PyBlock (alpha)

pyanodon's mods are here

Moderator: pyanodon

kingarthur
Smart Inserter
Smart Inserter
Posts: 1459
Joined: Sun Jun 15, 2014 11:39 am
Contact:

Re: PyBlock (alpha)

Post by kingarthur »

mrvn wrote:
Sun Jul 26, 2020 9:45 pm
kingarthur wrote:
Sun Jul 26, 2020 9:08 pm
mrvn wrote:
Sun Jul 26, 2020 8:39 pm
I just started a new game too.

I've run into a problem crafting automation science. For that I need incubated petri dish, which is unlocked by composting, which requires automation science.

Same problem with methanol being locked behind methanol processing.
No it isn't and you don't need methonal anywhere in the first two science packs
My bad. Helmod said I need methanol but that the "fuel" for the glassworks. I can change that.
Ya. You can also click the py button at the top left for a list of all the liquid fuels and there fuel values.

Did you find the petri dish? It should be unlocked from start unless so other mod broke something.

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Re: PyBlock (alpha)

Post by mrvn »

kingarthur wrote:
Sun Jul 26, 2020 10:12 pm
mrvn wrote:
Sun Jul 26, 2020 9:45 pm
kingarthur wrote:
Sun Jul 26, 2020 9:08 pm
mrvn wrote:
Sun Jul 26, 2020 8:39 pm
I just started a new game too.

I've run into a problem crafting automation science. For that I need incubated petri dish, which is unlocked by composting, which requires automation science.

Same problem with methanol being locked behind methanol processing.
No it isn't and you don't need methonal anywhere in the first two science packs
My bad. Helmod said I need methanol but that the "fuel" for the glassworks. I can change that.
Ya. You can also click the py button at the top left for a list of all the liquid fuels and there fuel values.

Did you find the petri dish? It should be unlocked from start unless so other mod broke something.
Yes, the incubated petri dish is indeed unlocked. It doesn't show up in the hand crafting menu and helmod shows it as locked too. But in the right building the recipe is there. You just have to find the right building.

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Re: PyBlock (alpha)

Post by mrvn »

Would it be welcome if I write a mini tutorial for getting started with pyblock? I'm thinking of 4 parts:

1) Seaweed generation (Build the plant, fish some seaweed out of the ocean, get it running)
2) Wood production
3) Landfill
4) Iron/Copper plate production

That basically covers the part where if you screw up you have to start over.

PS: You can also screw up at the start and use up all the landfill without reaching all the crashed spaceship parts.

kingarthur
Smart Inserter
Smart Inserter
Posts: 1459
Joined: Sun Jun 15, 2014 11:39 am
Contact:

Re: PyBlock (alpha)

Post by kingarthur »

mrvn wrote:
Mon Jul 27, 2020 12:35 am
Would it be welcome if I write a mini tutorial for getting started with pyblock? I'm thinking of 4 parts:

1) Seaweed generation (Build the plant, fish some seaweed out of the ocean, get it running)
2) Wood production
3) Landfill
4) Iron/Copper plate production

That basically covers the part where if you screw up you have to start over.
go for it. post it to reddit or something as well if you want so more people will see it.
mrvn wrote:
Mon Jul 27, 2020 12:35 am
PS: You can also screw up at the start and use up all the landfill without reaching all the crashed spaceship parts.
ya there have been dozens of tests run and so far it should be enough to get you going. if someone fail to take stock of their limited resources and plan for it thats on them.

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Re: PyBlock (alpha)

Post by mrvn »

kingarthur wrote:
Mon Jul 27, 2020 12:42 am
mrvn wrote:
Mon Jul 27, 2020 12:35 am
Would it be welcome if I write a mini tutorial for getting started with pyblock? I'm thinking of 4 parts:

1) Seaweed generation (Build the plant, fish some seaweed out of the ocean, get it running)
2) Wood production
3) Landfill
4) Iron/Copper plate production

That basically covers the part where if you screw up you have to start over.
go for it. post it to reddit or something as well if you want so more people will see it.
I mean an in-game tutorial. Like when you research trains you get a little pop-up telling you about the trains tutorial.
kingarthur wrote:
Mon Jul 27, 2020 12:42 am
mrvn wrote:
Mon Jul 27, 2020 12:35 am
PS: You can also screw up at the start and use up all the landfill without reaching all the crashed spaceship parts.
ya there have been dozens of tests run and so far it should be enough to get you going. if someone fail to take stock of their limited resources and plan for it thats on them.
It is enough unless you set the landfill size large and click badly. It's easy to use up over hundred landfill in a single click. Second time around I kept the default size and collected everything by building little paths. You end up with 1k landfill then. Still easy to spend it all and then not being able to reach seaweed when you learn you need to get some.

jindory94
Manual Inserter
Manual Inserter
Posts: 1
Joined: Sat Jul 18, 2020 3:21 pm
Contact:

Re: PyBlock (alpha)

Post by jindory94 »

With the recent pyal update, biomass fuel has been reduced to 1/4. Fuel consumption and power consumption are unbearable, but it is very difficult to play the game due to a power outage.

