XCom: Enemy Unknown (2012)

Paperboat

Edit history:

Paperboat: 2014-08-16 04:02:27 am

2014-08-16 03:40:59 am

XCOM Timesheet: http://goo.gl/BWDqbC

XCOM's only unskippable cutscene successfully skipped!

The final room is now safer, faster and completely consistent. Even a very sloppy proof of concept run allowed me to get from the muton despawn to the final cutscene in 3:05. For comparison, my pb takes 4:45 and Twyn's WR takes 5:10. Even a perfect clear would take just under 4 minutes using old strats so this is def faster.

http://www.twitch.tv/papers7/c/4916639

Twyn

2014-08-16 07:17:51 am

That's tedious as f**k, but better than waiting during a cutscene. And the final despawn gave me hard time, so I'm pretty happy with that.
After 100 tries, I never had the 1turn intro mission, but I dont give up ! Luck away !

KHANanaphone

2014-08-17 02:38:09 pm

Fiiive

Good job guys, I'm glad I no longer have to feel bad about my video getting corrupted.

secv

2014-08-18 01:23:46 am

So when's Enemy Within? thumbsup

Twyn

2014-08-18 04:52:57 am

Well, EW is pretty much the same thing but made longer with a couple of scripted mission. So it wouldnt be interesting imo. Maybe I'm wrong. But I'd rather never extend the RNG madness for longer than necessary Cheesy

Paperboat

2014-08-31 05:30:54 am

XCOM Timesheet: http://goo.gl/BWDqbC

I wasn't sure where to post this. It seems too specific an XCOM problem for any other thread so I guess I'l just leave it here and hope people other than me come here. Sorry for text wall.

There is a small paradox which I'm running into with segmenting XCOM. For those who don't know, XCOM has very imperfect save states. Any save made while somewhere on the base screen will reload to the fully zoomed out base and any save made in combat will reload with the camera on the selected soldier. These small inconsistencies between game state at save and game state at save load are completely irrelevant in an RTA run because the menu time required to reload a save would easily cancel out any advantages which could be accrued from using these save warps. I also know I have no interest in trying to create a segmented run which would abuse these small save warps to their full potential as such a run would be ugly as all hell and would be abusing timesaves which are actually impossible to use in RTA. I felt like this was contrary to what I believed a segmented run should be.

The obvious solution to this is to make myself include the menuing time required to reload a save while savewarping. This makes sense because it would more reflect what an RTA run would actually have to do in order to do a save warp. However, the deeper I get into the project, the more I run into segments whose natural ending point would create a small save warp upon reloading. And when I say small, I mean like a couple frames. Including the menuing time to reload the save in every single case, I realized, would create a run just as ugly but substantially slower than a run with full savewarp abuse.

I'm pushing on with the run for now without bothering with recording the menu reloading and also without any extensive exploitation of save warping. However, operating without a clear idea of the rules I'm running under is bothersome, and save warping has neccesarily started to play into decisions regarding where to segment and how often to segment.

If anyone has any advice or any experience with other games with similar problems, I'd love to hear about it. However, one way or another this run will continue. Thanks to anyone who actually read through that lol.

TLDR: If I accept savewarping absolutely, then the run will look terrible and defeat my purpose in creating the run. If I reject savewarping absolutely, the run will look terrible and be slow. The result is I'm walking on an uneasy middle path in an attempt to create a run which looks good and is also fast.

I wasn't sure where to post this. It seems too specific an XCOM problem for any other thread so I guess I'l just leave it here and hope people other than me come here. Sorry for text wall.

There is a small paradox which I'm running into with segmenting XCOM. For those who don't know, XCOM has very imperfect save states. Any save made while somewhere on the base screen will reload to the fully zoomed out base and any save made in combat will reload with the camera on the selected soldier. These small inconsistencies between game state at save and game state at save load are completely irrelevant in an RTA run because the menu time required to reload a save would easily cancel out any advantages which could be accrued from using these save warps. I also know I have no interest in trying to create a segmented run which would abuse these small save warps to their full potential as such a run would be ugly as all hell and would be abusing timesaves which are actually impossible to use in RTA. I felt like this was contrary to what I believed a segmented run should be.

