339 replies to this topic

[IceSerpent]

wolf74, on 11 January 2012 - 12:38 AM, said:

Pin Point fire w/Random Damage (RNG?): (Saw it didn’t really understand it) Sorry IceSerpent I just didn’t get it.

Let me try to elaborate a bit. First, a bit of a background for folks unfamiliar with BT. Contrary to the popular belief, the object of simulation is not an M1 Abrams that has legs attached to it and features multiple main guns, but a weapon platform (mech) described in BT books. A mech by definition lacks the ability to converge weapons on a single spot with any degree of accuracy and with regard to aiming is more similar to early 20th century naval vessel than to a tank.

This feature was simulated in TT game as RNG - a roll of dice for each shot determines both whether you hit or miss and where exactly you hit when you land that shot on target. This doesn't translate well into a simulation because we would end up with MMO-style "tab-targeting", i.e. designate the target, pull the trigger, and pray to RNG gods. So, the best solution (IMHO) should have both skill-based targeting where location of your reticle actually matters for aiming and less-than-perfect accuracy of the machine we're simulating.

Now, a lot of people come from FPS games and seem to think that random aiming (hit or miss) and random damage spread (what gets damaged when you hit) are one and the same. This is not the case. What I suggest is to have weapons converge perfectly on the spot under your reticle (either Pin Point Fire or Pin Point Accuracy with reticle movement in your list), with hit/miss being affected by projectile travel time, your movement, and movement of the target. The damage on the other hand is randomized across nearby hit-boxes in order to simulate inherent lack of accuracy.

This way whether you hit the target or miss it completely is determined by players' skills, but where exactly you hit (when you hit) is somewhat random. The spread should be limited to a certain area around the "projected point of impact" (where you would have hit if your weapon was 100% accurate), so that you still can try to exploit previous damage to the target. In other words, hitting left side damages something on the left side, hitting a leg doesn't result in a head shot, etc.

Hope this clears things a little.

[Vulpesveritas]

Foster Bondroff, on 13 January 2012 - 01:21 AM, said:

A system similar to the one used in WoT seems a good idea to me. It doesnt remove the skill element, while at the same time avoids pin point accuracy about extrem ranges.

Which is stupid. If you know math well enough and are a good enough aim and you want to use your light guass from more than a kilometer you should be able to do so without your well aimed shot moving awry from 'unknown' factors. Especially if it's the only weapon you are firing.


And oops. Meant to just edit this... >.> <.<

Edited by Vulpesveritas, 13 January 2012 - 09:24 AM.

[Vulpesveritas]

IceSerpent, on 13 January 2012 - 09:11 AM, said:

Let me try to elaborate a bit. First, a bit of a background for folks unfamiliar with BT. Contrary to the popular belief, the object of simulation is not an M1 Abrams that has legs attached to it and features multiple main guns, but a weapon platform (mech) described in BT books. A mech by definition lacks the ability to converge weapons on a single spot with any degree of accuracy and with regard to aiming is more similar to early 20th century naval vessel than to a tank.

This feature was simulated in TT game as RNG - a roll of dice for each shot determines both whether you hit or miss and where exactly you hit when you land that shot on target. This doesn't translate well into a simulation because we would end up with MMO-style "tab-targeting", i.e. designate the target, pull the trigger, and pray to RNG gods. So, the best solution (IMHO) should have both skill-based targeting where location of your reticle actually matters for aiming and less-than-perfect accuracy of the machine we're simulating.

Now, a lot of people come from FPS games and seem to think that random aiming (hit or miss) and random damage spread (what gets damaged when you hit) are one and the same. This is not the case. What I suggest is to have weapons converge perfectly on the spot under your reticle (either Pin Point Fire or Pin Point Accuracy with reticle movement in your list), with hit/miss being affected by projectile travel time, your movement, and movement of the target. The damage on the other hand is randomized across nearby hit-boxes in order to simulate inherent lack of accuracy.

This way whether you hit the target or miss it completely is determined by players' skills, but where exactly you hit (when you hit) is somewhat random. The spread should be limited to a certain area around the "projected point of impact" (where you would have hit if your weapon was 100% accurate), so that you still can try to exploit previous damage to the target. In other words, hitting left side damages something on the left side, hitting a leg doesn't result in a head shot, etc.

