Do The Majority Of Players Want To Get Rid Of Convergence?
#901
Posted 13 April 2015 - 10:41 AM
To argue that a standard deviation of a minute of angle or two (which by the way is quite a large estimate for a purely mechanical system) is enough to justify a mech-sized CoF is to commit the binary fallacy.
The fallacy goes like this: "You have a 99.99% chance of hitting reasonably close to where you aimed, and a 0.01% chance of being off by enough to hit the wrong component or miss the mech. There are two possible outcomes, you hit or you miss, therefore the probability of missing is 50%, therefore a mech-sized CoF is justified."
Have you ever heard of the term "statistically insignificant?" The reason realism doesn't hold as an argument for the type of CoF being suggested is that the amount of variation present is statistically insignificant for the size of target we are shooting at.
Another issue is just what your random element represents. In TT it primarily represents your pilot's gunnery skill, as that's the main thing that affects the roll. In infantry games like ARMA, it represents the infantry skills that your character has to perform but you don't, such as maintaining sight picture, holding the weapon steady, smooth trigger squeeze, and controlling breathing. Plus smaller variations are more noticeable on an infantry-sized target.
But we don't need to abstract pilot skill in MW because we are the pilots. The pilot is sitting in a chair pushing buttons just like we are. Some of us probably have joysticks that look an awful lot like the ones in the mech cockpits. So the pilot skill is our skill.
It's similar to how we don't need to give moving mechs an RNG shield to make them "harder to hit" (netcode aside), because moving makes them harder to hit already on its own.
You want to insist that every single possible mechanical deviation be simulated for "realism's sake"? Alright, the CoF has a standard deviation of one minute of angle in radius, which as I mentioned is rather large for just the mechanical parts (usually most of the error is human error). So it has about a 68% chance of landing in a circle with a radius of 2.9cm per 100m . It has a 99.7% chance of landing in a circle of 8.7cm per 100m.
So at 300m (roughly IS med laser range, with a module) all my lasers will land with 99.7% certainty in a 26.1cm circle. On a target that is 1200-1400cm tall and roughly 500-800cm wide. Units preserved to give a sense of scale. Push it all the way to 1000m and that circle only grows to 87cm.
See what I mean by "statistically insignificant"?
#903
Posted 13 April 2015 - 10:47 AM
Almond Brown, on 13 April 2015 - 10:24 AM, said:
But that will be nerfing my skillz!!! QQ! QQ! Rant! Ran!
On a more serious note, though ...
I'm not opposed to it. But, don't modern targeting systems do that already?
#904
Posted 13 April 2015 - 10:50 AM
#905
Posted 13 April 2015 - 10:58 AM
Kuritaclan, on 13 April 2015 - 10:17 AM, said:
I gave you math. Actual math, which you have failed to do yourself, and which you have elected to ignore. Here i'll post it again for you.
Quote
Kuritaclan, on 13 April 2015 - 10:17 AM, said:
What!? You do understand that the table top game came first right? Mechwarrior is a spin off of battletech, not the other way around. Get your facts straight.
Edited by pbiggz, 13 April 2015 - 10:58 AM.
#906
Posted 13 April 2015 - 10:58 AM
E Rommel, on 13 April 2015 - 10:41 AM, said:
To argue that a standard deviation of a minute of angle or two (which by the way is quite a large estimate for a purely mechanical system) is enough to justify a mech-sized CoF is to commit the binary fallacy.
The fallacy goes like this: "You have a 99.99% chance of hitting reasonably close to where you aimed, and a 0.01% chance of being off by enough to hit the wrong component or miss the mech. There are two possible outcomes, you hit or you miss, therefore the probability of missing is 50%, therefore a mech-sized CoF is justified."
Have you ever heard of the term "statistically insignificant?" The reason realism doesn't hold as an argument for the type of CoF being suggested is that the amount of variation present is statistically insignificant for the size of target we are shooting at.
There is a very good reason why I mentioned adopting something like R95.
#907
Posted 13 April 2015 - 10:59 AM
Mystere, on 13 April 2015 - 10:28 AM, said:
Quote
Which source materials say that the targeting/aiming systems are perfectly accurate and that all weapons converge perfectly?
The prove is in the pudding - see:
http://mwomercs.com/...ost__p__4358041
Aiming in MWO is "inaccurate" as i made clear with the ac20,ppc,laser example for alpha purpose:
http://mwomercs.com/...36#entry4358236
Auto convergance does not help you to alpha on ranges with a mixed loadout. Only bad enemy players help you to crush them with alphas of mixed loadout, which are the highest alphas.
