Jump to content

Dhs Effectiveness


183 replies to this topic

#161 Squidhead Jax

    Member

  • PipPipPipPipPipPipPipPip
  • 1,434 posts

Posted 09 November 2012 - 01:03 PM

View PostAmaris the Usurper, on 09 November 2012 - 11:50 AM, said:

I may start another thread on this topic, but for the moment I would appreciate your input. Can you think of any disadvantages?


It's certainly the more realistic mode, but it's harder to get a sense of the point at which you're driving yourself to overheat - since the overheat threshold will be reached after, perhaps long after (long in the sense of battlefield reaction times) that critical shot. And everything past that is another cobblestone in the road to shutdown hell.

Of course, it would be possible to add a "your heat will peak here" marker to the heat UI, which would actually be kinda neat because it would also be an advance warning to try and get out of the furball before you actually shut down (or suffer other heat penalties).

#162 Amaris the Usurper

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 100 posts

Posted 09 November 2012 - 01:13 PM

View PostSquidhead Jax, on 09 November 2012 - 01:03 PM, said:


It's certainly the more realistic mode, but it's harder to get a sense of the point at which you're driving yourself to overheat - since the overheat threshold will be reached after, perhaps long after (long in the sense of battlefield reaction times) that critical shot. And everything past that is another cobblestone in the road to shutdown hell.

Of course, it would be possible to add a "your heat will peak here" marker to the heat UI, which would actually be kinda neat because it would also be an advance warning to try and get out of the furball before you actually shut down (or suffer other heat penalties).

True, and the peak heat marker is a great idea.

With regard to "riding the heat scale" to prevent shutdown, I don't see it as much of a change from the present system. You already have to have an idea of where the heat scale will go before firing weapons.

Confusion would be more likely to result when running configurations that have a large alpha strike but little ability to sustain DPS due to low numbers of heat sinks. Never having seen this game mechanic in action, it might be a surprise to see the heat scale rise that quickly for that long without weapons fire. But the peak heat marker would take out the guesswork.

Edited by Amaris the Usurper, 09 November 2012 - 01:24 PM.


#163 Sephlock

    Member

  • PipPipPipPipPipPipPipPipPipPipPip
  • Legendary Founder
  • Legendary Founder
  • 10,819 posts

Posted 09 November 2012 - 01:20 PM

Has anyone charted the tipping points where, say, a mech that has an engine of size x with y number of internal heatsinks and z number of external heatsinks means you are better off with singles vs doubles?

I know heat generated is a factor but I'm trying to figure out the limits of dhs and regular hs so I can plan my builds around them.

#164 Squidhead Jax

    Member

  • PipPipPipPipPipPipPipPip
  • 1,434 posts

Posted 09 November 2012 - 01:22 PM

Except that what your heat will be is also influenced by how much heat is currently entering the system, and if a bunch of heat flows are about to expire, firing a given weapon might cause heat to never rise, or only slightly.

#165 Amaris the Usurper

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 100 posts

Posted 09 November 2012 - 01:30 PM

View PostSephlock, on 09 November 2012 - 01:20 PM, said:

Has anyone charted the tipping points where, say, a mech that has an engine of size x with y number of internal heatsinks and z number of external heatsinks means you are better off with singles vs doubles?

I know heat generated is a factor but I'm trying to figure out the limits of dhs and regular hs so I can plan my builds around them.

The number of internal HS is [engine rating]/25, rounded down. Below 250, you have to place at least enough HS outside the engine to make 10. For 275 and above, you get 10 HS in the engine for free, in addition to at least one engine HS slot. However, the DHS that you place in these slots only dissipate 1.4 heat/second, so the optimal rating (from a certain point of view) is 250.

Edited by Amaris the Usurper, 09 November 2012 - 01:31 PM.


