200 replies to this topic

[Antony Weiner]

gavilatius, on 14 February 2013 - 09:57 PM, said:

ok how bout a test that we can have in MWO
have EVERYTHING at TT values, from A-Z
keep the FPS/SIM skin and gameplay
see how balanced the game is.

no, just do it and then come back to me, I want to know if it is more or less balanced.
seriously, I freaking want to know.

I just did it, its less balanced.

[Yokaiko]

sC4r, on 14 February 2013 - 11:14 PM, said:

to op:
http://mwomercs.com/...l-dhs-whinners/

i just made it for you and the likes of you please read

Fundamentally wrong basis, we never had true DHS, the engine heatsinks were at 1.0 the first attemat DHS. Until players did the testing.
Even then double heatsinks for triple firing rates shouldn't be game breaking, the lack of logic in that statement is amazing.

Not to mention this is the same internal testing that missed Artemis causing 90 degree down angles and unstoppable full hit strikes.

[FiveDigits]

After a day of heated discussion I think it might be good to review the positions that emerged in the community regarding the DHS issue. I'll use fictional quotes of fictional people to illustrate what I obeserved and I'll commentate a little on each.

"True DHS would make heat management too easy. L2P!"

- Victor "N00bB4sh0r" Kerensky

• The discussion is not about heat management being "too hard". It is about the imbalances of the current implementation. Mechs with only the ten in-engine heat sinks (or close to it) operate under true DHS while mechs who invest more into heat dissipation get increasingly short-changed.
  
• Even with true DHS heat management would still be an issue because of the tripled rate of fire as compared to TT.

"The current implementation seems OK. Weapons and heat are quite balanced."

- Joe McWarrior

• Yeah, for players with fully customized mechs with unlocked heat efficiency pilot skills we are on average kind of at a 2.0 value for DHS. But shouldn't pilot skills be an additional benefit - not the required base line? What about new players and DHS stock mechs?
  
• After many many heat tweaks to many weapons most of them are kind of usable. Most of that tweaking would not have been necessary if the heat mechanics were sound. The current balance is built on shaky ground. It's collapse is foreseeable when the Clans come stomping around.

"PGI tested 2.0 DHS internally and found them to be game-breakingly strong."

- Whitney Knight

• There is no "three second Jenner". Neither are the 6 MLAS Cicada or 9 MLAS Hunchback bogeymen real threats to balance. They would of course be stronger, but not game-breakingly so. They can be and have been refuted mathematically.
  
• PGI's internal testing is shaky at best. It was the community who found that the first implementation of DHS had in-engine DHS working as SHS. Then PGI claimed to have tested all 2.0 DHS and found them game-breaking. We were supposed to get all 1.4 DHS. What we got were 2.0 in-engine DHS and 1.4 externals. This was again found by the community (and later labeled as intentional, yeah ... right). Then there was the Artemis LRM fiasco ...
  
• One possible balance issue is the boating of high alpha damage large energy weapons (6 PPC Stalker). Those could put out critical amounts of damage if DHS were changed to +2 heat threshold / +0.2 HPS dissipation. The real issue here is the heat threshold - not the dissipation. The solution is quite simple and has been presented numerous times: Make DHS increase heat threshold at the same rate as SHS, +1. A nice side effect of that change would be to give SHS a niche purpose (they currently have none) of being able to create a higher heat threshold than DHS. Another option is to set the heat threshold to a fixed value independant of the number of installed heat sinks.

"The current mixed DHS implementation is convoluted and imbalanced. It is the only system in the game with non-linear returns."

- Theo R. Crafter

• Adding another ton of ammo doesn't yield fewer shots than the previous. Adding another auto cannon doesn't decrease the range of all installed auto cannons. Adding another laser doesn't reduce the damage of each installed laser. Yet, adding an external DHS decreases the heat dissipation per installed DHS.
  
• Confused forum posts and questions prove time and again that a large portion of the player base does not understand the current implementation. Which DHS is an "engine DHS" which is "external"? Which kind of DHS cools how much? The system is complicated and people make bad build decisions based on it.
  
• The sweet spot of a 250 rated engine with 10 true DHS built in favors certain mech classes and builds. It also limits engine choice options. Downgrading an engine is often not feasible because of it and large mechs with energy-heavy loadouts get penalised.

We were told that the current DHS implementation was a first attempt, that PGI would monitor the situation and adjust the values. This was months ago. Nothing happend. This is a supposed beta and we are its testers. We have proven our worth numerous times in quickly identifying bugs and balance problems. Please change the DHS implementation and let us test it.

