Tanks - tracked vehicles driving and handling

[AveryTheKitty 2626]

Is it just me or has the precise-ness of the new driving system worsened since it was first implemented? At first it was super accurate and the AI stopped immediately when you stopped pressing forward.

 

Now they'll sometimes continue going forward and just seem more sluggish than normal.

[OMAC 254]

I think the new WASD driving system for commander was pushed to stable branch too fast.  It seems clear that the change was implemented to mitigate AI driving issues such as AI driver not backing up when ordered if obstacles are nearby.  IMO, AI driving issues should be fixed or otherwise mitigated, and new WASD commander driving change reverted.  I was fine with occasionally switching to driver position to get out of a tight spot.  I see no real advantage of the new system. 

 

This change caught me by surprise since I don't use dev branch:

 

 

Quote

There should be a penalty regarding fluidity and responsiveness of movement if you are commander.  This penalty was reasonably handled before.  The commander has many things to do other than drive the vehicle, and he/she should concentrate on those.  For example, commander needs to press another key to laze targets for his MG or GMG, and having to hold down W/S keys for movement makes lazing more difficult.

 

Issues on stable branch for all vehicles, not just tanks:

• No verbal STOP command by commander when W and S keys are released. 
• Vehicle does not stop immediately after W/S keys are released.  Vehicle sometimes just keeps moving, in which case key for opposite direction can be pressed, which results in abrupt stop, but voice command doesn't say STOP, it says to move in opposite direction.  There should be way to order complete STOP.  Note that Select subordinate>STOP Action Menu command does not work to stop vehicle when movement keys are pressed by commander (this is expected).
• No verbal commands for Fast or Slow Drive.
• No verbal STOP command by commander when handbrake key is pressed.
• Handbrake key shouldn't work at all for commander.
• Commander can't easily laze target while tank is moving (needs to release movement key or mouse to hit laze key).

 

Commander needs hands-free vehicle movement if he switches to gunner position and needs to control/fire cannon while tank is moving.   As commander doesn't actually have control of vehicle, all of his commands must be voice.   This new commander driving system is very arcadish given that commander does not have actual control of vehicle steering, accelerator, and brake. 

 

[teabagginpeople 398]

1 hour ago, OMAC said:

 

 

 

• Commander can't easily laze target while tank is moving (needs to release movement key or mouse to hit laze key).

 

 

 

Regards this issue omac. In the control settings for laze,You might just have to set up an additional bind w and what ever key you use for laze.  

 

Mine for example looks like this , lazer target- end, up arrow + end.

[OMAC 254]

I'm thinking of setting a mouse-4 button on my mouse to laze, so I can use the side of my thumb and thus not have to let up on the W or S keys, or my main mouse button.

 

My current laze button is End.

[SuicideKing 233]

The Kamysh and Tigris have very sensitive handling now, with 1.78 stable. They oversteer and stuff. Haven't tried the "Cat" APCs yet, but the wheeled vics and tanks handle fine.

[teabagginpeople 398]

3 hours ago, OMAC said:

I'm thinking of setting a mouse-4 button on my mouse to laze, so I can use the side of my thumb and thus not have to let up on the W or S keys, or my main mouse button.

 

My current laze button is End.

Iirc Might still conflict with the laze. 

 

If it is as I'm saying. The reason it does not laze sometimes is because you are holding down the w. Then when you press end it's sometimes reading as w+end. 

 

The same might happen with your mouse button 4.  if so just put as laze - mouse button 4,  w + mouse button 4. This covers all bases.

 

I came across this issue because I use end in the A10 ccip for lock. It sometimes didn't lock and I realised it was because I was holding airbrake while diving or trust at the time. So did above issue dissappeard.

 

Anyway hope it helps.

 

[lex__1 422]

It is necessary to put in order the system of commands for AI and the pressing of control keys of the vehicle, in accordance with the current state of the vehicle.
Under the spoiler - Keys and Commands, for different states of the vehicle.

Spoiler

All the commands W \ S \ A \ D sound right for a stationary vehicle.
The reason is to change the sound commands for the driver AI and their sequence for the vehicle in motion:

 - When the vehicle continuously moves forward / backward, commands to the left / right stop the vehicle after adjusting the movement left / right. This makes it necessary, after correcting the rotation, to repeat the command - "fast / slow forward"
 - When the vehicle continuously moves forward / backward, the back / forward commands stop the vehicle, but the AI command does not sound like "Stop", it sounds like backward / forward. It is distracting, it seems that you are giving a wrong command for AI.

https://feedback.bistudio.com/T127310

[OMAC 254]

4 hours ago, teabagginpeople said:

Iirc Might still conflict with the laze. 

