81 replies to this topic

[Riptor]

Joseph Mallan, on 06 November 2013 - 09:31 AM, said:

For me it was Large Pulses and SSRMS for the win in MW2.

Depends on wich MW2 you played.

Mercenaries didnt gave you alot of LV 2 tech so the normal medium laser was the go to weapon. Not saying the other weapons werent effective but who would use a gauss rifle if you can use 3 medium lasers for 3 tons and have the same effect on the enemy in terms of damage without the amoo limitations.. wich was 5 shots per ton so it was gone like in a second or two?

But for the clans.. yeah pulse lasers where even worse.. lots of laser dakka dakka

[ShadowedR]

I wonder if PGI ever evaluated the previous generation of MW games.

MW2 - 4
and the Living Legends Mod.

You know, what worked, what was wrong, etc...

[Krivvan]

ShadowedR, on 06 November 2013 - 09:57 AM, said:

I wonder if PGI ever evaluated the previous generation of MW games.
You know, what worked, what was wrong, etc...

One of the things they wanted to do different was make all the weight classes viable and usable.

It's definitely not as bad as in previous titles that were just races to the Assault mech. Mediums may be bad, but they can still be used somewhat effectively in PUG games.

[ShadowedR]

Krivvan, on 06 November 2013 - 10:01 AM, said:

One of the things they wanted to do different was make all the weight classes viable and usable.

It's definitely not as bad as in previous titles that were just races to the Assault mech. Mediums may be bad, but they can still be used somewhat effectively in PUG games.

Mediums only really shine once you factor in economic factors.

Look at traditional warfare.
The bulk of your army was made up of common foot soldiers.
Cheaper to train and maintain that knights.
Easier to train than archers.
We'll exclude crossbowmen - The Hackers of Antiquity .

Same with WW I and to a large extent II.

Yes you had your specialists and you had scouts etc.
But the bulk remained your common everyday soldier.

The same doesn't hold true for PUGS.
Not until PUG's start applying tactics and teamwork.

Look at DotA - I didn't really see a lot of organised play until the Dota2 International got broadcast and it became a big thing.
Now a PUB game is automatically structured with teamwork, coordination etc.

So, the more we educate the community, the better the quality of games?

[General Taskeen]

Riptor, on 06 November 2013 - 09:30 AM, said:

Are you kidding me?

MW 2.. Medium lasers where king.. no sense in using ANYTHING else.. you would just rapid fire down anything in seconds because they here using original armor values.

MW3: Again a nightmare in multyplayer.. no reason to take specific mechs because you would only go for the highest tonnage in a single weightclass.. nothing else mattered due to full customisations. Mechs where nothing more then glorified skins for a given tonnage. Also medium lasers where king once again.

MW4: Jump sniping up the bazoo.. it wasnt fun playing multyplayer because of the fragging coolant flushes + jumpjets + ER PPC spam.. atleast gauss rifles where so big that you had problems fitting more then 2 on a mech but that friggin black knight with 4 or 5 ER PPCs coring you out in two salvos... yeah.. once more it was ***** in multyplayer.

You are comparing games that where made first and foremost with single player in mind where balance doesnt mean jack ****. Aslong as the AI doesnt use cheeze no ones gonna cry foul in a single player game.

Whoa there buddy, a bit reactionary. You are listing all the fundamental issues that plagued previous Mech games without listing any of what actually did work in those games when "translating" ideas from TT.

For one, the heat systems in MW2/3 were phenomenal. None of this silly "put more heatsinks in, more heat threshold, derp!" balance of MWO where single heat sinks are Turds, unlike in MW3. MW4's went for arcade, and upped the pew-pew by making all Mechs have a threshold that was 60 or something like that. MWO is basically a high threshold like MW4, but raised to an extreme, to have all the "pew pew" arcade you could ever want with DHS.

