Why Do Hot Maps Seem To Punish Low Heat Mechs More Than Energy Boats?
#1
Posted 12 November 2012 - 11:22 PM
With my light srm mech, I go from being able to fire them all day long on a normal map, to barely even being able to cool down at all on the hot map. I heard one person that was running a light mech with no extra heat sinks and a small engine had their heat stuck at ~90%, but did not witness this first hand.
Meanwhile in a laser boat awesome, there doesn't seem to be a big reduction in heat dissipation.
My guess is that hot maps, instead of reducing the effectiveness of all your heat sinks, it is instead effectively subtracting a flat number of heat sinks from all mechs. Thus a mech with 8 heat sink -6 from the map would effectively have 2 heat sink, explaining the painfully slow dissipation, while a laser boat with 24 would be at 18, which isn't nearly as massive of a hit.
Shouldn't it just be a set reduction in heat sink efficiency? Such as 50% normal heat sink efficiency which would put the previous 2 examples at 5 and 12 respective. It just seems like a departure from the previous games where hot mechs had trouble on hot maps, but had an advantage on cold maps.
#2
Posted 12 November 2012 - 11:47 PM
Edited by Jules Gonzales, 12 November 2012 - 11:47 PM.
#3
Posted 12 November 2012 - 11:49 PM
DHS make it even easier to balance heat.
#4
Posted 12 November 2012 - 11:49 PM
Next question please.
Edited by xenoglyph, 12 November 2012 - 11:50 PM.
#5
Posted 13 November 2012 - 12:02 AM
It really becomes noticeable when you get to the shutdown threshold. Your 10 heatsink mech will take forever to cool off, while a 30 heat sink mech will shed those % points quickly. There is currently no circumstance where having few heat sinks is and advantage, more heat sinks is always better (which is why most people will convert to DHS, especially with a 250+ rated engine)
As an example, before the first (original) DHS patch, I had a mech with only 6 heat sinks. On Caustic Valley, my mech would only cool 1% heat per second, so if I ever overrode shutdown I had to wait 10 seconds for my mech to drop to 90% heat or risk blowing up. [of course now the HS minimum is 10 so the same mech now cools 500% faster (from 1 HS per 10s to 5 HS per 10s) on Caustic]
#6
Posted 13 November 2012 - 12:10 AM
With my light ERLLaser mech, I go from being able to fire them all day long on a normal map, to barely even being able to heat up at all on the cool map. I heard one person that was running a light mech with no extra heat sinks and a small engine had their heat stuck at 1%, but did not witness this first hand.
Meanwhile in a no heatsink boat awesome, there doesn't seem to be a big increase in heat dissipation.
My guess is that cool maps, instead of increasing the effectiveness of all your heat sinks, it is instead effectively adding a flat number of heat sinks from all mechs. Thus a mech with 8 heat sink +6 from the map would effectively have 14 heat sink, explaining the painfully fast dissipation, while a laser boat with 24 would be at 30, which isn't nearly as noticeable since they already run cool.
Shouldn't it just be a set increase in heat sink efficiency? Such as 50% normal heat sink efficiency which would put the previous 2 examples at 12 and 36 respective. It just seems like a departure from the previous games where cool mechs had trouble on cool maps, but had an advantage on hot maps.
Edited by ArcadiaNisus, 13 November 2012 - 12:19 AM.
#7
Posted 13 November 2012 - 03:49 AM
#8
Posted 13 November 2012 - 04:13 AM
As far as I can tell the latter would have the same visible results. I always assumed heat was more complex and relied on use of an ambient heat instead of changing the effectiveness of your heatsinks. The idea being that even before either a low or high heat build has fired a single shot, they are being heated up by the environment itself, Thus the caldara and higher base heat in Caustic.
#9
Posted 13 November 2012 - 04:32 AM
Shalune, on 13 November 2012 - 04:13 AM, said:
I seem to remember a dev describing it as such. There is a thread in the guides forum that discusses how heat on maps work more closely, based on experimentation. We'll have to check it out to figure out how things are working really perhaps?
