68 replies to this topic

[BR0WN_H0RN3T]

Krzysztof z Bagien, on 08 July 2013 - 07:21 AM, said:

It's actually the oposite: the higher the temperature difference between objects (like hot heat sink vs cold environment), the faster heat transfers. So hot heat sink should actually cool faster and dissipate MORE heat per second.

This is true BUT the heat sink is the cold object to start with. Heat transfers to it from the hot mech. So now the HS heats up and needs to dissipate the heat somewhere. As it's heating up it works harder drawing more current and surpasses its optimal heat dissipation efficiency. Think of how an AC unit works. Similar concept. PGI have already factored heat dissipation rates into map temperature so you see that the temperature of the external environment affects HS heat dissipation rate. What pgi have NOT done is factor in mech temperature into HS dissipation rate. This is what I'm saying.

Edited by Brown Hornet, 08 July 2013 - 03:34 PM.

[BR0WN_H0RN3T]

Lugh, on 08 July 2013 - 12:56 PM, said:

I stopped reading there. Heatsinks work LESS effectively the hotter they get. Basic physics. They weren't PUNISHING enough on the running hot piece on the last patch.

Traditionally, in past MW titles and TT both, anytime you ran hotter than 80% heat You started to fry up components of your mech.

It needs to be that way now. And they need to cap it with 150% HEAT = pILOT lifesigns null. Shutting down.

I agree with u and have clarified things in my post above. When HS get to a certain internal temp they will start being inefficient. It's that simple. This is real world physics.

[Catamount]

Brown Hornet, on 08 July 2013 - 03:27 PM, said:

This is true BUT the heat sink is the cold object to start with. Heat transfers to it from the hot mech. So now the HS heats up and needs to dissipate the heat somewhere. As it's heating up it works harder drawing more current and surpasses its optimal heat dissipation efficiency. Think of how an AC unit works. Similar concept. PGI have already factored heat dissipation rates into map temperature so you see that the temperature of the external environment affects HS heat dissipation rate. What pgi have NOT done is factor in mech temperature into HS dissipation rate. This is what I'm saying.

As has been explained on the last page, this is not correct.

Technically a cold heatsink will take from the reactor a little faster than a warm heatsink, but the heatsink will still conduct heat into the environment fastest when it is hot, and it's the environment that takes the heat ultimately, not the heat sink. The heat sink reaches its tiny capacity in a brief instant; it's the transfer environment that's important. The heat sink will never get as hot as the reactor, but if the heat sink is ten times hotter vs ambient than it was when the reactor was barely heating it at all, that heat sink will dump that heat into the environment ten times faster. If the heat sink is bleeding ten times faster, then basic physics says that it has to be being fed heat ten times faster because that energy is coming from somewhere, ergo the reactor is feeding heat into the heat sink ten times faster, and so, is cooling off a lot faster from, say, 10,000k over ambient than at 1,000k over ambient.

Additionally, the Stephan-Boltzmann Law dictates that the heatsink will bleed off that heat as IR radiation according to the 4th power of its temperature, so it's not only going to be conducting ten times as much heat into the environment, but it's going to be giving off ten thousand times as much heat as IR energy, and every single one of those joules of heat has to come from the reactor.



Maybe it would help you to understand thinking about it this way: As the heat sink gets hotter, it bleeds energy into the environment faster, right? That means that the hotter the heat sink gets, the harder the reactor has to work dumping heat into it to KEEP It at that temperature.
Your mistake is thinking of it as reactor -> heatsink, instead of reactor-> environment, and it's the latter that's ultimately happening. So yes, you're right that if the reactor is 1000k and your heatsink is at only 200k, that heatsink would suck energy from the reactor faster than if it was a 500k heatsink (assuming the heatsink isn't instead staying cool because of poor thermal contact with the reactor), because it means you're running a superior heatsink that stays closer to ambient temperature, which means it's ridding itself of the reactor's energy faster, but a GIVEN heatsink is not putting off less energy at 500k than 200k, it's putting off more, and since the reactor is what has to feed that energy from its waste heat, that means the reactor is losing energy faster, which it's doing when it runs hotter. The reactor is now continuously losing enough heat to keep that heatsink at 500k instead of only losing enough heat to keep the heatsink at 200k, and because it's losing that heat at a great rate to keep that heatsink that hot, it will cool down more quickly.

