Chevrolet Camaro Z/28 1969

[Cosmo°]

This is a re-release of Cube's '69 Z/28 that has been updated and quite heavily revised for current Racer beta versions, namely v0.8.33.

Since the original release more than four years ago, I've been working on it continously, but several significant changes in Racer's way of doing things delayed it's second showing.

The mesh, textures and sounds are largely untouched, although I did include minor alterations to accomodate shader changes and car.ini requirements. As such, some of the old shortcomings are still present - there's no driver model, the interior mirrors aren't really funcional and the lack of a separate interior glass material means the activated headlight texture is invisible from inside.

For me, this doesn't have a negative impact on the fun of driving it though, which is why I decided to put it up here for others to experience it once more.

Power delivery is quite peaky on this engine, but the steering rack ratio and period style tyre behavior encourage a smooth driving style with a good bit of throttle action to keep the nose pointing in the right direction.

The car.ini file contains two differential ratios which you can manually switch between if desired. The default setting is for the standard 3.73 axle, while a factory option 4.10 model can alternatively be selected to spice up the smoke production.

edit: New link with slightly updated archive added January 2013 - thanks a lot to Dave for hosting it!



Feedback is appreciated.





Click here to download

 

[KS95]

Very nice Cosmo, will test when I have time

 

[hierafas]

Thank you Cosmo! A few years ago this was one of my faves both in terms of looks and feel. I haven't touched it for a while but thanks to your update I'll take it around more often. No complaints from me, apart from ...that shadow texture, which is a little bit too narrow. But that is rather a matter of taste I suppose. Good one!

 

[Terden]

Wow, what a great car!!
Only thing is the speedo; when you brake it drops way below the speed you're driving.
When the brake is released it goes back up.
A bit strange but it doesn't hamper the driving experience.
Great job, thank you!

 

[Cosmo°]

Only thing is the speedo; when you brake it drops way below the speed you're driving.

I think that would be the exterior speedometer using wheel velocity instead of world coordinate velocity like the interior dial. You can change that in views.ini by adding the red part:

views
{
  	view0
  	{
    		elt1
    		{
...
      			var=velocity[COLOR="#ff0000"]_wc[/COLOR]
...

Thanks for the comments so far



hierafas: check your PM

 

[KS95]

I have to say, it really looks fantastic - hard to believe it's the same model And the physics are great, although the steering is a little hard on the mouse, but I can change that to my liking.

Great work dude, so glad to see some great oldies being updated

 

[Raphael Buthmann]

Awesome ^^ I can't help with sound now. My HD had broken and I lost EVERYTHING inside him. I wanna die. lol

 

[Splashonda]

Take it to a computer repair shop! If they swap the data plates (and if the problem isn't the plates, of course) to another HD, maybe the data is recoverable!

 

[Raphael Buthmann]

I can't do this lol . I don't know where here in Rio we have this kind of service . Anyway , it's better, My pc was with so many problems and now I'm without some **** on the new lol . I'll try to do all the thing I had better , a little motivation .

 

[boomer541]

Get an external hard drive for backups they are really cheap now.

 

[Raphael Buthmann]

Thanks boomer .
But getting back to the topic lol , this camaro is AWESOME.

 

[QuadCoreMax]

Great Cosmo & all who participated !

Here's 4 fixes done in seconds since I know where to mod ini files, so RC is same for front & back, steering lock reduced & car proxy collision mesh added & staller deactivated since I like them on by default.

Feedback :

After quick/short testing, when clutching fully engaged (G25) + throttling to max, you can hear the sound vanishing after a certain time (2s). Mirrors & some better textures (UV scaling/packing etc...) could be done apart from some more texture maps (Normal/Detail maps).

Other than that, +1, I love it !

body
{
    [B]model_collide
    {
        file=body.dof
    }[/B]
}

steer
{
      x=0.354
      y=0.065
      z=0.225
      radius=10
      xa=24.5
      [B]lock=240[/B]
      linearity=0.862
      model
      {
            file=s_wheel.dof
      }
}
engine
{
[B]       starter=0
      [/B]starter_torque=140[B]
      start_stalled=0
      enable_stall=0[/B]
}
susp_front
{
      y=0.000
      z=1.4865
    ; static ride height    0.430m
    ; bump travel        0.075m
    ; rebound travel    0.115m
      restlen=0.590
      minlen=0.355
      maxlen=0.545
    warplen=0.430
    ; sprung corner mass    405kg
    ; spring rate        380lb/in
    ; wheel rate        135lb/in
    ; critical damping    6184N/m/s
      k=23607
    damper_curve=suspension_damper_front.crv
    bumpstop_len=0.050
    bumpstop_k=100000
    bumpstop_damping=5000
    reboundstop_len=0.010
    reboundstop_k=100000
    reboundstop_damping=5000
    anti_pitch=0.25
      roll_center
      {
            x=0
            ; -0.067m
[B]                y=-0.464[/B]
            z=0
      }
}

 

[Cosmo°]

Thanks for the comment QuadCoreMax.

