362 replies to this topic

[Hillslam]

Lights at high speed should have extremely large turning radii. Physics.

[Khobai]

Quote

I'd say it's the heavies that needs slowing down, not the mediums that need speeding up.

Nope mediums definitely need speeding up. Because if you slowed down heavies youd still have a HUGE speed disparity between lights and mediums. Mediums need to go much faster in order to counter lights better.

Quote

Not really. Even on TT speed was the Light's best friend. A Panther v an Atlas was pretty much a known outcome. Outside of luck.

Not really, because getting the defensive bonus for movement required you to run in a straight line which was very difficult to do every single turn. The truth is JJs were a light's best friend. Lights had to end the turn in the rear arcs of other mechs in order to survive, and JJs were the best way of doing that by far. Running really fast was actually pretty useless in TT, because the vast majority of the time, JJs were outright better.

Edited by Khobai, 23 December 2013 - 10:29 AM.

[Joseph Mallan]

Khobai, on 23 December 2013 - 10:19 AM, said:

Nope mediums definitely need speeding up. Because if you slowed down heavies youd still have a HUGE speed disparity between lights and mediums. Mediums need to go much faster in order to counter lights better.



Not really, because getting the defensive bonus for movement required you to run in a straight line which was very difficult to do every single turn. The truth is JJs were a light's best friend. Lights had to end the turn in the rear arcs of other mechs in order to survive, and JJs were the best way of doing that by far. Running really fast was actually pretty useless in TT, because the vast majority of the time, JJs were outright better.

IF. A Panther was a 4/6 on TT, so getting to an Atlas was problematic. Depending on Terrain. I have never seen a Panther beat an Atlas 1 on 1 in 30 years.

With the tournament rules you are right, but add the Advanced rules where a Light could generate a +7 to hit due to speed.

[Trauglodyte]

I don't think that Mediums need to be sped up. They're about where they need to be. But, like we've said, if a larger mech and twist and keep you lined up, what good is the heavier engine that you've now got and the slightly faster speed than you had before? In most cases, a higher top speed actually gimps your payload. Or, you end up dropping in an XL engine and open yourself up to more issues.

I think acceleration/deceleration is a big sticking point. But, I've got no access to the numbers so what do I know?

[YueFei]

This thread, while well-meaning, is misguided. Even if we grant that a perfectly piloted Light mech can always defeat a perfectly piloted Assault mech in a one-on-one duel, Light mechs would *still* get crushed by Assault mechs in a team battle. This is because while you are dancing around staying perfectly out of the firing arc of 1 Assault mech, another Assault mech 150 meters away is gonna blow you away.

Ideally, Light mechs should be able to counter Assault mechs and slow Heavies by wearing them down or causing their demise *indirectly*. This means being able to spot for indirect fire like LRMs, or to drop air strikes and artillery strikes that attrition the Assaults down.

There are some issues with Light mechs doing this, however. The first is that the threat envelope that an Assault mech presents to a Light mech is too large. Projectile speeds are primarily to blame for this. A Light mech trying to close within TAG range to cut through an enemy ECM blanket must expose himself dangerously to return fire. I did the math in MavRCK's Mech Tier List thread, and a Light mech even past max PPC range cannot react fast enough to dodge PPC fire. At the moment, in a contest of perfect aim vs perfect dodging, perfect aim will win even at long range. Unless you have superhuman reflexes and can see, decide, and act to dodge a shot in less than 200 milliseconds.

Air and Artillery strikes are the safest way for Light mechs to drop damage on the Assault mechs, but they are consumables, and dodgeable by Assault mechs. Assault mechs have 4 seconds to react to an air or artillery strike. That is an eternity compared to the 200 milliseconds a Light mech pilot has to do-or-die against PPC and AC fire.

We could make Air and Artillery strikes on-map assets that can be called upon repeatedly, rather than a consumable item. Reduce the damage-per-shell, but increase the radius of the strike and increase the number of shells to saturate the area. This way, it is not a potential one-shot kill from hitting the cockpit, but the larger area makes it reliably deal damage even to Assault mechs who react immediately to clear the area. This gives Light mechs the ability to gradually wear down Assault mechs with repeated strikes.

