348 replies to this topic

[KraftySOT]

Alek Ituin, on 06 October 2014 - 11:25 AM, said:

Because BattleTech was balanced with the inaccuracy of the enemy in mind.

In TT, Mechs couldn't hit the broadside of a barn without a good pilot. In an FPS environment, everybody can hit their enemy with accuracy, and yet we still use a system balanced around inaccuracy!

So obviously, something has to change. As it happens, the easiest thing to do is make Mechs less accurate. Not only would that hold with lore, but be orders of magnitude easier than rebalancing an entire game because you don't want to lose PPFLD alphas.


Just saying.

Even with a good pilot...a light mech moving 15 hexes is still not a surefire hit if you moved...at all...even not moving youre looking at a 9+ for an aimed shot, 5+ (I think) for a standing still unaimed shot.

To be fair its abstracting these weapons firing faster...but then theres also no accounting for that with PGIs damage. You get the full 15 damage from a clan PPC even though it can fire 3 times every 10 seconds. Really it should be doing 5 if thats the recycle time.

Which of course would also fix the problem. 80 point alphas would be ~26 point alphas. If you can hit the same location with 3 alphas in 10 seconds...shame on the target.

And as said, I dont care how its fixed, just fix it. Change the damage to account for the recycle times, or eliminate convergence, or have convergence take more time, or have a fixed 1000m convergence, or a cone of fire....

Dont care. Just fix it.

Edited by KraftySOT, 06 October 2014 - 11:33 AM.

[Roadbeer]

KraftySOT, on 06 October 2014 - 11:33 AM, said:

Even with a good pilot...a light mech moving 15 hexes is still not a surefire hit if you moved...at all...even not moving youre looking at a 9+ for an aimed shot, 5+ (I think) for a standing still unaimed shot.

To be fair its abstracting these weapons firing faster...but then theres also no accounting for that with PGIs damage. You get the full 15 damage from a clan PPC even though it can fire 3 times every 10 seconds. Really it should be doing 5 if thats the recycle time.

Which of course would also fix the problem. 80 point alphas would be ~26 point alphas. If you can hit the same location with 3 alphas in 10 seconds...shame on the target.

And as said, I dont care how its fixed, just fix it. Change the damage to account for the recycle times, or eliminate convergence, or have convergence take more time, or have a fixed 1000m convergence, or a cone of fire....

Dont care. Just fix it.

Let me see if I can sum up what you said for the average layman.

Dump the damage of the weapons so, with their recycle time, they're doing the TT value damage over 10 seconds?

So, an ML would do 1.67 damage per shot over 1 second and an AC/20 would do 8 damage?

I'm a fan, it's elegant, simple, wonder why it hasn't been done?

Edited by Roadbeer, 06 October 2014 - 12:04 PM.

[stjobe]

Roadbeer, on 06 October 2014 - 12:02 PM, said:

Let me see if I can sum up what you said for the average layman.

Dump the damage of the weapons so, with their recycle time, they're doing the TT value damage over 10 seconds?

So, an ML would do 1.67 damage per shot over 1 second and an AC/20 would do 8 damage?

I'm a fan, it's elegant, simple, wonder why it hasn't been done?

I've been advocating this for the last two and a half years, claiming it to be one of PGI's biggest mistakes when converting TT to real time: They (roughly) tripled rate of fire, but let damage and heat per shot stay the same.

Which, of course, means that weapons do three times too much damage (which necessitated the doubling of armour) and three times too much heat (which... well, broke the already broken heat system even more).

What they should have done was (of course) to adjust damage and heat per shot by the exact same amount they increased rate of fire; then they could have kept us at TT armour levels and would have had at least a shot of having a sane heat system.

It's one of the more inexplicable gaffes they've ever made.

Edit: Also, for the TT purist, it's actually closer to TT than having an AC/20 do 20 damage per shot 2.5 times per 10 seconds; that's 50 damage per turn, not 20. If you want it to fire 2.5 times in 10 seconds, you have to have it only do 8 damage per shot, which adds up to 20 damage in 10 seconds.

Edited by stjobe, 06 October 2014 - 12:34 PM.

