a grenade drop for the locks that might be enough to lower the time unit. better boosts than old run etc...
Bunker- Simon made 57 here without having the door opened from the cutscene, 55 is the time to aim for I say.
Silo-
Frigate- I think I failed something like 54/55 without hostages escaping? The current route is probably the best one, if you're more patient than I was in manipulating them (or just flat out lucky) I.... don't really know what the max would be. Just be aware I spent like 30k do-overs on this stage alone before I just decided "whatever, 108 is good enough"
Surface- Yeah that camera through the trees thing. No horizontal aim in the cutscenes really sucks with that one cinema of it right there.
Bunker 2- Cell door warp + better clipboard lure = 44. Math at it's finest.
Statue- Probably some nuance with the Valstrafe I dont know of but as far as my testing went that seemed to be the fastest he'd move. I just used leans because I was lazy.
Archives- I still don't understand how your 52 worked out? But hey, good luck recreating it!!
Streets- I don't remember too much in particular about this stage. Outside the Valstrafe, I don't remember thinking that too much was improvable outside manipulatimg boosts and stuff.
Depot- Easy 4/5 seconds from cutscene glitch. Going into the computer warehouse, you should be able to take out the drone without losing time since the screen's already toast and no need to shoot it. That would stop a backboost in addition to the time gained from having the door open. I don't remember if guards pull grenades on this level or not, but if you could get one, you could pitch it under the Obj A door and not have to go all the way in there to start the chain reaction. If that works out too, this could be dangerously close to under 40 seconds.
Train- Simon has 143 here, and lower should be doable. You can get the laser out right away (switch to it during the cutscene) and for warping doors later switch between the standard guns since the laser takes too long to bring out. I don't think 141 or better is out of the question.
Jungle- Probably improvable with some cleaning up of things here and there.
Control- Still not sure on the quickest Nat escape. 353 at least should be doable. Oh, and you can elimanate the main screen in the cutscene, that might help with lag earlier and stuff. Open elevator's not as big a timesave you might think, the insta-warp I found for that elevator (same as Caverns) is well... instant.
Caverns- i'm not sure what exactly in the beginning you thought could be faster? Elaboration would be nice. I honestly thought this was the best level in the whole run. I made a tiny strafing error I didn't catch until a couple months after the publish (going back up the stairs in obj. B room) but I don't think that's enough to push the time unit lower... I'm convinced I maxed this level (but only this one... lol) but hey, if you can prove me otherwise, by all means!
Cradle- We know 33 is doable. Something I feel retarded for not trying (or maybe I did and forgot) was to throw the grenade into the hut ahead of trevelyan and blow up the console and him simultaneously. I think the best way to time the explosion would be right as you hit him with the double headshot (then just warp the explosion so you don't die)
Aztec- you got 129 or something right? At least that much should be doable. I remember in my 2.3/2.4 topic at t-e.n I went over a possible cutscene glitch strategy that would get an insane quick glass door, but I've yet to actually test it via Tool assistance.
Egypt- Simon got 42 recently, depending on the needed 00 shot on the last Baron that might be dupe-able here.
ANYWAY!
I'll watch the WIP tomorrow and say what I think of it. I know you well enough to assume that it's nigh unbeatable but there's still nothing wrong with a little constructive criticism.
Oh, and to anyone else, would you actually like to see my "cinematic" run? I certainly wont guarantee immediate results but more interest would probably motivate me to slug through the first couple levels again.
Joined: 6/13/2006
Posts: 3300
Location: Massachussetts, USA
Wyster: MHS (Memory Hacking Software) is your best bet for memory watching.
Laughing gas said he found the address; nfq said "009654A0 is the code" but I don't know if he uses MHS for that.
"47.9833333, 129.9833333"
Using this, I bet you can find exactly what address he was talking about. It seems to increment every 0.166667 every frame, according to what laughing gas said.
If I have time later, I might try to find the in-game time address.
XGeOmegaX: damn, a pretty genious idea for runway you posted if it works...
XGeOmegaX: Shooting nades i mean
Jackimus Wedge: it does work
XGeOmegaX: faster battery
Jackimus Wedge: yeah
Jackimus Wedge: well, i tried to do that in my run
Jackimus Wedge: err hm
Jackimus Wedge: I actually remember doing that and getting 34
Jackimus Wedge: ...what the hell?
XGeOmegaX: Lmao?
Jackimus Wedge: *checks youtube right quick*
XGeOmegaX: its 35 on youtube i believe
Jackimus Wedge: X FUCKIN D
Jackimus Wedge: I kept the wrong runway run!!
XGeOmegaX: Wow, this is unexpected
Jackimus Wedge: WOW
Jackimus Wedge: it took me two years to notice that? LOL!!!
XGeOmegaX: Hahahahahah
XGeOmegaX: Well this means i should redo end of runway as well..
XGeOmegaX: getting 33 i guess
Redoing ending as soon as i can...