 

If it is as I'm saying. The reason it does not laze sometimes is because you are holding down the w. Then when you press end it's sometimes reading as w+end. 

 

The same might happen with your mouse button 4.  if so just put as laze - mouse button 4,  w + mouse button 4. This covers all bases.

 

I came across this issue because I use end in the A10 ccip for lock. It sometimes didn't lock and I realised it was because I was holding airbrake while diving or trust at the time. So did above issue dissappeard.

 

Anyway hope it helps.

 

 

Oh, I see.  I'll try that.  Thanks.

 

----

 

Another strange thing with 1.78 stable:

 

Keybinds:

 

Car fast forward -> E, Lshift+W

Car slow forward -> Q, Lctrl+W

 

I get a verbal command "Fast" or "Slow" if I hit either of the two binds including W, but not when I hit just E or Q.  E and Q still work as normal, however.  If I set the binds as W+E or W+Q, nothing happens, and I don't think speed changes.  Odd.

[andersson 285]

Keybindings overall is very odd in this game. I have found quite a few examples of hardbound keys that mess up user bindings. Mostly in the editor though.

[reyhard 2082]

On 12/1/2017 at 12:27 PM, x3kj said:

If we could get details on what was unfavorable on dirt humps LOD that would be a help for us modders, to know what should be avoided on our terrain models.

There is one odd thing between editor placed & map objects after PhysX update - first one is using Geometry Physx & Roadway LOD for PhysX collisions were second one is using just Geo PhysX LOD. As you probably guess at this point, baking a PhysX LOD from Roadway LOD might not be always precise (especially with complex structures above 1,5k faces) and generate plenty of overlapping spots between two PhysX layers. Disabling of PhysX collisions with Roadway LOD is not an option since otherwise you would get crazy launches of soldiers if they i.e. get incapacitated ragdoll state - after ragdoll ends they would be pushed away to the top of roadway lod (or sucked into the object). Not sure when Roadway LOD baking into PhysX scene will be enabled once again, but it's better to assume that it might be fixed for i.e. next version.

Anyway, got a little bit far away from original question - in order to fix those Roadway quirks, I've tried to simplify Roadway LOD below 1k faces & pushed it a little bit above actual Geo Physx LOD. Seems to work in most cases

[Beagle 684]

On the other hand, he new tank commande rhandling system now finylly allows the full use of a HOTAS and Pedal system for all movement and targeting input. ArmA II Tankign was nevr so relaxing before with a HOATS+Pedal setup. I use Joystick fpr aiming, buttons for lasing and weapon switch and the pedals for throttle and brake and z axis for movement now. Viewmode and lase on the left hand, Waeapon control and triggr on the right hand, movement with feet...just perfekt like a snug suit.

 

fun fact...the same way steeringn a tank could be done in DID "Wargasm", a spiritual predecessor of the OFP/ArmA series.

 

A return to the old system after you adapt is like cutting a hand and a foot off.

[Grumpy Old Man 3545]

9 hours ago, Beagle said:

On the other hand, he new tank commande rhandling system now finylly allows the full use of a HOTAS and Pedal system for all movement and targeting input. ArmA II Tankign was nevr so relaxing before with a HOATS+Pedal setup. I use Joystick fpr aiming, buttons for lasing and weapon switch and the pedals for throttle and brake and z axis for movement now. Viewmode and lase on the left hand, Waeapon control and triggr on the right hand, movement with feet...just perfekt like a snug suit.

 

fun fact...the same way steeringn a tank could be done in DID "Wargasm", a spiritual predecessor of the OFP/ArmA series.

 

A return to the old system after you adapt is liek ecutting a hand and a feet off.

Great, now buy a new keyboard.

 

Cheers

[Beagle 684]

20 minutes ago, Grumpy Old Man said:

Great, now buy a new keyboard.

 

Cheers

I had to use the cheap one in the office. The one with all the coffee and crumb inside and missing key prints

[RforRush 6]

Direct control over the vehicle should also be applied to Qilin and Prowler LSVs.

[darkChozo 133]

Pretty sure it is. Or at least it was when I played through the TacOps content, because I was commanding a Qilin from the gunner's seat and it was driving normally.