Hope this clears things a little.

Yes but that detracts from a certain extent of realism, and I feel that should also be determined by realistic factors- like the fact that bipeds aren't nearly as stable as other platforms for firing weapons, especially multiple. plus lasers get bent by high humidity environments. By simply using physics, recoil, and dynamic environments, it would make everything fair without any real issue.

[IceSerpent]

Vulpesveritas, on 13 January 2012 - 09:23 AM, said:

Yes but that detracts from a certain extent of realism, and I feel that should also be determined by realistic factors- like the fact that bipeds aren't nearly as stable as other platforms for firing weapons, especially multiple. plus lasers get bent by high humidity environments. By simply using physics, recoil, and dynamic environments, it would make everything fair without any real issue.

Not really - there's no real weapon that is 100% accurate, so at the very least you will always have some damage spread even if you are firing a sniper rifle welded to a massive platform (so that weapon is always static and all recoil is absorbed). On top of that, when you use a mech, you don't actually aim the weapon - you tell the onboard computer where to aim it, and in BT a mech is not capable of accurately pointing weapon barrels in the right direction.

You certainly could use real life physics (humidity, cross winds, etc), but that would be cpu-intensive and why do that when the end result would be pretty much the same as RNG.

[Vulpesveritas]

Yes but if you know math you can compensate for it, an RNG is going to be random each time, you can't train your fire to the right slightly to hit your target should your aim be off a bit. And while no ballistic weapon is 100% accurate, they're unlikely to err more than 0.01% if that by where you aim naturally. Lasers will shoot where they aim, but in a liquid or gaseous medium they also err to the side, liquid moreso than gas. A PPC loses particles over distance, but for the most part would be dependent on the combined aim.

And it wouldn't be that CPU intensive. The original Ghost Recon had all of that save for environmental factors, which by the way are already supported by CryENGINE 3 for dynamic environments.

[DEVASTATOR]

Late to the party and i haven't read ALL the posts (there's 17 pages of them) but this debate has been raging for years and years as the BT purists try to find ways to make the video game version follow the tabletop rules. The fact is that they can't as the two games are just too different in so may ways.

I'm not in favour of weapons spread in any way at all as i believe its so unrealistic that it will deter from the game. If I aim at something I should hit it shouldn't I? Are we saying that the targeting computers of the far future are dramatically inferior to what we have today? The bottom line on the arguement for or against things like weapons spread is that with pipoint accuracy mechs die to quickly and there aren't any long, drawn out battles like you'd expect between 150 tons of armour and weapons. Ii think there's far better ways to accomplish this goal than with something as arbitrary and unrealistic as weapons spread. In fact, soke of this has already been done in previous games.

WEAPON RANGING:
A lot of this can be handled by making long range weapons less effective at closer distances. MW3 got it right when it came to LRM's for example. The way they were modelled was with a specific spead and turning radius. At close ranges, they couldn't turn fast enough to hit a target close by. So, they worked where they we supposed to work - at greater distances. Other weapons could be modelled to achieve the same basic goal.

Perhaps large lasers become more difficult to focus on a target that's close. As an example, anything that's within 500m can be targeted with a large laser but it may take a few seconds for tha large laser to center on the target. Pull the trigger before the laser has centered and you get a shot that's off target and might hit the arm instead of the torso you aimed at while all the other short range lasers do hit your target. I know that may shound like weapons spread but its not really as the weapons made for the range of the target you are firing at do hit the area you aimed at. Maybe its "selective" weapons spread.

ARMOR GRANULARITY:
Personally, I'd like to see a finer level of 'mech hitbox design. I mean, do we really believe that a mech's right torso is really one giant slab of armor and other tech? One would think that the only way you'd kill a mech in "real life" by hitting its CT would be to repeatedly hit the exact same point over and over again to punch through the armour and into the inner workings. If I hit the CT 20 times with a large laser in 20 different areas I'm just going to make that mech look like its got a bad case of acne and nothing more (again, I'm postulating a "real life" scenario so to speak).