Reduce projectile speeds for PPC and AC just a tiny bit, so that Light mechs can spot for LRMs and be able to dodge incoming fire. Either that, or increase Light mech acceleration and decelleration. Currently even if it takes a projectile nearly half a second to reach you, a Light mech does not accelerate or decellerate fast enough to fully dodge it even if the pilot reacts within 200 milliseconds.

Now Role Warfare emerges. Light mechs have the ability to gradually wear down and kill Assault mechs, by directing air and artillery strikes, and spotting for their team's LRMs. Mediums and Fast Heavies can fully dodge air and artillery strikes, and LRMs are inefficient against them.... these Mediums and Fast Heavies can form a screening element, to prevent enemy Lights from getting eyes on your Assault mechs, to chase off or kill those enemy Light mechs. And finally, those Mediums and Fast Heavies would get hammered by Assault mechs and slow Heavies if they get too close to them (so be careful how far you chase those Light mechs!).

Throw away those tonnage limit ideas. There would be no need for them anymore.

Edited by YueFei, 23 December 2013 - 01:22 PM.

[Sable Dove]

YueFei, on 23 December 2013 - 01:19 PM, said:

This thread, while well-meaning, is misguided. Even if we grant that a perfectly piloted Light mech can always defeat a perfectly piloted Assault mech in a one-on-one duel, Light mechs would *still* get crushed by Assault mechs in a team battle. This is because while you are dancing around staying perfectly out of the firing arc of 1 Assault mech, another Assault mech 150 meters away is gonna blow you away.

Ideally, Light mechs should be able to counter Assault mechs and slow Heavies by wearing them down or causing their demise *indirectly*. This means being able to spot for indirect fire like LRMs, or to drop air strikes and artillery strikes that attrition the Assaults down.

There are some issues with Light mechs doing this, however. The first is that the threat envelope that an Assault mech presents to a Light mech is too large. Projectile speeds are primarily to blame for this. A Light mech trying to close within TAG range to cut through an enemy ECM blanket must expose himself dangerously to return fire. I did the math in MavRCK's Mech Tier List thread, and a Light mech even past max PPC range cannot react fast enough to dodge PPC fire. At the moment, in a contest of perfect aim vs perfect dodging, perfect aim will win even at long range. Unless you have superhuman reflexes and can see, decide, and act to dodge a shot in less than 200 milliseconds.

Air and Artillery strikes are the safest way for Light mechs to drop damage on the Assault mechs, but they are consumables, and dodgeable by Assault mechs. Assault mechs have 4 seconds to react to an air or artillery strike. That is an eternity compared to the 200 milliseconds a Light mech pilot has to do-or-die against PPC and AC fire.

We could make Air and Artillery strikes on-map assets that can be called upon repeatedly, rather than a consumable item. Reduce the damage-per-shell, but increase the radius of the strike and increase the number of shells to saturate the area. This way, it is not a potential one-shot kill from hitting the cockpit, but the larger area makes it reliably deal damage even to Assault mechs who react immediately to clear the area. This gives Light mechs the ability to gradually wear down Assault mechs with repeated strikes.

Reduce projectile speeds for PPC and AC just a tiny bit, so that Light mechs can spot for LRMs and be able to dodge incoming fire. Either that, or increase Light mech acceleration and decelleration. Currently even if it takes a projectile nearly half a second to reach you, a Light mech does not accelerate or decellerate fast enough to fully dodge it even if the pilot reacts within 200 milliseconds.

Now Role Warfare emerges. Light mechs have the ability to gradually wear down and kill Assault mechs, by directing air and artillery strikes, and spotting for their team's LRMs. Mediums and Fast Heavies can fully dodge air and artillery strikes, and LRMs are inefficient against them.... these Mediums and Fast Heavies can form a screening element, to prevent enemy Lights from getting eyes on your Assault mechs, to chase off or kill those enemy Light mechs. And finally, those Mediums and Fast Heavies would get hammered by Assault mechs and slow Heavies if they get too close to them (so be careful how far you chase those Light mechs!).

