1 page
--
--
List results:
Search options:
Use \ before commas in usernames
Edit history:
LotBlind: 2015-11-02 11:37:43 am
LotBlind: 2015-10-12 08:57:52 am
LotBlind: 2015-10-11 01:08:29 pm
LotBlind: 2015-10-11 01:07:31 pm
LotBlind: 2015-10-11 12:54:47 pm
LotBlind: 2015-10-11 12:39:20 pm
LotBlind: 2015-10-08 01:13:06 pm
LotBlind: 2015-09-26 09:35:19 am
The previous thread I started about long wait times in runs made me think about RTS games and how there's stuff like StarCraft II WoL mission called The Dig (see AGDQ '13). In it, you can cheese the whole thing by blocking off some ramps after which you don't IIRC need to do anything for minutes to eventually win the mission automatically.

So basically I thought it would be a fun challenge for anyone who runs any RTS, or other game where you have such strategies available to you but they're not the fastest, to have you record a run where you try to reach a won state as fast as you can. So to be clear:

1) Play the level/scenario/whatever any way you like.
2) At some point you stop giving inputs. No more mouse clicks, hotkeys etc. (moving the view around for spectating or typing in the chat is okay)
3) You still win.

Your "Last Significant Input" is the time we count and what the run should be aiming to get as low as you can.  I know this would probably be more fun if you had someone to compete against, which I know not nearly everyone has, but if it inspires you and if you can think of a mission that lets you do this (preferably in a spectacular, skin-of-your-teeth way or otherwise interesting), then go ahead and post it in this thread!

In order to provide a wider berth for games that don't play themselves like Real Time Strategy games do, I'm willing to consider examples where instead of dropping the control device entirely you're holding a button or combination of buttons down instead. Button presses/releases are not really differentiated by the games themselves. It should still constitute a somewhat non-trivial case (i.e. not just Mario walking to the right a bit after defeating the last Bowser).

Please provide me with:
a) Time when you stop giving (significant) inputs.
b) How long the mission is idling after that until timing ends (an extra curio).

PS: Use the hidden tag when posting videos. (hidden) link (/hidden) replace () with [].

-----

HALL OF LAZE

BadJim - '15/10/07 - Hostile Waters - Mission 11 - Last Significant Input 1:00 - Idle Time 4:05
  Pushing a cart on the right tracks and waiting for a train to slowly shove it to its destination.

Zergreenone - '15/10/11 - Warcraft III - Undead Mission 3 (Hard) - Last Significant Input 1:26 - Idle Time 4:46
  A-move and AI does the rest.

Zergreenone - '15/10/11 - Empire Earth - Russian Mission 1 - Last Significant Input 1:20 - Idle Time 25:05
  1 infantry unit wrecks half the city until AI finally targets mission-critical structure.

jakeplissken - '15/10/12 - Thief II - Mission 1 (Normal) - Last Significant Input 0:03 - Idle Time 3:13
  Main character tells Basso to man up and get his girlfriend outta imprisonment himself.
Thread title:  
So are we going for some record, like the longest idle time, or just posting for fun? Anyway I can start out with one of many timed missions in SC/SC2. I stop playing at 4:20.

Edit history:
Onin: 2015-09-30 05:45:18 pm
I would make a slight change to this mission here. It wouldn't take me more than 5 minutes to build enough defenses to not lose a single transport ship, at which point it still takes over 10 minutes to actually beat the game.

Freezard: I wasn't thinking of setting records, just whatever is interesting.

Is that just the same optimized record run from the other thread? Can you please just read my posts first: "where you have such strategies available to you but they're not the fastest".

Onin: "Use the hidden tag when posting videos"

Otherwise you're saying 5 minutes would be the FASTEST time to make the rest be an automatic win from that situation? So you don't have to micro units at all or build structures or anything at all? I don't care how LONG it takes for the win to happen, so long as it happens. Feel free to record that and it qualifies but only if it's not the same as the speedrun strategy.
It would be practically the same as the speedrun strategy, just some actual risk taken. The speedrun strategy is "just build a shitload of units to ensure no transports die, then don't die of boredom before the mission is done". Finding out the cut-off point for not building units anymore wouldn't slow down the run (much), just make it risky.
Yes, you've got the right idea! Being more risky is almost a de facto quality of any such attempts.

Thanks for hiding the video! Feel free to be more verbose in describing the plays you make if you choose to upload actual Fastest Controller Down (FCD) runs. I for one will definitely watch all of them. Well maybe on fast-forward Smiley
TASvideos' automated timer is based on fastest controller down, so people typically aim for that as the category when it's close enough to other methods of timing. (Sometimes it isn't, in which case what timing mechanism to use is a lot more controversial; some players favour controller down in that situation, others use SDA timing (which would then have to be done by hand rather than via the automated timer).)