Anyways, taken in context though, those games had good ideas here and there:

- Consistent ECM balance across previous Mech games (never did shut off missiles, definitely useful for the 360 degree radar 'hiding' ambushes in those games)
- Heat Systems (as mentioned)
- Usually better MG's and Flamers (Flamers sort of acted like Napalm or something in MW4 and stuck to the Mech for a little bit, that was nice touch)
- All AC's were better translated from TT as bursts of bullets, rather than "one shell = TT damage" sillyness
- LB-X guns were better translated in those games
- Jump Jets were fun to use (yeah yeah, "poptart MW4 issues"), at least you could get airborne without hovering around like MWO's do (MW:LL actually expanded what those games did with JJ's, and added the 'balance' factor, while making them still highly maneuverable)
- SRMs were amazing ripple fire weapons in MW2, 4, LL rather than silly blob missiles like MW3 and MWO.
- SSRMs required skill
- NARC worked
- LRMs better balanced with the active/passive radar in those games

I could list more, but you'll get the picture.

Edited by General Taskeen, 06 November 2013 - 10:23 AM.

[Tombstoner]

I applaud the effort. I really do but this was covered in closed beta.
Using TT values as a starting point is a great place to start. but the conversion from TT to skill based targeting
fundamentally alters the relationship between tonnage allocated to speed, armor, weapons/heat dissipation, mech size (not part of TT), speed, range and dificulty to hit.
so survivability is drastically altered from TT where it could be looked at as a linear relationship, but in MWO its something very different that's not apparent on paper. large mechs got nerfed and lights got massive buffs that drop of very fast as mech size increases into the medium and heavy class.

The devs needed to have what i call a gunnery range where you gather data on the interaction between mech size, range, speed, hit boxes and i few other i cant remember. basically every thing thats needed to correct for the conversion from TT mech design hit and damage system to a live action skill based targeting and preserve the power per tone relation ship.
every mech should have location specific armor coeficants

But one thing to consider is the PGI wanted lights to have a 50-50 chance vs. a 100 ton mech. what we got was a cut and paste overlay of TT values with tweaked fire rates and double armor to fix it. the aggressive tuning was performed and tweaked some stats.

[ShadowedR]

Tombstoner, on 06 November 2013 - 10:30 AM, said:

I applaud the effort. I really do but this was covered in closed beta.
Using TT values as a starting point is a great place to start. but the conversion from TT to skill based targeting
fundamentally alters the relationship between tonnage allocated to speed, armor, weapons/heat dissipation, mech size (not part of TT), speed, range and dificulty to hit.
so survivability is drastically altered from TT where it could be looked at as a linear relationship, but in MWO its something very different that's not apparent on paper. large mechs got nerfed and lights got massive buffs that drop of very fast as mech size increases into the medium and heavy class.

The devs needed to have what i call a gunnery range where you gather data on the interaction between mech size, range, speed, hit boxes and i few other i cant remember. basically every thing thats needed to correct for the conversion from TT mech design hit and damage system to a live action skill based targeting and preserve the power per tone relation ship.
every mech should have location specific armor coeficants

But one thing to consider is the PGI wanted lights to have a 50-50 chance vs. a 100 ton mech. what we got was a cut and paste overlay of TT values with tweaked fire rates and double armor to fix it. the aggressive tuning was performed and tweaked some stats.

Which is cool. They started off with TT and started tweaking from there.

I'm not too sure, but I haven't read a patch note that said:
Armor buff to mech A, speed on mech B increased or tweak to mech C's load out configuration.

Not recently anyway.

I might be wrong here. I most certainly did not read EVERY patch note since inception.

In the event that I am right, it seems that the mechs themselves aren't being tweaked at all.
Just the modules and equipment.

Oh, hitboxes don't count. Those are fixes to broken features.

[Tombstoner]

ShadowedR, on 06 November 2013 - 10:44 AM, said:

Which is cool. They started off with TT and started tweaking from there.

I'm not too sure, but I haven't read a patch note that said:
Armor buff to mech A, speed on mech B increased or tweak to mech C's load out configuration.

Not recently anyway.

I might be wrong here. I most certainly did not read EVERY patch note since inception.

In the event that I am right, it seems that the mechs themselves aren't being tweaked at all.
Just the modules and equipment.

Oh, hitboxes don't count. Those are fixes to broken features.

mech specific coefficients dont exist. the same amount of armor protection on the commando would have to be stretched incredibly thin when over layed onto an atlas. to give the same amount of protection over all on the atlas would need way more then 19 tones of armor. but this is where the protection system completely comes off the rails. the system works for TT only where hit location is RNG. the volume of the mech is irrelevant, skill based and the armor is paper thin. then with hit boxes you could adjust the armor coefficient to reflect how art affect survivability.