A Double Heat Sink is a Double Heat Sink is a Double Heat Sink.

Edited by FiveDigits, 15 February 2013 - 03:58 AM.

[Yokaiko]

Good post.

[Agent of Change]

[Deathlike]

The 3-second Jenner must have been part of the initial attempt to set DHS to 2.0 across the board. I'm sure someone could write/link the numbers, but I can actually imagine a 6-med Jenner not really feeling the effects of overheating until a long period of time. If one were to consistently fire half of the weaponry and fire the other set of lasers afterwards (think constant stream of fire), it would actually produce a machine-gun like Jenner that potentially be overpowered. I've kinda done some of the same stuff with a cic-2A, so this actually isn't that far fetched. It could be done pretty easily on maps that aren't deserts, which is pretty much 3/4 of the maps we have (not factoring in that most maps are just a slightly different version of existing ones). The never-ending laser jenner (or cic) would possibly be worse than the Raven 3Ls we have already.

Edited by Deathlike, 15 February 2013 - 04:16 PM.

[FiveDigits]

The thing about the n MLAS of constant firing bogeyman is:
Even if you can fire them incessantly they do not do more damage than other builds. Revisit my comparison of the current HBK-4SP to the potential true DHS CDA-2A with 6 MLAS. While the Cicada can fire longer at full DPS it does not even catch up to the damage the Hunchback has done in the same time. And once both reached their full DPS heat limit they continue at their sustained DPS rates, where again the Hunchback is superior.
Nobody's argueing that energy boats wouldn't be better with true DHS. It's just that the premise that true DHS would make them superior to other builds or let their damage numbers skyrocket in a game-breaking way is fundamentally false.

[Yokaiko]

Thontor, on 15 February 2013 - 06:00 AM, said:

You are comparing a 4SP with 4ML and 2SRM6 with a Cicada with 6ML

I'd argue that despite the lower alpha, the Cicada puts much more effective damage on target...

Rare is the ideal situation wher you can hit with all your SRMs on the component you are trying to hit. For one they spread, a Lot, even with Artemis... Really requiring you to be at point blank range to reach their full theoretical potential.

Add to that the fact that they take a lot more skill to hit consistently with than Lasers... And Then when you do, unless you are at point blank range, they won't likely hit where you want them to if they even hit at the target at all.

So yes... 2.0 DHS allowing the "30 point, pinpoint, hitscan alpha strike" 6ML Cicada to do 3 extra alpha strikes for 90 more damage at 7.95 DPS full rate of fire before reaching heat capacity would be game breaking.

No frigging way.

First you would have NO armor, and I mean none. Second, it would be slow to fit the heatsinks. Slow plus no armor with a 40 ton mech equals dead.

...and I've mastered the Cicadas, including the -2A There is no way you would be able to stay there long enough to be "gamebreaking" the mech doesn't have close to enough armor, even capped out.

Its a hit and run mech, trying to layout 6 alphas would be suicide against anyone not pants on head stupid.

http://mwo.smurfy-ne...6723741fb9f4071

^^ That is about as close as you are going to get, It would go from 1.4 heat efficiency to 1.6 with true DHS.

Notice there is only space for 2 more HS, but not weight, and that mech has next to NO armor, furthermore, if you down the engine anymore you lose a heatsink there so dropping anymore is a false economy, to get the two extra tonnes out of the engines you lose two onboard heatsinks.

No FF, no endo, and stock armor levels.

The mech would be a joke.

Edited by Yokaiko, 15 February 2013 - 11:49 PM.

[arobilla]

Loving the graphs. Well done sir.

[AndyHill]

Deathlike, on 15 February 2013 - 04:16 PM, said:

The 3-second Jenner must have been part of the initial attempt to set DHS to 2.0 across the board. I'm sure someone could write/link the numbers, but I can actually imagine a 6-med Jenner not really feeling the effects of overheating until a long period of time.

The math isn't complicated at all, I'll be happy to demonstrat ewith this 6mll build: http://mwo.smurfy-ne...58670d929b22ed9 This is a build with for example no jump jets, to get those you'd have to sacrifice firepower, armor or heat dissipation.

With the current implementation, it has heat dissipation strength of 10 x 2 + 5 x 1.4 = 27
With true 2.0 sinks it has a heat dissipation strength of 15 x 2 = 30

It's a difference of three single heat sinks. The current implementation has 27 / 30 x 100 = 90% the heat efficiency of the dreaded 3-second Jenner. Both have less than 50% heat efficiency. The one we have now is able to non-stop fire at 3.38DPS, while the 3-second Jenner would do 3.75. In practice this means that The 3 second version can blast away with 3 of its 6 lasers, while the one we have now will have to hold fire 10% of the time. And that is probably the most affected Jenner build, for example this one: http://mwo.smurfy-ne...3dd50a3aed1491d with streaks and a few jumpies retains a whopping 95% of its heat efficiency compared to the true 2.0 Jenner.