#10
Posted 13 November 2012 - 04:36 AM
A heatstable build with only a few heatsinks suffers hard since he can not easily dissipate the heat once it is build up.
An energy boat has plenty of HS equipped -> it will still get rid of the heatb fast, just slower than used to.
#11
Posted 13 November 2012 - 04:44 AM
Kadreal, on 12 November 2012 - 11:22 PM, said:
I don't know if it is that way, but that would be canon. And therefore fine by me.
#12
Posted 13 November 2012 - 05:14 AM
Agent 0 Fortune, on 13 November 2012 - 12:02 AM, said:
It really becomes noticeable when you get to the shutdown threshold. Your 10 heatsink mech will take forever to cool off, while a 30 heat sink mech will shed those % points quickly. There is currently no circumstance where having few heat sinks is and advantage, more heat sinks is always better (which is why most people will convert to DHS, especially with a 250+ rated engine)
As an example, before the first (original) DHS patch, I had a mech with only 6 heat sinks. On Caustic Valley, my mech would only cool 1% heat per second, so if I ever overrode shutdown I had to wait 10 seconds for my mech to drop to 90% heat or risk blowing up. [of course now the HS minimum is 10 so the same mech now cools 500% faster (from 1 HS per 10s to 5 HS per 10s) on Caustic]
This is correct. Ambient tempuratures listed for maps, just act like you either have additional heat dissappation of .1-.6 Heat / sec (rainy / ice maps), or subtract off .1-.6 Heat / sec (jungle / volcano maps). This is basically the same thing as saying you reduce or increase the number of heatsinks on your mech.
Edited by CodeNameValtus, 13 November 2012 - 05:15 AM.
#13
Posted 13 November 2012 - 05:59 AM
#14
Posted 13 November 2012 - 06:04 AM
Purlana, on 13 November 2012 - 05:59 AM, said:
Actually, per my Thermodynamics class that I took many years ago, let me know if the laws of the universe have changed since I went to college. But I'm pretty sure they haven't.
This is NOT how things work. If it only just raised your heat a set amount, that's like saying, you just lower the heat threshold. And not do anything about heating/cooling. That simply isn't how the world works.
Being in a hotter environment would slow the rate of cooling on a mech, being in a cold environment would accelerate the rate of cooling on a mech. Neither would increase or decrease the maximum heat threshold.
HEY, that's exactly how they implemented it! Those PGI guys must have Thermodynamics backgrounds!
#15
Posted 13 November 2012 - 06:20 AM
CodeNameValtus, on 13 November 2012 - 06:04 AM, said:
Actually, per my Thermodynamics class that I took many years ago, let me know if the laws of the universe have changed since I went to college. But I'm pretty sure they haven't.
This is NOT how things work. If it only just raised your heat a set amount, that's like saying, you just lower the heat threshold. And not do anything about heating/cooling. That simply isn't how the world works.
Being in a hotter environment would slow the rate of cooling on a mech, being in a cold environment would accelerate the rate of cooling on a mech. Neither would increase or decrease the maximum heat threshold.
HEY, that's exactly how they implemented it! Those PGI guys must have Thermodynamics backgrounds!
Who cares about thermodynamics? Certainly not Battletech.
If we'd care about thermodynamics,wouldn't the dissipation rate not also be dependent on your heat level? The higher your heat, the faster it should dissipate?
That's not what is happenging, and not what "needs" to be happening either to make balance.
A high ambient heat level translated into the Battletech logic should more likely raise your ground heat level and overall give you less heat capacitiy to work with.
In the tabl top rules, it doesn't really matter much if you say "your heat level can never go below 3" or say "your effectiv heat sinks are reduced by 3." In a real time game, you can make a difference. And lowering the heat capacity of mechs would probably work better for low-heat-sink builds (even though it still hurts).
#16
Posted 13 November 2012 - 06:31 AM
MustrumRidcully, on 13 November 2012 - 06:20 AM, said:
If we'd care about thermodynamics,wouldn't the dissipation rate not also be dependent on your heat level? The higher your heat, the faster it should dissipate?
That's not what is happenging, and not what "needs" to be happening either to make balance.