#166 Bubba Wilkins

    Member

  • PipPipPipPipPipPipPip
  • 688 posts

Posted 09 November 2012 - 01:30 PM

View PostAmaris the Usurper, on 09 November 2012 - 12:35 PM, said:

I'm not much of an expert on how fusion reactors (don't) work, but the energy liberated by nuclear fusion in the engine would be converted into electrical work with less than 100% efficiency. Thus, heat would be generated there, and more load would imply more heat. At the same time, the electrical work would be converted into various forms of useful energy by lasers and PPCs, again with less than 100% efficiency, so heat would also be generated there. Furthermore, autocannons and missiles are driven by gasdynamic combustion processes, which produce a great deal of waste heat.

In other words, if you want to consider engine heat, just view the engine as another heat source, and apply the same argument as in my earlier post. The heat sinks are there to protect the internals from waste heat generated by the major systems. The waste heat is discharged gradually, not instantaneously.


Both Fusion and Fission reactors operate on the same basic principle when it comes to their use generating electricity. They both provide massive amounts of heat which is then harvested via a closed loop heat exchange system. The second half of that system utilizes steam to drive a turbine. See below:
Posted Image

Now, the turbine is just a device to convert the steam back to mechanical energy and thus has to be connected to an actual generator. Now here's the rub. The force required to turn the generator is directly proportional to the electrical load induced on it. The higher the load, the more force required to turn the generator, the more pressure required by the turbine to maintain that speed. The heatsinks in a mech are placed on the return line from the turbine ahead of the heat exchanger. Their functions is to bleed of the remaining heat so that the fluid entering the heat exchanger is sufficiently cool enough to undergo a larger change in temp and thus a more efficent transfer of energy while at the same time cooling the reactor core to manageable levels. In this house, we obey the laws of Thermodynamics!

Now, take your average energy weapon. It's operating at a relatively high voltage which increases overall efficiency and decreases the size of the conductors. Think of a modern day Tesla coil. The lightning created has massive heat and energy, but the coil itself remains relatively cool. The heat generated on the equipment is negligible, but the heat required at the point of generation is massive.

I agree, being instantaneous is a little bit too simplistic...there should be a ramp/spike/cool-off cycle, but for gameplay purposes that makes very little sense and over complicates things immensely.

Edited by Bubba Wilkins, 09 November 2012 - 01:33 PM.


#167 Like a Sir

    Member

  • PipPipPipPipPipPipPip
  • Bad Company
  • Bad Company
  • 589 posts
  • LocationUSA NW

Posted 09 November 2012 - 01:38 PM

View PostBubba Wilkins, on 09 November 2012 - 01:30 PM, said:


Both Fusion and Fission reactors operate on the same basic principle when it comes to their use generating electricity. They both provide massive amounts of heat which is then harvested via a closed loop heat exchange system. The second half of that system utilizes steam to drive a turbine. See below:
Posted Image

Now, the turbine is just a device to convert the steam back to mechanical energy and thus has to be connected to an actual generator. Now here's the rub. The force required to turn the generator is directly proportional to the electrical load induced on it. The higher the load, the more force required to turn the generator, the more pressure required by the turbine to maintain that speed. The heatsinks in a mech are placed on the return line from the turbine ahead of the heat exchanger. Their functions is to bleed of the remaining heat so that the fluid entering the heat exchanger is sufficiently cool enough to undergo a larger change in temp and thus a more efficent transfer of energy while at the same time cooling the reactor core to manageable levels. In this house, we obey the laws of Thermodynamics!

Now, take your average energy weapon. It's operating at a relatively high voltage which increases overall efficiency and decreases the size of the conductors. Think of a modern day Tesla coil. The lightning created has massive heat and energy, but the coil itself remains relatively cool. The heat generated on the equipment is negligible, but the heat required at the point of generation is massive.

I agree, being instantaneous is a little bit too simplistic...there should be a ramp/spike/cool-off cycle, but for gameplay purposes that makes very little sense and over complicates things immensely.


I haven't read any of this, but I like the picture...