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Re: PyBlock (alpha)

Post by mrvn »

jindory94 wrote:
Sun Aug 02, 2020 8:40 am
With the recent pyal update, biomass fuel has been reduced to 1/4. Fuel consumption and power consumption are unbearable, but it is very difficult to play the game due to a power outage.
On thursday I run into a brownout death spiral myself. Reinforced to me again the importance of building a dedicated, self sustaining and self powering power plant that is completley cut of from the rest of the electrical grid. The rest of the base has to use different steam engines or combustion turbines or whatever generators. That way a brownout of the base will not cause the power plant to death spiral.

I#m now designing an auog generator power plant. Looks promising from a size and power surplus direction.

kingarthur
Smart Inserter
Smart Inserter
Posts: 1459
Joined: Sun Jun 15, 2014 11:39 am
Contact:

Re: PyBlock (alpha)

Post by kingarthur »

I'm aware biomass was changed. Im the one that changed it. Unless your using biomass directly in steam engines for some godawful reason it shouldnt have had a major effect on power.

eatingburger
Manual Inserter
Manual Inserter
Posts: 4
Joined: Thu Aug 15, 2019 1:54 am
Contact:

Re: PyBlock (alpha)

Post by eatingburger »

Hey,

I've been trying to get into the modpack; I love pymods, I love seablock, I love spreadsheets!

I've been trying to plan out earlygame, but have noticed that powering glass furnaces is a ***massive*** bottleneck (I cannot emphasize how little coal gas is produced early game relative to the coal gas demand of red science). It seems like to solve this, you need to expand coal gas production (unless I'm mistaken, and there's a more efficient solution). It seems like the way to solve *that* is massively expanding wood or seaweed production (I prefer wood, but it seems like there honestly isn't much difference either way).

Both these require iron and copper. Iron oxide easily covers iron needs, but copper production through tar is just so ridiculously slow, and the tar production line elements (both the factories and inserters) are extremely copper hungry. I looked into using fish as an alternative, but it is also an equally slow, expensive solution.

Am I missing something? Is early game supposed to just be hours and hours of waiting for coal gas and copper to accumulate?

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Re: PyBlock (alpha)

Post by mrvn »

eatingburger wrote:
Fri Aug 07, 2020 1:56 pm
Hey,

I've been trying to get into the modpack; I love pymods, I love seablock, I love spreadsheets!

I've been trying to plan out earlygame, but have noticed that powering glass furnaces is a ***massive*** bottleneck (I cannot emphasize how little coal gas is produced early game relative to the coal gas demand of red science). It seems like to solve this, you need to expand coal gas production (unless I'm mistaken, and there's a more efficient solution). It seems like the way to solve *that* is massively expanding wood or seaweed production (I prefer wood, but it seems like there honestly isn't much difference either way).

Both these require iron and copper. Iron oxide easily covers iron needs, but copper production through tar is just so ridiculously slow, and the tar production line elements (both the factories and inserters) are extremely copper hungry. I looked into using fish as an alternative, but it is also an equally slow, expensive solution.

Am I missing something? Is early game supposed to just be hours and hours of waiting for coal gas and copper to accumulate?
It's a grind for sure. I build out my wood production to 3 sets of 5 forestries and a wood cutter, which maxes them out. All of that wood gets turned into coal and coal into coke. The coke then fires the boilers and steam engines to power my base. Once I had logistics researched I used a priority splitter to turn coke into ash for a third source of caol gas. And still my glass production is starving for fuel.

Using coal gas to max out one glass furnace (just the liquid glass) takes:
- 5 Automated screeners (quartz)
- 5 steam powered ashers (sand)
- 10 steam powered sil extractors (soil)
------
- 15 crude distructive distilleries (going all the way to ash)
- 16 steam powered wood cutter
- 79 Slowwood forestries
- 47.43 coal/s for the necessary steam and some more for power (1.5MW)

That doesn't even count in the 3 glass furnaces needed for glass, petri dishes and bottles.