You see this in CPU temperatures when working with computers. Make a modern Intel processor that idles as 30C run Prime95 and it might get up to 70C. When you shut off prime95, the temperature initially drops VERY quickly because the hot heatsink is bleeding energy into the ambient environment so quickly, so you might drop from 70 to 40 in like 3 seconds, but those last ten degrees back down to the 30C might take you another 30, because the heatsink isn't very hot, which means it's not bleeding off its energy very quickly. Don't take my word for it; try it yourself. A hot computer heatsink cools way faster than one that's merely a little warm, and you see it beacuse the chip underneaths cools way faster when it's really hot, and visibly comes down from big highs really quickly.


Edit: rephrased for more clarity

Edited by Catamount, 08 July 2013 - 08:25 PM.

[Ilithi Dragon]

Ah, I see, you're talking about the heat sink taking heat from the mech, not dumping it into the environment.

As the temperature difference between the heat sink and the internals of the mech decrease, the heat sink is able to pull less heat from the internals of the mech.

But, at the same time, as the heat sink heats up, it's able to dump heat into the surrounding environment faster and cool off faster, increasing the temperature difference between the heat sink and the internals, and increasing the rate at which the internals are cooled again.

As I understand the heat bar, it is a measure of the internal temperature of the mech, not the heat sink capacity, so if we were to model MWO after real-world physics, when that spikes up, it should initially decrease at a faster rate, because the internals are suddenly a lot hotter than the heat sinks, and then the rate should decrease over a set amount of time, proportional to the size of the heat spike, as the temperature difference between the heat sinks and the internals decreases (the heat sinks warm up and the internals cool).

As that temperature difference declines, the rate at which heat is transferred to the heat sinks would then become proportional to the rate at which the heat sinks can dump heat into the environment, which would increase the hotter the heat sinks become.

So the observational effect would be that high-heat alpha strike would see a large spike in heat when they fired that decreased rapidly at first, but slowed over a given period of time,proportional to the size of the spike and the number of heat sinks equipped on the mech.

Lower heat sustained fire would see heat levels normally at first, but at a decreasing rate as the mech got hotter, because they would be bleeding heat at or near the rate at which heat sinks could dump heat into the environment, since the rate at which the temperature difference between the internals and heat sinks increased would be slow.

I think this would still work.

The high-heat alpha builds would not be affected so much at first, because the rapid shift of heat from internals to heat sinks from the rapid spike in heat would largely negate any reduction in the heat pool, at first, but after the mech fires a few times, and the heat sinks and the internals are near the same temperature at elevated heat, the high-heat alpha builds would run into problems once their heat sinks got to the same temperature that the internals have at the build's fire-and-shut-down threshold, because that threshold would be low enough that their heat sinks wouldn't be hot enough to dump heat into the environment at a significantly accelerated rate. So, basically, the high-heat alphas could dump out fire at an accelerated rate initially, and then at a reduced rate if the fight became sustained.

The low-heat DPS builds would also not be affected at first, their heat would start to build as normal as we currently see it, but the hotter their mech got, the slower it would build, and some mechs, with the right build and/or fire management, and/or environmental conditions, would be able to effectively sustain their fire indefinitely while running at high very heat levels.

That would have to be balanced carefully to keep certain builds from becoming OP, and would require some tweaking over time to get it right, but I think it could work.

Edited by Ilithi Dragon, 08 July 2013 - 04:16 PM.

[Bloody Moon]

@OP 10-15 would be a bit too big of a change. Starting with 9-12 then checking the results from there is much better, but to be honest we'd need balance changes much more often. 2 weeks with 9-12 heat then change to 10-13 if necessary, not letting it stay for months...

[Ilithi Dragon]

Bloody Moon, on 08 July 2013 - 04:26 PM, said:

@OP 10-15 would be a bit too big of a change. Starting with 9-12 then checking the results from there is much better, but to be honest we'd need balance changes much more often. 2 weeks with 9-12 heat then change to 10-13 if necessary, not letting it stay for months...

Well, 10 and 15 is what they used to be at. I played with PPCs and ERPPCs when they were at 10 and 15 heat, respectively, and they were quite playable, from a heat management perspective. At the time, latency was horrible, so they weren't very effective weapons because the heat penalty was too high for their ability to actually hit targets, but now that latency is fixed, that's not an issue anymore.

