704 replies to this topic

[Spades Kincaid]

dimstog, on 15 June 2013 - 12:02 AM, said:

There's still hope then !?



That is a valid argument, however you base it on the presumption that the Targeting Computer keeps constant convergence and missile lock on the target.

Since the TC is not "psychic" as you put it, it does not even converge the weapons until you actually pull the trigger. That might not sound right, but since we accept that maintaining lock and converging weapons on the target taxes the TC, maintaining lock and keeping constant convergence on the target with ALL weapons until you decide to fire one or all of them, would mean that the TC is constantly taxed with maximum penalties! So, if we accept Bill's system, the only way for it to work, is to accept that the convergence is calculated after you pull the trigger, not before.

Thus, overtaxing or not the TC and detrimental effects are calculated after you pull the trigger, and causality is maintained !

When it comes to missiles, the theory is the same, the TC keeps the target locked, but calculates flight paths after you pull the trigger. Depending on previous load and added load of the second shot, it either makes it or overtaxes itself. The loss of lock in this case is logical since the LRM guidance system is semi-active. The missiles while in flight are guided to the target by your mech, not by their on-board guidance system.

So the system is not theoretically unsound or illogical by any means.

Science fiction mumbo-jambo aside, it is a bit un-intuitive, in that the player has to actually know if the shot they take is going to overtax the TC by "memorizing" the TC values of their weapons - which is not bad imho, you have to know your mech buddy! There is no way of showing that before they actually take the shot. And in order to show the player whether their shot would overtax the TC we would need actually need either a "psychic" interface or a yes/no box : "Your shot is going to overtax the targeting system, do you want to take it ?"

You make a good point, I can understand what you are trying to say.

But, it's premised on 'accepting Bill's system' as a default and that's the trouble I have. Fact is, at least currently, that -is- how it works. Your weapons converge based upon what you aim at, they don't wait until you pull the trigger. Probably because that would be clunky and inefficient, causing a delay from the trigger pull to your actual fire, for the weapons to converge. It just wouldn't make sense for the targeting computer to do that.

Same with the LRM's, a lock basically -is- a firing solution. That's rather the point of needing a lock. You can't fire until the computer has acquired a target -and- has calculated a flight path to that target. It's the latter that you are waiting on for a missile lock. It wouldn't fire the missiles without being ready to tell them where to go.

I get what you are saying, it has to continue providing updates and changes to that flight path. Though I'm not sure how that would be more taxing than calculating it in the first place. Of course, I'm no computer expert so maybe it would be.

Anyway, you are correct lore fluff isn't important. And lore fluff wasn't ever my real point. But to me, an intuitive feel is and that's often correlated to 'does this make sense?'. The examples I gave of instant complete loss of convergence or loss of missile lock, don't make sense to me. They seem clunky and 'off' for the reasons above.

Again, I really like the general concept of the system. I think perhaps it's a bit too ambitious however, in trying to solve -every- problem with one fell swoop. Like some things got wedged in that don't really fit, just to address a certain issue. And I think it suffers for it.

Problem is, if it doesn't do that then despite it's originality and thoughtfulness, it doesn't have as much going for it over using the measures already in game (like changing heat management and just changing the way convergence works in itself).

Edited by Spades Kincaid, 15 June 2013 - 03:43 PM.

[Lightfoot]

Just shrink the torso hitboxes and increase the arm hitboxes. See what happens.

[Tombstoner]

Lightfoot, on 15 June 2013 - 03:56 PM, said:

Just shrink the torso hitboxes and increase the arm hitboxes. See what happens.

you start getting hit when people shoot just over your head or if the arm box sticks out around the corner.
on the mech is makes aiming for the ct harder. that i like, but its an inelegant solution. this game has good hit detection for placment of damage. its just the system is just way to easy to link fire a high damage alpha into your CT. adjsuting the area i think is too old school and kinda deminishes what skill in aiming there is. do able but cheep. other MW games used this apporach. they also had pin point damage and that is the main issue.

[Rhakhas]

So a very simple alternative (that could be used as a temporary solution?) could be to have pinpoint accuracy for single / chain fire and use fixed convergence (at weapon max effective range) for group fire. It's not as flexible or robust as Bill's solution, but would likely be much easier to implement.

[Zaptruder]

It's good. It seems like a workable version of the much needed divergence solution for the high pin point damage woes the game is currently suffering.

Those numbers can obviously be tweaked, but I really like the idea of combining what is essentially an energy system with a spatial divergence system.