The engine has no rev limiter, so when you keep the throttle down I guess there could be a bang sound playing for when something decides it had enough torturing - instead it just fades to silence right now, as you noted. It can deal with 7500rpm in stock form, but anything more than that is asking for trouble. The fade out happens beyond 9000rpm, so actually there's room left for overrevving on aggressive downshifts, just not for no-load-violence

I didn't use the body mesh for the collision model because it produces qlog entries as it's not a purely convex shape and because it's not a very efficient method (waste of polygons).

Not sure why you would want to change the roll centre heights - they should be pretty close (within a few mm) to the static situation in real life already.

The steer.lock value is fixed per car and I would suggest not to change that. For visual reasons on the one hand, but mostly because it has an influence on the force feedback effects too.
If you use a G25 you should really be in 900° mode anyway (and have that setting in your controller .ini file, as well) and let Racer do the linearity scaling automatically (car.ini steer.linearity is ignored nowadays, I just left it for backwards compatibility and to keep qlog happy).
If you always use 240° mode, probably best to change your controller file lock setting to 240 and leave car.ini steer.lock as it was meant to be - otherwise it really messes up each vehicle in a different way and you can never really compare your experiences to others'.

 

[QuadCoreMax]

Thanks for the comment QuadCoreMax.

The engine has no rev limiter, so when you keep the throttle down I guess there could be a bang sound playing for when something decides it had enough torturing - instead it just fades to silence right now, as you noted. It can deal with 7500rpm in stock form, but anything more than that is asking for trouble. The fade out happens beyond 9000rpm, so actually there's room left for overrevving on aggressive downshifts, just not for no-load-violence

Oh I understand...

I didn't use the body mesh for the collision model because it produces qlog entries as it's not a purely convex shape and because it's not a very efficient method (waste of polygons).

I didn't check the Qlog, but from what I know in game productions flows, there should be always a proxy collision mesh no matter what. I debugged your car with 'show carpoints' & 'show carbbox', & I instantly done the tweaking, I'm sure some here will appreciate.

Not sure why you would want to change the roll centre heights - they should be pretty close (within a few mm) to the static situation in real life already.

Same story here, + from Ruud/Mitch advices, you shall NEVER put your RC (front or back) outside your carbbox (collision mesh) & that's exactly what KS felt when playing with your car. So I 'equalized' both heights.

The steer.lock value is fixed per car and I would suggest not to change that. For visual reasons on the one hand, but mostly because it has an influence on the force feedback effects too.
If you use a G25 you should really be in 900° mode anyway (and have that setting in your controller .ini file, as well) and let Racer do the linearity scaling automatically (car.ini steer.linearity is ignored nowadays, I just left it for backwards compatibility and to keep qlog happy).
If you always use 240° mode, probably best to change your controller file lock setting to 240 and leave car.ini steer.lock as it was meant to be - otherwise it really messes up each vehicle in a different way and you can never really compare your experiences to others'.

The fact here, especially on G25, is its 900° rotation, I tend to have less, so yeah somehow it's get more precise.
Sure I forgot to mention, I lowered the ff_factors, so now it feels better...just try it out, you'll see !

Anyways, I admit there's a lot of ways of getting the same results...or better said, an infinity of different behaviors

 

[Cosmo°]

No problem at all, I'm just responding to the changes you mentioned because they seem to alter the experience significantly and not necessarily in a realistic manner in some points. I'm always open for a discussion on the how's and what's behind the releases I'm involved with though

I didn't check the Qlog, but from what I know in game productions flows, there should be always a proxy collision mesh no matter what.

There is a collision mesh, it's the plain box shape that I trimmed longitudinally so the vehicle doesn't "bottom out" in odd places. I've been using the body mesh before, when in a pinch, but that's not a clean way to do it so the car wasn't released that way. Like I said, firstly it creates a qlog entry because of the complex geometry, secondly it's just not optimized at all. Not a big deal really, just saying that it wasn't an oversight

Same story here, + from Ruud/Mitch advices, you shall NEVER put your RC (front or back) outside your carbbox (collision mesh) & that's exactly what KS felt when playing with your car. So I 'equalized' both heights.