You are right, though, that balance changes do need to come a lot faster.

Honestly, I think that's part of the reason why PGI is going to drastic measures to correct the symptoms of what are really small problems, like the OP poptart and PPC meta, because they go for so long without doing any balance tweaks that the frustration in the community becomes so high, and the community perspective of the issue becomes so skewed, that the input PGI is receiving indicates to them that drastic measures are required.

If, instead, they would just do a balance overview every two weeks and make small balance adjustments every two weeks, we wouldn't see nearly as many problems as we're seeing, and they wouldn't be pushed into taking nearly so drastic measures like they are.

[Catamount]

Ilithi Dragon, on 08 July 2013 - 04:40 PM, said:

Honestly, I think that's part of the reason why PGI is going to drastic measures to correct the symptoms of what are really small problems, like the OP poptart and PPC meta, because they go for so long without doing any balance tweaks that the frustration in the community becomes so high, and the community perspective of the issue becomes so skewed, that the input PGI is receiving indicates to them that drastic measures are required.

If, instead, they would just do a balance overview every two weeks and make small balance adjustments every two weeks, we wouldn't see nearly as many problems as we're seeing, and they wouldn't be pushed into taking nearly so drastic measures like they are.

Oh god, this. A thousand times, this. The fact that PGI let's problems fester and frustration build and don't bother to address things until after it all boils over into riots in the streets, and then massively over-react in their balance decisions to the by-that-time angry and emotional feedback over those balance problems, is exactly what leads to the problems here.

PGI ends up with things three steps too far to the left, but let's it grate on us for weeks and weeks until we all explode over it, and they panic and rush 67 steps to the right.

Edited by Catamount, 08 July 2013 - 06:42 PM.

[Asmosis]

I might disagree with everything else in the thread, but i agree with that

[BR0WN_H0RN3T]

Ilithi Dragon, on 08 July 2013 - 04:14 PM, said:

Ah, I see, you're talking about the heat sink taking heat from the mech, not dumping it into the environment.

As the temperature difference between the heat sink and the internals of the mech decrease, the heat sink is able to pull less heat from the internals of the mech.

But, at the same time, as the heat sink heats up, it's able to dump heat into the surrounding environment faster and cool off faster, increasing the temperature difference between the heat sink and the internals, and increasing the rate at which the internals are cooled again.

As I understand the heat bar, it is a measure of the internal temperature of the mech, not the heat sink capacity, so if we were to model MWO after real-world physics, when that spikes up, it should initially decrease at a faster rate, because the internals are suddenly a lot hotter than the heat sinks, and then the rate should decrease over a set amount of time, proportional to the size of the heat spike, as the temperature difference between the heat sinks and the internals decreases (the heat sinks warm up and the internals cool).

As that temperature difference declines, the rate at which heat is transferred to the heat sinks would then become proportional to the rate at which the heat sinks can dump heat into the environment, which would increase the hotter the heat sinks become.

So the observational effect would be that high-heat alpha strike would see a large spike in heat when they fired that decreased rapidly at first, but slowed over a given period of time,proportional to the size of the spike and the number of heat sinks equipped on the mech.

Lower heat sustained fire would see heat levels normally at first, but at a decreasing rate as the mech got hotter, because they would be bleeding heat at or near the rate at which heat sinks could dump heat into the environment, since the rate at which the temperature difference between the internals and heat sinks increased would be slow.

I think this would still work.

The high-heat alpha builds would not be affected so much at first, because the rapid shift of heat from internals to heat sinks from the rapid spike in heat would largely negate any reduction in the heat pool, at first, but after the mech fires a few times, and the heat sinks and the internals are near the same temperature at elevated heat, the high-heat alpha builds would run into problems once their heat sinks got to the same temperature that the internals have at the build's fire-and-shut-down threshold, because that threshold would be low enough that their heat sinks wouldn't be hot enough to dump heat into the environment at a significantly accelerated rate. So, basically, the high-heat alphas could dump out fire at an accelerated rate initially, and then at a reduced rate if the fight became sustained.