Alphas out of boating have other problems - high heat - if this get punished - there you see the alphas go away - and since beam duration is a problem to translate the full damage into a specific part on a high range + decrease in damage/heat ratio it is not that problematic with lasers as some think it is. It only becomes a problem between bad players and players who at least can aim on enemys which are standing nearly still.
Why having perfectly accurate and that all weapons converge perfectly? You ask? I say common sense. If an engineer have to construct a targeting assistant system which do not provide "perfectly accuracy" he get fired or to say it in the speach of Kurita or Liao - beheaded, for bringing up lackluster tech, that does not help you on the task!
Mystere, on 13 April 2015 - 10:28 AM, said:
No thx i do not look up your 7522 post. But i give you this: http://mwomercs.com/...ost__p__4115437
Quote
So there is the side who claim, pinpoint convergence with 60-80 damage alphas is a problem. I say pinpoint convergance is not a problem. Why? - You have to maintain your heat, observe the minmap/radar, watch your little mech readout, find the weak spots on targeted enemy mechs, move yourself (just a little note torso movement + leg movement seems to be a hurdle for new players) into positions so you have options for retreat or atk and aim to kill other mechs with weapons who do not at all behave the same way and new to the party VoiP for the players who did not played in coordinated groups before. I guessed you are on the side of those who claim that pinpoint/instant/auto convergence is a problem mixed up with high alphas. Anyhow since you got rude to me saying i want to have a "low skill game" to clearly mean that i am a low skilled player, i don't know why it is usfull to discuss this point any further.
Edited by Kuritaclan, 13 April 2015 - 11:33 AM.
#908
Posted 13 April 2015 - 11:01 AM
Simultaneously, anything that is implemented right now to address convergence/pinpoint would likely break the game given other issues, most notably hit reg. Agitators demanding a "fix" do so for a problem that doesn't exist and at the risk of jeopardizing gameplay quality. In my opinion, it is a short-sighted desire stemming from their own lack of skill in aiming or care in positioning themselves during battle.
Edited by Nightmare1, 13 April 2015 - 11:04 AM.
#909
Posted 13 April 2015 - 11:04 AM
E Rommel, on 13 April 2015 - 10:41 AM, said:
To argue that a standard deviation of a minute of angle or two (which by the way is quite a large estimate for a purely mechanical system) is enough to justify a mech-sized CoF is to commit the binary fallacy.
The fallacy goes like this: "You have a 99.99% chance of hitting reasonably close to where you aimed, and a 0.01% chance of being off by enough to hit the wrong component or miss the mech. There are two possible outcomes, you hit or you miss, therefore the probability of missing is 50%, therefore a mech-sized CoF is justified."
Have you ever heard of the term "statistically insignificant?" The reason realism doesn't hold as an argument for the type of CoF being suggested is that the amount of variation present is statistically insignificant for the size of target we are shooting at.
Another issue is just what your random element represents. In TT it primarily represents your pilot's gunnery skill, as that's the main thing that affects the roll. In infantry games like ARMA, it represents the infantry skills that your character has to perform but you don't, such as maintaining sight picture, holding the weapon steady, smooth trigger squeeze, and controlling breathing. Plus smaller variations are more noticeable on an infantry-sized target.
But we don't need to abstract pilot skill in MW because we are the pilots. The pilot is sitting in a chair pushing buttons just like we are. Some of us probably have joysticks that look an awful lot like the ones in the mech cockpits. So the pilot skill is our skill.
It's similar to how we don't need to give moving mechs an RNG shield to make them "harder to hit" (netcode aside), because moving makes them harder to hit already on its own.
You want to insist that every single possible mechanical deviation be simulated for "realism's sake"? Alright, the CoF has a standard deviation of one minute of angle in radius, which as I mentioned is rather large for just the mechanical parts (usually most of the error is human error). So it has about a 68% chance of landing in a circle with a radius of 2.9cm per 100m . It has a 99.7% chance of landing in a circle of 8.7cm per 100m.
So at 300m (roughly IS med laser range, with a module) all my lasers will land with 99.7% certainty in a 26.1cm circle. On a target that is 1200-1400cm tall and roughly 500-800cm wide. Units preserved to give a sense of scale. Push it all the way to 1000m and that circle only grows to 87cm.
See what I mean by "statistically insignificant"?
Two big big points you ignore.
Nobody is talking about a large cone of fire for firing one gun. The numbers you cite would be correct for one mechanical gun platform firing one weapon at a stationary large target, and...
THESE mechanical systems (Battlemechs) are not nearly that accurate.
