Improvements for vehicles

[Erdem]

I've brought this up before in LO forums, but I think this is time to discuss it again.

 

The problem is, since we're closing down to the ground level and vehicles becoming our main focus, we need to think about some improvements over them. Of course, this could be in form of new skins, models and movement improvements and they are being done as I've seen. But for a dedicated ground-pounder sim, this is not enough in my opinion.

 

The vehicles in LO have very simple damage models. They have hitpoints, and some properties that allow them to take damage from different weapons, or not get affected at all. When you do enough damage to them with your weapons, their hitpoints run out and they just pop up. Now, this is an ideal modelling for a sim like Lock-On. But for Blackshark, dealing with vehicles may get boring quickly if no improvements made for vehicle damage modelling.

 

What I'm trying to say is, make some improvements on vehicles. Let us be able to cripple them with machine gun fire, destroy their tracks, disable their engines, overkill(well, vaporise :) ) them with bombs and rockets, pop their turret up with our Vikhr's, disable their weapons and so on. Of course this will need some different damage models for all these situations and require extensive work, but it doesn't have to be done for all vehicles. It'd be logical if this was applied to our very common enemies like M1A1, T-80, Shilka, Avenger and trucks from both sides in initial BS version. Rest could be updated later with other add-ons.

 

Other things that bog my mind is, how the vehicle AI behave in combat. In current LO it doesn't look bright, they don't have any tactical behaviour at all. They still follow their waypoints senselessly. Blackshark do need some spice in this area as well. Here are some examples for what can be done:

• First, if they cannot fight back yet, SAM's and AAA's should disband from the convoy and seek protection and a nice firing position in the surroundings.
  
• Tanks, trucks and other transport should speed up to make targeting harder for the pilot.
  
• Convoys should leave greater spaces between the vehicles to minimize the damage taken.
  
• Tanks should seek cover, change position in cities or in forests under fire.
  
• AAA's and SAM's assigned to defend certain areas should change positions rapidly to make targeting harder.
  

 

This kind of vehicle AI will make the mission replayable and much more fun and challenging to deal with. It'll also make the game much more interactive with the player. I know I may be "dreaming", but this is what I've expected from BS from the first minute. ED may have other plans, but if this was done, BS will be one of the best ground-pounder sim ever.

 

Please give your feedback :)

[Vekkinho]

I see Your point Erdem, however don't you think that destroying Shilka ASAP is much better approach than slowly crippling it just to see panels falling off while hovering inside the range of it's weapons?!

[Malleus]

The AI improvements has been mentioned/suggested a few times already, and I agree with those, but I think the damage model doesn't need that much detail. You wouldn't engage these vehicles with guns IRL anyway, and it would take up too much development resources that could be used for more important things (like the AI).

[Erdem]

I see Your point Erdem, however don't you think that destroying Shilka ASAP is much better approach than slowly crippling it just to see panels falling off while hovering inside the range of it's weapons?!

 

That's not the point. It's not a crucial part of the game but, a leap forward for Blackshark.

[britgliderpilot]

It'd be absolutely fantastic, no argument there . . . but you have to apply some realistic development time constraints here ;)

 

 

I don't know a first-person-shooter that has that kind of damage model for armoured vehicles. Hell, does Steel Beasts even have that good a damage model?

 

Justifying the time and cost of that level of input is very, very difficult.

 

Just doing the 3D models to that level of damage detail is challenging. AFAIK even the brand-spanking-new models don't have damage models or parts to fall off, so they'd all have to be reconstructed . . . to say nothing of the code to allow all that to happen.

Imagine the poly counts!

 

ED don't have infinite resources, and it makes more sense to focus them on . . . well . . . the flying bits of the flightsim . . .

 

 

Some AI improvement is more likely, but again, very time-consuming. And as a tester, it's just as time-consuming to figure out when they're doing it right and when they're doing it wrong . . .

 

 

Black Shark is already going to be far and away the most detailed combat helicopter flight sim there's ever been - perhaps you could even dispense with the "combat" tag - but there's got to be a limit to how good it can be made!

 

It's probably as good a time as any to remind everyone that pure system simulation doesn't automatically mean great immersion - those who've flown Flanker 2 and Falcon 4 should know this.