The low-heat DPS builds would also not be affected at first, their heat would start to build as normal as we currently see it, but the hotter their mech got, the slower it would build, and some mechs, with the right build and/or fire management, and/or environmental conditions, would be able to effectively sustain their fire indefinitely while running at high very heat levels.

That would have to be balanced carefully to keep certain builds from becoming OP, and would require some tweaking over time to get it right, but I think it could work.

I would support such a mechanic. We're approaching it from different perspectives of how heat dissipation should work, but I think we're onto the same outcome. I think the answer might be to do with dissipation drop-off that doesn't follow a linear curve. e.g. 10 heat sinks dissipate normally up to between 76-100% heat, then dissipate at lower capacity at say 51-75% heat, then lower again at 26-50 and perhaps are half as effective at 1-25%. Using your aproach this could be one answer, but I'd rather HSs work normally at lower tolerance levels to penalise alpha striking with high heat weapons i.e. 100% HE @ 1-25% heat, 90% HE @ 26-50% heat, 80%HE @ 51-75% heat, 70% HE @ 76-100% heat and 50% HE between 101-125% and there is an elevated risk of HS failure (destruction) above 125%.

Catamount, on 08 July 2013 - 06:41 PM, said:

Oh god, this. A thousand times, this. The fact that PGI let's problems fester and frustration build and don't bother to address things until after it all boils over into riots in the streets, and then massively over-react in their balance decisions to the by-that-time angry and emotional feedback over those balance problems, is exactly what leads to the problems here.

PGI ends up with things three steps too far to the left, but let's it grate on us for weeks and weeks until we all explode over it, and they panic and rush 67 steps to the right.

I call this "pendulum balancing." Seriously tho, I really wonder if they play their own game. You'd have to be a complete ignoramus to not see the effects of any proposed change before it's implemented. In the early days this could be forgiven, but now, we know what will happen. We have foresight.

[Asmosis]

Well the only changes that really contributed to the current situation were HSR and projectiles speed. Heat and cooldown really have nothing to do with the current situation, PPC's were complete junk prior regardless of their heat because you couldnt hit anything past about 400m anyway. Now people can lead a light mech at 1km where that used to be a dream unless you lived next to the server.

Prior the projectiles were slow enough you could walk out of the way at longer ranges, they were about as easy to dodge as lrms. 1200ms i think it was originally making ER PPC's useless for long range. Combine faster velocity with HSR letting the greater playerbase actually hit stuff with PPC's, and also ac2/5/10/gauss led to the current situation, not heat/cooldown adjustments.

best solution involves adjustments to convergence and projectile speed for all long distance things. Besides, they only seem so great right now because the main brawling weapons suck (s/srm's) limiting brawling to ac20's and the other long range item lrms is somewhat neutered in comparison (outside of organised team play).

*edit*
Possibly a +1second cooldown just to give LPL a DPS edge at close range compared to PPC's. ER's already run hot enough.

Edited by Asmosis, 08 July 2013 - 09:33 PM.

[Kenyon Burguess]

ppc/erppc are fine. lower the damage to reactor to 100%, mech death at 110% heat

[Sharp Spikes]

So, people cry "Nerf PPC! Nerf PPC!!! Q.Q"... Guys, I'll tell you what: if devs listen to you and actually nerf PPCs, next time you'll cry "Nerf LgL!" and then "Nerf dual Gauss!" Because, as it was already stated multiple times by different people, PPCs (and other weapon types) are not the problem. Heat system and convergence is. "Nerf X, buff Y" approach without major changes to game mechanics can't solve anything.

[BeezleBug]

Ilithi Dragon, on 06 July 2013 - 06:45 AM, said:

reduce the travel speed of the PPC projectile to 1200-1400, down from the current 2000.

jup that is right the PPC´s are way to fast. There is no reason why the PPC should be faster then a Gaus. It is way to easy to hit with the PPC, pointy, and fast as hell not good for the gameplay.

Ilithi Dragon, on 06 July 2013 - 06:45 AM, said:

reduce the travel speed of the PPC projectile to 1200-1400, down from the current 2000.

jup that is right the PPC´s are way to fast. There is no reason why the PPC should be faster then a Gaus. It is way to easy to hit with the PPC, pointy, and fast as hell not good for the gameplay.

[Lugh]

Ilithi Dragon, on 08 July 2013 - 01:52 PM, said:

..............