Think about how a commando has the exact same space in its arms as the atlas. Mech design works for a 1980's TT game not a 2012 llive action shooter. cut and paste development and your wondering whats going on. cant blame you. personally i recommend not analising this game you will go mad... try comprehending ghost heat and the laws of thermodynamics. i call it magic and i for one only want that in my sci fi when its called shadowrun.

Just about everything in the game is balanced by feel. from what i have seen. How the game feels being played by the devs is very different then the new player and the highly skilled competitive player.

[Lykaon]

sokitumi, on 06 November 2013 - 07:10 AM, said:

you're right. Personally I'd go a step further and just say trying to adhere to TT is daft. This game could be balanced so much easier/better if TT was just left behind where it belongs.. on a table... or in some dusty basement... next to the axis and allies box... that your parents keep threatening to throw away unless you take it asap.

There are a handful of TT features that should be kept to retain a Battletech/Mechwarrior licence/feeling.

Mech names and shapes
Weights and weight classes
Tonnage and crit slots of mechs and weapons/items
Some sort of heat mechanic

That is it.There are no reasons at all to translate TT mechanics into a shooter game that can not cope with the translation to the new format (I'm looking at you TT armor mechanics).

[Davers]

ShadowedR, on 06 November 2013 - 08:00 AM, said:

Sure, but a CN9 was designed to be a cavalry unit: OK-ish weaponry, focusing mainly on short/medium fights. But the main idea is to be speedy about supporting your flanks / flanking other lances.
A Hunchback wasn't designed to be all that in terms of speed, but to fit an AC/20.
So, in terms of differences, I'd suggest raising the armor cap on hunchies and let centurions get away with a bit more speed on lower rated engines.

I might be wrong about their intended design purpose, but you get the idea

CN9 and Hunchbacks are the exact same speed in TT. They weren't any speedier than Hunchbacks. Not sure where you get your 'calvary' example from.

[ShadowedR]

Davers, on 06 November 2013 - 02:54 PM, said:

CN9 and Hunchbacks are the exact same speed in TT. They weren't any speedier than Hunchbacks. Not sure where you get your 'calvary' example from.

It was suggested earlier in the thread that a CN9 was used as a cavalry unit.
Went with the context

[ShadowedR]

Tombstoner, on 06 November 2013 - 02:05 PM, said:

...

Just about everything in the game is balanced by feel. from what i have seen. How the game feels being played by the devs is very different then the new player and the highly skilled competitive player.

You hit it on the nose - the game is balanced by feel.

As for analyzing the game, I made peace with a broken system a long long time ago.
I do the analysis more for fun than expecting my post to alert the devs to problems.
It might, but w.e

[Karl Streiger]

Lykaon, on 06 November 2013 - 02:38 PM, said:

Mech names and shapes
Weights and weight classes
Tonnage and crit slots of mechs and weapons/items
Some sort of heat mechanic

Crit Slots are based on 2d6 - i really would like real hitboxes of criticals instead. Becaues most players tend to say: I want to hit were i shot and it is completely ignored that the Crits are random - you shoot for the SRM in an Atlas but your shot knocks out the LRM.

So you have to translate every descission that is made in TT by using dice into another system - reducing speed - fall damage. erattic movements - flickering hud and much more to reflect for example Pilot Saving Rolls.

Edited by Karl Streiger, 06 November 2013 - 11:56 PM.

[ShadowedR]

Karl Streiger, on 06 November 2013 - 11:55 PM, said:

Crit Slots are based on 2d6 - i really would like real hitboxes of criticals instead. Becaues most players tend to say: I want to hit were i shot and it is completely ignored that the Crits are random - you shoot for the SRM in an Atlas but your shot knocks out the LRM.

So you have to translate every descission that is made in TT by using dice into another system - reducing speed - fall damage. erattic movements - flickering hud and much more to reflect for example Pilot Saving Rolls.

One of the main challenges is how do you translate a system that was based on random's to a skill based system.