Throw away those tonnage limit ideas. There would be no need for them anymore.

There's no one way to fix everything wrong with the game, but there's no reason your ideas and mine can't coexist. However...

I'm actually opposed to reducing projectile speeds though. It plays hell with hit detection, and it's why the AC20 has such horribly inconsistent hit detection (and I would wager it's part of why SRMs have so many problems). I would support reducing ballistics to double-range instead of triple though (and reducing the ERPPC's range to what it's supposed to be).


Maybe a better way would be to reduce (base) sensor range, and increase TAG range and the benefits from the sensor modules/BAP. I also suggested some time ago that mechs have a delay before they are automatically detected that scales with their size/weight. ((Here)) Similar to how movement archetypes were implemented. This would allow light mechs to out-range assaults' sensors and mark targets without the game instantly telling said assaults exactly where the light is.

Perfect aim isn't so bad by itself. But it is when the game automatically identifies targets for you, and you have the agility get them in your sights and shoot them before they can get back out of sight.

There are many ways to introduce role warfare. If only we could get PGI to do any one of them, it'd be an improvement.

[Myomes]

You're treating the symptom, not the cause,.

The cause is that Engine size modifies agility, such as turn and twist speeds.

A assault mech with a 320 engine and full levels up in pilot "skills" is going to turn on a dime.

The cause is that engines change agility at all, rather than just velocity.

Treat the cause, divorce engine from agility, don't treat the symptom and bloat and distort the game even further.

[Diego Angelus]

Myomes, on 23 December 2013 - 03:18 PM, said:

You're treating the symptom, not the cause,.

The cause is that Engine size modifies agility, such as turn and twist speeds.

A assault mech with a 320 engine and full levels up in pilot "skills" is going to turn on a dime.

The cause is that engines change agility at all, rather than just velocity.

Treat the cause, divorce engine from agility, don't treat the symptom and bloat and distort the game even further.

Yeah but I would go one step further and tune down speed I mean some mech builds are just going to fast for their class.

[Sable Dove]

Myomes, on 23 December 2013 - 03:18 PM, said:

You're treating the symptom, not the cause,.

The cause is that Engine size modifies agility, such as turn and twist speeds.

A assault mech with a 320 engine and full levels up in pilot "skills" is going to turn on a dime.

The cause is that engines change agility at all, rather than just velocity.

Treat the cause, divorce engine from agility, don't treat the symptom and bloat and distort the game even further.

An assault mech with a stock engine moves and turns too fast with the efficiencies. Engine rating changing agility isn't the issue, if any one thing is. If they removed or reworked the mech efficiencies so that they didn't benefit heavies and assaults most, these problems would be much smaller.

[Myomes]

Diego Angelus, on 23 December 2013 - 03:21 PM, said:

Yeah but I would go one step further and tune down speed I mean some mech builds are just going to fast for their class.

had I been in charge of both PGI and IGP, I'd have told them to implement the pilot skill trees so that your mech underperforms compared to TT, and then you level it up so that it performs like TT.

Would've made more sense instead of 172 kph lagshielding jenners

[mark v92]

i fully support reducing agility. maybe more in the lines of MWLL?

[Myomes]

honestly, I've given my opinion a dozen times over the last few years, all of them serially ignored.

Lights were never meant to tackle mediums, heavies, or assaults 1v1, or even 2v1. Their speed in BT didn't do that, it helped them find cover. They were meant to scout or to hit hard and fast and ****. Not to run around in circles forever and ever being untouchable by the bigger mechs.

Edited by Myomes, 23 December 2013 - 04:48 PM.

[Diego Angelus]

Myomes, on 23 December 2013 - 04:48 PM, said:

honestly, I've given my opinion a dozen times over the last few years, all of them serially ignored.

Lights were never meant to tackle mediums, heavies, or assaults 1v1, or even 2v1. Their speed in BT didn't do that, it helped them find cover. They were meant to scout or to hit hard and fast and ****. Not to run around in circles forever and ever being untouchable by the bigger mechs.

Yeah if we had large maps and mechs slowed down lights would be real scouts/harassers and game overall would be more tactical.