Personally, I think the energy system is easier to go with, but if the community is more accepting of the fluff behind this solution, then that's perfectly fine for me too.

An addendum suggestion for the idea proposed in the OP though: In the weapon grouping list, groups that exceed 100 'energy' are highlighted in a different colour to show the player that weapon group will suffer from inaccuracy if fired in a single shot.

So 2xGauss groups = highlighted in red.

Or even better, highlight the groups that aren't going to experience TC shake with green.

And on the 6 weapon group boxes surrounding the reticle; a weapon group that is either on chain fire or will fire as normal will be in green, a group on cool down will be in red and a group that will have accuracy penalties will be in yellow.

As for the effect of the TC, I think the less TC 'energy' you have, the greater the inaccuracy could be. So just peaking over 100 energy you get deconvergence, while pushing it over 150 would cause a COF penalty to shots (not dissimilar to the JJ shake COF... without the cockpit shake - maybe some reticle shake to indicate the load on the TC).

YOU HAVE MY SWORD HOMELESS BILL.

Edited by Zaptruder, 15 June 2013 - 11:42 PM.

[Zaptruder]

Soy, on 14 June 2013 - 07:09 PM, said:

Point was, Clans shouldn't be 'balanced'. They were never supposed to be 'balanced'. The deus ex machina for them is supposed to be, well, Inner Sphere zerg strat. Obviously that will take some hand of god mechanics in CW and stuff to get right. But right isn't balanced, hehe...

Clans will still be far more powerful than IS mechs... but allowing them to engage in the current high pin point damage metas would just lead to an amplification of the current metagame woes.

[HRR Insanity]

Pater Mors, on 14 June 2013 - 02:47 PM, said:

Hmmm. After reading through the OP I still feel that a properly implemented heat system will solve almost all these problems.

Yes, you still get that first monster Alpha but you die or severely cripple yourself instantly. No more cheesing every Mech you meet because you're dead or can barely move/fire.

Drop heat capacity drastically. Raise heat dissipation drastically. Suddenly it's infeasible to Alpha because you're going to cook yourself in the process 9 times out of 10. Firing discipline comes back. Chain fire becomes the norm. Versatile builds become the norm. Boating still works but requires a lot more skill to do properly. Alpha is put back in it's place as a last resort tactic.

Edit: I'd like to see penalties for constantly running hot as well. Say for more than a few seconds above 80% heat. Convergence could certainly be a part of these penalties.

Ballistics require more skill to use than lasers, PPC's and missiles and I think that the increase in skill is a good enough trade off for not having as much in the way of heat penalties. The AC40 Jag is only dangerous if the pilot can shoot and get within effective range and they're very squishy anyway. If it's really that bothersome, mess with the projectile speeds and reload times a bit to increase the skill needed to use ballistics. Also the chance of ammo explosion should be upped (20 - 30%) when taking damage to a stripped section with ammo in it.

Incorrect. The 2xGR has been and will be dominant again if heat is used as a way to 'fix' alpha builds. GR builds are completely unaffected by heat modifications and when used well are completely dominant.

[Zyllos]

HRR Insanity, on 16 June 2013 - 08:20 AM, said:

Incorrect. The 2xGR has been and will be dominant again if heat is used as a way to 'fix' alpha builds. GR builds are completely unaffected by heat modifications and when used well are completely dominant.

So true.

Balancing "alpha strikes" solely by heat changes does not fix the problem 100%.

It will take some heat changes, convergence changes, and hardpoint changes to get everything in the correct order.

[Homeless Bill]

Pater Mors, on 14 June 2013 - 09:04 PM, said:

1 - New players. Firstly I agree that a heat system is punishing to them, but it was punishing in TT and in all the original MW games and people still got on fine. I think that if the trial Mech system is scrapped and tutorials are implemented properly than this will become a moot issue. Thinking man's shooter remember? Getting punished a few times is how you learn to be a better pilot.
2 - I've always thought as Alpha Strikes as a last resort option but that's just my opinion. That last bit of payback when you know you're screwed anyway. But that being said, there isn't any reason that the heat cap has to be so low that Alpha Strikes aren't feasible. Just low enough to prevent the Super Alpha Meta that's currently in-game.

The thing about ballistics is they require a lot to use.

In the future we will have the bigger, badder Mechs that can pack huge amounts of ballistics and I think that some form of recoil penalty on convergence should be applied when firing multiples of large ballistics.

But, I worry that this is yet another system on top of what we already have and I feel like it can all be achieved using the systems currently in place.

