Damage model - CBU-87 on Tents?

[sgibson]

I generated a cenario with some tents and dropped two CBU-87 on them. One or Two tents were damaged. Should a CBU-87 flattend them all????

[Witchking]

From what I read, the CBU 87 was more meant for anti-personnel while the CBU 97 was meant against unarmored/light armored and few armored vehicles. Its hard to say because apparently all objects in lomac have a "health bar" thingy. For aircraft, u can see the stages of damage...but ground objects completely lack this. So u could have damaged a lot of them...but not fully. We see only a "functional" and a "dead" state. Another aspect that needs attention for the next DCS major update.

[MeerCaT]

Don't forget that the 87's are completely 'dumb' bomblets that randomly scatter across the area.

 

As Witchking mentioned, objects have an amount of 'health' that must be depleted before it will be considered 'dead'. It is possible that it takes more than 1 CBU-87 bomblet to destroy a tent (though I doubt that would be the case for tents).

 

But either way, it is entirely plausible that a random scattering of bomblets would only score a direct hit on 1 or 2 tents within the area of effect. And if it really does require more than one direct hit, that would reduce the kill count even more.

 

As for the 97 variant, as far as I understand these use 'smart' skeets that search for infrared heat signatures of armoured vehicles. I wouldn't have thought a tent would necessarily show as a viable target.

[sgibson]

Thats all true ... but a CBU-87 is a cluster bomb and flattens every soft target beneath it which includes a tent.

 

I think it is simply not modelled right.

[MeerCaT]

flattens every soft target beneath it

 

Every soft target - not necessarily. Only the ones that actually get hit. :)

 

But no I agree, we would all love to see a site of total devestation emerge from the huge blanket of explosion and dust that these things inflict on an area.

 

 

Just thought of another good point: All of the CBUs contain a setting (accessed through the Inventory page) that determines how high above the ground they will 'shed their load'. ("HOF" - Height Of Function).

 

If you use the default setting without changing it then this is usually something like 1500 ft (...or metres?).

The higher the 'spreading' happens then the further apart each little bomblet is likely to be by the time they reach the ground, and if there is more space between each one then that leaves a much higher chance that one of your intended targets manages to 'slip through the net' and avoid getting hit directly.

 

If you don't already, try changing this setting to 500 or 700 for a much tigher grouping of your little day-ruining parcels.

[sgibson]

I know the HOF settings ... it doesn't change the fact that if I drop a cluster bomb over 6 tents and 4 of them are still there something is wrong.

 

Even if the spread is larger, such soft targets are ripped apart. If not directly from the bomb then from other ricochets.

 

I see these kind of strange modelling all over the place. Especially with bombs such as Mk-82s. Those things are dumb and most of the time you don't hit a target directly. But that is not necessary.

 

If I drop 3 Mk-82s 20ft around light- or non-armored targets they are gone! Not in DCS!

Edited September 29, 2012 by sgibson

[MeerCaT]

...such soft targets are ripped apart. If not directly from the bomb then from other ricochets.

 

Yes sure, I would love to see 'splash' damage be much more effective, esspecially against soft targets.

I get the impression the current damage modelling implementation is relatively lightweight, incorporating a simple Hit Points Vs Damage Dealt system; with Damage Dealt probably sharply decreasing with distance away from the epicentre of the area of effect.

 

The issue of modelling damage can become very complex very quickly.

It is entirely possible to concieve of a 'damage model' that is as detailed and involved as the flight model with its incredibly complex aerodynamics calculations.

 

The physics and mathematics invovled is immense the closer you attempt to model the real world processes and interactions between objects.

I think many strategy games over the years have come up with some elegantly simple ways of approximating the complexities involved in the concept of 'damage'.

 

They often divide the concepts into a few basic categories, such as:

 

• sharp
  
• blunt
  
• explosive
  
• shock wave
  
• heat
  
• (radioactive!)
  
• etc.
  

Then each 'object' in the virtual world has an 'armour' value for each category representing its ability to resist such damage, while each 'weapon' has a value for each category representing its ability to deal damage of that type.

 

Then apply algorithms to calculate the NET effect of the coming together of these two entities to determine how much underlying damage is done to the target object, whilst also factoring in a sliding scale of effectiveness to incorporate distance from 'epicentre'.

 

There is almost no limit to how deep and complex you could take this modelling if you wish. It is just a reflection of how truely complex the real world we live in is.

 

The daring developers of a 'simulator' have to prioritise the many real world aspects they wish to incorporate into their virtual world and what level of detail they wish to take each aspect to.

For a military flight sim what kind of priorities are reasonable?

 

1. Flight model (aerodynamics etc)
   
2. Aviation systems (cockpit bells and whisltes)
   
3. Weapon systems (target tracking, ballistics modelling)
   
4. Artificial Intelligence (friendly and enemy activities of many different units)
   
5. Weather system
   
6. Damage modelling
   
7. 3D object modelling ('smoothness' and realism of object shapes)
   
8. Visuals (ground and object textures, lighting effects)
   
9. Audio (sound samples, sound effects)
   
10. Networking (multiplayer)
    
11. ...the list is endless
    

 

Just a few thoughts for the day :)

[badger66]

Those tents shoulda been wiped out , end of .

[marcos]

You probably need to lower the HOF and RPM but the damage modelling in DCS is flawed. Land a Mk-82 5 yards from a parked fighter and no damage is visible.