So, what if the CT was still the CT from the perspective of overall armor and weapons and engine placement, etc but from a hit box perspective it had several sub sections. Maybe the CT is made up of 5 different hitboxes that all have to be destroyed in order to actually destroy the CT? And, if you destroy one hitbox there's a critical hit associated with it that affects engibe speed, torso twist speed, weapons, ammo reserves, electronics, etc. That would make killing the mech a longer affair but still reward for an accurate shot to the same place on the CT.

[Vulpesveritas]

DEVASTATOR, on 13 January 2012 - 10:19 AM, said:

Late to the party and i haven't read ALL the posts (there's 17 pages of them) but this debate has been raging for years and years as the BT purists try to find ways to make the video game version follow the tabletop rules. The fact is that they can't as the two games are just too different in so may ways.

I'm not in favour of weapons spread in any way at all as i believe its so unrealistic that it will deter from the game. If I aim at something I should hit it shouldn't I? Are we saying that the targeting computers of the far future are dramatically inferior to what we have today? The bottom line on the arguement for or against things like weapons spread is that with pipoint accuracy mechs die to quickly and there aren't any long, drawn out battles like you'd expect between 150 tons of armour and weapons. Ii think there's far better ways to accomplish this goal than with something as arbitrary and unrealistic as weapons spread. In fact, soke of this has already been done in previous games.

WEAPON RANGING:
A lot of this can be handled by making long range weapons less effective at closer distances. MW3 got it right when it came to LRM's for example. The way they were modelled was with a specific spead and turning radius. At close ranges, they couldn't turn fast enough to hit a target close by. So, they worked where they we supposed to work - at greater distances. Other weapons could be modelled to achieve the same basic goal.

Perhaps large lasers become more difficult to focus on a target that's close. As an example, anything that's within 500m can be targeted with a large laser but it may take a few seconds for tha large laser to center on the target. Pull the trigger before the laser has centered and you get a shot that's off target and might hit the arm instead of the torso you aimed at while all the other short range lasers do hit your target. I know that may shound like weapons spread but its not really as the weapons made for the range of the target you are firing at do hit the area you aimed at. Maybe its "selective" weapons spread.

ARMOR GRANULARITY:
Personally, I'd like to see a finer level of 'mech hitbox design. I mean, do we really believe that a mech's right torso is really one giant slab of armor and other tech? One would think that the only way you'd kill a mech in "real life" by hitting its CT would be to repeatedly hit the exact same point over and over again to punch through the armour and into the inner workings. If I hit the CT 20 times with a large laser in 20 different areas I'm just going to make that mech look like its got a bad case of acne and nothing more (again, I'm postulating a "real life" scenario so to speak).

So, what if the CT was still the CT from the perspective of overall armor and weapons and engine placement, etc but from a hit box perspective it had several sub sections. Maybe the CT is made up of 5 different hitboxes that all have to be destroyed in order to actually destroy the CT? And, if you destroy one hitbox there's a critical hit associated with it that affects engibe speed, torso twist speed, weapons, ammo reserves, electronics, etc. That would make killing the mech a longer affair but still reward for an accurate shot to the same place on the CT.

Simple solution, make the models for the mechs be divided into two layers, an inner layer, where each part of the skeleton and inner parts is a hitbox, and an outer layer, which is the armor, and is made up of a set size of triangles all the way around the mech. A weapon would hit a certain number of hitboxes, and if the armor is destroyed, it allows the inner parts to be targeted. This also means while smaller weapons are great in close range and the like, a larger weapon can destroy more hitboxes, so you can deal more internal damage as well. Just a thought.

Also if you're going to say "that's too CPU intensive" no it's not. It just takes longer to code the models. That's all. Really, if you have any system 6 years old or less with at least a 2ghz processor speed and a dual core you'll be fine with that.

[verybad]

DEVASTATOR, on 13 January 2012 - 10:19 AM, said:

ARMOR GRANULARITY:
Personally, I'd like to see a finer level of 'mech hitbox design. I mean, do we really believe that a mech's right torso is really one giant slab of armor and other tech? One would think that the only way you'd kill a mech in "real life" by hitting its CT would be to repeatedly hit the exact same point over and over again to punch through the armour and into the inner workings. If I hit the CT 20 times with a large laser in 20 different areas I'm just going to make that mech look like its got a bad case of acne and nothing more (again, I'm postulating a "real life" scenario so to speak).