1. I'm all for a thinking man's shooter, but I think accessibility is something to shoot for. Being able to nuke yourself so quickly is a bit harsh, but that's just my opinion. It's not that bad, it's just a touch too far for my taste.
2. Again, we just have a difference of opinion here. I like them as a desperate measure, but not a vengefully suicidal one. I love watching huge blasts of damage sail between two assault 'mechs.

I'll agree that ballistics have a lot of inherent disadvantages, but I think the pinpoint damage and low heat offset its tonnage and dependence on ammunition. I think raising the chance of ammo explosions would actually be to harsh to ballistic 'mechs; anyone that knew what they were doing would go for the legs and score an explosion far more frequently.

See, that was my initial thought: heat for everything else and recoil for ballistics. It makes perfect sense. But then it doesn't really touch missiles, and you have two sets of numbers solving a single problem. The more I thought on it, the more I saw the need for a totally separate set of numbers.

I won't respond to this now, but I'll update the OP when I publish the article on Monday. I adamantly believe de-coupling these systems is the way to go. It's not a small problem, the dynamics will be altered by future weapon additions, and I think a robust and flexible solution is the only logical way to go.

Spades Kincaid, on 14 June 2013 - 10:19 PM, said:

I want to like this system, I really do. But I just can't get behind preemptive consequences to an action that hasn't been taken yet.
No matter how you slice a loss of missile lock that occurs because you tried to fire before you had enough TCS, it makes no sense. You had the lock. How did pulling the trigger on that lock now overtax the targeting computer?

Trying to preemptively penalize a player for an action is never going to make more sense than the punishing afterward that you dismissed as inferior earlier in this thread.

The quick dissipation of TCL and the HUD will solve this problem. The weapon group icons around the reticle will change colors based on what TCL will be at: green for <=100, yellow for >100, and red for >200. Players will be able to know before they fire what the consequences will be.

Furthermore, it dissipates quickly enough that heat and weapon cooldown are still, by far, the limiting factors in DPS. This system might confuse players for a round or two, but the way it works is pretty intuitive, and the HUD will do an adequate job of conveying the consequences before the action is taken. Once players are used to slightly staggering their fire, this will not be an issue.

Sable Phoenix, on 14 June 2013 - 11:40 PM, said:

I started reading this post feeling extremely skeptical. By the end, you had convinced me totally.

Mahws, on 15 June 2013 - 12:42 AM, said:

Completely unconvinced at the start, completely convinced by the end.

The only thing I'd suggest differently is dropping the lock on loss for SSRM/LRM and increasing the TCL cost for all missiles. Instead of losing your lock it increases spread, LRM/SSRM would bunch up less the more you fire (A LRM5 at short range would hit mostly the CT, 5xLRM20 would rain all over a 50m square area) and SRM would become dramatically less accurate when fired in mass. Mechs like the Catapult could have a missile TCL cost reduction quirk to help preserve their role as dedicated missile boats.

There are few things more satisfying than hearing that.

That's actually a pretty good idea, but I still like it functioning as the boolean punishment that it does for all other weapon types. It would be much more work for PGI to implement (they'd need to change quite a few things in the missile code), and it would be hard to communicate to the player, "When you went over 150TCL when you fired those missiles five seconds ago, it made the spread huge." It's too delayed for it to be intuitive in the case of LRMs.

BigMekkUrDakka, on 15 June 2013 - 12:02 AM, said:

TL DR but i like how much effort you put into it, but please shorten it or make it easier to read (ie less formatting like this)

Ironically, the bunderlined text is a stand-alone, abridged version of the OP - not emphasis.

Zyllos, on 15 June 2013 - 08:27 AM, said:

On regarding introducing removed convergence before CoF, why is that more effective than CoF first?

But that is the problem, is that CoF for weapons at 800m or 200m? A CoF has to be centered around a specific distance because a CoF tuned at 800m will provide less spray up close while a CoF at 200m will provide spray at basically all distances. Also, at the brawler range, having a CoF would still effect them because they will be hitting locations they are not aiming at.

So, basically, what I am asking is that I think it would be better to have just an expanding CoF up to the point where you just lose convergence, along with the CoF, which would make fighting basically impossible to achieve effectively, at any ranges.

I would also suggest that as the TCL goes up, and CoF increases, I would suggest making the lockon time last longer and time to lose lockon when not aiming at the selected target decreases down to instantly. Which would basically make it where if your ridicule ever leaves that box for any amount of time, you lost the lock and it will take even longer to gain the lock again. The only time I would make missiles impossible to lockon is if you lose convergence.