#168 VoidConductor

    Member

  • PipPipPip
  • 72 posts

Posted 09 November 2012 - 01:45 PM

some month ago I proposed more realistic but still simplified heat mechanic.
for ppl who is interested:

mwomercs.com/forums/topic/3263-heat-management-shutdown-heat-cap/

In my opinion there should be weapon balance between energy weapons, before HS balance. That means more heat for SL and slightly increase for ML . Heat decrease of LL and PPC (+ER). Current PPC is like ballistic slow crap and doesnt feels like BT PPC. The speed should be increased.
All tweaks and changes need to be made not by BT values but by trends ...

- Standard version more heat efficient : 0.8 - 0.9 heat per 1 dmg
- ER versions less efficient + extr range: 0.9 - 1.0 heat per 1 dmg
- similar DPS value, wich increases with tonnage of the weapon
- SL filling weapons, additional boost. ML towards main dmg. LL and PPC main dmg
- consideration of synergy aspects . fast mech speed of lights and small range of SL

DHS 2.0 should be ok.

Edited by VoidConductor, 09 November 2012 - 01:46 PM.


#169 Amaris the Usurper

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 100 posts

Posted 09 November 2012 - 02:23 PM

View PostBubba Wilkins, on 09 November 2012 - 01:30 PM, said:


Both Fusion and Fission reactors operate on the same basic principle when it comes to their use generating electricity. They both provide massive amounts of heat which is then harvested via a closed loop heat exchange system. The second half of that system utilizes steam to drive a turbine. See below:
Posted Image

Now, the turbine is just a device to convert the steam back to mechanical energy and thus has to be connected to an actual generator. Now here's the rub. The force required to turn the generator is directly proportional to the electrical load induced on it. The higher the load, the more force required to turn the generator, the more pressure required by the turbine to maintain that speed. The heatsinks in a mech are placed on the return line from the turbine ahead of the heat exchanger. Their functions is to bleed of the remaining heat so that the fluid entering the heat exchanger is sufficiently cool enough to undergo a larger change in temp and thus a more efficent transfer of energy while at the same time cooling the reactor core to manageable levels. In this house, we obey the laws of Thermodynamics!

Now, take your average energy weapon. It's operating at a relatively high voltage which increases overall efficiency and decreases the size of the conductors. Think of a modern day Tesla coil. The lightning created has massive heat and energy, but the coil itself remains relatively cool. The heat generated on the equipment is negligible, but the heat required at the point of generation is massive.

I agree, being instantaneous is a little bit too simplistic...there should be a ramp/spike/cool-off cycle, but for gameplay purposes that makes very little sense and over complicates things immensely.

WARNING! ENGINEERING DEGREE DETECTED.

(OK, I'll admit it, I have one too.)

I said that I didn't know how fictional fusion engines work in the BattleTech universe, not that I don't understand how present-day engines exploit heat reservoirs to produce mechanical work. The fusion reactor in your diagram could be replaced by a pile of burning logs, and the diagram would still be valid. However, to my knowledge, it has never been stated in a canon source that the fusion engines in BT are heat engines in the conventional sense. They may not be even remotely analogous in working principle to (say) present-day nuclear power plants. (Correct me if I'm wrong.) Given that faster-than-light communication and travel are both possible in this fictional context, it is just conceivable that they have some whiz-bang method of generating electrical power from fusion processes without a turbine or even a working fluid.

But this is beyond the point. The point was to eliminate heat "front loading" (excessively high heat threshold with large numbers of HS)--which runs counter to expectations and has never been a part of BT or MW games--while minimally affecting gameplay balance. You have given a mechanical engineering lecture and then changed the subject, stating that my fix isn't realistic enough to be acceptable, and that in order to be realistic enough, it would have to be too complicated to be acceptable.