Note that coal gas has a fuel value of 0.2MJ and tar has a fuel value of 0.2MJ as well. The above produces 50 coal gas per second and 78.3 tar per second. You could more than half the build by using tar to make glass as well.


But I think a better approach is to research Steel (50 packs) and Coal Processing 1 (30 packs). This opens up a recipe for tar using stone and gives you a smaller factory for glass:
- 5 Automated screeners (quartz)
- 5 steam powered washers (sand)
- 10 steam powered soil extractors (soil)
------
- 1 (0.4 actually) crude distructive distilleries (tar)
- 4 steam powered washers (stone)
- 7 steam powered soil extractors (soil)
- 15.8 coal for the necessary steam and some for power (1.5MW)

That doesn't sound so bad, does it?

So my recommendation is to A) set up a glass furnace and quartz as soon as possible. You can buffer quite a bit of liquid glass and there is no point in throwing coal gas away, and B) really just use that to research Steel + Coal Processing. Keep a coal gas powered furnace, add a second tar powered one and switch the other glass work over to tar.

PS: electric versions for the washer and soil extractor are twice as fast, distilleries too.

PPS: Since you have no overflow walves build the factory in a line (as far as the tar pipe goes): furnaces, ore production, coal from wood, some bit of pipe to hinder flow, tar. That way ore production will suck up all the tar it can get and your coal production can get rid of all its tar. The furnaces will mosty just get leftover tar and the tar production won't be able to swamp the pipe blocking coal production.

eatingburger
Manual Inserter
Manual Inserter
Posts: 4
Joined: Thu Aug 15, 2019 1:54 am
Contact:

Re: PyBlock (alpha)

Post by eatingburger »

mrvn wrote:
Fri Aug 07, 2020 5:26 pm
It's a grind [to get to red science] for sure. I build out my wood production to 3 sets of 5 forestries and a wood cutter, which maxes them out. All of that wood gets turned into coal and coal into coke. The coke then fires the boilers and steam engines to power my base. Once I had logistics researched...[snip]

[unsnip]...
Using coal gas to max out one glass furnace (just the liquid glass) takes:
- 16 steam powered wood cutter
- 79 Slowwood forestries
...[snip]

[unsnip]...But I think a better approach is to research...[snip]

That doesn't sound so bad, does it?
Yes, that does sound bad.

The coal gas requirements of the glass furnaces (a requirement for basic research) are currently insanely skewed relative to the balance of this mod, as your calculations imply.

One good solution might be to add a more efficient early game way to convert Coke into coal gas, or to allow the glass furnaces to be fueled by Coke.

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Re: PyBlock (alpha)

Post by mrvn »

One problem is that you need glass sheats, glass bottles and glass petri dishes. Glass bottles aren't so much a problem because they require stoppers but the rest has a stacksize of 500. They will just keep producing and producing eating up all your liquid glass unless you disconnect them. Or have just one furnace and switch recipes? But then I'm always afraid of loosing liquid glass and fuel.

The 80 red science to get better tar aren't so bad. That's 13.333 recipes for the packs or 53.333 bottles. Ok, 533.333 recipes for the liquid glass. Maybe it is bad. Still once you have those things get better.

Maybe an alternate two step process: quartz -> mallable glass (furnace) and mallable glass -> sheet / bottle / petri dish (handcraft only)? Mallable glass would not be so hot as liquid glass so it could use less fuel. And you save the fuel to turn liquid glass into the finished product.

Or just unlock the tar recipe from the start?

eatingburger
Manual Inserter
Manual Inserter
Posts: 4
Joined: Thu Aug 15, 2019 1:54 am
Contact:

Re: PyBlock (alpha)

Post by eatingburger »

mrvn wrote:
Fri Aug 07, 2020 6:46 pm
The 80 red science to get better tar aren't so bad. That's 13.333 recipes for the packs or 53.333 bottles. Ok, 533.333 recipes for the liquid glass. Maybe it is bad. Still once you have those things get better.
From what I recall from my calculations, each red science requires around 1,700 units of Coal Gas.

During casual play, over the course of three hours, I have accumulated enough copper for about five forestry facilities (during normal experimentation I naturally used Copper to try out other facilities after getting a basic power loop down).