Liked your post, especially the Armor Granualarity idea. I've seen that mentioned before, but yes, it is a good idea. The one problem with it is that the more stuff goes through the network, the more lag you can get. So while the idea has a lot of merit, it is not without some potential drawbacks.

Edited by verybad, 13 January 2012 - 01:00 PM.

[Vulpesveritas]

verybad, on 13 January 2012 - 01:00 PM, said:

Liked your post, especially the Armor Granualarity idea. I've seen that mentioned before, but yes, it is a good idea. The one problem with it is that the more stuff goes through the network, the more lag you can get. So while the idea has a lot of merit, it is not without some potential drawbacks.

Depends on your internet connection. But if you have anything more than a megabyte/second you should be fine. If you have a 3G/4G phone/hotspot, any DSL/cable/FIO, or the like, you should be fine in that regard.

If you're still running dialup/2G..... umm... not sure what to say.

[DEVASTATOR]

Hi vb, yes there "could" be some issues with performance if this system had numerous hit boxes to track. However, one would "hope" that if we kept it simple it wouldn't be too much of a drag on your system. We essentially have 8 hit boxes (RA, RT, CT, LT, LA, RL, LL and head) today and, to make things seem logical, one might go with:

Arms = 3
Side Torso = 4
CT = 4
Legs = 3
Had = 2

which would increase it from 8 to 26. I say this is "logical" only because in the mech lab of old you'd be able to put two or three weapons in the arm of some mechs thus the 2 or 3 hitboxes. I have to believe that this wouldn't become too CPU intensive. Ideally, each hitbox would be associated with a piece of hardware like a weapon, ammo storage, engine critical, electronics, etc. such that destroying the hitnox destroys the associated piece of hardware giving the effect of the "critical hit" we experieinced in MW2.

If you wanted to ge real fancy you could dynamically create hitboxes based on the mech design. If you stuck 2 weapons in the side torso the system would dynamically create 2 hit boxes for you to destroy. If you had 2 weapons and ammo storage there'd be 3 hitboxes. Going this far might be putting a strain on the system though.

p.s. I've been touting this idea since the MW 3/4 days although I'm sure others have suggested similar stuff too. in fact, maybe I'll start a post on just this idea.

Edited by DEVASTATOR, 13 January 2012 - 01:21 PM.

[plodder]

plodder, on 08 January 2012 - 10:48 AM, said:

This is awesome!!!!!I can't read it all, but you fellas seem to know your stuff, or at least one of you do ......

Now.let me start with the hit points placement if you have multiples of hit point placements on each limb torso etc. at every angle, then a hit from even a lg. laser at any range would have some but a limited splash over several hit point placements(if there was enough of them). So the likehood of hitting the same points at long range less realistic if programmed correctly. the cone pictured above should not be how the damage is spread me thinks, but the probable area in will hit. even at extreme range, a lg laser'* *** will not be more ten about 8' of damage area, and if the target is moving that is spread over up to 26" of damage (moving speed while torso twist limb hit) .
can the MWO support a mech with 400-700 hitting points/ places that are hit that may have 10 actual value, meaning 10 x 400-700.....

concentrated fire is difficult, that is why the mech is the baddest fighting weapon any pilot can have. Thanks uncle danno

DEVASTATOR, on 13 January 2012 - 10:19 AM, said:

Late to the party and i haven't read ALL the posts (there's 17 pages of them) but this debate has been raging for years and years as the BT purists try to find ways to make the video game version follow the tabletop rules. The fact is that they can't as the two games are just too different in so may ways.

I'm not in favour of weapons spread in any way at all as i believe its so unrealistic that it will deter from the game. If I aim at something I should hit it shouldn't I? Are we saying that the targeting computers of the far future are dramatically inferior to what we have today? The bottom line on the arguement for or against things like weapons spread is that with pipoint accuracy mechs die to quickly and there aren't any long, drawn out battles like you'd expect between 150 tons of armour and weapons. Ii think there's far better ways to accomplish this goal than with something as arbitrary and unrealistic as weapons spread. In fact, soke of this has already been done in previous games.