The issue is not consistency with thermodynamics; the core gameplay mechanics of BT already aren't. Why do 20 and 100 ton mechs have the same heat capacity? Why do they have the same internal volume? Why does heat dissipate faster when the environment is colder but not when the mech is hotter? If you obey the laws of thermodynamics in your house, you won't be playing BT or MWO there. These are fun games about mechs blowing each other to pieces, not technical computing products. The BT universe may not be--strictly speaking--realistic, but it is definitely a fun place to spend some time.

In any case, if you have any thoughts about how the proposed system would actually work in this computer game, I would like to hear them.

Edited by Amaris the Usurper, 09 November 2012 - 02:50 PM.


#170 Bubba Wilkins

    Member

  • PipPipPipPipPipPipPip
  • 688 posts

Posted 09 November 2012 - 02:47 PM

View PostAmaris the Usurper, on 09 November 2012 - 02:23 PM, said:

WARNING! ENGINEERING DEGREE DETECTED.

(OK, I'll admit it, I have one too.)

I said that I didn't know how fictional fusion engines work in the BattleTech universe, not that I don't understand how present-day engines exploit heat reservoirs to produce mechanical work. The fusion reactor in your diagram could be replaced by a pile of burning logs, and the diagram would still be valid. However, to my knowledge, it has never been stated in a canon source that the fusion engines in BT are heat engines in the conventional sense. They may not be even remotely analogous in working principle to (say) present-day nuclear power plants. (Correct me if I'm wrong.) Given that faster-than-light communication and travel are both possible in this fictional context, it is just conceivable that they have some whiz-bang method of generating electrical power from fusion processes without a turbine or even a working fluid.

But this is beyond the point. The point was to eliminate heat "front loading" (excessively high heat threshold with large numbers of HS)--which runs counter to expectations and has never been a part of BT or MW games--while minimally affecting gameplay balance. You have given a mechanical engineering lecture and then changed the subject, stating that my fix isn't realistic enough to be acceptable, and that in order to be realistic enough, it would have to be too complicated to be acceptable.

The issue is not consistency with thermodynamics; the core gameplay mechanics of BT already aren't. Why do 20 and 100 ton mechs have the same heat capacity? Why do they have the same internal volume? Why doesn't heat dissipate faster when the environment is colder but not when the mech is hotter? If you obey the laws of thermodynamics in your house, you won't be playing BT or MWO there. These are fun games about mechs blowing each other to pieces, not technical computing products. The BT universe may not be--strictly speaking--realistic, but it is definitely a fun place to spend some time.

In any case, if you have any thoughts about how the proposed system would actually work in this computer game, I would like to hear them.


Not an engineer, but my background is power delivery. :)

Our primary disagreement pertained to where heat was generated and how it was handled. You proposed relocating it outside of the engine which is very much against lore and generally against physics. My response explained why the existing system generally made more sense.

As for which enriches gameplay the most, the KISS method would imply the current system to be superior.

#171 Gristle

    Member

  • PipPipPipPipPipPip
  • 484 posts
  • LocationN. E. Kentucky

Posted 09 November 2012 - 03:04 PM

View PostBubba Wilkins, on 09 November 2012 - 02:47 PM, said:

... My response explained why the existing system generally made more sense....



No it doesn't - ballistics and missile heat.

#172 Zyllos

    Member

  • PipPipPipPipPipPipPipPipPip
  • 2,818 posts

Posted 09 November 2012 - 03:48 PM

Maybe instead of adding the heat over the 10s, maybe add it over the CD of the weapon?

That means PPCs will add 3.0 HPS instead of 0.9 HPS. What this does is give the players a chance to react to each shot in the heat management scale. Meaning that each weapon fired will not generate any additional heat after the CD is up.

3x PPC = 27 heat in MWO (Ohm's charts)

With 28 SHS, this means 9.0 - 2.8 = 6.2 HPS for 3s, which is 18 heat, thus will not overheat based on the 30 point scale but still give players a chance to react during combat by not over dumping HPS that will last long into the fire and not knowing when they are tipping themselves over the scale.

Of course, I did not notice, but doing this as HPS instead of all at one time, will cause mechs to shutdown after they fired their shot, not during their shot, which may or may not be an issue.

