Mass discrepancy (Listed vs experimental)

[CodingKitteh]

So I have been experimenting why objects are so easily cooled with liquids/gases. Basically, a few kg of water or gas shouldn't cool a humongous machine that weighs 800 kg quickly. The answer is because the listed mass and the actual mass of the object used in calculations (derived through Q=mC∆T) is different by several factors less. I repeated the same experiments over and over again using different temperature liquids and amounts of liquids to determine the mass. Basically, when using the experimental mass, the observed temperatures can all be calculated with 0.1°C accuracy (possibly due to game rounding numbers)

Here are some of the results:

Item	Experimental Mass (g)	Listed Mass (g)	Factor Difference
Battery (Large)	80,000	400,000	5
Thermo Regulator	40,000	200,000	5
Battery (Small)	20,000	200,000	10
Hydrogen Generator	160,000	800,000	5
Natural Gas Generator	160,000	800,000	5
Copper Ore	80,000	160,000	2
Hydro Switch	40,000	200,000	5

 

So, is this intended behavior? Or are masses of objects going to be changed in future patch? Why use a smaller mass for calculations vs. listed mass?

[Kasuha]

A gas pipe segment acts as if it has 5 kg instead of 25.

Since all the factors are nice integers, it looks like it's intentional balancing touch. There's difference between material used and the thermal capacity. Though I would probably prefer if instead of hacking thermal mass, the digging yield and actual usage of materials was changed. The reported specific heat capacity values have little sense if the mass of the object doesn't match.

[CodingKitteh]

 

1 hour ago, Kasuha said:

A gas pipe segment acts as if it has 5 kg instead of 25.

Noted. This will save me time when I try to designed the most efficiency heat cooler or polluted water purifier (at little to "no" energy cost). Now it starts to make sense why gas cooling is so ineffective.

 

1 hour ago, Kasuha said:

Since all the factors are nice integers, it looks like it's intentional balancing touch. 

Hence, why I made this post instead of going straight to bug report. I do hope this consistency issue would be addressed instead of ninja fixed. I am guessing there may be a re-balancing of masses to be planned. 

 

____

I dislike seeing 800 kg as the mass of generators when only 160 kg is really used to calculate heat transfer. 

[rezecib]

Looks like this is indeed intentional, finally found the code responsible in BuildingTemplates.cs:

    instance.MassForTemperatureModification = construction_mass[0] * 0.2f;

I believe this affects all buildings, but I guess not tiles.

[CodingKitteh]

9 minutes ago, rezecib said:

Looks like this is indeed intentional, finally found the code responsible in BuildingTemplates.cs:

    instance.MassForTemperatureModification = construction_mass[0] * 0.2f;

I believe this affects all buildings, but I guess not tiles.

Nice find!

 

I did even more testing and found some more weird mass discrepancies depending on the medium of heat transfer. Liquid-vacuum-solid, liquid-gas-solid, gas-solid. I still haven't come up to conclusion yet because I am either going crazy from all that testing or the mass of objects are changing with respect to setup and transfer mediums.

If the masses are changing too much like 40 kg for a thermo regulator in a liquid-solid setup vs 160 kg for a thermo regulator in a liquid-gas-solid setup, then that's a huge problem. I still haven't concluded though because I need to do more testing and reproduce the result.

Right now, less mass is observed to be used (aside from pipes 5 kg vs 25 kg) in buildings as part of the internal calculations.

[Nidhoggur]

16 hours ago, Kasuha said:

 Though I would probably prefer if instead of hacking thermal mass, the digging yield and actual usage of materials was changed.

The digging yield is probably subject to game balance as well. Maybe that's the actual reason they made real and listed mass different.
 

5 hours ago, CodingKitteh said:

 I still haven't come up to conclusion yet because I am either going crazy from all that testing or the mass of objects are changing with respect to setup and transfer mediums.

Judging from the code above

instance.MassForTemperatureModification = construction_mass[0] * 0.2f;

Any mass should be 5 times smaller for temperature calculations, regardless of medium.

On a side note, I feel like you guys have reverse-engineered half of the game already...

[CodingKitteh]

5 hours ago, Nidhoggur said:

 

Any mass should be 5 times smaller for temperature calculations, regardless of medium.

If that's the only formula, yes. However, it doesn't explain why I got a 10 factor mass difference for the small battery. It could be my testing error (testing can get quite repetitive trying to control so many variables) however, I do try to reproduce the result before I post my observations.

I hope that it's my mistake but depending on the amount of phases present, it affected my results. For example a 5 component system, spanning 2x2 size surrounded by abyssalite tiles. Top left (TL) block is hydrogen, Top right (TR) is hydrogen, BL and BR is water. The battery starts at 20°C when you place it. 

The theory is that the system is neither gaining (the battery itself is unplugged) or losing heat (surrounded by abyssalite walls). Therefore heat loss of one component is heat gain from another. Equlibrium is reached when temperature stops fluctuating.

The beauty is that the heat can be calculated (q=m * c * dT) and because it's a computer results should be 100% accurate.

_____________

Now, I was hoping some other posters also test the thermodynamics. Perhaps maybe enough to build momentum to have Devs add input, making the game better.