230 replies to this topic

[The6thMessenger]

Der Geisterbaer, on 03 May 2021 - 10:06 PM, said:

Let's just say that 3^3 = 27 and seems to assume that something has grown 3 times in all three dimensions in relation to something else. I won't comment on why that thought is flawed or why your cube comparision graphic ultimately makes a very similar mistake.

If you won't justify why it's wrong, I don't really see the point then.

[TheAbsoluteMadman]

The6thMessenger, on 03 May 2021 - 08:45 PM, said:

About that 27 times, I don't understand where you got that number from.

Square Cubed. Granted, it's just a really general example that would assume the atlas is proportionally bigger in every way, which it's not. It was an example that even if the atlas was 3 times bigger in every dimension than the jenner, It would not be "3 times the volume". The volume would be the multiplier cubed (3^3). Just showing that It would be awkward to balance and scale mechs accurately because it would be ridiculous.

The6thMessenger, on 03 May 2021 - 08:45 PM, said:

The components are already interchangeable and it's just a matter of slots and tonnage, that means it's already accounted for.

Yeah I know that, I was just saying that it is a weird thing to relate weight or mass directly to volume unless density was involved. In this case, we'd only know the volume based on in game measurements. Either way, realism and scaling aside, I definitely have fun with the game and I'd like to see how these changes mix things up.

[TheAbsoluteMadman]

Der Geisterbaer, on 03 May 2021 - 10:06 PM, said:

Let's just say that 3^3 = 27 and seems to assume that something has grown 3 times in all three dimensions in relation to something else. I won't comment on why that thought is flawed or why your cube comparision graphic ultimately makes a very similar mistake.

The thoughts are in the right place and promote discussion. The flaw is relating real world physics to a video game in the future. The principle is constant, the graphic is only flawed in the sense that realistically the atlas would look really disproportional compared to the commando since the cross sectional area can't really keep up with the tonnage increase and the internal structure would have a hard time keeping shape. The design of the atlas would need to change or the materials utilized in construction would have to have really good properties. But uh you know.. BATTLETECH.

[Bowelhacker]

D A T A, on 03 May 2021 - 03:56 PM, said:

pitch has been put to 25, further increases would delete vertical play.
that result can be achieved via rescale only, as right now you would need 35 pitch to get there (current is 15-20)

Bring on the rescale!

PS: thanks for actually answering.

[Der Geisterbaer]

TheAbsoluteMadman said:

The thoughts are in the right place and promote discussion.

It's more of a personal matter as to why I was and still am reluctant about going into details why both the 3^3 calculation and that cube comparision are flawed.

TheAbsoluteMadman said:

The flaw is relating real world physics to a video game in the future.

I can't quite agree there. Trying to relate certain real physics and math (basic geometry) to a video game (the "future" qualifier being mostly irrelevant) is not by default flawed ... particularly not once the idea is to "simulate" certain things to with reasonable accuracy. The seriously flawed parts here are on the one hand the over-simplifications that are being made and on the other hand the game's source material never having bothered with such details in the first place.

TheAbsoluteMadman said:

The principle is constant, the graphic is only flawed in the sense that realistically the atlas would look really disproportional compared to the commando since the cross sectional area can't really keep up with the tonnage increase and the internal structure would have a hard time keeping shape.

That cube graphic incorrectly assumes both an identical density of both objects as well as an equal amount of length increases in all dimensions (just like the 3^3 comparison did). While simplifications certainly are necessary to a degree, the visual comparison of two cubes instead of e.g. a cuboid that is more reasonably well suited to outline the supposedly humanoid contours of both mechs involved in this comparison is just as deceptive as using those cubes to visualize a three times larger volume on a 2d plane in general. An honest statistician might tell you how and why the choice of a visualization is so important in terms of spinning the human perception towards a desired conclusion.

TheAbsoluteMadman said:

But uh you know.. BATTLETECH.

So ultimately you agree that both the 3^3 calculation as well as that cube comparision are ill-suited to properly explain the difference between a 30t and a 100t BattleMech and subsequently are also bound to fail when it comes to properly portraying them in the game? Not to mention what the ramifications for "balance" are ...

[Dogstar]

The only oddity I noticed is that the Night Gyr is going to be the least agile heavy mech of all with a turn rate of only 40, making it substantially slower than all the other heavies. It also has the slowest acceleration and torso turn rate, although the difference is less noticeable there.

Is there a reason for not buffing it to the same levels as most heavies? It was never a particularly agile mech.

p.s. the reason why the cube analysis doesn't work is because mechs aren't cubes (well except for the dire wolf), mechs are mostly humanoid shaped meaning they are roughly twice as tall as they are wide or deep. Which manes that .63 height multiplier is both completely wrong and highly deceptive. I don't know what the precise number is but it's probably more like .8ish and would be a different number for each and every mech based on their actual shape rather than generalising it.

