21 replies to this topic

[Vermaxx]

My friend says even SHS have diminishing returns. He didn't do math, so it was posted here or somewhere else already.

Where do these secret numbers start, how rapidly do they diminish, ect?

[Quxudica]

Vermaxx, on 31 October 2012 - 09:15 AM, said:

My friend says even SHS have diminishing returns. He didn't do math, so it was posted here or somewhere else already.

Where do these secret numbers start, how rapidly do they diminish, ect?

Each normal Heat Sink is 0.1 heat per second (if I remember correctly), there is no diminishing return. Every sink will always add that much to your cooling as far as I am aware. You realistically cannot get enough sinks to reach a point where a DR mechanic would be necessary anyway. At least, not and still have room/weight for a load out that actually needed the cooling.

[Vermaxx]

Yeah it was starting to sound like he was speed reading things to argue with me on anyway.

[stjobe]

Here's a post from the devs about how heat sinks (and double heat sinks) work - or are supposed to, there's a bug which is also outlined in that post.

AFAIK there's no diminishing returns on heat sinks.

[Lonestar1771]

there was a nice post from some guys in the beta forums just the other day but since we can't access that stuff anymore....

[Vermaxx]

Yeah but the Doctorate Level Math thread was all about doubles. My mate just claimed SINGLES were secretly forking us all too, as a reason why I should buy DHS right now.

[Sign]

Lonestar1771, on 31 October 2012 - 09:27 AM, said:

there was a nice post from some guys in the beta forums just the other day but since we can't access that stuff anymore....

Yep, they basically did several tests and concluded that HSs dissipate less heat per unit the more of them you mount, irregardless of DHS.

I didn't catch the later pages though, so I don't know if they reached another conclusion afterwards.

[gregsolidus]

That's not what many players posted about on the beta page, some guy named something Helmner and a few of his friends sat down and manged to rangle out the exact formula for the diminishing returns. I don't remember it but I do remember that the gains for more than 20 heat sinks is so negligible that your just wasting space.

[Vermaxx]

"During the testing of the DHS bug we uncovered a long standing heat related bug. This is also going to be addressed in the next patch. Expect some widespread changes. I will fill you later when the exact numbers are tested.?

From Ekman's post (in which he also admits that DHS are SUPPOSED to double everything in the engine too, so no waffling PGI), apparently there are major issues with the heat system like the thread where someone said pulse lasers aren't adding the extra heat like they're supposed to.

If half the weapons in the game are too cool right now, I wonder if people will still oppose the doubling of engine sinks?

[Twosteps]

During the testing of the DHS bug we uncovered a long standing heat related bug. This is also going to be addressed in the next patch. Expect some widespread changes. I will fill you later when the exact numbers are tested.


I'm really curius what that second long standing bug is. Can't wait for details.

[Vlad Ward]

Bugs aside,

The effectiveness of each individual heatsink has constant returns.

The total heat effectiveness of your mech has geometrically decreasing returns.

All of this is basic arithmetic and doesn't require a doctorate.

Oh, and as I recall the thread originally proposing a logarithmic reduction in heat efficiency was only taking into account the Mechlab Heat Efficiency rating rather than the heat efficiency in-game, which is why they were getting roughly geometric (logarithmic would look very similar at this scale) results.

Edited by Vlad Ward, 31 October 2012 - 09:36 AM.

[Vermaxx]

Vlad Ward, on 31 October 2012 - 09:35 AM, said:

Bugs aside,

The effectiveness of each individual heatsink has constant returns.

The total heat effectiveness of your mech has geometrically decreasing returns.

All of this is basic arithmetic and doesn't require a doctorate.

Oh, and as I recall the thread originally proposing a logarithmic reduction in heat efficiency was only taking into account the Mechlab Heat Efficiency rating rather than the heat efficiency in-game, which is why they were getting roughly geometric (logarithmic would look very similar at this scale) results.

Care to elaborate?