[Trip Hammer]

I may be wrong here but I seem to recall that Russ had said that de-syncing the convergence of weapons was a problem due to the load that it put on the server. I think it had to do with tracking the number of weapons all firing at the same time from potentially 24 players in matches.

Like I said, I could be wrong here but I believe that that was the reason that they have not done something about instant convergence, and likely wont.

[stjobe]

The Faceless, on 06 October 2014 - 12:33 PM, said:

I may be wrong here but I seem to recall that Russ had said that de-syncing the convergence of weapons was a problem due to the load that it put on the server. I think it had to do with tracking the number of weapons all firing at the same time from potentially 24 players in matches.

Like I said, I could be wrong here but I believe that that was the reason that they have not done something about instant convergence, and likely wont.

Yep. It breaks HSR.

Back before HSR, we had non-instant convergence, but it put too much load on the servers with HSR, so they removed it.

And it won't be back, so everyone thinking non-instant convergence is a solution to our little problem needs to think again.

[Roadbeer]

stjobe, on 06 October 2014 - 12:29 PM, said:

I've been advocating this for the last two and a half years, claiming it to be one of PGI's biggest mistakes when converting TT to real time: They (roughly) tripled rate of fire, but let damage and heat per shot stay the same.

Which, of course, means that weapons do three times too much damage (which necessitated the doubling of armour) and three times too much heat (which... well, broke the already broken heat system even more).

What they should have done was (of course) to adjust damage and heat per shot by the exact same amount they increased rate of fire; then they could have kept us at TT armour levels and would have had at least a shot of having a sane heat system.

It's one of the more inexplicable gaffes they've ever made.

Edit: Also, for the TT purist, it's actually closer to TT than having an AC/20 do 20 damage per shot 2.5 times per 10 seconds; that's 50 damage per turn, not 20. If you want it to fire 2.5 times in 10 seconds, you have to have it only do 8 damage per shot, which adds up to 20 damage in 10 seconds.

Were we ever given a reason for this? Or did it fall into "The Forumites be crazy, yo!" period of development?

[stjobe]

Roadbeer, on 06 October 2014 - 12:36 PM, said:

Were we ever given a reason for this? Or did it fall into "The Forumites be crazy, yo!" period of development?

Dunno, it was already like this when I joined in August 2012. I guess they built the game that way, because although I've heard they had TT armour values in Friends & Family, I've never heard they had any other system of firing rates/damage/heat than the one we have now.

[Roadbeer]

stjobe, on 06 October 2014 - 12:39 PM, said:

Dunno, it was already like this when I joined in August 2012.

Same, was hoping there might have been some known history about it.

Add that to the Big List-o-Questions for Russ.

[jarien13]

Roadbeer, on 06 October 2014 - 07:57 AM, said:

So, while enjoying my coffee during my morning constitutional, I whipped up a simple convergence system, that I don't believe would require a lot of work to code.

...[edited for space]...

3. Add in speed modifier, say .02 seconds/100 meters over 30 KPH

Just putting it out there.

Oh, and look, you can add in a new Mech Adjustment like ARTEMIS called "Improved Gimbals" that add a modifier to convergence speed and increase the cost/weight/slot of "Improved Gimbals" weapons.

Roadbeer, on 06 October 2014 - 09:54 AM, said:

% of max speed?
% of ?

I see where you're going, just curious about what would trigger the targeting computers to start to loose effectiveness.

Indirectly make it tied to speed by having it directly tied to throttle percentage. No worry about scaling for speeds based on currently installed engine or speed tweak, only to the throttle scale. It would be as if the power draw from the throttle level for movement is impeding the power requirement for converging the weapon systems. Think about the lights in your house momentarily dimming when your air conditioner or refrigerator compressor kicks in.