None of these builds core an Atlas in 3 seconds. However, the current Jenner's chances of coring this Atlas: http://mwo.smurfy-ne...6f886bea3e13002 have been improved slightly, since compared to the 90% or 95% heat efficiency retained by the Jenners, it has only 83% of its heat dissipation capacity remaining compared to what it would have with true doubles.

[Glythe]

armyof1, on 14 February 2013 - 03:32 AM, said:

Non-linear heat dissipation aside, I find the heat balance is actually quite good. If we changed to 2.0 for all HS then we'd need to look over every weapon's heat again, otherwise we'd make all mechs with huge burst damage too powerful.

With the current setup DHS is an upgrade in heat efficiency for 95% of mech builds 50 tons or less. All light mechs perform better with DHS. Meanwhile the DHS upgrade is actually a downgrade in heat efficiency for many assault mech builds.

See the problem?

Light and medium mechs rarely use up extra heat sink slots due to weight restrictions so ALL of their heat sinks are 2.0 while most assault mechs only get half of their heat sinks as doubles. Let's also remember that ES and FF are almost always upgrades for lights and mediums but they don't work well on heavies and assaults most of the time.

So what you end up with is a system where the upgrades favor lights and mediums. If things were fair light and medium mechs would only get DHS for heat sinks they place in the extra engine slots and external heat sinks. Generally speaking that would make most lights have about 2 DHS and would prevent them from using both the ES and FF upgrades if they want to make good use of DHS.

A real solution is to have scaled values for DHS in heavier mechs (1.5 for mediums, 1.6 for heavies and 1.8 for assaults).

Edited by Glythe, 16 February 2013 - 04:50 AM.

[AndyHill]

I'd be bold enough to claim that DHS are currently an upgrade for 95% of all 'mechs. Can you post some examples of assaults that are better off with SHS (in smurfylab format), I'd be interested in taking a look?

[Deathlike]

6 med JR7-F:
link

17DHS, optimized for DHS over speed

6 med CDA-2A:
link

18 DHS, optimized for DHS over speed

These are pretty good for what I can tell (50% heat efficiency), and if it ever was the case that DHS was @ 2.0 across the board.. these would be on the field a whole lot more.

When referencing SRMs, you can't always guarantee full damage.. neither can you for lasers, but the damage is more likely to be spread out vs other lights.

In very rare cases SHS would be better than DHS (probably anything that consumes a lot of slots like ballistics or LRMs).

Edited by Deathlike, 16 February 2013 - 05:43 PM.

[Deathlike]

Yokaiko, on 15 February 2013 - 11:42 PM, said:

No frigging way.

First you would have NO armor, and I mean none. Second, it would be slow to fit the heatsinks. Slow plus no armor with a 40 ton mech equals dead.

...and I've mastered the Cicadas, including the -2A There is no way you would be able to stay there long enough to be "gamebreaking" the mech doesn't have close to enough armor, even capped out.

Its a hit and run mech, trying to layout 6 alphas would be suicide against anyone not pants on head stupid.

http://mwo.smurfy-ne...6723741fb9f4071

^^ That is about as close as you are going to get, It would go from 1.4 heat efficiency to 1.6 with true DHS.

Notice there is only space for 2 more HS, but not weight, and that mech has next to NO armor, furthermore, if you down the engine anymore you lose a heatsink there so dropping anymore is a false economy, to get the two extra tonnes out of the engines you lose two onboard heatsinks.

No FF, no endo, and stock armor levels.

The mech would be a joke.

You forgot endosteel, so you could easily buff the armor or speed... so, please build a better mech example.

[FiveDigits]

Deathlike, on 16 February 2013 - 05:47 PM, said:

You forgot endosteel, so you could easily buff the armor or speed... so, please build a better mech example.

You can get a little more armor with Endo, but still ... meh.

[DrBlue62]

FiveDigits, on 16 February 2013 - 06:27 PM, said:

You can get a little more armor with Endo, but still ... meh.

Drop two heat sinks and add armor.

The survivablitly gained is well worth it.

[Heeden]

I'm started to get confused about how people are upset at the "fairness" of DHS.