The environments you're flying in need to be well-populated, changing, and filled with AI who respond in both a challenging and believable manner - how well they fall apart is of little import compared to that.

 

If your Shilka is going to be destroyed by a single hit anyway, it's more important that it's placing and behaviour is appropriate than bits fall off it in a realistic manner. That's down to mission/campaign building (and to some extent a powerful enough editor), AI behaviour, and unit performance.

 

To that end, the ME's completely new and AI behaviour has been improved - but until progress on those has been completed there's no indication as to how much better it'll be.

You'll have to have faith that ED will get it properly sorted before release - and now they're self-publishing it's in their best interests to do just that.

 

 

On a different note - if you're in a position to only damage the subsystems of a ground unit, you can bet your bottom dollar it's in a position to do some SERIOUS damage to you ;)

 

At which point it's the damage model of your aircraft you need to worry about, rather than the damage on the ground unit . . . .

[JaNk0]

Maybe a task for the sim-moders?

[brewber19]

BGP, I just need to say...damn I'm glad there's a real ground-pounder as a beta-tester.

 

Couldn't agree more with your post and I feel alot happier about ME and AI behaviour with input like yours :D

 

Rep inbound. (edit: stupid forum won't let me :()

 

It'd be absolutely fantastic, no argument there . . . but you have to apply some realistic development time constraints here ;)

 

 

I don't know a first-person-shooter that has that kind of damage model for armoured vehicles. Hell, does Steel Beasts even have that good a damage model?

 

Justifying the time and cost of that level of input is very, very difficult.

 

Just doing the 3D models to that level of damage detail is challenging. AFAIK even the brand-spanking-new models don't have damage models or parts to fall off, so they'd all have to be reconstructed . . . to say nothing of the code to allow all that to happen.

Imagine the poly counts!

 

ED don't have infinite resources, and it makes more sense to focus them on . . . well . . . the flying bits of the flightsim . . .

 

 

Some AI improvement is more likely, but again, very time-consuming. And as a tester, it's just as time-consuming to figure out when they're doing it right and when they're doing it wrong . . .

 

 

Black Shark is already going to be far and away the most detailed combat helicopter flight sim there's ever been - perhaps you could even dispense with the "combat" tag - but there's got to be a limit to how good it can be made!

 

It's probably as good a time as any to remind everyone that pure system simulation doesn't automatically mean great immersion - those who've flown Flanker 2 and Falcon 4 should know this.

The environments you're flying in need to be well-populated, changing, and filled with AI who respond in both a challenging and believable manner - how well they fall apart is of little import compared to that.

 

If your Shilka is going to be destroyed by a single hit anyway, it's more important that it's placing and behaviour is appropriate than bits fall off it in a realistic manner. That's down to mission/campaign building (and to some extent a powerful enough editor), AI behaviour, and unit performance.

 

To that end, the ME's completely new and AI behaviour has been improved - but until progress on those has been completed there's no indication as to how much better it'll be.

You'll have to have faith that ED will get it properly sorted before release - and now they're self-publishing it's in their best interests to do just that.

 

 

On a different note - if you're in a position to only damage the subsystems of a ground unit, you can bet your bottom dollar it's in a position to do some SERIOUS damage to you ;)

 

At which point it's the damage model of your aircraft you need to worry about, rather than the damage on the ground unit . . . .

[britgliderpilot]

BGP, I just need to say...damn I'm glad there's a real ground-pounder as a beta-tester.

 

Couldn't agree more with your post and I feel alot happier about ME and AI behaviour with input like yours :D

 

Rep inbound. (edit: stupid forum won't let me :()

 

 

No need to single me out - there's plenty of seasoned simmers on the beta team who can and do suggest some very desirable features and changes :)

 

Of course some changes make it, some don't . . . but generally things work out pretty well!

[Shaman]

Two diffrent wreck models would be cool per each vehicle, at least.

 

First one simmilar to the one we remember from Lock On.

 

Second... so in case we do this "overkill" we just see some tracks, and scattered twisted steel bars.

 

Oh, I forgot to add, the we should be able to hit the 1st wreck model again, and then see 2nd..totalled model. Should be easy solution, would save CPU cycles (no super complex damage modeling for each vehicles), and development time is shorter here.