Their maintenance is done by techs that are good at putting parts together, but they do not really know HOW they work. The parts are usually fabricated nearby, and not from the factory with factory tolerances. On top of that, most of the 'mechs WE use are altered from their factory state with Large Pulse lasers cobbled into medium laser mounts, and Gauss Rifles jury-rigged to fit in a Machine gun slot.
How about we let stock builds have a 2 minute angle of deviation, but customized build have more.
#910
Posted 13 April 2015 - 11:11 AM
Therefore I really cringe when you talk about "skills" and "aiming" when it comes to laser weapons...actually I think it is harder to have an impact and do well with LRMs than with lasers.
High heat is also not that much of a problem because a lot of maps simply let you "alpha, alpha, back into cover, cool down, alpha, alpha rinse and repeat".
Too bad that there are not "soft drawbacks" for excess heat like in the TT. Getting slower or making your HUD flicker would make you think twice about your heat management.
#911
Posted 13 April 2015 - 11:11 AM
Ragtag soldier, on 12 April 2015 - 07:53 PM, said:
Would you be so kind as to show us a graphical (with pictures) break down of your "degrading convergence" methodology please. Using dem big words is hard on us kids. Thanks.
#912
Posted 13 April 2015 - 11:12 AM
Kuritaclan, on 13 April 2015 - 10:59 AM, said:
In other words, you have nothing.
Kuritaclan, on 13 April 2015 - 10:59 AM, said:
Let us be perfectly accurate here. What I said was:
Mystere, on 13 April 2015 - 09:25 AM, said:
And even then, I do at times do get rude to someone who chooses to paste a whole wall of text, especially if the intent is to obfuscate, rather than just placing a concise argument that goes straight to the point.
#913
Posted 13 April 2015 - 11:14 AM
"I believe that a reasonably skilled pilot, with a couple of hundred hours logged in the cockpit, should not be able to reliably hit a Centurion-sized target at 270m with more than one weapon at a time, regardless of the weapon type(s) or either 'Mech's state (heat, movement, etc.)."
Strongly Agree---o---o---o---Strongly Disagree
(1453-R's answer: Strongly Disagree)
"I am perfectly willing to accept serious problems with hit registration, provided those problems arise from some sort of alpha-disabling convergence which forces the server to make many multiple times extra calculations per shot taken."
Strongly Agree---o---o---o---Strongly Disagree
(1453-R's answer: F*** that noise)
"I don't particularly care if missiles become the sole viable weapon type due to unreliability of both convergence and hit registration of all other weapons in the face of my desired convergence fix."
Strongly Agree---o---o---o---Strongly Disagree
(1453-R's answer: Disagree)
"Alpha strikes are the Devil."
Strongly Agree---o---o---o---Strongly Disagree
(1453-R's answer: slightly Disagree)
"I can actually describe to other players what I mean by the term 'alpha strike', why they are the Devil, and how ensuring that any player who fires an alpha strike is punished by both crippling heat spikes and missing ninety percent of the shots fired in that alpha strike is going to make MWO a better game."
Strongly Agree---o---o---o---Strongly Disagree
(1453-R's answer: Disagree)
"No, really. I mean it. Alpha strikes are the Devil."
Strongly Agree---o---o---o---Strongly Disagree
(1453-R's answer: Disagree)
#915
Posted 13 April 2015 - 11:18 AM
E Rommel, on 13 April 2015 - 10:41 AM, said:
To argue that a standard deviation of a minute of angle or two (which by the way is quite a large estimate for a purely mechanical system) is enough to justify a mech-sized CoF is to commit the binary fallacy.
The fallacy goes like this: "You have a 99.99% chance of hitting reasonably close to where you aimed, and a 0.01% chance of being off by enough to hit the wrong component or miss the mech. There are two possible outcomes, you hit or you miss, therefore the probability of missing is 50%, therefore a mech-sized CoF is justified."
You bring up excellent points, and the binary fallacy is exactly what most of the critics of a CoF system have been exclaiming.
Now, when you cite 1 MOA as a huge error for these purely mechanical systems, I have to ask, where you get that to be true? It may be valid, but do you have some rationale for it? 1 MOA is about the accuracy of precision small arms today - a modern sniper rifle is decent if it's better than 1 MOA (the best ones are mechanically about 1/4 MOA). I don't know enough about tank guns to say if 1 MOA is too precise or not. I would expect lasers to be significantly more accurate, but again, when dealing with a 4 ton laser on articulated, moving mounts, with significant heat effects, I'm not sure 1 MOA is going to be so outside the realm of possibility.
Who argued for a mech-sized CoF, and at what range? I think you're picking out your own numbers and calling the whole idea bad because of it.