No.

Catamount beat me to explaining the heat transfer coefficient, and did it better.







I agree that heat needs to be increased for PPCs, that's what I'm arguing for here, but I disagree that we need to punish players for overheating that severely.

I've played MW3 and MW4:Venge and Mercs, and none of them inflicted internal damage for running hotter than 80% heat. In fact, I'm pretty sure none of them inflicted any damage for running to hot, unless you crossed the reactor failure threshold and blew your reactor. I am also not aware of any TT rules that caused internal damage from exceeding 80% heat capacity.

http://d20battletech.wikidot.com/heat

There are minor penalties as low as 30% heat scale.

[Lugh]

Catamount, on 08 July 2013 - 01:50 PM, said:

I'm sorry, but this is incorrect.

Review the page on heat transfer, here

http://en.wikipedia....fer_coefficient

if h= Q / A * ∆T

where
Q = heat flow rate or heat transfer rate, J/s = W
h = heat transfer coefficient, W/(m²K)
A = heat transfer surface area, m²
ΔT = difference in temperature between the solid surface and surrounding fluid area

then you can reorder that equation to get

Q = h * A * ∆T

[font=Times New Roman, serif]So Q is directly proportionate to ∆T, the difference in temperature between the object in question and its surroundings. Increase the heat of the heat sink relative to its surroundings, and Q, the rate of heat transfer into the surrounding, goes up.

Why are you quoting me heat transfer coefficients for Fluids rather than air?

https://www.google.c...tml%3B510%3B335

Furthermore why are you ignoring the fact that metal heat sinks can only get SO hot before they reach a melting point of their own and cease to function?

[Catamount]

Lugh, on 09 July 2013 - 05:48 AM, said:

Why are you quoting me heat transfer coefficients for Fluids rather than air?

https://www.google.c...tml%3B510%3B335

I'm not quoting you a coefficient for one type of fluid or another, and air is a fluid, genius; I'm using the scientific definition, not the layman definition that makes it synonymous with "liquid", but I digress.

The basic equation for heat transfer doesn't change regardless. As ∆T increases, the rate of heat flow increases rather than decreases. You've said nothing that contradicts that, so I have no idea what you're bothering to respond since you didn't address my point. You claimed heatsinks get less efficient as they get hotter based on "simple physics", but when the heatsink gets hotter relative to the surrounding environment, that's an increase in ∆T. Please bother to read what I'm posting next time before responding.

Quote

why are you ignoring the fact that metal heat sinks can only get SO hot before they reach a melting point of their own and cease to function?

I haven't ignored the fact that anything has a melting point. Do you have evidence that Battletech heatsinks get hot enough to melt under normal operating conditions? Even if you did, it doesn't justify the "basic physics" relationship you claimed. Generally speaking, an increase in ∆T means an increase in Q. You didn't identify some special condition for your claim, like the heatsinks melting off.


If that's how you want to play it, then I guess I can make both of us happy: heatsinks get more efficient as they get hotter, unless they somehow got so hot that they take serious damage, which you've provided no evidence of the occurrence of with BT heatsinks.

Edited by Catamount, 10 July 2013 - 07:43 AM.

[Lugh]

Catamount, on 10 July 2013 - 07:34 AM, said:

I'm not quoting you a coefficient for one type of fluid or another, and air is a fluid, genius; I'm using the scientific definition, not the layman definition that makes it synonymous with "liquid", but I digress.

The basic equation for heat transfer doesn't change regardless. As ∆T increases, the rate of heat flow increases rather than decreases. You've said nothing that contradicts that, so I have no idea what you're bothering to respond since you didn't address my point. You claimed heatsinks get less efficient as they get hotter based on "simple physics", but when the heatsink gets hotter relative to the surrounding environment, that's an increase in ∆T. Please bother to read what I'm posting next time before responding.



I haven't ignored the fact that anything has a melting point. Do you have evidence that Battletech heatsinks get hot enough to melt under normal operating conditions? Even if you did, it doesn't justify the "basic physics" relationship you claimed. Generally speaking, an increase in ∆T means an increase in Q. You didn't identify some special condition for your claim, like the heatsinks melting off.