Five forestry facilities produce:

Code: Select all

3 Coal Gas/Sec
.05 Copper/Sec
Waiting time for new forestry facility:

Code: Select all

Forestry Facility: 75 Copper / .05 Copper/Sec = 25 minutes
2 Inserters: 8 Copper / .05 Copper/Sec = 3 minutes
Total: 30 minutes for new forestry facility
Waiting time for 80 red science:

Code: Select all

1700 Coal Gas / 3 CG/Sec = 10 minutes per red science
80 * 10 = 800 / 60 = 13 hours waiting 
If I get one new forestry facility on average every ~20 minutes or so, production will be doubled by the two hour mark, and quadrupled by the three or four, so it's more like six hours waiting for enough coal gas to accumulate. Note that *tonnes* of unusable Coke is accumulating during this time.

So yeah, some way to better use Coke to produce coal gas, or a better way to power Glassworks would be nice.

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Re: PyBlock (alpha)

Post by mrvn »

eatingburger wrote:
Fri Aug 07, 2020 9:13 pm
Note that *tonnes* of unusable Coke is accumulating during this time.
Coke can be distilled again into more coal gas, more tar, more oxidized iron and ash. You don't get a lot out of it, even less ash and ash stacks to 1000. So you can just put it in a chest for hours and hours without problems.

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Crash when other surfaces are used

Post by mrvn »

The event handler for events.on_chunk_generated does not check the surface but assumes it is nauvis. This leads to it trying to change tiles in chungs that are eigther old or don't exist at all.

In control.lua: 264 add:

script.on_event(defines.events.on_chunk_generated, function(event)
+if event.surface.name ~= "nauvis" then return end


By the way, why aren't you using a noise expression to generate a map with just water except in the origin? Would avoid having to remove tiles and entities completly.

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Re: PyBlock (alpha)

Post by mrvn »

I just took a look at early power generation.

First generation power will be slowwood forestry -> steampowered woodprocessing unit -> Crude destructive distilation column -> boiler -> steam engine. You can build this right from the start.

Second generation comes with researching Energy generation 1. You turn Coal gas into Combustion mixture in a Power house and then burn that in a Combustion turbine. The conversion needs either coke or biomass. You can get biomass from the left over steam from the power house. So that sounds really great.

But I just checked out the numbers in helmod and wasn't impressed. I've made a setup with 4 power houses for a total of 200 Combustion mixture per second at 500°C. A steam turbine uses 60/s to produce 12.2 MW of energy (500°C is only half the max temp).

200 / 60 * 12.2 = 40.666 MW

To produce all the coal gas 23.81 wood are turned into 23.81 coal, 23.81 coal into 14.29 coke and 14.29 coke into ash. The thing is that if you put 23.81 coal into boilers you get:

23.81 * 4 (MJ per coal) / 1.8 (MJ consumption per boiler) * 0.9 (MW per steam engine) * 2 (per boiler) = 95.240 MW

So you get more than twice the power by sticking with boilers and steam engines compared to the more advanced Combustion turbines. And then you still have some coal gas left over (23.81/s) that you could put into a power house for another ~7.2MW.

So for power generation the Energy generation 1 seems to be a big bust Only benefit I can see is that converting coal gas to electricity and using that to power a "soil -> stone -> tar -> glass" chain with the extra goal gas. You get more glass that way.

Seems a bit unbalanced to me.

Note: 23.81 wood/s is produced by 80 Fastwood Forestry mk1 producing logs from water. Which is the best you can do at that tech level. Costs you ~30MW. So the actual energy gain is 10MW vs. 70MW.

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Re: PyBlock (alpha)

Post by mrvn »

I found a better way to make energy (or rather coal-gas). Creating it from wood simply is a bad idea. But you can separate tar into coal-gas and flue-gas:
power.png
power.png (539.07 KiB) Viewed 4445 times
I waste the steam because the steam powered soil extractor and washer have half the speed, so you would need 33 + 20 respectively. That part is already the biggest part of the power plant and I don't want to double it's size. So that's why the cooling towers at the bottom (which should be 5, not 10.82 so they consume 1000 steam).

Half the list is just a mini wood plantation. You probably have a big one somewhere where you can steal the 1.47 wood/s. Probably a source for 0.31 coal/s too. But I wanted to make it self contained hence the long list.