[Yeniceri]

I am try'n to make a new damage model to hog as you all know..

So i learned a thing or to about damage models..

There are two ways a damage model mechanic works..at least in lockon

 

1- You can contain a damage model IN the original model.

 

When you do that lets say the Hog's model file loads up with the damage model poly's loads with it. When you fly undamaged you already fly with the damage model also. It is just not visible. When you get hit from the engine, the undamaged engine model becomes invisible and damaged model becomes visible. So.. The more detailed damage means, more poly, less fps.

 

2- You can contain the damage model Externally.

 

If we move from the same example, when you get shot from the engine, the undamaged engine becomes invisible and game engine loads the damaged model into the scene. You wont carry the damage model with you in normal flight, but when you get hit the system should instantly load the damage part in. Now as game players we all know what that means during the game play.

 

After all these explained. Imagine that you and your wingman are fly'n low near a city. There are buildings and a 10 vehicle convoy. You fired missiles, they fired missiles, buildings got hit, you got hit, wingman and conwoy got hit. Now all these damage model polys started to fly around. Plus there is flight dynamics, eyecandys and countless little details that should be calculated and processed (damage smokes, flames, etc.)

 

These should be well balanced in game. When you try to reach the perfect game, if you lose focus, you can end up having an unplayable one..

 

This is what i think..I think a heli sim should have priorities. And i think a killed vehicle's damage model should not be at the upper lines of the list..

 

But thats just me :)

 

Cheers

Yeniceri

[MBot]

I kind of agree that detailed vehicle damage is not a primary concern for the moment. Perhaps two different destroyed model for a mobility kill or total destruction would be nice for the start.

 

What is much more important first is how the vehicles will react in combat, since those vehicles will be the primary way how the player will interact with his environment in a helicopter sim. A realistic visual scan and detection model, surprise effect, reaction times, able to get suppressed, realistic engagement routines, ability to search cover etc. This is all stuff that will have an immediate impact on the player and reward to use of realistic tactics.

[JEFX]

I kind of agree that detailed vehicle damage is not a primary concern for the moment. Perhaps two different destroyed model for a mobility kill or total destruction would be nice for the start.

 

What is much more important first is how the vehicles will react in combat, since those vehicles will be the primary way how the player will interact with his environment in a helicopter sim. A realistic visual scan and detection model, surprise effect, reaction times, able to get suppressed, realistic engagement routines, ability to search cover etc. This is all stuff that will have an immediate impact on the player and reward to use of realistic tactics.

 

I cannot agree more!!!

 

JEFX

[Avimimus]

Just doing the 3D models to that level of damage detail is challenging. AFAIK even the brand-spanking-new models don't have damage models or parts to fall off, so they'd all have to be reconstructed . . . to say nothing of the code to allow all that to happen. Imagine the poly counts!

 

Now this isn't entirely true. One could, for instance,:

- have smoke pour out from where the engine is located indicating engine damage

- have the cargo on a truck catch fire (with the driver ignoring it and keeping moving)

- have the crew abandon the vehicle (with smoke coming out of the open hatch)

- have the turret jam at a specific angle and be unable to move

- depress part of a wheeled vehicle so that the lower section of a wheel is below the terrain (indicating a blown tire)

- spawn a 3d model for a length of tread behind one of the tracks (the track would still have a full tread on it - but who would really notice? In combat you would barely have time to note the length of tread behind the vehicle and figure out that it is immobile).

 

So you could do all of the things listed here without modifying the 3d models.

 

If ED wanted to get fancy in a few years they could write some code so that damaged parts of vehicles will have their textures modified (indicating abraising to armour, puncture holes, loss of ERA etc.)

[Witchking]

I would personally love to have an "improved damage model". By this I don't mean placement specific damage: turret of tank flying away or something like that. I just mean some random damage. the moment it gets it.... flip it to a damage model which has a canned animation) of a few POLYGONS flying away (the ammo bunker explosion in FC). Then, just make the fire and smoke realistic as in IL2 or any other game out there and I will be very very satisfied. :) These small effects give the "TAKE THAT"... "OOHH YEAHHH!!!" kind of feeling when u are evading sams and flying low to take out a target. ;)

[colubridae]

Definitely would be enjoyable to see a tank brewing-up.

 

please