Edited by Zyllos, 09 November 2012 - 03:50 PM.


#173 Bubba Wilkins

    Member

  • PipPipPipPipPipPipPip
  • 688 posts

Posted 09 November 2012 - 03:54 PM

View PostGristle, on 09 November 2012 - 03:04 PM, said:



No it doesn't - ballistics and missile heat.


okay, you got me there. Still, seems kind of late to reinvent the heat wheel when it's worked this way since I can remember.

#174 Amaris the Usurper

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 100 posts

Posted 09 November 2012 - 04:48 PM

View PostBubba Wilkins, on 09 November 2012 - 02:47 PM, said:



Not an engineer, but my background is power delivery. :)

Our primary disagreement pertained to where heat was generated and how it was handled. You proposed relocating it outside of the engine which is very much against lore and generally against physics. My response explained why the existing system generally made more sense.

As for which enriches gameplay the most, the KISS method would imply the current system to be superior.

From the "Heat Sink" article on the BattleTech Wiki,

Quote


Heat sinks operate by collecting heat with coolant distributed to heat sources (weapons, engines, myomers, electronics, etc.) and delivering that to a radiator.

Waste heat generation by electrical devices is definitely not against physics. By discussing heat generation in the weapons, I did not mean to imply that the fusion engine does not produce waste heat, or that this is not a major factor. In fact, I was not considering the engine at all. In any case, heat generated by the weapon itself would be lumped together with induced waste heat as the "total heat" of the weapon, so the distinction is not very relevant from a gameplay standpoint.

I don't know the operating principle of BT lasers, except that they are definitely not chemical lasers (these exist but wouldn't be used on vehicles with fusion engines). Current solid-state lasers generate a great deal of waste heat, and this is a major obstacle to scaling up such systems for military purposes. The waste heat is deposited into the solid-state gain medium, i.e., into the laser itself; it does not represent losses in the transmission lines or at the power plant. On the other hand, free electron lasers (to my limited knowledge) jettison their own gain medium--an electron beam--at nearly the speed of light, so waste heat is not much of an issue.

Regardless of the operating principle(s), phrases like "the heat generated by the laser" are frequently used in BT canon, and most people (including, probably, the authors) would take them at face value. When we talk about the heat generated by a light bulb, for example, we don't mean the waste heat generated by a distant power plant in producing the electricity that powers the light bulb, we mean the heat flowing out of the bulb itself, and it shouldn't be different with lasers or other devices.

And there is still the issue of missile and autocannon heat, which you haven't addressed.

Anyway, I was merely trying to illustrate that, even if heat is generated quickly in a particular component--be it a weapon or the engine--it would not instantaneously spread throughout the chassis and begin to affect other systems before the HS have had a chance to work. Instead, the heat would bleed out gradually, just as it is removed gradually by the HS.

Yes, the rate of heat transfer out of a given weapon would be larger initially (when it is much hotter than the surrounding medium) and smaller later on. Assigning an average rate of heat production for 10 seconds (or some other interval) should be adequate for a computer game. But the point is not to faithfully replicate real world behavior; instead, it is to put weapon heat generation behavior on the same basis as heat dissipation behavior.

Increasing the heat threshold was a kludgy solution that made alpha strikes possible without overheating but introduced the heat "front loading" issue. In fact, it introduced the complication of a variable heat threshold, where most people would have expected a uniform one. Putting heat generation on the same basis as heat dissipation would elegantly fix the problem.

The KISS rule is applicable when a solution is more complicated than would be necessary to adequately solve a problem. A broomstick is undeniably simpler than an airplane, but that does not make it better for flying.

Edited by Amaris the Usurper, 09 November 2012 - 05:30 PM.


#175 Amaris the Usurper

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 100 posts

Posted 09 November 2012 - 04:57 PM

View PostBubba Wilkins, on 09 November 2012 - 03:54 PM, said:


okay, you got me there. Still, seems kind of late to reinvent the heat wheel when it's worked this way since I can remember.