(Ahem, everyone take your time reading this, it's a lot to soak in):

CoF = CEP =/= "wild spray 360 everywhere RNG"
http://en.wikipedia...._error_probable
To whom it may concern:
Trained snipers are more than capable of making up for the inherent ballistic imprecision that still exists in their high spec, tight tolerance, high accuracy, high precision firearms. Even "non-snipers" are capable of making accurate shots on target with practice. If you are really in possession of "elite aiming mad-skillz" then you too can overcome the known and quantifiable precision deviation of any weapons system. Yes, all weapons have imprecision, some just less so than others. It is dependent upon too many factors to eliminate completely, but with training and experience it may be calculated and accounted for and overcome to make incredible shots. Just so long as you understand the difference between qualitative error and quantitative error. One you "feel" and "think it's not right", the other one you can measure and statistically prove and thus mathematically account for. In any case, not everyone, sniper or otherwise, is Carlos Hathcock.

Also, I believe someone made a silly remark about lack of guidance of missiles is what makes that weapon spread. Last I checked, AC rounds don't have guidance systems on them and are ballistic once they exit the muzzle, and therefore unable to adjust their flight trajectory. What phenomenon do exist then to reduce the spread deviations of an AC shell?
(Don't reply to that right away, think on it for a bit and maybe read a physics book.)


(Don't worry, I have more)
As far as the "pinpoint damage is only a phantom problem of bads and noobs that can't torso twist", the ONLY way to soften a 100 point alpha is by NOT BEING HIT BY ANY OF IT AT ALL. This leads to the other prevalent issue in the game of no one wanting to engage because they're afraid to scuff their paint. (we've all seen the lone assault mech still alive at the end of a match with pristine armor...) Whether you twist in time to not get cored is moot if it takes off your full armored, full health arm in one blast, like some of you say arms are there for (guess you don't use arm weapons, ever?). Guess next shot you'll twist and lose your other arm, then what, somehow use your leg to block the next 100 point alpha? How about saving us all the trouble and use your face to block it instead.

Sure, torso twisting and arm shielding is great and is a trick many long time players of this game agree is what you should be doing. You should also realize the meta has changed since that tactic was first employed. "Proper" twisting in my Centurion with max armored left arm these days just lets the enemy pad their damage for a few more seconds before I die from massive enemy focused alpha damage that usually results in my left leg and left torso blowing off the same time as my CT melts as I'm frantically trying to twist my right side into the instant shot to (lol) spread the damage. I'm just as survivable running an XL as a STD due to how fast all the damage transfers through my instantly vaporized arm into my side torso (usually destroying my weapons in that process) and overflowing into my legs and CT when my LT gives way on the next salvo. Most of the time though, I'm just instantly cored out, front or rear, with my armor all pretty on all the rest of my mech (if I could only move unused armor points from the arms to my CT, or if FF actually boosted your max armor, instead of its pitiful tonnage "bonus")

If you are in anything less than a full armored assault mech, the damage transfer into the next component will be more than an irritating itch. In a medium or light, or anything with an XL it will be just as fatal to twist and take a 100 point only to the arm as it will be to take it to the center or either side torso while running away trying to dodge. Let's discard the hyperbole of FLPPA all hitting the same pixel, cause the only thing that matters is 50, 75, 100 points of near instant damage hitting the same component. At the end of the day, the only way you will ensure surviving massive front-loaded pin-point alphas is by completely dodging them. It's not even a matter of being "too scared to scratch your mech" it's simply a matter of people want to play longer than the first time they get shot.


(Watch out, here's where I get rolling)
This means, from the high-damage alpha firing mech perspective, appropriately balanced TTK pretty much has to result in more than 50% of their shots missing. And I don't mean your individual experience, which could be a statistical outlier just as easily as the mean or median, I mean the statistical average across the WHOLE PLAYER BASE OF USERS. How is that better than alternative ideas that have been presented? You have already stated you don't want your "aiming skillz" to be negated by a "hit or miss" mechanic, yet the only way under the current system the game balances itself is when people run lights with their messed up hit-boxes everyone complains about so they can spoof the game into not being hit ever. The reasons the light queue doesn't run 50% to take advantage of that is because everyone wants to have enough weapons to alpha with to remove components with every shot, and because lights especially have to play the dodge or dead game to a lower threshold of alpha strike. It's like flipping a coin to decide if you survive a match or not.