WEAPON RANGING:
A lot of this can be handled by making long range weapons less effective at closer distances. MW3 got it right when it came to LRM's for example. The way they were modelled was with a specific spead and turning radius. At close ranges, they couldn't turn fast enough to hit a target close by. So, they worked where they we supposed to work - at greater distances. Other weapons could be modelled to achieve the same basic goal.

Perhaps large lasers become more difficult to focus on a target that's close. As an example, anything that's within 500m can be targeted with a large laser but it may take a few seconds for tha large laser to center on the target. Pull the trigger before the laser has centered and you get a shot that's off target and might hit the arm instead of the torso you aimed at while all the other short range lasers do hit your target. I know that may shound like weapons spread but its not really as the weapons made for the range of the target you are firing at do hit the area you aimed at. Maybe its "selective" weapons spread.

ARMOR GRANULARITY:
Personally, I'd like to see a finer level of 'mech hitbox design. I mean, do we really believe that a mech's right torso is really one giant slab of armor and other tech? One would think that the only way you'd kill a mech in "real life" by hitting its CT would be to repeatedly hit the exact same point over and over again to punch through the armour and into the inner workings. If I hit the CT 20 times with a large laser in 20 different areas I'm just going to make that mech look like its got a bad case of acne and nothing more (again, I'm postulating a "real life" scenario so to speak).

So, what if the CT was still the CT from the perspective of overall armor and weapons and engine placement, etc but from a hit box perspective it had several sub sections. Maybe the CT is made up of 5 different hitboxes that all have to be destroyed in order to actually destroy the CT? And, if you destroy one hitbox there's a critical hit associated with it that affects engibe speed, torso twist speed, weapons, ammo reserves, electronics, etc. That would make killing the mech a longer affair but still reward for an accurate shot to the same place on the CT.

I like your ideas, you can read my above post. I would like to combine these ideas somehow. What you think? Thanks uncle danno

[IceSerpent]

Vulpesveritas, on 13 January 2012 - 09:55 AM, said:

Yes but if you know math you can compensate for it, an RNG is going to be random each time, you can't train your fire to the right slightly to hit your target should your aim be off a bit. And while no ballistic weapon is 100% accurate, they're unlikely to err more than 0.01% if that by where you aim naturally. Lasers will shoot where they aim, but in a liquid or gaseous medium they also err to the side, liquid moreso than gas. A PPC loses particles over distance, but for the most part would be dependent on the combined aim.

You can only compensate for some factors like bullet drop for example. Try firing a 38 special at a target 300m out and your rounds will be all over the place no matter how hard you try to compensate.

Quote

And it wouldn't be that CPU intensive. The original Ghost Recon had all of that save for environmental factors, which by the way are already supported by CryENGINE 3 for dynamic environments.

Something tells me that you're guessing here
Try to do the math for a single round, assuming that you know all the factors (i.e. actually wirte down the equations) and see what it looks like. Then scale it up to a 12 v 12 brawl per match instance (note that single mech tends to have multiple weapons it can fire at the same time) and to however many instances you want to be played simultaneously on a server.

[IceSerpent]

DEVASTATOR, on 13 January 2012 - 10:19 AM, said:

Are we saying that the targeting computers of the far future are dramatically inferior to what we have today?

This is exactly what we are saying. Targeting capabilities in BT universe are much, much worse than what we have today IRL. It's actually a fairly important "feature" and is well described in the novels.

[Drakor]

I find it hard to restrain myself from a long period, filled with facepalm, in reaction to this thread.

To all those whom go for the "Cone of Fire" as the One, and ONLY methond of preventing alpha-strikes of instant pwnage... And insist that it won't be that serious- Pick on or the other. It has to be pretty serious to prevent it all together. But having the cone of fire completely random is... Well, perhaps technically 'accurate' to tabletop, but not very intelligent in it's rendering of such.