Lighter mechs don't gain more from the system, a heavier chassis carrying the same weapons and heat-sinks will have the exact same heat efficiency as a lighter chassis. Smaller mechs seem less restricted by the the heat-sink system not as an act of favouritism but because they are already restricted by weight and space requirements.

Laser-boating lights don't seem too powerful, and high-heat heavy and assault builds don't seem too weak, so the only problem I see is DHS being a mandatory upgrade for pretty much everyone, rather than a considered side-grade or possible downgrade depending on your particular chassis and build.

[FiveDigits]

DrBlue62, on 16 February 2013 - 06:34 PM, said:

Drop two heat sinks and add armor.

The survivablitly gained is well worth it.

Nice and dandy. The point of the whole exercise to build a 6 MLAS Cicada with the maximum amount of DHS you can cram into it was to show how it would not even remotely be "game breaking" under true DHS mechanics form a DPS perspective - not to create a viable build one would actually run.

Heeden, on 16 February 2013 - 06:41 PM, said:

I'm started to get confused about how people are upset at the "fairness" of DHS.

Lighter mechs don't gain more from the system, a heavier chassis carrying the same weapons and heat-sinks will have the exact same heat efficiency as a lighter chassis. Smaller mechs seem less restricted by the the heat-sink system not as an act of favouritism but because they are already restricted by weight and space requirements. [...]

So, you equip the same 4 MLAS and 2 SRM4 with 12 total DHS on a AS7-D as on a JR7-D? You then proceed to fill up the Atlas's remaining tonnage with MG ammo to make a nice fireworks display when you go down?
You'd probably not. A heavier mech has more weapons, more armor and - most likely - more heat sinks along with it. And the current implementation creates increasingly worse returns on each additional installed "crapsink" DHS. That is the imbalance.

[Deathlike]

Just FYI, if you are not fond of the semi-silly 6 PPC Stalkers, just imagine what tweaking DHS does for them.

I think overlooking how powerful 6 meds on a jenner/cic is understating things. I've done 400-500+ damage with the cic-2A on a good day (not with the cfg I linked, since it's more of a theoretical build - I run something slightly hotter anyways). I'm not a good pilot or shooter by any means, but people that don't see the value of the effectiveness and DPS of the aforementioned mechs is turning a blind eye to heat balance.

[Vapor Trail]

Muffinator, on 14 February 2013 - 04:00 AM, said:

If my hottest heavy and assault builds were much more heat efficient with weapons than they are now, they would be completely broken. For example my 6 LL stalker can get off 2 shots with the third leading to shutdown. Most of the time 2 good hits with 6LL can kill a medium and 3 can kill a heavy. Imagine if I could pop off 4, 5, 6 shots without overheating?

True doubles on a mech with 6 Large (basing this off a Stalker build).
6 Large Lasers @ 7 heat per shot.
42 heat per salvo. 4.25 second recycle time. 9.88 heat per second generated @ max RoF.

Mech can mount 22 DHS total (strip a bit of armor to fit the 300 Std and 2 extra heat sinks in the engine).
That's 4.4 heat per second with true Double Heat Sinks (compared to 3.68 with 0.14's).

Max number of salvos out of the mech.
Shutdown threshold is 74, Heat dissipated per salvo time is 4.25 * 4.4 = 18.7
74-42 = 32
32 / (42-18.7) = 1.37
So you can fire twice@ max rate without shutdown.

Leaving you at 88% heat ( or 65.4 if you prefer the numerical).

You need to cool 33.4 heat before you can salvo again. That's 7.59 seconds. (3.34 seconds longer than normal recycle).
Time between salvos to ensure no heat buildup: 9.54 seconds.

To compare with current heat:

Same heat generation.
42*2 - 15.64 = 68.36 heat (so two salvos @ max rate before overheat).
Time to wait to ensure no shutdown on third salvo, 9.88 seconds.
Time to wait between salvos for no heat buildup: 11.41 seconds.

So... the best you can get out of an increase to .2 heat per second doubles is a two second decrease in the time between salvos.

"Imagine if I could pop off 4, 5, 6 shots without overheating?"
To quote the stereotypical blonde bimbo. PUH-LEEEEEEEEZE.

TL; DR:
Before you start making outrageous claims:

Do.
The.
Freaking.
Math.

Deathlike, on 16 February 2013 - 08:04 PM, said:

Just FYI, if you are not fond of the semi-silly 6 PPC Stalkers, just imagine what tweaking DHS does for them.

And this is what exactly?
Firing once, waiting nine and a half seconds, firing again, then firing every thirteen and a half seconds thereafter?

Still seems semi-silly.

Edited by Vapor Trail, 16 February 2013 - 08:29 PM.