Let's not forget, there are more people out there playing than just you and your "mad skillz". One mans "elite" gaming experience tends to be balanced by at least one man's terrible experience that does in fact lead to them uninstalling. GGCLOSE. Who you gonna kill to pad your stats if all us bads and noobs leave? Let's try to drop the "holier than everyone not myself" act and "L2P" bullcrap that ignores the learning curve of this game and remember this game needs to be balanced across all levels of skill, not just the elitist snobs, tryhards, and competitives.

I know, I know, it's not all as bad as I make it out to be. It's really not as bad as either side makes it out to be, or there would be no one still playing. And, no matter what anyone thinks, at the end of the day, this game will live or die based on it's communities desired involvement and ideas for improvement as well as its devs willingness to listen to and respect that community while being able to filter and not cater to the outright "nerf everything cause it just killed me" QQ whiners. If you don't believe me, just go ask someone about the difference between the FF11 and FF14 development teams and WoW.

(I'm spent, hope you all actually read and enjoyed this and didn't just skip down here to the end)
/salute
~end transmission~

Edited by jarien13, 07 October 2014 - 01:30 AM.

[Mcgral18]

stjobe, on 06 October 2014 - 12:36 PM, said:

Yep. It breaks HSR.

Back before HSR, we had non-instant convergence, but it put too much load on the servers with HSR, so they removed it.

And it won't be back, so everyone thinking non-instant convergence is a solution to our little problem needs to think again.

Would going from instantly pinpoint to instantly parallel put excess strain on the server?

[Joseph Mallan]

stjobe, on 06 October 2014 - 12:29 PM, said:

I've been advocating this for the last two and a half years, claiming it to be one of PGI's biggest mistakes when converting TT to real time: They (roughly) tripled rate of fire, but let damage and heat per shot stay the same.

Which, of course, means that weapons do three times too much damage (which necessitated the doubling of armour) and three times too much heat (which... well, broke the already broken heat system even more).

What they should have done was (of course) to adjust damage and heat per shot by the exact same amount they increased rate of fire; then they could have kept us at TT armour levels and would have had at least a shot of having a sane heat system.

It's one of the more inexplicable gaffes they've ever made.

Edit: Also, for the TT purist, it's actually closer to TT than having an AC/20 do 20 damage per shot 2.5 times per 10 seconds; that's 50 damage per turn, not 20. If you want it to fire 2.5 times in 10 seconds, you have to have it only do 8 damage per shot, which adds up to 20 damage in 10 seconds.

Once again I dislike this LESS than what PGI has done. Solaris 7 boxset style rules died a silent death for a reason. A majority of players hated it!

With standard armor lvls... I could accept this I think, But would like to test it before I commit one way or the other.

[Karl Marlow]

Roadbeer, on 06 October 2014 - 12:36 PM, said:

Were we ever given a reason for this? Or did it fall into "The Forumites be crazy, yo!" period of development?

Are you implying that we are no longer crazy?

[Lily from animove]

Roadbeer, on 06 October 2014 - 07:57 AM, said:

So, while enjoying my coffee during my morning constitutional, I whipped up a simple convergence system, that I don't believe would require a lot of work to code.

1. Remove the convergence bonus from the skill tree. This makes no sense, as convergence is mechanical and no amount of skill can make a computer adjusting gimbals move faster.

2. These are example maths so you can see where I'm going with it, obviously these numbers are open to debate (and more than likely will be debated rather than the premise of the post /lesigh)
CT/Head weapons: 0 (It's where you mech is pointing, no convergence modifier)
RT/LT .01 seconds * .01 seconds/100 meters (time it takes adjustments to align with CT weapons)
RA/LA .02 seconds * .01 seconds/100 meters (Fixed arm or "locked arms" adjustments to align with CT weapons)
RA/LA "Unlocked" (Arm movement speed) * .01 seconds/100 meters.

3. Add in speed modifier, say .02 seconds/100 meters over 30 KPH

Just putting it out there.

Oh, and look, you can add in a new Mech Adjustment like ARTEMIS called "Improved Gimbals" that add a modifier to convergence speed and increase the cost/weight/slot of "Improved Gimbals" weapons.

now think further and tell me how a US spider should ever be able to be hit by any european wielding lasers, or even further any uassie or russian. Those gamers convergence would be a 100% luck based aim on how the server situation is compared to what they see. I cna yet already keep visually my lasers on a light and yet only a fraction of damage is dealt to the HSR. Add convergence and the game is unplayable with soem builds. This is in the todays time of netcode latency and stuff not a tehcnically practical solution. It would work in a singleplayer game where no synchronisation between 24 individuals is needed. But not in any Online game. Probably also on a lan with an extremely low latency for all participants.