Edited by Dogstar, 04 May 2021 - 12:39 AM.

[Storming Angel]

Im interested too see, what kind of buffs the gargoyle will get in terms of quirks. If they won't buff its mobility any further (Max speed, torso yaw etc). They will need some good quirks to bring them up to speed, and/or give them more options.

Otherwise, looks very promising so far.

Edited by Storming Angel, 04 May 2021 - 12:46 AM.

[TheAbsoluteMadman]

Lego

Der Geisterbaer, on 03 May 2021 - 11:46 PM, said:

It's more of a personal matter as to why I was and still am reluctant about going into details why both the 3^3 calculation and that cube comparision are flawed.

Still haven't given me anything good so I'll ignore this

Der Geisterbaer, on 03 May 2021 - 11:46 PM, said:

I can't quite agree there. Trying to relate certain real physics and math (basic geometry) to a video game (the "future" qualifier being mostly irrelevant) is not by default flawed ... particularly not once the idea is to "simulate" certain things to with reasonable accuracy. The seriously flawed parts here are on the one hand the over-simplifications that are being made and on the other hand the game's source material never having bothered with such details in the first place.

The future qualifier is not irrelevant at all, bring me detailed schematics of these mechs and material properties of components and alloys (endo-steel, ferro fibrous armor, etc) Show me sims and stress analysis that prove a functioning mech of that size can handle repeated impacts and load forces without collapsing (let alone stand or walk) Better yet, I'll do it myself if you bring those specifications.

Der Geisterbaer, on 03 May 2021 - 11:46 PM, said:

That cube graphic incorrectly assumes both an identical density of both objects as well as an equal amount of length increases in all dimensions (just like the 3^3 comparison did). While simplifications certainly are necessary to a degree, the visual comparison of two cubes instead of e.g. a cuboid that is more reasonably well suited to outline the supposedly humanoid contours of both mechs involved in this comparison is just as deceptive as using those cubes to visualize a three times larger volume on a 2d plane in general. An honest statistician might tell you how and why the choice of a visualization is so important in terms of spinning the human perception towards a desired conclusion.

The cube graphic is absolutely incorrect, but the principle remains the same. Bring me a 20 ton vehicle the size of a honda civic that functions. Like I said, it's fun for discussion, but there's absolutely no way to accurately assume density or material properties. The statistician can tell me whatever he wants, bring me the engineer that designed the machine and the discussion is over.

Der Geisterbaer, on 03 May 2021 - 11:46 PM, said:

So ultimately you agree that both the 3^3 calculation as well as that cube comparision are ill-suited to properly explain the difference between a 30t and a 100t BattleMech and subsequently are also bound to fail when it comes to properly portraying them in the game? Not to mention what the ramifications for "balance" are ...

I think I've explained why both are ill suited to explain the difference, in fact I believe i said "BATTLETECH". If you understood the meaning of the examples, you would not need ask me about the ramifications or the difficulty to portray them in game.

Der Geisterbaer, on 03 May 2021 - 11:46 PM, said:

While simplifications certainly are necessary to a degree,

So ultimately you agree that given the lack of detailed information, specifications, material properties, or working examples, simplifications must be made to explain why a product may fail and why a different approach may be necessary. Keep in mind, I was just pointing out the issue with the other guy's initial statement, there's no way to have a playable game with realism in this case.

[TheAbsoluteMadman]

Dogstar, on 04 May 2021 - 12:26 AM, said:

p.s. the reason why the cube analysis doesn't work is because mechs aren't cubes (well except for the dire wolf), mechs are mostly humanoid shaped meaning they are roughly twice as tall as they are wide or deep. Which manes that .63 height multiplier is both completely wrong and highly deceptive. I don't know what the precise number is but it's probably more like .8ish and would be a different number for each and every mech based on their actual shape rather than generalising it.

The initial simplification was crazy, but just to say it was not a flat multiplier. I agree with this, thanks.

[Der Geisterbaer]

TheAbsoluteMadman said:

Still haven't given me anything good so I'll ignore this

But you have given me more than enough confirmation with regards to my personal reasons for not delving too deep into the specifics. So this will end here with this final note: You very literally confirmed twice now that my original comment on the cube picture and your math being "wrong" yet you feel the need to defend both against me and then move goal posts and commit other fallacies ~shrug~
Note - among other things - that I already made the argument concerning incorret stipulations concerning density and then ask yourself how relevant "future" tech truly is or how misguided an argument against me about unknown density becomes.
Take a step back and note: you're barking up the wrong tree for the wrong reasons.