A high ambient heat level translated into the Battletech logic should more likely raise your ground heat level and overall give you less heat capacitiy to work with.
In the tabl top rules, it doesn't really matter much if you say "your heat level can never go below 3" or say "your effectiv heat sinks are reduced by 3." In a real time game, you can make a difference. And lowering the heat capacity of mechs would probably work better for low-heat-sink builds (even though it still hurts).
Actually, this is incorrect. The higher your heat, you don't disappate % any faster. That would be like saying, if I take a pot of water that is 99 degrees out into the winter, that it would freeze faster than a pot of water that is 87 degrees. Sure you might see a slight delta in the degrees per minute, but as an overall percentage of heat, it would be the roughly the same. Nice try though!
Time to reach equilibrium tempurature is about the only factor on ambient heat on a map. Which would be why you would see yourself standing and sitting at 3% on Caustic Valley, as well as disappating heat slower. Equilibrium heat is higher, and the rate at which you disappate is based on ambient compared to your mechs heat.
#17
Posted 13 November 2012 - 06:36 AM
If they reduce the effectiveness of each heatsink: low heat, low heatsink builds aren't affected much and high heat, high heatsink builds are hurt more. Hot builds suffer more on hot maps than cold builds.
The second makes more sense. Which is currently in use?
#18
Posted 13 November 2012 - 06:42 AM
Blaank, on 13 November 2012 - 06:36 AM, said:
If they reduce the effectiveness of each heatsink: low heat, low heatsink builds aren't affected much and high heat, high heatsink builds are hurt more. Hot builds suffer more on hot maps than cold builds.
The second makes more sense. Which is currently in use?
The first. The second doesn't make sense at all. Because then, people would only play low heat, low heatsink builds, because it gives them an advantage regardless of which map they go on.
Oh wait, Gausscats don't care much about heat anyways!
#19
Posted 13 November 2012 - 07:03 AM
however if you look at it clearly you will see (and we will use an extreme example) how this works:
Mech with 20 heat sinks:
- normal = 20 cooling, weapon heat = 13, efficiency =1.5
- -.6 = 12 cooling, weapon heat = 13, efficiency =.9
- +.6 = 32 cooling,weapon heat = 13, efficiency =2.4
- normal = 10 cooling, weapon heat = 6.6, efficiency =1.5
- -.6 = 6 cooling, weapon heat = 6.6, efficiency =.9
- +.6 = 16 cooling, weapon heat = 6.6, efficiency =2.4
So some basic math there, now regardless of the actual amount of heatsinks if you mech is sitting balanced for your heat sink load (say an efficiency rating of 1.5) you will be affected identically regardless of how many heat sinks you have because you generate X heat to HS all things being equal. to be clear a mech of any weight class with a heat efficiency of 1.5 will be affected the same as any other mech in any weight class with the same efficiency.
The point is there is no benefit to "high heat mechs" the higher your heat efficiency the less affected you will be by high heat environments and the better you make out in low heat environments, and it works the same in reverse. If you try to min/max HS/Weapons you are going to be hurt by high heat and it's your own fault because heat management is part of the game.
Edited by Agent of Change, 13 November 2012 - 07:11 AM.
#20
Posted 13 November 2012 - 07:10 AM
CodeNameValtus, on 13 November 2012 - 06:31 AM, said:
Actually, this is incorrect. The higher your heat, you don't disappate % any faster. That would be like saying, if I take a pot of water that is 99 degrees out into the winter, that it would freeze faster than a pot of water that is 87 degrees. Sure you might see a slight delta in the degrees per minute, but as an overall percentage of heat, it would be the roughly the same. Nice try though!
Time to reach equilibrium tempurature is about the only factor on ambient heat on a map. Which would be why you would see yourself standing and sitting at 3% on Caustic Valley, as well as disappating heat slower. Equilibrium heat is higher, and the rate at which you disappate is based on ambient compared to your mechs heat.
Your example of freezing water is incorrect http://math.ucr.edu/.../hot_water.html. I understand what you were going for...but I thought it funny that I actually knew your example was not always correct. Mpemba effect.
7 user(s) are reading this topic
0 members, 7 guests, 0 anonymous users