CoF is an easy way to simulate inaccuracy influenced by only your acts. way more easy way better to design technically reliable.

[LordKnightFandragon]

stjobe, on 06 October 2014 - 12:29 PM, said:

I've been advocating this for the last two and a half years, claiming it to be one of PGI's biggest mistakes when converting TT to real time: They (roughly) tripled rate of fire, but let damage and heat per shot stay the same.

Which, of course, means that weapons do three times too much damage (which necessitated the doubling of armour) and three times too much heat (which... well, broke the already broken heat system even more).

What they should have done was (of course) to adjust damage and heat per shot by the exact same amount they increased rate of fire; then they could have kept us at TT armour levels and would have had at least a shot of having a sane heat system.

It's one of the more inexplicable gaffes they've ever made.

Edit: Also, for the TT purist, it's actually closer to TT than having an AC/20 do 20 damage per shot 2.5 times per 10 seconds; that's 50 damage per turn, not 20. If you want it to fire 2.5 times in 10 seconds, you have to have it only do 8 damage per shot, which adds up to 20 damage in 10 seconds.

3x to much heat? Our heat scale is 3x higher...its what? 80? with perks, over 100?

[Mcgral18]

LordKnightFandragon, on 07 October 2014 - 05:44 AM, said:

3x to much heat? Our heat scale is 3x higher...its what? 80? with perks, over 100?

3x heat, 1x dissipation, rising heat cap.

Dissipation is still TT, for the most part. Buffed if you have 17 or fewer DHS, nerfed if you have 18+ DHS due to the 15% elite efficiency, and the buffed 2.0s making up for the gimped 1.4s.

[Almond Brown]

Alek Ituin, on 06 October 2014 - 10:50 AM, said:

Quirks.

Give Lights a Quirk that drastically boosts convergence time. Plus, they have tiny arms with tiny weapons, they could naturally have a faster convergence time.

Quirks really could solve at least 75% of the issues this game has, it's a fantastic concept.

Agreed, Quirks are/will be great BUT giving the Lights the ability to Converge quickly would make them even more deadly versus the larger, slower Mechs. It is bad now, imagine if when in an Atlas, your weapons could never converge fast enough to actually get a (decently potent) shot on a FAST Light Mech.

Look what happened when they Quirked the Victor (so to speak). It went from really good to a Garage Mech, almost over night. Quirks will have to be done gently, or many units will get "garaged".

Edited by Almond Brown, 07 October 2014 - 06:48 AM.

[Almond Brown]

Roadbeer, on 06 October 2014 - 12:36 PM, said:

Were we ever given a reason for this? Or did it fall into "The Forumites be crazy, yo!" period of development?

Pace of play. The first likely decision made pre-code building. How fast do we want a Match to last? The 15 minute time selection, at the time, was OK, 30-45 minute Matches would not float with the Xbox generation.

One can assume that everything after that was done as a result. The one constant in MWO and has never changed, not even tried or tested, has been the 15 minute Match. Almost "everything" else has changed.

It could be catastrophic to the Game to change a fundamental component, like Match time, at this point in the games life. The Dev's know this as it has never been given any consideration. And I do not blame them.

To slow MWO's game play down now, to try and satisfy...some. Too late.

Edited by Almond Brown, 07 October 2014 - 07:05 AM.

[Roadbeer]

Lily from animove, on 07 October 2014 - 05:41 AM, said:

CoF is an easy way to simulate inaccuracy influenced by only your acts. way more easy way better to design technically reliable.

Russ and Forumites have a 'distaste' for CoF, Forumites have a 'distaste' for PPA. Servers have a 'distaste' for Convergence.

Might as well close up shop on the topic then.

[stjobe]

Mcgral18, on 07 October 2014 - 04:34 AM, said:

Would going from instantly pinpoint to instantly parallel put excess strain on the server?

