Factorio Forums

ili wrote:If mods have a problem of local editing on multiplayer games making desyncs
maybe it will be beneficial to add Signatures/checksum to the mod that will be verified against the server version of the mod to prevent inconsistencies and corruptions

That's what we do now.

EdTheMad wrote:I cant find anywhere all the changes they plan to do at 0.15.

Anyone knows where I can find that?

There isn't one as we aren't done with 0.15 yet and add things every day that we thought of the previous. Sometimes we don't like something and it gets removed or altered.

Rseding91 wrote:

EdTheMad wrote:I cant find anywhere all the changes they plan to do at 0.15.

Anyone knows where I can find that?

There isn't one as we aren't done with 0.15 yet and add things every day that we thought of the previous. Sometimes we don't like something and it gets removed or altered.

Can you quickly list things that will probable be in 0.15?

@EdTheMad
The roadmap thread contains the plans they have, should be up to date:
viewtopic.php?f=3&t=678

Wow, major props to you and everyone else for those performance improvements! That is a massive improvement going from 30 to 55 or 60. I think when 0.15 hits, a new era of mega bases will begin. I am so damn excited for this stuff!

The train looks magnificent as well. Albert always does good work. Will be interesting to see how it actually ends up in game.

Out of curiosity, what parts of the game do you still need to test for performance improvement? Would help with knowing what type of maps or saves to send in.

Rseding91 wrote:I took a few larger save files and started looking at them and every time I would re-run the profiler some small piece of code would show up taking 1-2% of the time. I'd spend 20~ minutes re-writing it and then it takes < 0.1% of the time (not accurately measurable by the profiler). I did that for 3 days and when I finished with the saves I had the end result was almost a 2x speedup over 0.14. Our test suite was invaluable during all of this to make sure I didn't break anything in the name of performance.

You're awesome. That is fervent commitment I can genuinely appreciate.

Don't forget, that belt optimisations and inserters still aren't there.

kovarex wrote:Don't forget, that belt optimisations and inserters still aren't there.

Inserters already take far less time than they used to with the optimizations I've done Belts are going to be the real noticeable improvement once that's done.

"a test render to see how the new train models will hold up in higher resolution settings"

Myep, looks like they're holding up

Can I has that sample picture as a background preaseeeeeee!!!!

Jonathan88 wrote:AMAZING!!!
Imagine if you (the devs) had seen this picture from the future when you first started Factorio... Factorio has progressed an almost incomprehensible amount!

Please developers make wallpaper from that image

Great job!

Not sure if you noticed but that picture is a link to very big version of the picture
You can do save as in most browsers and rescale it a bit to your resolution.

Marconos wrote:Can I has that sample picture as a background preaseeeeeee!!!!

Even though the general composition of this image is not made for such format, I've prepared a fast version for 1920x1080 px screens.
Please, do not take this image as an official Factorio wallpaper, because the official ones will be (hopefully) nicer. Anyway here's the link:
https://dl.dropboxusercontent.com/u/603 ... 1080px.jpg

EDIT: image with real 1920x1080px size. (sorry)
https://dl.dropboxusercontent.com/u/603 ... 1080px.jpg

what about mods and performance?!

me and few friends are playing factorio on vps (at wedos). when we were only playing vanilla (0.13 back then), our limit was basicly only memory. cpu usage at about 20 %.
after that i changed vps: same cpu performance, double the memory, updated to 0.14 and added bobs mods. now we have about 50 % memory used, but the cpu can't keep up anymore.

i know that bobs mods is a large modpack, but it mostly only contains new items, liquids, researches and few buildings. i would have hoped for it not to be such a show stopper
i have tried other mods, like anthroposomething, which changed bitter's ai. i can understatnd why this mod uses a lot more cpu, but i dont understand it with the bobs mods.

or is this problem entirely on the mod's developers?
or is it a lua limitation?

Really looking forward to the mod settings API

I love the fact that the fastest game on my (crappy) computer is getting even faster!

The train graphics look awesome - I just hope 0.15 still runs on my crappy 512MB graphics card (even if it means using the lower resolution graphics).

I assume we are allowed to plaster that amazing render all over the internet?

Do you have any thoughts on the mod config system's ability to turn features on and off? That is, I have seen quite a few mods that have "This mod has five different functions, turn the ones you want on/off in the config file". As far as I know, the game's resources are more or less fixed upon startup, so being able to (for example) remove a technology at run-time seems impossible?

Not saying that it wouldn't still be really useful, even if it's "just" limited to providing mods with a runtime list of key-val pairs -- just wondering about your thoughts.

The ability to specify settings for the mods sounds awesome! I also can read on the roadmap the following:

Allow loading games with different mod settings (and automatically download the mods if necessary)

The automatic downloading of lacking mods for online multiplayer is definitely a must. I am really too lazy to join parties with mods since it requires me to manually install all the needed mods, one by one.

Also, about the optimizations, one thing that would be really worth checking is the train mini-map. When you open it everything is so laggy!

Mod settings is a really great idea, I really love it!

I just had a couple of related thoughts. I'm pretty sure that you already considered this or even have this planned, but here goes.

1. Support for having multiple version of the same mod installed and automatic switching of those when required.

For example: you play two different multiplayer sessions and they both have the mod X, but of the different versions - 0.x.a and 0.x.b because they started half a month apart. And also you have a singleplayer session with the same mod of version 0.x.c (because its the latest one and there is THAT feature you always wanted).

You obviously can manually switch them when required and can get older mod versions from mod portal. But how about we automate those things? Just keep different versions in the mods directory and have an additional drop-down menu/list in the mod selection screen to select one of the installed versions to play.

Also making mod settings separate from the mods will make this change a lot less messy.

2. Automated mod downloading.

Right now game just tells us about mod mismatch (if there is any) and does not join the MP session. How about we make it try to deal with that? It can:

- Try to download missing mod of required version from the mod portal
- Try to enable/diasable mismatching mods
- Download required mods from the server itself if there is no such mod (or mod version) on the mod portal.

About the last one: the probably should be an option to restrict big mods/loads of mods from being downloaded to save server bandwidth. At such event joining player will be provided with corresponding message and (if the server manager specified them) with the directions to how to obtain required mods otherwise. This will also encourage server starters to add server-specific mod changes as a separate mods and not a modifications to huge base mods.

I believe that those two changes (three with already planned mod settings) will bring joining any MP game as close as it possible to simple selection from list/IP copy-paste and a single click on the Connect button at any circumstances. This will truly be awesome.