Factorio Roadmap for 1.0.

[Candunc]

Darn, kinda sad that multiplayer has been postponed.

What do you think about electrified railroads?

I like this idea, but the electric trains should be much more expensive to build.

[Aza-Industries]

Factorio 0.5.0 (expected release at the beginning of June):

[*] (Maybe) Speed improvements [/*]
- Thread separation of render and updater[/list]
[/color]

Just a quick and possibly ignorant question.

If you split those up into separate threads. Will that really offer much performance improvement?

I say this because I look at the overlay and the updates seem to be already using 90% of the thread, so if you split it won't it still be maxing out and be causing performance drop?
The game doesn't seem to be putting much strain at all on rendering. Unless you're (possibly) planning giving updates more then 1 thread to itself?

I don't know much about this stuff so sorry if this sounds really silly.


Sorry about all the questions I bombard you guys with in all my posts. >.>

[FreeER]

Just a quick and possibly ignorant question.
If you split those up into separate threads. Will that really offer much performance improvement?
I say this because I look at the overlay and the updates seem to be already using 90% of the thread, so if you split it won't it still be maxing out and be causing performance drop?
The game doesn't seem to be putting much strain at all on rendering. Unless you're (possibly) planning giving updates more then 1 thread to itself?
I don't know much about this stuff so sorry if this sounds really silly.
Sorry about all the questions I bombard you guys with in all my posts. >.>

Two links to help explain, cause I'm lazy and don't want to explain myself (and I looked it up to make sure I was right lol)
Stackexchange question "what can multiple threads do that a single thread cannot"
and a short article which might sum it up better
but basically with a single threaded application you can not utilize more than one core and since most people have at least dual-core processors now...

[Aza-Industries]

Two links to help explain, cause I'm lazy and don't want to explain myself (and I looked it up to make sure I was right lol)
Stackexchange question "what can multiple threads do that a single thread cannot"
and a short article which might sum it up better
but basically with a single threaded application you can not utilize more than one core and since most people have at least dual-core processors now...

I'll have a read through those.
I was aware that it would allow the game to use more cores but my question was.
If updates are already using 90% of the 1 thread and slowing down the game, than will giving it's own thread improve the performance much?
Because it will only gain 10% more headroom, in my basic understanding of this.

I read their (possible) future implementation of multithreading as they were just going to split render and updates into 2 separate threads.

It seems to me the only intense processing in the game is the updates, so I thought it would be good (if possible) to make updates gain more then just one of its own threads but multiple.

But anyway.. I shan't talk anymore on the subject before reading those links, thanks for them.

Here is quick picture I drew about how it looks in my mind. =P

[kovarex]

a) For most people the render is higher than 3%, when I zoom out in big factory, it is something like 30% for me.
b) The division of render/update is just the first step, we also plan to divide the update into separate threads, not all actions could be divided this way, but some, that works just locally can (like moving items on transport belts, inserters and assembling machines, furnaces etc.), it won't be easy, as we would have to make sure, that the game is still deterministic, but this si the plan
c) Optimisations, by spending time with the profiler and looking at the code, we can still optimize the raw demand of the game update processing, the last time I did this, I made transport belt logic run few times faster ...
PS. Save like the one you have could be handy for this task

[Aza-Industries]

a) For most people the render is higher than 3%, when I zoom out in big factory, it is something like 30% for me.
b) The division of render/update is just the first step, we also plan to divide the update into separate threads, not all actions could be divided this way, but some, that works just locally can (like moving items on transport belts, inserters and assembling machines, furnaces etc.), it won't be easy, as we would have to make sure, that the game is still deterministic, but this si the plan
c) Optimisations, by spending time with the profiler and looking at the code, we can still optimize the raw demand of the game update processing, the last time I did this, I made transport belt logic run few times faster ...
PS. Save like the one you have could be handy for this task

Thanks for the clarifications, it cleared up a lot!

I'll upload my save for ya in a few minutes. =P
I can't play on it anymore on my laptop it's just a tad too slow.

It's weird because it was running at 30fps earlier today and I didn't change much and now it won't go above 10.

Holy hell 6mb compressed 250mb uncompressed, here it is:
http://www.cerberusmc.net/001.7z

And for some reason this guys save runs better and I swear its a way bigger factory than mine.
https://forums.factorio.com/forum/vie ... ?f=8&t=173
In his world update time is around 40-60%
My world is 95%
No fair, why can't my world run properly.

Maybe I found a bug?

Note: My factory will be non-functioning in it because I was halfway through deconstructing to be rebuilt better when the performance dropped badly. I say this because I don't want you to see the mess it is atm and think I always build like that. >.>

[Aza-Industries]

I don't want to sound pushy, and I know you guys have plenty of more important stuff to do.