[Karl Streiger]

ShadowedR, on 07 November 2013 - 12:52 AM, said:

One of the main challenges is how do you translate a system that was based on random's to a skill based system.

OK 2d6 is at least not really random: 1d6 is random or 1d12 is random. Or in case of critical hits its also random each critical system could be hit with a chance of ~8.3 %.
Thats random.
But to hit distribution is not: its 2d6 - with a better chance of hitting CT or other facing torso sides.

It is true that you have to revamp the system from scratch - this would be a kind of vanilla solution:

Instead of TT ablative armor system with 8 hit zones - you have to use much more - with different armor values. (for example instead of a hugh RT location for the Atlas - you have a lightly armored shoulder area - shots may hit the shoulder and blockade the arm - damage power supply lines for the arm - even cut the arm of.
following is a best armored location - penetrating hits can damage XL engine or ammunition storrage
and in the lower part you have the AC 20...shots from the front can only damage the barrel - while shots from the flank my destroy ammo supply or hit the ammunition storrage.
Its more likely that you can destroy heatsinks with shots at the back - because the heatsinks are at the surface of the hull.

The modelling of a Mech has to follow the rules of TT TRO. When you say - the AC 2 is to dense lets make it for example 3 crits size - you have to add 4 additional side torso crits for the Mauler.
When you say that the Awesome 8Q need 56 SHS instead of 28 to work properly you have to bring huge heat dissipating blocks (good reason for the big torso of the Awesome) and 2 crits housing 3 SHS - for 2 tons.

So the desing prozess
-Mech Sketch -> Performance in Battle -> Solution for internal layout -> canonizing that layout.

I know I'm little off topic and i have to excuse. But the standard component Mech Lab of CBT - made for simplissity and speed of development from a time were only few people had an calcultor shouldn't be necessary in a modern computer game.

[ShadowedR]

If you look at something like EVE Online, it is impossible to fit Torpedo Launchers on a Frigate sized hull.
You do not have enough CPU and Power capacity to fit.



What they did, to get around this, was to add Ship Specific Role Bonuses.
For the Manticore, it reads:

Role Bonus:
-99.65% reduction in Torpedo Launcher powergrid needs
-50% reduction in Cloak CPU Use
-100% targeting delay after decloaking

Translating that to MWO, an Awesome should get a role bonus such as:
-30% Heat Generation from PPCs
Ghost Heat when fitting more than 3x PPCs
After all, a canonized build of the AWS is 3x PPCs

I think this could work to allow both a balanced game while allowing for specific builds that tie in with the TT universe

Edited by ShadowedR, 07 November 2013 - 01:23 AM.

[ShadowedR]

Karl Streiger, on 07 November 2013 - 01:12 AM, said:

I know I'm little off topic and i have to excuse. But the standard component Mech Lab of CBT - made for simplissity and speed of development from a time were only few people had an calcultor shouldn't be necessary in a modern computer game.

You're not that far off topic. I just highlighted that mechs of a different weight class get treated differently, but mechs of the same weight class get the same treatment.
I suggested that PGI look at tweaking individual mechs and not just the weight classes.
Your post follows in that line -> instead of keeping everything linear, to some extent, make every mech a unique beast

[Furball42]

ShadowedR, on 07 November 2013 - 01:22 AM, said:

If you look at something like EVE Online, it is impossible to fit Torpedo Launchers on a Frigate sized hull.
You do not have enough CPU and Power capacity to fit.



What they did, to get around this, was to add Ship Specific Role Bonuses.
For the Manticore, it reads:

Role Bonus:
-99.65% reduction in Torpedo Launcher powergrid needs
-50% reduction in Cloak CPU Use
-100% targeting delay after decloaking

Translating that to MWO, an Awesome should get a role bonus such as:
-30% Heat Generation from PPCs
Ghost Heat when fitting more than 3x PPCs
After all, a canonized build of the AWS is 3x PPCs

I think this could work to allow both a balanced game while allowing for specific builds that tie in with the TT universe

^ all of this.