Probably. Instant pinpoint is just one point to track, parallel is multiple points to track - which, if I understand it correctly, is what mucks up HSR; it takes to much time to track (and back-track) all the aiming points for 24 'mechs with 5-10 weapons each.

LordKnightFandragon, on 07 October 2014 - 05:44 AM, said:

3x to much heat? Our heat scale is 3x higher...its what? 80? with perks, over 100?

Yeah. As McGral18 said, capacity is way higher. Dissipation is still at TT values. Which means 'mechs heat up too fast and cool down too slow - which, not incidentally, is a large contributor to the "Alpha, alpha, all the time" game play we suffer under. You have to make your limited heat count, because it takes forever and a day to cool down, and while cooling down your DPS is zero.

The game would be a lot better with a lower capacity and higher dissipation. Less alpha-striking, more chain-firing - and lasers would get a boost, perhaps even to the degree that ballistics weren't the no-brainer choice it is today.

Edited by stjobe, 07 October 2014 - 08:08 AM.

[CompproB237]

stjobe, on 07 October 2014 - 08:07 AM, said:

Probably. Instant pinpoint is just one point to track, parallel is multiple points to track - which, if I understand it correctly, is what mucks up HSR; it takes to much time to track (and back-track) all the aiming points for 24 'mechs with 5-10 weapons each.


Yeah. As McGral18 said, capacity is way higher. Dissipation is still at TT values. Which means 'mechs heat up too fast and cool down too slow - which, not incidentally, is a large contributor to the "Alpha, alpha, all the time" game play we suffer under. You have to make your limited heat count, because it takes forever and a day to cool down, and while cooling down your DPS is zero.

The game would be a lot better with a lower capacity and higher dissipation. Less alpha-striking, more chain-firing - and lasers would get a boost, perhaps even to the degree that ballistics weren't the no-brainer choice it is today.

We did some basic calculations on a 30 hard cap for heat. We determined that the 'Dragon Slayer meta-build' from a while back (2x PPC, 2x AC/5 3t ammo, max Jump Jets, Standard 275, 14 'total' DHS). Lets assume DHS does 0.2 HPS giving 2.8 HPS sinking.
First Alpha: 22 Heat (73 1/3%)
Cooldown on AC/5s of 1.66s
Next 2xAC/5 shot you're at 17.352 heat, adds 2 heat, 19.352 heat (64.51%)
The pilot could fire one PPC but probably want to wait especially if any other factors in this (dissipation rate modifiers like location temperature)
At this point the Dragon Slayer pilot could fire the 2xAC/5s indefinitely and slowly cool off.
It would take roughly 5 shots of both AC/5s from the initial alpha before the heat has been reduced enough to where the 2x PPC could be fired again.
That's roughly 8.3 seconds using current values.

Lets say this Dragon Slayer pilot was jump jet sniping instead of standing still.
Jump jetting: 3.41 HPS * 3.5 seconds = 11.955 Heat - 2.8 HPS sinking = 0.616 HPS generated by JJ = 2.155 heat at peak of jump
Initial Alpha: 10x2 + 1x2 = 22 heat + 2.155 from JJ = 24.155 heat (80.52%)
They fall to the ground.
It would take them 8.63 seconds after firing before they've sinked off the heat they generated and be able to fire again.

That's a hell of a lot better than it is now:
66.72 Threshold on temperate map on same 'Mech
This allows the 'poptart' to jump and fall, rinse and repeat 5 times before overheating on the 5th time. This is over the span of 35 seconds. Obviously I'm kind of guessing on JJ heat generation and burn time/cooldown time numbers (I just assumed 3.5 seconds). I've watched a video of this build and it's pretty close. During that 35 seconds they've dished out 150 Damage or 4.286 DPS.

Compare that to the 30 Hard Cap same build. It is only able to do this just under 3 times in that same duration. 36.4 seconds is 3 times it can Alpha. We'll use this for our 'DPS' calculation and we get: 90 Damage in 36.4 seconds or 2.473 DPS. This is roughly a 42.3% reduction in DPS. Without changing any damage values on weapons.

Edited by CompproB237, 07 October 2014 - 12:03 PM.