But I was wondering if you've had the chance to look into my save?
I can't figure out why it doesn't run as well as another save that has a much bigger factory.

I've been wanting to play foctorio for the past few days but I don't want to start a new factory just yet, and my only established factory's save runs at <10fps.

Any advice/info on this would be greatly appreciated.

[Dakkanor]

I don't want to sound pushy, and I know you guys have plenty of more important stuff to do.

But I was wondering if you've had the chance to look into my save?
I can't figure out why it doesn't run as well as another save that has a much bigger factory.

I've been wanting to play factorio for the past few days but I don't want to start a new factory just yet, and my only established factory's save runs at <10fps.

Any advice/info on this would be greatly appreciated.

well.. i do know that vel-masters save has either no logistics robots, or none at all, i also seem to remember that he had little to no defences, possibly playing a custom scenario to avoid creeper attacks....

[grobyc]

Problem is in storage chest probably same as with requester chest.
If requester chest can contain 3072 units of material and u set request in chest on 4000 and more logistic robots will be fly over that chest with material till they will put it inside, if they fly over it they start freeze the game (I'm not a game programmer and I don't knew how to translate this for more technical language), probably if storage chest is full of materials is not set as full and logistic bots must count each of nearest storage chest and if he count nearest is 273 chest then he start fly, if you have over 700 bots each bot must count it for himself before start fly to correct one.

I was burn all of storage chest and problem pass out.

Solution for you is to not store materials in storage chest but in provider chest on the end of production line like I do this with iron plate and copper plate and wait till problem will be solved. I too waiting for better future for storage chest system.

[kovarex]

I already responed to him, I opened the save and found out, that 73% of all the time is spent in the logistic manager
I'm just writing code to optimise the whole logic, the algorithm used there is quite "naive".

[Dakkanor]

I already responed to him, I opened the save and found out, that 73% of all the time is spent in the logistic manager
I'm just writing code to optimise the whole logic, the algorithm used there is quite "naive".

=D yay, that'll let me spam more bots

[Aza-Industries]

Thanks a lot Kovarex!

It was running so well for 10 hours of 700 robots and tons of chests and then performance plummeted.

I guess it didn't like the couple extra I added.

I'm upgrading my preorder, because you guys have been so patient with me, and just great generally. =P

[ficolas]

* Additional resources and electricity sources- titan and uran
- nuclear energy
- advanced mining drills

I noticed that oil is not there, will it come any time soon?

[Balinor]

Updated the roadmap for two months ahead. For details behind this have a look at the post https://forums.factorio.com/forum/vie ... ?f=3&t=962

Factorio 0.6.0 (expected release in the July 2013):

• Better looking terrain.
• Improvements to the map generator.
• Map of discovered part of the world.
• Speed improvements (multithreading).

Factorio 0.7.0 (expected release in the August 2013):

• Combat mechanics.
• More enemy variety and inteligent enemy behavior.
• Standard Campaign and Scenario Pack Extension.

Both of these will include small features and improvements (often) based on the Ideas & Suggestions forum.

We would love to add more content to the game (namely for the late game stage). However at the moment we are limited by the graphics. Our artist Albert is working hard on putting the game in the current stage together to have presentable consistent looks. And we want to avoid putting any more items with provisional graphics into the game.

While I am looking forward to all of these changes to come, the biggest issue for me currently is the balance of the game throughout the entire game length. So while it is good to see improvements coming for the late stage of the game (such as the advanced energy later on that has been removed from the roadmap for now) will there be alternatives also for early parts? Ie, if you create a map and there is very little coal in the vicinity of your area, do you plan to make alternative energy sources available at this point? Or will the only solution be to create a new map?

Thanks for all the patches and hard work.

[RawCode]

titan and uran removed from plans or just from roadmap?

[slpwnd]

We are definitely not dropping the plans for adding more ores and processing. The problem for now is the art side. The graphics are behind the programming so it is difficult to squeeze in the new content. The bigger content upgrades will come after the game becomes visually more consistent. In spite of this we try to push pieces of new content in every update.

[Balinor]

Thanks for the updates. While I can certainly understand your desire for a unified look to the games graphics, this does not seem as important (to me) as content for the experimental versions but I'll try to be patient. It is very refreshing how honest you guys are with your customers. I think we all appreciate that very much.

[RawCode]

adding something with placeholder sprite is OK, not adding reactor or something similar only cos there is no sprite is joke.

[DRBLN]

End of August is coming

[zer0t3ch]

End of August is coming

And.......?