#916
Posted 13 April 2015 - 11:22 AM
Mystere, on 13 April 2015 - 10:58 AM, said:
There is a very good reason why I mentioned adopting something like R95.
Doesn't change how absolutely miniscule the effect is. Congrats, your odds of hitting dismounted infantry at 1000m is very low! Too bad we don't have infantry in this game.
If you are saying that you are willing to accept that 1 MoA (or reasonably, less) variation being modeled for completeness despite its triviality then whatever, but I don't see the point of bothering.
#917
Posted 13 April 2015 - 11:23 AM
Nightmare1, on 13 April 2015 - 11:01 AM, said:
Actually, adding fixed or even zero convergence does not in any way force a random change, nor make it chance-based. But, it can be argued that it will increase the skill cap.
Adding CoF is obviously a different issue altogether.
Nightmare1, on 13 April 2015 - 11:01 AM, said:
Yes, I myself am aware that hit registration might be an issue, especially because it was probably what caused PGI to remove the delayed convergence that existed during closed beta. But, at the same time, I also see it as a technical issue that will need to be discussed at a later date, just not right now in this thread.
#918
Posted 13 April 2015 - 11:27 AM
Dino Might, on 13 April 2015 - 10:10 AM, said:
I just want to know why our proposed ideas aren't legitimate because we are, 'a minority whining on a forum,' when the counterargument used is predicated entirely on another game's, 'minority whining on a forum.'
It would appear to be based on the common denominator "a minority whining on a forum" And a typical response is to cherry pick from the thread, ones that fit your sides needs.
No one has "shown" any of these fine CoF models yet? Why not. Scared? How does Range and Multiple weapons fired interact in this CoF?
How does the CoF react when a small Light Mechs buzzes around (gets Small-Big-Small-Big-Small-Big) versus an Assault Mech?
Show us some examples so we can better understand how this system that is proposed would function inside MWO. Is that such a difficult ask?
#919
Posted 13 April 2015 - 11:28 AM
pbiggz, on 13 April 2015 - 10:58 AM, said:
Quote
What!? You do understand that the table top game came first right? Mechwarrior is a spin off of battletech, not the other way around. Get your facts straight.
That does not prove it false - i quoted aiming/targeting assistant systems out of a BATTLETECH®-source book. It just does not matter what was first. it is the BATTLETECH®-Universe what is the anchorpoint - what is made out of it (TT,Novels,other things) have to be conform with BATTLETECH®-Universe. But there is no interdepance as some do claim between TT and MWO - if they have some chared basics, they come from the BATTLETECH®-Universe.
pbiggz, on 13 April 2015 - 10:58 AM, said:
I gave you math. Actual math, which you have failed to do yourself, and which you have elected to ignore. Here i'll post it again for you.
Quote
This is not much away from what i said a couple pages away (http://mwomercs.com/...ost__p__4349588) to quote myself #566:
Quote
As said back then, Your proposal is not thought out, when it comes down to higher count weapon mounts in one location, like the 6 Laser Hardpoints in the HBK or the 4 Mounts of the arm in the GRF-1S or the 4 to 5 mounts in the BNC or the arms of the Dire Prime or the Nova Prime with 6 Lasers. - Those Laser Arays are set on the mech parts to converge ultimative. Everything else your system would do to a HBK Shoulder is making the idea behind such a Weapon-array usless. And then I would ask myself - ok we now have more simulation, but instead of simulating the intend of creatinga an array we get absurd convergences with not to mention some people who wanna ad CoF onto lasers.
Maybee this is the time for some people to come into this thread and post some facepalm pictures.
Hotthedd, on 13 April 2015 - 11:04 AM, said:
You mean the other way around. A gauss that is fit into mg slot has nearly to no angle to converge, while the stock mg could converge rather quike since it is stock. This would have had hand and foot. But anyway this needs remodeling so much, that you nearly will break the aiming, and hitreg system of the game. Maybee this "reality effect" will be introduced via patch in 50 years. Harr harr.
Edited by Kuritaclan, 13 April 2015 - 11:47 AM.
#920
Posted 13 April 2015 - 11:30 AM
E Rommel, on 13 April 2015 - 11:22 AM, said:
If you are saying that you are willing to accept that 1 MoA (or reasonably, less) variation being modeled for completeness despite its triviality then whatever, but I don't see the point of bothering.
Where did I state that I wanted 1 MoA? I mentioned adopting something like R95 specifically because it will create "Oh, ****!" moments, and hopefully at the time you want it least to happen.
6 user(s) are reading this topic
0 members, 6 guests, 0 anonymous users