A cone of fire should be balanced for long-range (500m+). Balancing a cone of fire for <100m makes the cone of fire way too harsh for any medium- or long-range engagement.

And because it's balanced for long-range, at 100-200m, it hardly has any effect at all. Here's why convergence loss at 201 instead of 101 would not fix the problem: 4xPPC Stalker or 2xAC/20 Jagermech at 150m. Even the 6xSRM6 Catapult will need a penalty once SRMs get some love. A cone of fire will have basically no effect on SRMs, while convergence loss will.

It's jarring, but it's meant to be. You're not supposed to go over 100, just like on the heat scale. To me, 40xLRMs/second is extremely generous. I favor simply losing lock the minute you're greedy enough to hit 101, just like convergence.

Rhakhas, on 15 June 2013 - 07:27 PM, said:

So a very simple alternative (that could be used as a temporary solution?) could be to have pinpoint accuracy for single / chain fire and use fixed convergence (at weapon max effective range) for group fire. It's not as flexible or robust as Bill's solution, but would likely be much easier to implement.

Fixed convergence would just push combat to a certain range and set of weapons. We've had enough bandaids, and I think it's time for a robust solution.


Also, OP updated with HUD mock-up.

Edited by Homeless Bill, 16 June 2013 - 01:13 PM.

[Pater Mors]

Homeless Bill, on 16 June 2013 - 01:12 PM, said:

1. I'm all for a thinking man's shooter, but I think accessibility is something to shoot for. Being able to nuke yourself so quickly is a bit harsh, but that's just my opinion. It's not that bad, it's just a touch too far for my taste.
2. Again, we just have a difference of opinion here. I like them as a desperate measure, but not a vengefully suicidal one. I love watching huge blasts of damage sail between two assault 'mechs.

I'll agree that ballistics have a lot of inherent disadvantages, but I think the pinpoint damage and low heat offset its tonnage and dependence on ammunition. I think raising the chance of ammo explosions would actually be to harsh to ballistic 'mechs; anyone that knew what they were doing would go for the legs and score an explosion far more frequently.

See, that was my initial thought: heat for everything else and recoil for ballistics. It makes perfect sense. But then it doesn't really touch missiles, and you have two sets of numbers solving a single problem. The more I thought on it, the more I saw the need for a totally separate set of numbers.

I won't respond to this now, but I'll update the OP when I publish the article on Monday. I adamantly believe de-coupling these systems is the way to go. It's not a small problem, the dynamics will be altered by future weapon additions, and I think a robust and flexible solution is the only logical way to go.

That's a fair enough response.

I think ballistics should suffer an inherent convergence penalty anyway because it's simply physics. What we have now are arcade 100% accurate gunz from Quake and I think coupled with heat penalties, realistic ballistics removes issues. Again, with the ammo stuff; thinking man's shooter. Everyone dumps ammo in the legs now because there's no penalty for it. If you actually had to consider where you put your ammo more thoughtfully it would make for a more interesting game and CASE would be used a lot more often (I literally never use it now, no reason to).

As for missiles, just bump the heat slightly, problem solved. I think you should be able to spam a couple of LRM40 shots or SRM 18 shots anyway and with the current state of AMS the balance there is feeling really good to me lately.

Edited by Pater Mors, 16 June 2013 - 01:20 PM.

[Lazy Eye]

While I do not agree with your solution, I whole-heartedly support your analysis!

Yours is the most comprehensive treatment of the reasons for the problems and why solutions won't work that I've seen to date.

There is, however, one issue that you have missed: rate of fire. Until rate of fire is addressed, no change can fix the Gauss.
The point of the Gauss in the TT game, was that it had a minimum range and that you'd only get 1 or 2 shots against most units before closed on you, because of the relative speed of 'mechs.

It is the more-than-doubling of the RoF of all the weapons that has created so many problems.

For some reason, the devs think that having a recycle time of about 10 seconds is too high. I cannot see how this is the case: If you're not confident in your accuracy, take more than one weapon and use chainfire. If you are, take a smaller number and/or group fire.

Right now, with a recycle time of 4 seconds, there is 0 incentive to take multiple weapons. Push-up recycle time for most weapons and you'll also be able to reduce the armour values back to their originals.

Also, I believe that convergence SHOULD be removed.
There are 3 types of convergence:

• Torso weapons
  
• Arms
  
• Arm-mounted weapons

Torso weapons should not converge. Simple. They're fixed in the torso and fire forward.