That would be news to me, pretty strange news to be honest. As far as I know, there's no connection between the collision model and the roll centre location. Got a quotation from Ruud or Mitch on that maybe? Certainly I'm not the only who would like to know if there was such an oddity around

By relocating the roll centre you directly influence the handling balance. In this case, where you moved the front up by about 0.4m, it basically increases understeer tendencies in a big way, which means we're not driving the same car anymore.

I don't know exactly what KS95 meant when he said steering was hard with the mouse, but I'll assume that the vehicle will feel a bit unresponsive if his controller file doesn't have a fairly low maximum lock setting. He'd end up having to move the mouse quite far to get reactions and then again to catch slides etc. Roll centre location is not very high on the list for possible reasons or fixes to that - it's likely more a matter of input/output ratio, ie linearity.



The thing about steer.lock and your controller lock is quite important in my opinion:

steer.lock is not meant to be adjusted by the end user, because it's a fixed setting for a particular vehicle (like body.mass or wheel.radius, for instance). It determines the visual motion range of the steering wheel mesh, but for whatever odd reason, it also acts as a factor in the force feedback calcuations. This means that if you change steer.lock, what you feel through the FFB of your controller wheel will be different from what everybody else on the same controller will feel. If you lower it far enough, nasty oscillations will occur around the centre spot, thus motivating people to lower the ff_factor to compensate. Of course, that moves your FFB experience even further away from what was originally intended.

Now, controller lock is a matter of taste. If you like low lock angles for a more pronounced response, that's fine. Personally, I try to stay as close to a 1:1 ratio between controller and virtual steering rack as possible.
As such, my wheel is set to 900° mode and Racer automatically cuts the input on vehicles where steer.lock value is <=900° for a 1:1 ratio, and it calculates and applies a linearity value smaller than 1.0 for steer.lock values >900° (ie 900°/1080°=0.833). In the old days, we had to do this calculation manually for each vehicle and controller combination, so there was a lot of confusion and frustration about it. Nowadays this is done correctly behind the scenes already.

Racer assumes your actual controller lock limit is the angle specified in your controller .ini file. For example, the presets for the G25/27 are already set at 900°. Hence, if you always want a 240° limit in your case, you'd set this accordingly and Racer does the rest for you - it applies the correct linearity and leaves the vehicle specific settings intact, as the author intended them to be.



Now I'm not having a go at you by any means. Just getting the impression that there are some misunderstandings around, concerning the motivations for the changes you made. At the end of the day, I won't tell you what to do - I know why I set things the way they are and that's all I'm trying to explain, because I know it's not always intuitive in the Racer world

 

[KS95]

I don't know exactly what KS95 meant when he said steering was hard with the mouse, but I'll assume that the vehicle will feel a bit unresponsive if his controller file doesn't have a fairly low maximum lock setting. He'd end up having to move the mouse quite far to get reactions and then again to catch slides etc. Roll centre location is not very high on the list for possible reasons or fixes to that - it's likely more a matter of input/output ratio, ie linearity.

Correct! Just the linearity.

 

[Splashonda]

I think that this car is just plain Muscle perfection, it behaves exactly like it should! I downloaded v033 just to test it, but even on v0810 it feels fantastic!

About the steering talk up here, these cars are not suposed to behave responsively like a modern sports car, they don't like corners or drifting very much, remember!

But better sounds are needed... part of the charm of these V8's is the soundtrack!

Great work Cosmo!

 

[Alex Forbin]

Hi Cosmo,
I would suspect that anyone driving your car with a mouse would be just as unhappy driving a real car with a mouse. So that means you must be doing it right.

Quadcoremax,
FWIW Roll centers are almost never at the same height, this causes handling issues.

Alex Forbin

 

[Cosmo°]

If only it was that easy, drawing conclusions based on the "two evils make a good" principle.
Kidding aside, I'd be glad to hear some more of your impressions as well

 

[Mr Whippy]

As far as I know, there's no connection between the collision model and the roll centre location. Got a quotation from Ruud or Mitch on that maybe? Certainly I'm not the only who would like to know if there was such an oddity around

RC have to be outside the bounding box in plenty of cases.

RC is just a notional point, it really represents nothing except a mathematical helper for Racer's physics engine to calculate weight transfers with.

I've got RC set at ground level on loads of cars where that is the case.

Cosmo is indeed right to have that way under the floor at the front, as that is the nature of that suspension

Dave