You are probably right. As it stands, the core gameplay is very good. I'm happy to be getting a MW game that feels like a MW game. It's been a long time! Earlier I mentioned the possibility of introducing a "hardcore" gameplay mode with certain game elements re-tuned for greater realism; we can hope.

#176 Amaris the Usurper

    Member

  • PipPipPipPipPip
  • Ace Of Spades
  • Ace Of Spades
  • 100 posts

Posted 09 November 2012 - 05:12 PM

View PostZyllos, on 09 November 2012 - 03:48 PM, said:

Maybe instead of adding the heat over the 10s, maybe add it over the CD of the weapon?

That means PPCs will add 3.0 HPS instead of 0.9 HPS. What this does is give the players a chance to react to each shot in the heat management scale. Meaning that each weapon fired will not generate any additional heat after the CD is up.

3x PPC = 27 heat in MWO (Ohm's charts)

With 28 SHS, this means 9.0 - 2.8 = 6.2 HPS for 3s, which is 18 heat, thus will not overheat based on the 30 point scale but still give players a chance to react during combat by not over dumping HPS that will last long into the fire and not knowing when they are tipping themselves over the scale.

Of course, I did not notice, but doing this as HPS instead of all at one time, will cause mechs to shutdown after they fired their shot, not during their shot, which may or may not be an issue.

It could also be done this way. Mainly, I was interested in ways to get rid of the variable heat threshold while still permitting alpha strikes. Once we have the concept of "spreading the heat," the best implementation should (and can only) be determined by testing. Based on TT, 10 seconds would be my baseline, but I'm not attached to that value if another (or varying) values would enhance gameplay.

Edited by Amaris the Usurper, 09 November 2012 - 05:31 PM.


#177 197mmCannon

    Member

  • PipPipPipPipPipPip
  • Go-cho
  • Go-cho
  • 265 posts
  • LocationCincinnati, OH

Posted 10 November 2012 - 09:21 AM

I know they are called double heatsinks but I don't really care if they actually dissipate twice the amount of a single.

I just want the game to be balanced.

Right now heatsinks feel pretty good to me.

I think guass should make more heat and PPCs should make a little less heat. Other than that it feels good.

It's nice knowing the math now though. Even if your build has no crits for heatsinks it is still worth upgrading to double because of the engine sinks.

#178 Ultrabeast

    Member

  • PipPipPipPipPipPipPip
  • Elite Founder
  • Elite Founder
  • 992 posts
  • Twitter: Link
  • LocationDas Amerikas (The US)

Posted 10 November 2012 - 09:30 AM

I get like 5 kills per game with my 9M with DOUBLE HEATSINKS. Find something else to cry about.

#179 Wreck0000

    Member

  • PipPip
  • 26 posts

Posted 10 November 2012 - 10:54 AM

Wow...

Thanks to the OP for testing, and others for the XML checks.

#180 shabowie

    Member

  • PipPipPipPipPipPipPip
  • 877 posts

Posted 10 November 2012 - 04:38 PM

View PostAmaris the Usurper, on 09 November 2012 - 12:35 PM, said:

Furthermore, autocannons and missiles are driven by gasdynamic combustion processes, which produce a great deal of waste heat.


Much of the heat in a ballistic weapon firing cased ammunition is absorbed by the case and ejected with said case. Missiles and rockets most of the heat produced would be vented out as back or side blast. We can look at what we get in BT or MWO as the leftovers from that process.

I liked your idea in a previous post of dealing with Heat Over Time and suggest that as a name if anyone wants to support implementing it as a mechanic. Also at higher heat levels since you could fire again before the heat from the first salvo has built up and bled off under many configurations your heat buildup could be doubling or even tripling for short periods, making high heat thresholds and override dangerous - as it should be.





7 user(s) are reading this topic

0 members, 7 guests, 0 anonymous users