Friday Facts #248 - Not Saturday Facts

[AntiElitz]

We'd really appreciate a nice Graphic for the atomic bomb, if you have some capacity left - maybe i good place to test shaders too!

[Light]

Holy shit.

It's time to start sending bags of dicks to quite a few people I know. Thanks to Wube for both Factorio and promotion for sending bags of dicks to people you hate... plus yourself.

[MeduSalem]

There are of course other small projects that are ongoing, such as improved pipe-fluid physics and improved map generation, but more on those when they are fleshed out.

Already curious how the fluid physics will turn out.

We thoroughly enjoyed shoving them down our throat.

... mkay.

[AngledLuffa]

0.17 when?

When it's ready.

Stupid question. (Stop asking the developers to make promises!!)

I don't think we're looking for a day and time. Just an estimate would be nice: days, weeks, months, end of year, etc

[flapje]

are things like FFF-228 and FFF-191 still going to be implemented or are they in the same place as spidertron?

[AndrolGenhald]

A closed-beta branch was sent to a few players to test that rendering works correctly across different hardware.

I'd love to help you guys test this out. I have the same problem as here, which isn't Factorio's fault but it'd be nice to know if it's avoided now. I have a 5 monitor setup with a GTX 780 Ti and a cheap Radeon 5450, which I'm guessing isn't all too common. I also have a GTX 970 in a Windows machine I can swap with the 780 for further testing.

[Omnifarious]

are things like FFF-228 and FFF-191 still going to be implemented or are they in the same place as spidertron?

I'm nearly positive 228 will be implemented because most of it is already available as a mod by Klonan.

[Gergely]

0.17 when?

When it's ready.

Stupid question. (Stop asking the developers to make promises!!)

I don't think we're looking for a day and time. Just an estimate would be nice: days, weeks, months, end of year, etc

I can give you one:

This year.

Wanna bet on it?

I don't think we need anything like that.

[loup-vaillant]

So they are working on the fluid physics. it's quirks made it one of the few subsystems in Factorio that didn't live up to the dev's overall standards. I'm quite delighted to learn they're working on it for real—and not just announcing they may take a look.

[Bassad]

While I appreciate that you always have the modding community in the back of your head, I would like to know what the impact of the new GUI will make on current and future mods? For example the small icons on the top left corner or other windows. Does that mean modders will need to reinvent themselfs and re-write their whole designs to adapt it to the new GUI or will it come naturally without having to redo everything (in terms of mod-gui design) ?

[5thHorseman]

I don't think we're looking for a day and time. Just an estimate would be nice: days, weeks, months, end of year, etc

One of those, for sure. Hopefully not "etc." but who knows.

[Jap2.0]

are things like FFF-228 and FFF-191 still going to be implemented or are they in the same place as spidertron?

Those should both be in 0.17.

[Jupiter]

D*cks? I only see boobs, a belly and a mons pubis.

[quyxkh]

3GB for high res? I run high res on a 2GB card and it does not lag or anything, nice steady 60/60

At 1920x1080 with no compromises or backoffs, everything on, nvidia-smi reports 2116MiB used without compression, 1809MiB with. You want a little left over for running other things, right?

Plus I'm betting they're going to use about half that last gig-ish for 0.17.

[triffid_hunter]

Anyone else irritated by the zoom glitch video having content-disposition header so the browser downloads it instead of playing it?

Code: Select all

Content-Disposition: attachment; filename=fff-248-ui%20zooming.webm

[BlueTemplar]

3GB for high res? I run high res on a 2GB card and it does not lag or anything, nice steady 60/60

I have 2GB of video memory, and it crashes on startup if I select high res graphics.
(which I guess should technically be called ultra-high res, as it seems to be designed for ultra-high resolution monitors?)
That video memory is shared though, and I only have 4GB RAM - might explain the crash?

[BlueTemplar]

As for taking mod requests,
and where fluids are concerned :
I'd like to make a "more realistic" boiler/steam engine mod, and for that I need several things :
- Be able to control when a liquid becomes a gas : check, I think? (as long as it happens in a building, and not a pipe/tank)
- Be able to heat fluids : check, I think, with boilers?
- Be able to take energy for the phase change : check, I think, with chemical plants?
- Be able to take energy for the phase change using fuel rather than electricity - this doesn't seem to be possible right now?
(- It would be nice to be able to cool fluids too, for full cycle heat engines for instance - I don't know if it can already be done)
- Be able to vary efficiency depending on fluid temperature and ambient temperature
- Be able to raise ambient temperature in a pollution-like fashion

- Maybe, with some luck, I won't need to deal with fluid volume/pressure ? (these designed to be fixed for each factory/pipe)

Some physical background for those interested - from Discord

I was wondering about making a mod that follows more closely the laws of thermodynamics, for instance having boilers/steam engines follow the Carnot efficiency of η = 1 - Tc/Th

(Tc and Th in Kelvin)
so for Tc=15+273 Th=165+273 (boiler + steam engine), that's η=34%(edited)
for Tc=15+273 Th=500+273 (nuke + turbine), that's η=63%(edited)
Or maybe this, which seems closer to real-life?
https://en.wikipedia.org/wiki/Endorever ... modynamics
(might be a bit too complex though, thermodynamics are already hard enough!)