If that's how you want to play it, then I guess I can make both of us happy: heatsinks get more efficient as they get hotter, unless they somehow got so hot that they took serious damage, which you've provided no evidence of the occurrence of with BT heatsinks.

http://www.sarna.net/wiki/Heat_sinks

So you believe that that radiators don't crack and break at extreme heat levels?

You further believe that coolant rods inserted into reactors to radiate heat into surrounding cooling tanks don't melt under extreme temperatures?
Methinks you are overthinking the 'basic physics'

[Catamount]

Lugh, on 10 July 2013 - 07:45 AM, said:

http://www.sarna.net/wiki/Heat_sinks

So you believe that that radiators don't crack and break at extreme heat levels?

You further believe that coolant rods inserted into reactors to radiate heat into surrounding cooling tanks don't melt under extreme temperatures?
Methinks you are overthinking the 'basic physics'

You've provided no evidence, whatsoever, that Battletech heatsinks heat to sufficient temperature to incur significant damage.

You're also changing the subject now that you've lost the debate on how objects behave when they heat up.


I also already covered this possibility when I offered the caveat in my last post:

Quote

heatsinks get more efficient as they get hotter, unless they somehow got so hot that they took serious damage, which you've provided no evidence of the occurrence of with BT heatsinks.

Do you have something to present that's contradictory to that statement, or don't you?

Edited by Catamount, 10 July 2013 - 08:11 AM.

[BR0WN_H0RN3T]

Catamount, on 10 July 2013 - 08:06 AM, said:

You've provided no evidence, whatsoever, that Battletech heatsinks heat to sufficient temperature to incur significant damage.

You're also changing the subject now that you've lost the debate on how objects behave when they heat up.


I also already covered this possibility when I offered the caveat in my last post:



Do you have something to present that's contradictory to that statement, or don't you?

No, I actually disagree and support Lugh on this one. I can't think of anything mechanical or electrical that gets progressively more efficient the hotter it gets. That might be true at a single point in time on an optimisation curve, like a car engine starting from cold but all materials have resistance and there's optimal efficiency at certain temperatures and temps when they get too hot they eventually fail. The only way they could maintai efficiency as they heat up is through an active cooling process that forces heat transfer from the heat sink back to the environment.i.e. cools the HS down the hotter it gets and maintains it at an optimal operating temp. This is why AC units or fridges (for that matter) are very hot at the compressor when they're hard at work. So, HSs heat up and become less efficient, so something is keeping them cool which in turn, is heating up. This is not an indefinite process. Either the HS or the cooling mechanism will fail with heat.

Edited by Brown Hornet, 10 July 2013 - 08:37 PM.

[Catamount]

Brown Hornet, on 10 July 2013 - 08:35 PM, said:

No, I actually disagree and support Lugh on this one. I can't think of anything mechanical or electrical that gets progressively more efficient the hotter it gets. That might be true at a single point in time on an optimisation curve, like a car engine starting from cold but all materials have resistance and there's optimal efficiency at certain temperatures and temps when they get too hot they eventually fail. The only way they could maintai efficiency as they heat up is through an active cooling process that forces heat transfer from the heat sink back to the environment.i.e. cools the HS down the hotter it gets and maintains it at an optimal operating temp. This is why AC units or fridges (for that matter) are very hot at the compressor when they're hard at work. So, HSs heat up and become less efficient, so something is keeping them cool which in turn, is heating up. This is not an indefinite process. Either the HS or the cooling mechanism will fail with heat.

None of this has anything to do with how a heatsink works. Heatsinks don't "force" the heat into the surrounding environment; the heat flows because of the Second Law of Thermodynamics. The reactor is hotter than the surrounding environment, and by extension the heatsink is hotter than the surrounding environment, and the heat travels to the cool surrounding environment until themodynamics equilibrium is achieved. The hotter the heatsink, the faster it transfers heat into the surrounding environment, as per the previously provided equation.

Now, heatpumping mechanisms may exist to do things like cool the cockpit or other components to below ambient temperature, since pilots don't spontaneously die in 50C+ environments, and THOSE might well become less efficient as their components heat up (or they might not, we don't know), but the actual heatsinks and ultimate cooling of the reactor do not, because a heatsink simply dumps heat to the outside environment. Saying they do is a violation of the second law of thermodynamics.

Edited by Catamount, 11 July 2013 - 10:19 PM.