[Vlad Ward]

Vermaxx, on 31 October 2012 - 09:37 AM, said:

Care to elaborate?

If you have 1 heatsink and you add 1 more, you just doubled your heat dissipation ability.

If you have 2 heatsinks and you add 1 more, you just increased your heat dissipation ability by 50%.

If you have 3 heatsinks and you add 1 more, you just increased your heat dissipation ability by 33%.

Geometrically decreasing returns on total heatsinks. That is pretty standard for constant per-heatsink returns.

[Sign]

Vlad Ward, on 31 October 2012 - 09:35 AM, said:

Bugs aside,

The effectiveness of each individual heatsink has constant returns.

The total heat effectiveness of your mech has geometrically decreasing returns.

All of this is basic arithmetic and doesn't require a doctorate.

Yes, but the tests these guys conducted showed, iirc, that for example 10 hs dissipate 10h/s, while 20 hs dissipate 15h/s instead of 20h/s as they should. I think the cut-off was around 15. They might have reached another conclusion later in the thread as i missed it, or it may be the "long standing heat bug".

[TehArgz]

gregsolidus, on 31 October 2012 - 09:32 AM, said:

That's not what many players posted about on the beta page, some guy named something Helmner and a few of his friends sat down and manged to rangle out the exact formula for the diminishing returns. I don't remember it but I do remember that the gains for more than 20 heat sinks is so negligible that your just wasting space.

That is what I gathered as well. IT was all very well done and heavily researched, but I am still very unsure as to how SHS or DHS really work out on the battlefield.

[TehArgz]

Vlad Ward, on 31 October 2012 - 09:35 AM, said:

Oh, and as I recall the thread originally proposing a logarithmic reduction in heat efficiency was only taking into account the Mechlab Heat Efficiency rating rather than the heat efficiency in-game, which is why they were getting roughly geometric (logarithmic would look very similar at this scale) results.

As I recall, they did some pretty thorough field testing to arrive at their numbers

[Vlad Ward]

Sign, on 31 October 2012 - 09:39 AM, said:

Yes, but the tests these guys conducted showed, iirc, that for example 10 hs dissipate 10h/s, while 20 hs dissipate 15h/s instead of 20h/s as they should. I think the cut-off was around 15. They might have reached another conclusion later in the thread as i missed it, or it may be the "long standing heat bug".

Are you talking about their later threads or their first one?

As I recall, their first thread didn't actually use heat dissipation in-game. If it did, I'd assume that has something to do with the recently uncovered heat bug. I'm only talking about how the system is supposed to work.

TehArgz, on 31 October 2012 - 09:41 AM, said:

As I recall, they did some pretty thorough field testing to arrive at their numbers

Yes, in their later threads, which were addressing the difference between DHS and SHS rather than the functionality of the heat system as a whole if I recall correctly. If not, as I said, I'd attribute the discrepancy to bugs.

[Vermaxx]

Well, I want to live in a world where every HS does either .1 or .2/sec like it claims and I will happily accept that adding one heatsink doesn't do more than add one of those values.

I'm also very interested in how broken the heat system actually is. I've been defending the tabletop heat exchange ratio this entire time under the assumption that heat generation was just a function of how fast you fire. If there are buggy additional factors at play I may look the fool.

[TehArgz]

Thontor, on 31 October 2012 - 09:43 AM, said:

For their first post, which concluded the diminishing returns, they were only using the mechlab to test, looking at the heat efficiency scale there

They had done later posts, in which they tested in game, and found there was no diminishing returns.

Quoted for (as far as I can tell) accuracy.

[Vlad Ward]

Vermaxx, on 31 October 2012 - 09:46 AM, said:

Well, I want to live in a world where every HS does either .1 or .2/sec like it claims and I will happily accept that adding one heatsink doesn't do more than add one of those values.

This is currently the case. All is well in the world.

It's just important to point out that this can be considered a geometrically decreasing return as far as total heat dissipation ability is concerned for the reason I showed last page. It just depends on your frame of reference.