(funny, in the example they give, the coal powerplant has a higher temperature than the nuclear one!)
(I guess because coal is safer?)
Now, if it's the steam engine/turbine that uses this efficiency calculation, wonder how then to calculate boiler efficiency (to be close to real life) ?
Otherwise there's also the issue that Factorio's Steam engines don't lose efficiency as the steam they receive gets colder...
(and there doesn't seem a way to change efficiency on-the-fly?)

----

for a "realistic" boiler, I'm probably going to run into the issue that :
In Real Life :
Water :
Specific Enthalpy at Atmospheric pressure at 1,01325 bar, i.e. normal atmospheric pressure on the sea level at 0°C.:
100°C : 0.419 MJ/kg , 0.958 kg/L => 0.401 MJ/L
15°C : 0.063 MJ/kg, 0.999 kg/L => 0.063 MJ/L
Heating 1 L of water from 15°C to 100°C => 0.338 MJ
(on average a Specific Heat of 3.98kJ/°C/L or 3.56kJ/°C/kg)
https://www.thermexcel.com/english/tables/eau_atm.htm

*Factorio : heat_capacity : 0.2kJ per degree per liter (was same for all fluids, doubled for steam and water in 0.15.0)
https://wiki.factorio.com/Version_history/0.15.0#0.15.0 *

Heat of Water Vaporization : 2.26 MJ/kg => 2.25 MJ/L
(yes, that's indeed several times more than what is required to heat it from 15 to 100 °C!)
* No equivalent in Factorio *

Steam :
Well, there's an issue here, as when Water is vaporized, its volume (and pressure) increases **significantly** (1 694 times for volume at 1 atm.!), typical Rankine cycle boilers seem to run at 30 atm. pressure?
(This is actually what creates mechanical work in a steam engine.)
https://en.wikipedia.org/wiki/Rankine_cycle
So to keep some consistency, I'm going to rather work in kg.
At constant pressure, at 1 bar (~1atm) the Specific Heat of Steam is around 2 kJ/°C/kg
With the Specific Enthalpy of Superheated at 165°C Steam of 2.81 MJ/kg. => +2.75 MJ/kg from 15°C water
With the Specific Enthalpy of Superheated at 500°C Steam of 3.49 MJ/kg. => +3.43 MJ/kg from 15°C water (1.25 times more)
http://www2.spiraxsarco.com/esc/SH_Properties.aspx
*(Again, Factorio's steam heat_capacity is now 0.2KJ per degree per... kg??
So, total, that would be :
for 165 °C steam : 0.2KJ/°C/kg*(165°C-15°C) = 30 kJ/kg from 15°C water
for 500 °C steam : 0.2KJ/°C/kg*(500°C-15°C) = 97 kJ/kg from 15°C water (3.23 times more)
)*


You can also notice that the Rankine cycle injects the water under such pressure that it boils only at ~270°C !
*(Another way in how Factorio differs...)*
Yeah, I'm going to have to redo these calculations...
The normal Rankine cycle also sends the steam to the turbine as soon as the water has been copletely vaporised (no superheated steam)
The other Rankine cycles (with superheat, reheat, regenerative) seem to have higher efficiencies, and would allow for higher tier boilers (like the current nuke unlocked turbines?) that would work *both* for realism and gameplay !

P.S.: Thanks a lot for Bilka for adding the steam engine efficiency in its tooltip and description !

[Meddleman]

Oh, and our coding always goes as planned without any problems.

Still, got a nice smooth 60fps there, so we got that going for 0.17, which is nice I guess. ^^

[nuhll]

I would really like more research, more vanilla stuff to build, more recipes, more tiers of stuff.

I know there are mods, but still, they have another "swag"?!?!! then the vanilla recipes and tiers.

Also you should really integrate this SNEI (is it corret spelled?!) in the end, just an INGAME overview over all recipes, buildings, etc. Maybe a little bit more user friendly.

[eradicator]

• Having multiple versions of a mod installed and being able to switch between them.
• Letting mods set entities to not require power.

Wohoo, awesome. One great unexpected present and finally power-free things :D.

I think the most common "hack that might be fixable" is spawning additional entities when entity x is build (i.e. invisible inserters, chests, pipes, poles etc). If it was somehow possible to specify that (including automatic removal handling) it could remove lots of scripting, especially blueprint handling for that type of hack is pretty prone to bugs.

Code: Select all

data:extend {
 type = 'assembling-machine',
 name= 'my-modded-special-machine',
 associated_entities = {
  { name  = 'invisible-inserter', offset = {-0.5,-0,75},
  { name  = 'invisble-chest', offset = {-0.5, -1.25},
  }