[Myomes]

alpine peaks is a very large map but with PUGs it tends to just be a tactless slugfest over a single ridge for some reason. Also, it murders my frames per second.

[Fooooo]

I really don't mind the idea of slowing or reducing the agility of assaults / heavys, well.....at least some of them...
(dragon / quickdraw would stay agile and some others if they ever get in game).

Without all the eff and elite I still think an atlas is too agile. Mainly in torso turning speed. I think reducing this alone for most of the heavier mechs may also acheive the same results as doing torso and leg agility........or at least, close to......

I also believe that you should not be able to make your mech any better than stock in terms of agility etc via pilot skills, mech skills or modules or engines.....engines for speed...ok...but not turning speed etc......

(ok maybe....MAYBE modules or skills but only if there are 3 or 4 lines in the new skill revamp, and you can only go down 1 line for each mech etc.....sort of like WoW talent trees of old or most RPG's.....)

Edited by Fooooo, 23 December 2013 - 08:55 PM.

[Myomes]

Fooooo, on 23 December 2013 - 08:51 PM, said:

I really don't mind the idea of slowing or reducing the agility of assaults / heavys, well.....at least some of them...
(dragon / quickdraw would stay agile and some others if they ever get in game).

Without all the eff and elite I still think an atlas is too agile. Mainly in torso turning speed. I think reducing this alone for most of the heavier mechs may also acheive the same results as doing torso and leg agility........or at least, close to......

I also believe that you should not be able to make your mech any better than stock in terms of agility etc via pilot skills, mech skills or modules or engines.....engines for speed...ok...but not turning speed etc......

(ok maybe....MAYBE modules or skills but only if there are 3 or 4 lines in the new skill revamp, and you can only go down 1 line for each mech etc.....sort of like WoW talent trees of old or most RPG's.....)

Well...

The reason some heavies and assaults have such "agility" is due to how they tied engine size into it.

I already spoke a little about how this can make a fully trained assault turn on a dime with a large engine. If you've seen the special Awesome mech with the biggest engine possible, that thing can literally spin around in less than half a second between turn and twist so it's facing right behind it.

In contrast, try taking the smallest engine possible on an assault mech. You should be going about 32 Kp/h. Now try moving around and shooting. You'll notice that you turn and twist slow as {Scrap}, and you'd be practically useless in a fight due to the rate of turn and twist.

Let's take the http://www.sarna.net/wiki/Annihilator

It moves 32 Kph as well. But in TT, it, like all other mechs of any size, has enough turning speed to turn a certain radius in a single move. It didn't have any problems in previous Mechwarrior incarnations of which it was a part.

You think it would survive or last in this game? It is my belief that if you cant drop in a mech from other titles or BT and have it function properly that you have, somewhere along the line, a fundamentally flawed system.

The agility tied to engine is such a flaw. That heightened agility you're seeing in those atlases? That's because they took biggest XL engine possible. It's not the engine's fault, nor the mech, it's the fault of introducing agility bonuses based around the engine.

Im spectating a Jagermech with an 82.2 kph engine and the torso twist rate on it is near-instant, practically related to only how fast can the person playing move their mouse..

Edited by Myomes, 23 December 2013 - 09:40 PM.

[Myomes]

Update for some numbers to give people an idea of what agility can do.

I chose a fully Tier 1 leveled Thunderbolt.
No modules.

Engine 130
Speed 32 kph
360 degree turn time = 15 seconds.
180 degree torso twist time = 5 seconds.

Engine 300
Speed 74 kph
360 degree turn time = 7 seconds.
180 degree torso twist time = 2 seconds.

For an Awesome.
Fully trained tier 1 skills.

Engine 160
Speed 32 kph
360 degree turn time = 15 seconds
180 degree torso twist time = 5 seconds.

Engine 300
Speed 60.8 kph
360 degree turn time = 9 seconds.
180 degree torso twist time = 3 seconds.

You can see how lower engine size is punishing even though it'd be a relevant and valued alternative in game to modify a mech build so it could be slower but carry more weapons, as in the annihilator. It's doubly punishing to not only be slow and unable to escape enemies with some heavy or assault mechs' stock engines, but also have a much reduced turning and twisting speed as well.