TheAbsoluteMadman said:

Keep in mind, I was just pointing out the issue with the other guy's initial statement, there's no way to have a playable game with realism in this case.

I would say you didn't keep that in mind youself when reacting to my original comment

Can we agree to not actually disagree and be done here?

[TheAbsoluteMadman]

Der Geisterbaer, on 04 May 2021 - 01:13 AM, said:

Can we agree to not actually disagree and be done here?

Yeah I think we're done here. Clearly we're not communicating effectively if you feel the need to assert yourself and please understand I don't care who you are or what you do. I was satisfied with the conversation to a point though, I think there's always something to learn even in a pointless exchange.

On that note, I'm still looking forward to these changes. They look interesting.

[D A T A]

Darian DelFord, on 03 May 2021 - 06:09 PM, said:

What about the legs man.... what about the legs

If your gonna make that argument ya better include more facts :>

You can spec survival tree as you stated if you do not wish Back armor, on at all, you would have 5 points left. To my knowledge the only chassis that reliably run no back armor are assault and then come here and whine when they get cored out from the back. I dabbled with this when the re-scale initially happened, so did most light pilots. Only a few mechs really benefit from it and you really have to be quirked for it, most lights are not.

By fully speccing in the survival tree, you severely limit your firepower / sensor / mobility tree.... and I mean severely. For what an extra 6 armor and 15 or 16 structure. Which means your less mobile, have to be even closer to your target and your DPS drops off. Besides, most lights die more from a legging than they do a torso shot. The torso shot "usually" only happens after they are legged

If what you say is true, then every light would be doing it....... hmmm they are not.... it makes you wonder why?

But you also have to look at different aspects.... and this I think is where PGI went severely wrong. FP does not equal QP.

Every comp pilot is doing it already.
The only ones that do not use survival tree on the lights are the noobs, easy.
There is no competitive pilot in this world that uses a vulcan/wolfhound/commando and so on without survival tree.
And the funny thing is that people who complain getting one shot are usually the noobs that are not doing that.

Edited by D A T A, 04 May 2021 - 02:21 AM.

[The6thMessenger]

Der Geisterbaer, on 03 May 2021 - 11:46 PM, said:

That cube graphic incorrectly assumes both an identical density of both objects as well as an equal amount of length increases in all dimensions (just like the 3^3 comparison did). While simplifications certainly are necessary to a degree, the visual comparison of two cubes instead of e.g. a cuboid that is more reasonably well suited to outline the supposedly humanoid contours of both mechs involved in this comparison is just as deceptive as using those cubes to visualize a three times larger volume on a 2d plane in general.

So basically it's because atlas and commando is of different shape, and even more so the contours -- and that is also the difference of mech shapes like a bushwacker ain't humanoid. And in addition are different materials that may have different density.

Of course, there's interchangeability between tech. Why interchangeability and standardization matter? Because logistics, we do that right now. Rifles have to be made the same weight, tanks has to be made the same way, built with the same parts. There are specific techs made of specific stuffs that are standard like Ferro-Fiber, Endo-Steel, Myomers. So there's good reason to assume that there's a degree of standardization of materials that would mean simmilar materials, simmilar densities.

But at the meat of the issue, so what though? Why can't we just assume simmilar densities in the first place? Wouldn't it be easier to start from similar density? After all, it is "volumetric rescale", it's just about volume.

I think the both of you are overthinking it. If anything it's not "wrong", it's just incomplete. We can only derive based on the information we have. It's not going to be exact, but we have some idea of how it could be, just ballpark is enough.

The Atlas and the Commando seems to me the closest we could get for an analog due to the humanoid shape, so we can only make use of those to gauge information. And the takeaway of that is that the current mechs like the Atlas are nowhere near to their simmilar-density-volumetric-scale.

Der Geisterbaer, on 03 May 2021 - 11:46 PM, said:

An honest statistician might tell you how and why the choice of a visualization is so important in terms of spinning the human perception towards a desired conclusion.

Go ask Nightbird. He's the one that actually came up with that.

https://mwomercs.com...metric-scaling/

Edited by The6thMessenger, 04 May 2021 - 03:04 AM.

[The6thMessenger]

D A T A, on 04 May 2021 - 02:15 AM, said:

Every comp pilot is doing it already.
The only ones that do not use survival tree on the lights are the noobs, easy.
There is no competitive pilot in this world that uses a vulcan/wolfhound/commando and so on without survival tree.
And the funny thing is that people who complain getting one shot are usually the noobs that are not doing that.

Yep. Maxed it out on the Urbie.

It's the little mech that could.