"role warfare" was supposed to be part of MWO from the start and they way the described it sounded awesome. buuuut that kinda didn't realize the way we expected.

doing what CCP did with EVE would be a great move. would force certain chassis into the actual roles they were intended. make the Raven more intelligence orientated. spiders and commandos as proper scouts/harassers. catapults as fire support etc.

that would fix the problem of one chassis being another chassis to some extend. even variants in the same chassis.

maybe it's my pipedream for something more complex than slower Counter-strike. i'd LOVE subsystems that need to micro'd if something goes wrong, helping you extend your battle time just that bit more. i mean we are piloting advanced machines of war.. but i dunno, maybe complexity isn't the flavour for the mass gamer these days. ( and i don't mean mechlab complexity - i mean in-game). got this thing stuck in the back of my head where one can target 2 different mechs. once locked, one can still fire missiles at Target B while focusing on Target A (both locked).

eh *goes back to work*

[Ixios]

Great post ShadowedR.

Your post got me thinking on the topic of armor values, mech variants and balancing.

I think the issues you're highlighting here boiling down to:

In the conversion of a table top game into a FPS, your mech size didn't effect enemy accuracy.

Mediums suffer because they have large hitboxes respective to their armor values. Lets assume PGI will never rescale mech sizes as this would require too much work. What are the solutions to making mediums and more generally, inter tonnage mech variants unique?


A loadout Perks system:

One that helps increase diversity of loadouts by promoting particular weapons for specific inter mech variants, whilst having more general perks to mech chassis.

Just an example:

• CPLT - ALL: 20% increase in Arm max armor - (Flavor fluff: Sturdy construction and large)
  
• CPLT - C1: 15% increase in missile lock speed - (Flavor fluff: Improved navigation avionics )
  
• CPLT - C4: -1 ton weight for LRM15/20 weapons - (Flavor fluff: Advanced weapon mounts )
  
• CPLT - A1: 15% increase in Missile reload speed - (Flavor fluff: Rapid reloader tubes )
  
• CPLT - K2: -2 Heat to ERPPC/PCCs- (Flavor fluff: Thermally efficient arms )

These are just quick suggestions. The point being that a Perk system can encourage a wide variety of builds, and a reason to choose some of the weak variants.



Unique Armor per ton and Max Armor per Mech body location.

The purpose is to have a dynamic armor system that takes into account the physical size of the mech and its body locations, based on hitbox sizes and Mech tonnage. The then means that larger, easier to hit mechs have slightly higher armor, and thus the same effective survivability of a slightly smaller mech in it's same class. These bonuses can be rolled in with the perk system like I did with the CPLT.

Rules:

• Per Mech tonnage: Denser mechs have a Lower Max armor. ( Less space to mount the armor )

What this means is a 50 ton Trebuchet would have slightly higher Max armor then a 50 ton Centurion. This would probably be around 10% more: Both have 338 so a 10% increase would result in 371, which is the same as the base armor for a 55ton mech.

• Max Armor distribution is unique per body part. Large body parts have more armor.

If a mech has 60% of its body as CT, 60% of it's Max Armor should be in the CT. This would be another great way to fundamentally balance particular mechs hitboxes, whilst rewarding those who aim for the weak spots.




Conclusion:
I believe a combination of the above ideas implemented into the game would massively increase balance and diversity. It also introduces another way for the developers to balance particular overpowered builds and chassis.

Ix

Edited by Ixios, 07 November 2013 - 03:23 AM.

[ShadowedR]

Maybe this thread wasn't as aptly titled as I initially thought.

It seems that the discussion has migrated more towards:
1. Break away from the Table Top mechanics that simply doesn't work
2. Start tweaking the mech chassis themselves
3. Create unique and challenging mechs to pilot while still aiming to keep things balanced out for the competitive community.

PGI has gone into terminal tunnel vision - tweaking the modules and equipment like there is no tomorrow.
A mech is comprised of 3 main things:

1. Chassis
2. Loadout
3. Pilot
PGI has no control over the pilot's abilities. So, lets skip it.
Why is the main "balance" focus only on the equipment? A sizable portion of a mech's performance is based on the chassis' ability.

I'd LOVE to see the EVE-esque role bonuses come into MWO.
Along with a few armor/speed tweaks that break from the TT standards and the inclusion of a few modules (such as the Modular Armor mentioned earlier).

One can dream, can't they?