Technologies needed:
- Steel processing
- Coal processing 1
- Energy generation 1
- Cooling tower
- Composting + Biotechnology + Xenobiology + Mining machines + Engine)
- Botany (optional, Slowwood Forestry works fine too)

Power: production: 61 MW, internal consumption: 11.7 MW, usable gain: ~50 MW

Sopel
Long Handed Inserter
Long Handed Inserter
Posts: 61
Joined: Mon Sep 24, 2018 8:30 pm
Contact:

Re: PyBlock (alpha)

Post by Sopel »

I think it's possible to do better with soil -> stone -> tar -> copper/iron ore + tailings -> coal-water slurry -> aromatics + benzene -> combustion mixture (with biomass provided by logs made from water and co2). I think it's less buildings and uses less power.

fractalman
Long Handed Inserter
Long Handed Inserter
Posts: 80
Joined: Fri Aug 05, 2016 10:07 pm
Contact:

Re: PyBlock (alpha)

Post by fractalman »

eatingburger wrote:
Fri Aug 07, 2020 1:56 pm
Hey,

I've been trying to get into the modpack; I love pymods, I love seablock, I love spreadsheets!

I've been trying to plan out earlygame, but have noticed that powering glass furnaces is a ***massive*** bottleneck (I cannot emphasize how little coal gas is produced early game relative to the coal gas demand of red science). It seems like to solve this, you need to expand coal gas production (unless I'm mistaken, and there's a more efficient solution). It seems like the way to solve *that* is massively expanding wood or seaweed production (I prefer wood, but it seems like there honestly isn't much difference either way).

Both these require iron and copper. Iron oxide easily covers iron needs, but copper production through tar is just so ridiculously slow, and the tar production line elements (both the factories and inserters) are extremely copper hungry. I looked into using fish as an alternative, but it is also an equally slow, expensive solution.

Am I missing something? Is early game supposed to just be hours and hours of waiting for coal gas and copper to accumulate?
Produce more coal gas from doing wood->coal instead of seawead->raw coal, and use coal->coke to sacrifice a tiny bit of fuel value to get more coal gas and tar. OR, use raw coal to coal to sacrifice a significant amount of fuel value in exchange for more coal gas and tar. (you can also, as mentioned, sacrifice excess coke for more coal gas and tar, but I find this to generally be a waste, what with how many things are hungry for fuel, esepcially in the early game...)


Stone-> tar is better than wood to coal coke is better than seaweed to raw coal, in terms of the tar/coal gas/fuel production you get from a given amount of iron and copper you put into the buildings. BUT, of course, stone->tar requires research, and the wood to coal to coke method has more things that can be screwed up with it. Seaweed is the worst, but its incredibly reliable. As long as you have a single piece of fuel *somewhere* in the base, you can reboot this method by hand. logs->wood->coal->coke is the best you can do right from the get go, but it's got a lot of cyclic dependencies that can go wrong.



By best and worst, I mean in terms of fuel/tar per second compared to the input ingredients-iirc wood gives something like 2-3x as much?



And as others repeatedly mentioned, you can also use TAR for fuel instead of just coal gas.

>I think it's possible to do better with soil -> stone -> tar -> copper/iron ore + tailings -> coal-water slurry -> aromatics + benzene -> combustion mixture (with biomass provided by logs made from water and co2). I think it's less buildings and uses less power.
I think THAT requires science 2 (tailings->coal water slurry), or at least an awful LOT of red science, by which time you probably have plenty of iron and copper and can pick and choose whatever power method you please without concern for how much raw mats this'll take. (hydrogen combustion mixture isn't great but its sheer simplicity gives it a lot of appeal, for instance.)


I found a better way to make energy (or rather coal-gas). Creating it from wood simply is a bad idea. But you can separate tar into coal-gas and flue-gas:
If you just want liquid energy for your glass furnaces, as the original comment that started this was about, and have that kind of research, your best option is to turn coal gas into tar and *syngas*, which more than doubles the fuel value you get out of it.

mrvn
Smart Inserter
Smart Inserter
Posts: 5682
Joined: Mon Sep 05, 2016 9:10 am
Contact:

Re: PyBlock (alpha)

Post by mrvn »

We started talking very early game glass production, just at the start. It's crucial to consider the research cost to get to any better method. I think stone -> tar is the best option there to aim for, even just as stepping stone to better ways.


Then I mentioned power production through combustion fluid. Looks similar but it's a different topic. You need something that can be turned into electricity.

What do you need to split coal gas into tar and syngas (and then syngas into combustion fluid to get electricty)? Isn't that Energy generation 2? That would be the next step up from splitting tar into coal gas + flue gas then.

Post Reply

Return to “PyMods”