The obvious solution to this is to make myself include the menuing time required to reload a save while savewarping. This makes sense because it would more reflect what an RTA run would actually have to do in order to do a save warp. However, the deeper I get into the project, the more I run into segments whose natural ending point would create a small save warp upon reloading. And when I say small, I mean like a couple frames. Including the menuing time to reload the save in every single case, I realized, would create a run just as ugly but substantially slower than a run with full savewarp abuse.

I'm pushing on with the run for now without bothering with recording the menu reloading and also without any extensive exploitation of save warping. However, operating without a clear idea of the rules I'm running under is bothersome, and save warping has neccesarily started to play into decisions regarding where to segment and how often to segment.

If anyone has any advice or any experience with other games with similar problems, I'd love to hear about it. However, one way or another this run will continue. Thanks to anyone who actually read through that lol.

TLDR: If I accept savewarping absolutely, then the run will look terrible and defeat my purpose in creating the run. If I reject savewarping absolutely, the run will look terrible and be slow. The result is I'm walking on an uneasy middle path in an attempt to create a run which looks good and is also fast.

Paperboat

Edit history:

Paperboat: 2014-09-12 05:01:17 pm

2014-09-12 05:00:08 pm

XCOM Timesheet: http://goo.gl/BWDqbC

Funny glitch. I managed to build in a square which I shouldn't have access to. Not particularly useful though and extremely difficult/inconsistent.

Paperboat

2014-09-13 03:56:57 pm

XCOM Timesheet: http://goo.gl/BWDqbC

A quick update for those interested:

The XCOM segmented run which I started in March of this year finished the first month and was then scrapped due to a number of problems ranging from video quality to just suboptimal gameplay. I restarted the project in August with the help of Twyn and the run has been going strong since then. We've just finished the first month again, this time in 5:31 (5:16 loadless) putting us about 80 seconds ahead of the old run.

Timesheet: http://goo.gl/BWDqbC

NMS

2014-09-14 07:33:46 am

Would it save much time to not have to build Firestorms? Because in my first playthrough (on Classic) I shot down the Overseer with a Raven with EMP cannon using tracking boost.

Paperboat

2014-09-14 10:31:20 am

XCOM Timesheet: http://goo.gl/BWDqbC

Its a neat idea NMS, I hadn't looked into Tracking Boost before.

My biggest concern is how exactly we would pick up the two cyberdisk kills since you can only get disks June or later. Getting a quick mission with two cyberdisks could be near impossible. There are a lot of other small ups and downs to think about as well regarding soldier promotion and the like if we are to skip the 5-7 aliens in the first two months we would otherwise kill for weapon fragments.

However, your idea opens up a number of possibilities related to an early SHIV laser route which we originally threw out because it required too many weapon fragments over those we needed for the Firestorm. This is definitely something I will look into.

Tseralith

2014-09-14 01:35:03 pm

Wouldn't the structure placement glitch you found be useful to get a "simple and quick" adjacency bonus? Like for the Satellite Uplink? Or maybe just build two Access Lifts at the same time.

By the way, can you make use of this building, or does the game prevent you from even doing anything to it when it's finished? Also, can you build in spaces where you haven't excavated yet as well?

I must say I enjoy seeing this kind of game being beaten so skillfully! Well done!

Paperboat

2014-09-14 05:18:26 pm

XCOM Timesheet: http://goo.gl/BWDqbC

Thanks for the interest Tseralith! Its nice to know people enjoy seeing this kind of run after all the hundreds of hours I've put into the game in the last few months.

The main limitation of the glitch is you can only build in places which have already been excavated. It's possible you could build 2 access lifts at the same time but we only build 1 access lift in segmented anyway and doing this glitch in SS I don't think would ever be worth the number of attempts it would take. (the timing is VERY precise) Also the glitch only works when you move from an unbuilt square to another unbuilt square which further limits its usefulness. Of course it COULD come up if someone were to make another segmented run with a different base layout seed, but for now it won't see any use.

Also about the tracking boost route: Although XCOM is too variable a game to make these kinds of exact statements, it appears to me that the tracking boost route would not save time due to the necessity to do missions in June to get Cyberdisk wrecks. It also could introduce problems with panic. It was a really neat idea and proof that there is too much stuff in this game for me to know that I have explored every available option. Thanks for pointing it out anyway NMS!

