Jump to content

Late Game Lag Kills this Game


Recommended Posts

I know my complaint is tempered, at least partially, the fact that I am playing on a Mac Mini.  But when you get to the late game, and I am between 1000 and 1500 cycles in, this game is unplayable.  I would open the game and check what cycle I am on but it takes several minutes to load a save game.  I was playing yesterday and I going to work on moving some things around my base.  I set dupes to destroy some tiles.  They come over, take about 5 seconds to stare at the tile before destroying it.  Then, after destroying it, they stand there mid animation.  I am left with time to make up stories in my head about how the dupe is contemplating the meaning of life with the destruction of stone tile as analogy for life.  The game is just unplayable.  So now I sit and wait until the next release so I can have a reason to start a new base.  If anyone ever asked my opinion about buying this game, I would definitely say, "It is fun, BUT...".  Don't worry Klei, I don't have people beating my door down to get my recommendations.  But this is kinda silly. Even for early release.  

Yunru is correct. The game is early access and time spent optimizing systems that could change or be removed before beta is time wasted. They will do some optimization for the sake of playability but don't expect any dedicated overhauls at this stage of development.

34 minutes ago, Yunru said:

That said, if they optimise it too much, too early, it makes it significantly harder to change things, let alone add new things in.

I expect the last couple of updates before launch proper will be heavy optimisation.

That is just it: Early optimization is often wasted, because you will replace the code again. The one exception is selection of the algorithms, but there Klei cannot have done too badly or everything would grind to a halt much earlier.

2 hours ago, BlueLance said:

I hope they do fix the fact that the last patch has been a kick in the teeth in terms of performance though

Well I mean they did add in a LOT of new stuff, including a new biome. Give them some time to refine the code and finalize the major stuff that was added in this upgrade, and then they'll start working on optimizing it.

4 minutes ago, watermelen671 said:

Well I mean they did add in a LOT of new stuff, including a new biome. Give them some time to refine the code and finalize the major stuff that was added in this upgrade, and then they'll start working on optimizing it.

The biome was in the files though a couple of patches ago ;) but there must be a memory leak or something because adding a new biome etc would not contribute so heavily to the performance loss 

9 minutes ago, BlueLance said:

The biome was in the files though a couple of patches ago ;) but there must be a memory leak or something because adding a new biome etc would not contribute so heavily to the performance loss 

Spoiler

Also blame the critters, they still need a lot of cleaning up with their pathing

As I said in the spoiler, they still need to polish the coding for the critters, because Oxyg1701.thumb.png.e30411482830d465d565cd7302a1fd28.png, they cause a lot of lag still. That in tandem with all of the fresh code introduced, performance loss should be expected.

Spoiler

Also, IIRC @RoboPhred pointed out that they changed something about the parsing, which I have no idea what that means. :drunk:

On 6/8/2018 at 12:21 AM, RoboPhred said:

I took a peek into a CU save.  The immediate compatibility issue is the introduction of a new serialization type flag IS_VALUE_TYPE (64), which only affects certain custom objects inside arrays.  There might be more, but I cannot discover them until this one is dealt with.

I played around with it, but it turns out I made some pretty bad design decisions around the multi-version support stuff that prevent it from actually being workable.  I could get it to work, but it would break compatibility with the current stable.

Since we are so close to the CU release, I will just wait for the CU release, refactor the parser to be far more sane, and declare CU the new minimum supported version.

Work will be happening in the cu-versioning branch if you want to follow along.

I am rethinking my plans to try for multi-version support.  Since Klei only has one stable release at a time, there's not much point in keeping support for older stable versions around.  Multi-version would be able to handle the test branches, but I don't believe that's a good idea as it interferes with the actual testing.  The only real use-case I can see is for trying to inspect or copy data from old saves, but that doesn't seem like it would come up too often.  Does anyone have any opinions on this?

 

13 minutes ago, watermelen671 said:

As I said in the spoiler, they still need to polish the coding for the critters, because Oxyg1701.thumb.png.e30411482830d465d565cd7302a1fd28.png, they cause a lot of lag still. That in tandem with all of the fresh code introduced, performance loss should be expected.

Critters existed in MkI and MkII and the game was fine, Cosmic didnt add more to critters etc.

Also I wish! I drop to about 12FPS on triple and 26 on single!

Just now, BlueLance said:

Critters existed in MkI and MkII and the game was fine, Cosmic didnt add more to critters etc.

Also I wish! I drop to about 12FPS on triple and 26 on single!

I dunno, maybe I just need to stop making shine bug bombs...

20180624123337_1.thumb.jpg.2e83f10bf760ee28fa0ad43e8b59294c.jpg

1 hour ago, Yunru said:

So I haven't experienced any lag, but ONI does push my laptop into the 80+*C range D:

I actually have a small fan that I bought to blow over my mac when I play the game.  I just figured it was the cost of "gaming" on a mac mini.  The mini was designed to be small (in 2012 terms) not for cooling.   

1 hour ago, old_fantom said:

I actually have a small fan that I bought to blow over my mac when I play the game.  I just figured it was the cost of "gaming" on a mac mini.  The mini was designed to be small (in 2012 terms) not for cooling.   

why not just submerge it in liquid nitrogen?

1 hour ago, Yunru said:

The buttons might crack when you press them, duh.

Also if exposed for too long, there's a risk of hypothermia.

Psh. With that attitude.

1 minute ago, Cilya said:

Put one wheezwort near your laptop then !

Precisely.

1 hour ago, Yunru said:

The buttons might crack when you press them, duh.

Also if exposed for too long, there's a risk of hypothermia.

It has a lot more to do with the fact that I have no means to make or procure liquid nitrogen and that my analog mind never went past a small desk fan.  This is why my bases are not full of complicated automation.

17 hours ago, Yunru said:

That said, if they optimise it too much, too early, it makes it significantly harder to change things, let alone add new things in.

I expect the last couple of updates before launch proper will be heavy optimisation.

As a programmer, this make no sense to me. Proper optimization brings *further* ease to make changes, fix bugs and overall makes life easier. They should damn well be aiming to optimize, the sooner the better.

I think it has more than a little to do with the fact that the game is made in Unity. Unity's performance just isn't all that good, particularly with memory-intensive games like this. I genuinely have no idea why they chose it. That said, cleaning up debris seems to make a HUGE difference, so build a warehouse full of storage compactors.

42 minutes ago, Megouski said:

As a programmer, this make no sense to me. Proper optimization brings *further* ease to make changes, fix bugs and overall makes life easier. They should damn well be aiming to optimize, the sooner the better.

"Premature optimization is the root of all evil." If you write clean, modular code, you can optimize it later; no sense wasting your time optimizing something that will need 10 rewrites anyway.

I'm no coder, but I somehow doubt everything that could be optimized is also something that could still be radically changed. If the game is becoming too computationally cumbersome they should optimize it now, rather than wait until it's completely unplayable.

Reading articles and such written by actual developers, pushing back Q&A style issues as well as optimization until the end of the development cycle often leads to bad results.

Archived

This topic is now archived and is closed to further replies.

Please be aware that the content of this thread may be outdated and no longer applicable.

×
  • Create New...