One dramatic example is Rosenkreuzstilette, in which the route changes between RTA and controller down (in controller down, you go out of your way to collect an item that allows you to defeat the final boss without inputs and thus put the controller down earlier). Such examples aren't common, but do exist.
Right. But you're talking about TASing. I guess you meant to bring out that even TASes sometimes go by "last hit on boss" type timing. I hadn't thought of that, but I guess that's still a small minority of cases. I really should know.

But yes, a TAS strat real-time run could meet this challenge's criteria.
Hostile Waters. Mission 11. You must destroy the train control center, drag a canister onto the track and protect it while an automated train pushes it into the heart of the Cabal base. But since the Cabal doesn't make any attempt to destroy the canister, you only have to get the canister onto the track and you can then wait for the train to do its' work.



The mission can be beaten faster in a regular speedrun by dragging the canister along the track into the Cabal base, a bit like this
Edit history:
LotBlind: 2015-10-08 01:12:42 pm
Congratulations BadJim! You're the first to complete the challenge (with video). I guess I'll start listing all the runs and runners in the first post...

Feel free to give me a more exact "idle time" count.

Remember guys it's okay to give inputs to spec what's happening but in BadJim's case I agree with what he did.
Edit history:
Zergreenone: 2015-10-10 11:44:12 pm
Zergreenone: 2015-10-10 11:44:08 pm
Zergreenone: 2015-10-10 11:43:11 pm
Zergreenone: 2015-10-10 11:32:46 pm
Zergreenone: 2015-10-10 07:58:10 pm
Warcraft III Undead Mission 3 "The Dark Lady" Hard Difficulty last input 1:26 after loading.


I forgot to show the ending screen, but it said that the final time was 6:12, which is consistent with the amount of real time that passes in the video.

It is possible to get a faster ending time if you micro better during the wait time, or indeed micro at all, and I also placed priority on getting to the bandit lord quickly over keeping all my units alive, which may be detrimental to an ordinarily timed run (but may not, who knows).

I start with the same items and approximately the same experience as in the current sda run and I end with the same items and approximately the same experience as in the current sda run.

Also, why is the idle time mentioned in the "hall of laze" and not the ' "win state time" '? I could just shift click my units all over the map if I wanted a large idle time, and if it is supposed to be a short idle time then it would be going against the point of this challenge
Empire Earth Russian Mission 1 "The Crocodile" last input 1:20 after loading. Victory is 26:25 after loading (39:16 ingame time).


For some reason the video is slightly flickery, possibly because I had to use screen capture as opposed to window capture or game capture. Also I suck at typing.

Uses a very different strategy to the one used in the actual route

I think that you could possibly sneak a citizen (pyotr) into the enemy base using this strategy in an ordinary speedrun. More on this in the empire earth thread.
Zergreenone: hmm... I suppose you're right. If we count by idle time it assumes it's an optimized or nearly optimized run. Idle time is mostly a curiosity I guess. So yeah I'll start listing "controller down" times as the main thing.

Added your two runs, grats! The WCIII one looks much as I envisioned. A-move to win! As for exp going into missions... Seeing as these are just demonstrations it shouldn't matter too much. The Empire one... maybe you could have left him more to the east so he couldn't have targeted the airports?
Edit history:
jakeplissken: 2015-10-12 02:31:22 am
I'm an Olympic Straferunner
Thief 2: The Metal Age, Running Interference with the last input 5.02 seconds after "start mission," which is standard T2 timing. Lot tipped me off to this thread and since he's contributed so much theory to this game, I figured I'd try it out.



In Running Interference, you're supposed to enter the mansion and clear the way for Basso to rescue Jenivere. Normally if you just let him go on his own before clearing the way, he gets killed by the guards. But if we move back slightly, the we are far enough away from the guards that their AI isn't fully loaded, so they can't stop him. Then Basso runs off and completes the mission for us. Timing stops after my jump because jumps carry your momentum and it lets me make the last input a little bit sooner. The time it takes for him to complete the mission varies because his AI sometimes behaves strangely around doors, but here he finished in 3:16, giving us an idle time of 3:11.
Edit history:
LotBlind: 2015-10-12 08:57:21 am
LotBlind: 2015-10-12 08:55:19 am
So you wouldn't have done it if I hadn't been so useful before? Sad

Can't include the loading time in timing actually, so I took off 2 seconds. And the real idle time should be 3:13 shouldn't it technically? Hmm... glad to hear my theory about the guard AI worked.

Yeah as Jake said, the final time has a big variance and I've sometimes gotten much lower times, but the AI was improved between my version and Jake's.

Grats! Smiley

Wait are you headbanging to the stridulations?