The individual weapons on the arms should not converge; if you have two lasers on one arm they should fire parallel to each other.

The only convergence should be from arms with lower-arm actuator. This means that Jagers and Jenners do not get convergence.

When firing on chainfire, each group has a reticle, which accurately reflects the trajectory of the selected weapon - this would also solve the problem of not knowing whether or not you're going to hit terrain. When firing a group, the reticle reflects the centre-point of the group, but there would still be no convergence. Arms would still converge normally, but not the individual weapons on them.

Implementing both of these with proper treatment of heat WOULD address the 3 'mechs you highlight, because they would NEVER get pin-point accuracy and even if they did, pilots would HAVE to be sure of their accuracy.

[Homeless Bill]

Lazy Eye, on 16 June 2013 - 05:38 PM, said:

Rate of fire.

Convergence only on things that make sense.

The rate of fire definitely is a big change from tabletop, but I think it's necessary for realtime. I think the game would play too slow and drive the mainstream crowd off. Aside from that, it still doesn't solve a single click putting out 60 points of damage on a single location.

Basically, that solution for convergence has all the downsides of wholesale convergence removal (irritating, hard for new players, hard to represent on the interface), half of the solvency, and it introduces additional confusion and chassis favoritism. You say it solves the three 'mechs I gave as examples, but I have an entire universe at my disposal: http://www.sarna.net/wiki/Warhawk

Your proposed change makes that 'mech an 85-ton god, while making 'mechs like the Jagermech and Stalker totally useless because their arm mounts are so far apart.

[Zaptruder]

I do like that the "Convergence only on things that make sense" idea provides a much needed natural buff to mechs with lower arm actuators over mechs with upper arm actuators (who have a large advantage in terms of the firing position of their arms).

Is there a way to split the difference? Unfortunately, not naturally as a part of the system that HB has suggested, but it could definetly be added on top to help fine tune the balance of the game.

[Noosemane]

It's not a bad idea.

[Homeless Bill]

Zaptruder, on 16 June 2013 - 06:24 PM, said:

I do like that the "Convergence only on things that make sense" idea provides a much needed natural buff to mechs with lower arm actuators over mechs with upper arm actuators (who have a large advantage in terms of the firing position of their arms).

Is there a way to split the difference? Unfortunately, not naturally as a part of the system that HB has suggested, but it could definetly be added on top to help fine tune the balance of the game.

Believe me, I'd like to have perfect balance and have everything make perfect sense. But if I have to pick one, I'll pick perfect balance every time. I played with the idea for a while, but it ultimately just needlessly plays favorites with certain chassis. Typically, 'mechs with lower arm actuators have lower mounts, and that makes them quite a bit better in a brawl than Stalker/Jager-like high mounts.

I'm definitely not against fun stuff like this being tried after the core system is in and balanced. I think it's flexible and allows for any number of interesting additions.

[cyberFluke]

As a simplistic stopgap, how harmful do you think just removing convergence unless chain firing would be?

[DeadlyNerd]

After browsing through a front page worth of pointless threads this one's a jewel. Bill hit the nail on the head with this one.
A coffin nail to PGI's balancing solutions too.

[MustrumRidcully]

Homeless Bill, on 16 June 2013 - 05:57 PM, said:

The rate of fire definitely is a big change from tabletop, but I think it's necessary for realtime. I think the game would play too slow and drive the mainstream crowd off. Aside from that, it still doesn't solve a single click putting out 60 points of damage on a single location.

Would it still be to low if you couldn't group-fire?

If you're equipped with just 4 weapons, you'd fire an average of every 2.5 seconds.

AC/40 Jagers or Quad PPC Stalkers fire only every 4 seconds.
An AC/20 + 4 ML + 2 SRM Atlas would need to fire 3 weapons seperately inside a 4 second window.

[Ronin Starwalker]

I support Bill's proposed solution.

The proposed heat fix from PGI is clunky and will not solve the ballistics issue.

We will just see AC and Gauss boats.

A new mechanic needs to be introduced that sits under both direct fire systems, energy and laser

[BR0WN_H0RN3T]

PGI have never ever redone anything from scratch because their egos write cheques their bodies can't cash. No seriously, it's a well thought out post but the solution needs to be expressed in fewer words. Firstly, armlock needs to go. Weapon reticle needs to jiggle when moving and heat penalties need to be meaningful. That will be enough for me.