Conversely, it's extremely powerful to go max engine on things such as atlases, as you pointed out. Talk about a light-flyswatter. Additionally, this is why heavier trial mechs are worse for new players. Heavy-Assault trial mechs that go 64 kph or 52-48 kph, respectively, and especially without any skills, become deathtraps for new players against people with bigger engines in their mechs, such as maxed out medium or light swarmers. I only add this last to claim that not only does it weaken gameplay flexibility and diversity, but it also harms new players coming into the game, which may not be good for player retention, if PGI cares about such a thing.

Edited by Myomes, 23 December 2013 - 10:15 PM.

[Roland]

Hillslam, on 23 December 2013 - 08:12 AM, said:

Lights at high speed should have extremely large turning radii. Physics.

Bipedal movement doesn't have the same restrictions that wheeled movement does... that's generally why you would want to have legs instead of wheels.

The chief difference is that legs are able to direct force at different angles, instead of perpendicular to the ground's surface. This is how you're able to do stuff like shuttle runs.

[YueFei]

Sable Dove, on 23 December 2013 - 02:48 PM, said:

There's no one way to fix everything wrong with the game, but there's no reason your ideas and mine can't coexist. However...

I'm actually opposed to reducing projectile speeds though. It plays hell with hit detection, and it's why the AC20 has such horribly inconsistent hit detection (and I would wager it's part of why SRMs have so many problems). I would support reducing ballistics to double-range instead of triple though (and reducing the ERPPC's range to what it's supposed to be).

You're right that reducing projectile speed could make issues with HSR. I mentioned another solution, which is to *increase* Light mech acceleration/deceleration. Gotta be able to move about 3 meters in 300 milliseconds. That's an acceleration of ~55 meters/sec/sec. Basically, you have to do 0 to 150 kph in 0.75 seconds.

Increasing accel/decal for Mediums would also help the short-range Medium strikers. They could step laterally out of cover, and if someone's already looking at them, they can step back into cover more quickly. As it is now, I feel safer peeking over hills in my Hunchback than I do stepping around a building, and that's just not right. With hills, if I maneuver properly, I can stick only my head out for a split second for a quick look by running a parabolic arc along the edge of the hill. I can't do that at the edge of a building.

Quote

Maybe a better way would be to reduce (base) sensor range, and increase TAG range and the benefits from the sensor modules/BAP. I also suggested some time ago that mechs have a delay before they are automatically detected that scales with their size/weight. ((Here)) Similar to how movement archetypes were implemented. This would allow light mechs to out-range assaults' sensors and mark targets without the game instantly telling said assaults exactly where the light is.

This is absolutely a great idea. Detection range must scale with mech size. The larger the mech, the farther out sensors should be able to detect it. And the smaller it is, the closer you have to be to pick it up. Increasing TAG range would also help.

Quote

Perfect aim isn't so bad by itself. But it is when the game automatically identifies targets for you, and you have the agility get them in your sights and shoot them before they can get back out of sight.

I agree that perfect aim by itself isn't so bad, but the LRM+spotter mechanic forces the spotter to be exposed for such a long time... dropping air and artillery is a different story, since you can pop out, make sure no one is looking directly at you, spend a split second to line up your air or artillery strike, drop it, and run away. With your sensor range categories, it might actually be possible for a light mech to spot for extended periods of time without being spotted and shot at. Maybe also make the TAG beam invisible? Or rework ECM so it doesn't act as a null-signature system, then TAG wouldn't be absolutely necessary to cut through an ECM umbrella. =/

Quote

There are many ways to introduce role warfare. If only we could get PGI to do any one of them, it'd be an improvement.

Yeah, a lot of great ideas have been suggested by folks in the community. Unfortunately even the simple ones that involve a few minutes of time to edit an XML file seem to never happen.

[Grrzoot]

honestly this would be the easiest thing to do to increase the overall ttk of mechs and give more meaning to role warfare.

question is, could it be abused?

Edited by Grrzoot, 23 December 2013 - 10:36 PM.