[FupDup]

Dogstar, on 04 May 2021 - 12:26 AM, said:

The only oddity I noticed is that the Night Gyr is going to be the least agile heavy mech of all with a turn rate of only 40, making it substantially slower than all the other heavies. It also has the slowest acceleration and torso turn rate, although the difference is less noticeable there.

Is there a reason for not buffing it to the same levels as most heavies? It was never a particularly agile mech.

Probably because it carries more firepower than a lot of assault mechs while having enough speed to keep up with most heavies. It's gotta have some kind of drawback so that there's a point in picking other heavies.

[Der Geisterbaer]

The6thMessenger said:

I think the both of you are overthinking it.

I explicitly avoided doing that ... and it got me expected reactions nonetheless.

The6thMessenger said:

It's not going to be exact, but we have some idea of how it could be, the Atlas and the Commando seems to me the closest we could get for an analog due to the humanoid shape. We can only derive based on the information we have.

If anything it's not "wrong", it's just incomplete.

And it's not even the incompleteness that bothered me there but rather the deceptiveness of chosing a cube instead of a cuboid when trying to contrast those two humanoid mechs against each other. Since I have no suitable drawing program at hand I'll leave it up yo your imagination on how different the result would look - even in a 2D projection - when tripling the volume of a cuboid with dimensional ratios of 7u in height, 3u in width and 1u in depth (very rough ratio approximation for a cuboid outlining a humanoid shape) instead of making the visual comparison with a cube of 1u in all three dimensions. The required length multiplier per dimension is the exact same in either case but the overall result is very different even when accepting that we have to dial down expectations concerning physics like differences in material composition / density, etc.

The6thMessenger said:

Go ask Nightbird. He's the one that actually came up with that.

There's a reason I referenced both honesty and statisticians and their choice of graphical visualization for any given subject.

[PocketYoda]

Darian DelFord, on 03 May 2021 - 05:39 AM, said:

That's Kinda Ash's point. NO light mech is armored buff like the Urbie is. There is no sense for any other light mech to spec in the surviveability tree. The returns are just not worth it. Not even on a mech like the Oxide.

Maybe the adder or kitfox.

[The6thMessenger]

Der Geisterbaer, on 04 May 2021 - 03:07 AM, said:

I explicitly avoided doing that ... and it got me expected reactions nonetheless.

Well, you people kinda ARE when you're adding density in the equation. Yeah it's simplistic to assume simmilar density, but so what though?

Der Geisterbaer, on 04 May 2021 - 03:07 AM, said:

And it's not even the incompleteness that bothered me there but rather the deceptiveness of chosing a cube instead of a cuboid when trying to contrast those two humanoid mechs against each other.

Yeah sure you can technically make it the same volume if you made the Commando just as tall or just as wide -- basically deformed, but why though? Why can't we just assume uniform size increase in all direction that maintains the overall shape of the object? Wouldn't it be easier to start off from a simplistic assumption because it's easier to understand and to convey towards the audience? I mean we're not necessarily engineers here, we're just trying to piece together information to create a consistent world here. Hell, the writers of BattleTech, to my knowledge, didn't even bothered that far enough to provide the different densities of different mechs, why would I?

Lots of work in theorizing is assumption linking together whatever information we have, but there must be common ground for any suitable processing of information to work.

And this talk about volumetric rescale just happened because PGI is claiming volumetric rescale ( https://mwomercs.com...21jun2016#scale ), which is not and it led to oversized mechs. If anything, they set the cube-standard, so go complain to them, they started this mess.

Edited by The6thMessenger, 04 May 2021 - 03:42 AM.

[Absaint]

Bring it on, I trust the experience of the Caldron Collective to make something solid out of this.

And any problems we can always iterate and correct over time.

The weapons change worked great and I trust mobility and quirks will put things in a nice baseline.

The thing i suggest keeping an eye on, especially on mobility and quirks, is to not balance so that all are equal, but make "Flavor niches". Some mechs can be tipically slower to accel or rotate than others to bring flavor, but have an advantage in quirks per example. Its a great way to use the variety of mechs we have so that they mean more and afford more choices.

Keep up the great work guys.

[Steve Pryde]

Overall I like the agility changes alot but I dont understand why the Night Gyr still has to suffer in terms of agility? Give it similar agility like the Nova Cat because they are very close in terms of a weapon plattform. Just make it fun again to pilot it. With the actual comp changes it will still be unfun to pilot. 17 accleration for a heavy mech is still garbo when other 75t heavys have almost double.

edit: and give all assaults at least 90 degree torso twist range. No mech should have less than 90 in my opinion.

Edited by Steve Pryde, 04 May 2021 - 05:03 AM.