[RforRush 6]

I checked on 1.80.143869 and Prowler works, but Qilin doesn't.
 

[x3kj 1247]

Quote

Tweaked: Friction properties of surfaces were adjusted (WIP) 

that is indeed a happy new years start

[R3vo 2654]

6 hours ago, x3kj said:

that is indeed a happy new years start

What's that for?

[xxgetbuck123 945]

4 hours ago, R3vo said:

What's that for?

 

My first guess was that it maybe makes surfaces actually behave more like the actual surface you're trying to portray, so a dirt road might be a little more slippery due to it being loose rocks compared to a sand surface which, well, would behave like if you drove through sand. 

 

But I have no idea really haha

[oukej 2910]

At this point we couldn't afford really any drastic changes. But there should now be a bigger difference between tarmac, dirt roads, grassy fields or beach sand. And the frictions should be slightly more consistent across similar surfaces on vanilla terrains.
Also when two objects collide it should be slightly easier to slide along each other (e.g. sliding along guiderails) at a low angle or bounce off instead of stopping on a spot.

Further tweaks are possible and the feedback is welcome - be it about the general handling feel (sliding, drifting, braking distance, understeer, loss/gain of grip) or directly bugs and issues (boxes and ragdolls suddenly sliding down the hills....)

[en3x 209]

There 2 notable troubles that I experience. 

 

Stopping at low speeds. (makes tanks impossible to move with infantry with M+Keyboard)

 

And spinning on the spot on any angled slope makes you move forward or backward without your input. (so staying still and turning right without any throttle on the slope would move tank backwards down the slope)

[xxgetbuck123 945]

22 hours ago, oukej said:

At this point we couldn't afford really any drastic changes. But there should now be a bigger difference between tarmac, dirt roads, grassy fields or beach sand. And the frictions should be slightly more consistent across similar surfaces on vanilla terrains.
Also when two objects collide it should be slightly easier to slide along each other (e.g. sliding along guiderails) at a low angle or bounce off instead of stopping on a spot.

Further tweaks are possible and the feedback is welcome - be it about the general handling feel (sliding, drifting, braking distance, understeer, loss/gain of grip) or directly bugs and issues (boxes and ragdolls suddenly sliding down the hills....)

 

From my quick but exciting testing the improvements are great! Feel a lot more natural, especially with tanks, as it feels as though the weight carrys through a turn rather than just turning in a nice circle. 

 

One question though, do these surfaces impact acceleration? I.e hard surface vs soft surface acceleration?

[oukej 2910]

On 1/4/2018 at 5:18 PM, en3x said:

Stopping at low speeds. (makes tanks impossible to move with infantry with M+Keyboard)

The "autobrake" can be annoying, but atm we can't remove it in order to prevent vehicles from moving on their own (vehs atm are always "in gear"). You have to ask your dismounts to sweat a bit and run. And then use the limited speed (Ctrl+W) :/

 

On 1/4/2018 at 5:18 PM, en3x said:

And spinning on the spot on any angled slope makes you move forward or backward without your input. (so staying still and turning right without any throttle on the slope would move tank backwards down the slope)

This one is a bit tricky, the differentials don't seem to simulated well enough so once you release brakes one of the tracks will go loose. I'm afraid we won't be able to solve that.

 

6 hours ago, xxgetbuck123 said:

One question though, do these surfaces impact acceleration? I.e hard surface vs soft surface acceleration?

Yes, it can - depending on the wheel / track slip.

[dragon01 902]

How about using the "combat pace" button as the special "toggle speed limiter" or something like this? With KB+M, it could function as a very low limiter for precise movement with infantry (the speed could be normalized across vehicles to make things even easier), and with an analog controller, it could lock the vehicle in the first gear, allowing finer speed control. Essentially combat pace for vehicles.

[en3x 209]

3 hours ago, oukej said:

The "autobrake" can be annoying, but atm we can't remove it in order to prevent vehicles from moving on their own (vehs atm are always "in gear"). You have ask your dismounts to sweat a bit and run. Then use the limited speed :/

 

This one is a bit tricky, the differentials don't seem to simulated well enough so once you release brakes one of the tracks will go loose. I'm afraid we won't be able to solve that.

 

Yes, it can - depending on the wheel / track slip.

 

Yeah I get around autobreak by using controller with analog triggers. Is not perfect (pc users without peripher.) but it works.

Thanks for elaborating on the issues as to why they are there.