Thanks for the interest Tseralith! Its nice to know people enjoy seeing this kind of run after all the hundreds of hours I've put into the game in the last few months.

The main limitation of the glitch is you can only build in places which have already been excavated. It's possible you could build 2 access lifts at the same time but we only build 1 access lift in segmented anyway and doing this glitch in SS I don't think would ever be worth the number of attempts it would take. (the timing is VERY precise) Also the glitch only works when you move from an unbuilt square to another unbuilt square which further limits its usefulness. Of course it COULD come up if someone were to make another segmented run with a different base layout seed, but for now it won't see any use.

Also about the tracking boost route: Although XCOM is too variable a game to make these kinds of exact statements, it appears to me that the tracking boost route would not save time due to the necessity to do missions in June to get Cyberdisk wrecks. It also could introduce problems with panic. It was a really neat idea and proof that there is too much stuff in this game for me to know that I have explored every available option. Thanks for pointing it out anyway NMS!

Paperboat

2014-10-01 11:18:01 pm

XCOM Timesheet: http://goo.gl/BWDqbC

strats

Twyn

2014-10-18 08:41:47 am

1h32m24

Still slow and sloppy. But the temple has the latest strats aka Berzerker one shot and cutscene skip.

(twitch highlight on twitch.tv/twyn_o)

Paperboat

2014-10-18 10:14:40 pm

XCOM Timesheet: http://goo.gl/BWDqbC

1:31:07!

http://www.twitch.tv/papers7/c/5323576

Local record broke and highlight is muted rip.

Twyn

Edit history:

Twyn: 2014-10-21 06:26:18 am

2014-10-20 10:44:50 am

1:20:45 !

http://www.twitch.tv/twyn_o/c/5333531

I have a local recording that starts 5min into the run. It looks way better :

No silly talking Cheesy

Paperboat

2014-10-20 02:23:44 pm

XCOM Timesheet: http://goo.gl/BWDqbC

The first actually good run of xcom has finally happened. I don't expect massive improvements after this but I doublt this will be the last improvement.

Paperboat

2014-10-21 09:46:53 pm

XCOM Timesheet: http://goo.gl/BWDqbC

Got a 1:14:13 earlier today. I'd say that we shouldn't expect major improvements after this, but I've said that before.

Twyn

2014-10-29 06:35:37 pm

Got a 1:16:24 then a 1:14:10 tonight.

then

Paperboat

2014-11-08 11:49:55 am

XCOM Timesheet: http://goo.gl/BWDqbC

An update on the segmented run:

Segmented base assault is complete in 6:54!

The run is now in pre-Overseer menuing with 17:29 on the clock.

Paperboat

2014-11-13 04:55:22 pm

XCOM Timesheet: http://goo.gl/BWDqbC

Got a 1:13:41 today. I'm done with full game runs until someone beats this.

http://www.twitch.tv/papers7/c/5491158

Also we are done with pre-overseer menuing in the segmented run and are now working on the overseer seed.

Twyn

2014-11-22 07:27:36 am

You asked for it ! :p 1:10:43 <3

http://www.twitch.tv/twyn_o/c/5545488

Paperboat

Edit history:

Paperboat: 2014-11-25 08:03:41 pm

2014-11-25 07:08:55 pm

XCOM Timesheet: http://goo.gl/BWDqbC

1:10:22

I'm gonna keep going until we have sub 70 but I'm gonna be focusing on the segmented run for a few days.

KHANanaphone

2014-11-26 01:47:03 am

Fiiive

Ooh, getting really close to 1:10 now!

One thing I remember now is that you can sell stuff really fast by both holding down right-arrow/d button AND mashing the arrow with your mouse at the same time. Gotta save those frames!

Paperboat

2014-12-17 05:37:58 pm

XCOM Timesheet: http://goo.gl/BWDqbC

Segmented run update: Overseer mission done in 4:02.95 (for reference my gold split there is a 10:17). All that's left is a bit of menuing and temple ship. Looks like it'll be done some time in January.

@khan haha that's cool. That'll probably be used in the next segment.