To all those whom have been insisting that the weapon will remain pinpoint accurate, it'll just sway, so that'll spread it out... Go check out testing of weapons, done by various militairies. You'll find a weapon pointing at the same spot... will not hit the same spot. Look up MoA accuracies. That's right folks, to an extent, Cone of Fire is Real life!

So here's the only LOGICAL idea, as far as it goes- A combination of the two. (It's been mentioned before, more then once in this thread, but still, seems to need to be mentioned again.)

Each weapon has a SMALL cone of fire, equating to a very little spread over it's entire range. What do I mean? Instead of a crosshair of "+", you get a crosshair of "o". Which at close range (even out to about 200-300~M) would hit reliably within a small area on a target mech... But at far range would still produce enough of a spread to... not stop entirely, but severely hamper Alpha-Strikes.

Then add in the effects of movement sway- Go ahead and say battlemech's are stable. Look at every last rendering of them so far, The arms sway, the torso bobs, the head bounces... You can say that those Gyro's stabilize it all you want, but the Quite Cannonical renderings of those battlemechs dissagree with you. Violently so.

Try and take a shot with a PPC, Running at 64~KM/h. About average heavy speed. Your arms are converging on the target, but wait! You're moving, so that convergence distance is constantly changing... Odds are that computer system is just a fraction of a second behind you- Your aim likely isn't perfect enough to stay on target at a closure speed of about 100~KM/h, assuming you're not both running straight at eachother...

You let loose with the PPC... A Flash that would have blinded you is rendered moot by your battlemech's auto-tinting screen... But the shot went wide, you were running. Your lead was fine, but your timing was off. Just as your crosshair got into position to fire, the arm was jostled- Forgot to wait for the midpoint between steps. Oh well, it'll recharge quickly enoudh anyway, right? You're at 600m, and closing.

So your system's getting conflicting data... Urban environment? Occasional lamp-post, or small building inbetween you two... and the occasional Larger building, obstructing your sight (and aim) completely, and throwing off your targeting computer completely. It converges too close... Building's gone... but your aim's still off- It starts going too far... you're now about 400m from the target- your not even in Medium Laser range yet. Your crosshair passes over him, but you haven't been recalibrated to that range yet...

You let loose with MRM's... set to converge too far. Spread is all over his torso, right? Because your timing is so perfect, that at 400m, Closing at 100km/h, you can hit dead center in a target that doesn't want to be hit at all. Wrong-o! More likely you're gonna slam some into an arm, because we BOTH know he'd twitch the stick, (or keyboard) a little on seeing you fire. Remember- Urban environment... and moving shot... Chances are, your missiles get halfway there in their corkscrew... and some hit a building! A couple hit the arm, and a few more pass by harmlessly.

By the time your MRM's have recycled, and are ready for the next shot, you've closed within 300m. Medium laser time! You were hit by some MRM's from that guy too, so your aim was thrown off for a moment- Not long, but enough to converge your guns even FURTHER out from the target- no wonder, he rocked you towards looking up. You swing him back in the middle of your reticule... It flashes gold, your targeting computer says you'll hit... He pulls off his shot first, and unluckily for you, it was an autocannon. Your shots go wide, once more, and worse, his shot hit your MRM launcher.

The damage isn't critical, you've got a mech with CASE after all... but the rocking from the shot, and the destruction of a good part of your punch? Let's try with that PPC again then, right? Nice and close, should be pretty simple. Remembering the earlier wide shot, you pull DOWN on your crosshair, and aim at the ground, infront of him. Ease off on the throttle, and bring your crosshair to bear on the center of his mech. FINALLY a good hit.

Get my point in that long text-demonstration? By the way, Check the MRM bit, if you missed where the small cone comes into play. If your wondering why I did weapon by weapon... I personally dislike MW4's rendering of heat. It's too... soft, I guess, is the closest I can say. I'm surprised that it renders your mech as running at 0k (Also known as ABSOLUTE ZERO, the temperature at which ALL MOVEMENT CEASES) when Powered on, and Running!

And the 10k heat limit seems excessive too... I'd think 5k would be quite high, considering a fair bit of that temperature, According to the lore is dumped into the cockpit. Hell, by that bit of the lore, Alpha-striking with a laser boat that could core a medium in one shot would be akin to suicide.

But yeah... You get my points, hopefully.

[Liam]

In my opinion arms act as weapon platforms (torsos too), this means weapons aiming is not coupled with arm movement.
All weapons (except missiles with non-direct fire) would have twist and nick mechanics allowing better aiming while running.

Check this out and take care on the bore of the tank (only the first 20 seconds of this video):


However there should be some natural cone of fire for ballistic weapons (but at the end this would have negative effect in balancing case compared to energy weapons). Very low cone of fire can be accepted but again (+/- 0.5 at 1000 m wouldn't be much).
Because of the lag I would prefer to compensate the accuracy of the weapons (non cone of fire) by further hit zone segmentation, this will made weapons like LBX and LRM more interesting allowing to strip down a wide armor area of the enemy mech. And finishing it by Lasers, ACs and gauss rifles.
But well its up to developers.

If this game became a short range playground with RNG hit chance mechanic I won't play this game ... so simple is that.

[Liam]

Jump and shoot


EDIT: Ground impact indicator mechanic

btw to make everyone happy we could keep perfect accuracy and introduce step indicators (foot impact indicator).
So if a mech is running each step will be indicated (incoming step + step)
During leg impact pilot suffers in accuracy (RNG based cone of fire)
This skill could be improved in pilot tree allowing better run and gun ability.

So good pilots could watch out for non impact periods to shoot (increasing accuracy to ~95 %) while running/ walking
Shooting during impact could be improved by pilot skill ( 0/5 points, each point improves accuracy during impact to + 10 %)
Standing still: 100 % accuracy

I would like to see something similar in MWO

Edited by Liam, 15 January 2012 - 06:12 AM.

[djuice1701]

Cone/Dispersion + Aim time = Win.

Each weapon should have a set base accuracy/dispersion and aim time based on targetting computer/player's skills.

Here an example, Diverse Optics Sunbeam Large Laser has a base accuracy/dispersion 0.1m/100m (radii) with a aim time of about 1.5 seconds. So, if you are stationary (not moving) your maximum dispersion will be 0.6m @ 600m, you will hit whatever is within that 1.2m diameter cone. It will still allow for pinpoint aiming.

Now if you are moving, there should be a movement penalty modifier, let say @ 80kph there is a 70% accuracy penalty, so it's 0.17m/100m, this still allows for fairly accurate hits @ 400m. Now if you stop, there is a slight delay before it reachs maximum accuracy again due to targeting computer/acuators etc adjusting itself, that's where the aim time comes into play, this is prevent abuse of faster light mechs instantly stopping and gaining 100% accuracy firing and then instantly moving again.

Different weapons will have different accuracy/aim time modifiers, like pulse lasers been more accurate then normal lasers,

Edited by djuice1701, 15 January 2012 - 10:02 AM.

[Vulpesveritas]

I will say no to a cone of fire. again and again and again if need be. Use physics for it, not a random number generator.

a random number generator still takes away skill from the game, no matter how it is implemented.

[Hyzoran]

I think certain projectiles (like bullets) should curve slightly or alot into a ramdom direction based on the weapon's accuracy. and beams, depending on the beam weapon should become less and less focued as the distance increases. (Laser spreads out and does less damage farther away)

[UncleKulikov]

Vulpesveritas, on 11 January 2012 - 08:24 PM, said:

Yes but you have to remember light diffraction and that CryENGINE 3 has weather effects- lasers aren't much good in the rain. at all. Should they decide to have random weather effects you should take it into account. Oh and having actual real penalties like exploding for overheating.

I'm fine with weather affecting weapons. But it should be static alterations, not dynamic ones.

Quote

And it won't be random. merely give it on your sensors the airspeed and direction. then if you know how to compensate for it you'll do fine.

Wind direction can change between you and the target, and there is no way to check that without actually firing and seeing where it lands. And it would be random, for unless the wind is going the same direction the same every time that map is loaded, it will be randomly generated. And randomly generated wind can alter how effective some cover is and some isn't.

Basically the game isn't made better by adding wind interference, only more complicated.

Edited by UncleKulikov, 15 January 2012 - 04:05 PM.