1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Released Porsche 914-6 (v0.9.0 update)

Discussion in 'Racer Cars' started by JustinMartin, Jul 15, 2012.

  1. [​IMG][​IMG]



    I've been updating my old cars for v0.9.0... and here's a beta of the first one. The model is over six years old now, so it's not the most detailed. I'm not planning on doing anything about that - this is just a physics, shader, and feature update. The headlights, windows (W key) and targa top (T key) are animated, and it has reverse lights. It has six different colors, set up as Raven variants. (do people still use Raven? I've been gone too long...) I'm still tweaking the physics and shaders, but they're good enough to share.

    Suggestions definitely welcome, I'm still trying to relearn everything I forgot and learn all the new stuff. :O_o:
    • Like Like x 9
  2. KS95

    RACER Moderator

    Very nice to see this again, I was thinking about it not so long ago, it was a favourite of mine. It sounds like a very good update so I can't wait to try when I have more time ;)

    And yes, some people do still use raven although some cars have came with basic programs to choose options. (Whippy did iirc)
  3. Nice job Justin! I like the pop-up lights and windows, excellent. Liked the readme but I had to unzip it to read it. You should have the contents in a folder as somebody might just unzip it and scatter the files hither and yon. Also the tires need to cast a shadow.

    Also nice to see no Qlog errors!!!!
    • Like Like x 1
  4. Looking good :D

    I'm not sure if it's active on the Lambo or not but there was a shader that made emission on that specific material alter depending on the view normal, so the headlight glow would shine more when looking dead ahead vs at an angle like real reflector headlights.

    It does add a lot of realism imo and should be a standard shader, but not sure if it is or not it's ages since I looked :D

    Yep, good to see some nice content that although not as flashy and fancy as Forza 4 or something, it just works and has some real heart and soul feeling to it :D

    • Like Like x 1
  5. Added it, tire shadowing looks good on the 914, and does wonders for the Unimog. :D Checking for QLOG errors is one thing I definitely remember.

    vf_emission? It's still in the Lambo... works very nice.:thumbsup:
  6. Awesome update to the little car. only real issue I found with it is one of the views for me was in the passengers seat of the car. not that it mattered much to me anyways. :thumbsup:
  7. A fun drive :)
  8. It's nice to see such a Racer classic being revived. The base vehicle still is as solid as I remember and updating didn't hurt the experience at all.
    Overall, controlling weight transfer is the challenge when driving this quickly, but it's also a nice reward when a section of road really comes together.

    A few points I noted down:

    - Still has the proper flat-six sound, but unfortunately no horn.
    - clutch.max_torque is quite high - makes it very hard to start off slowly and smoothly, around 125% of engine peak torque is usually a good starting point unless you have specific data.
    - In the pacejka section, parameter a8 was causing force reversal issues, so I changed it to -0.03. Since a111 was not set to zero yet, the vehicle was also artificially unstable over kerbs and such - basically launched by magic forces.
    - Personally, I disabled ff_damping and ff_friction (set to zero), because the sensation feels fake to me and also covers actual Mz torque effects that I rely on for driving.
    - There seems to be no taking into account of drivetrain losses - normally, we use the differential friction parameter to introduce them because lowering engine.max_torque would have side efffects on the revving response etc.
    - If you use RLE texture compression for the .tga files, the folder size can be reduced from 32.5 to 9.5MB with no visual losses.
    - On the other hand colour banding on the dashboard is visible because global texture compression is enabled by default in racer.ini. We use compression=0 in the shader files themselves to override this. In order to avoid repeating this line for every material, you can put it globally into the vf_gloss/glass/matte sections.
    - You might also want to enable shadow casting globally, ie. cast_shadow=1 - Ruud decided to disable it by default for performance reasons.
    - In the racer.ini.dev section you can find selections for note_ini_fallback and note_ini_unused - set to one, these two give you the full QLOG listing instead of just major errors (there are a few minor items listed as missing or obsolete).

    Regarding your script related questions in the other thread, yes, these two documents list the available parameters. You can use "maxtorque factor" and "wheel X powergearing" via the system scripts to change the torque split on the fly. AWD in Racer is as weak as it was when introduced though, we're waiting for Ruud to give us control over multiple differentials still.

    For the window and targa top scripts, you could have the coefficient of drag change depending on the state - you have the data already, after all :)
    • Like Like x 1
  9. Indeed... struts in the front with very little camber and toe change combined with semi-trailing arms in the rear with lots of camber and toe change definitely makes it feel less planted than modern cars. I've never driven a 914, but my rwd Celica has the same suspension arrangement, and it responds to mid-cornering throttle changes much more than my 350Z, despite having 1/3 the power.

    Thanks for all the suggestions, they are really helpful.:thumbsup:

    I had tried setting powergearing=0 with powered=1, but the front wheels still rotate slightly when one is in the air. It doesn't seem to be getting much power, so it probably doesn't matter from a physics perspective, but visually it annoys me.:cautious:

    One nice improvement since I originally made the Unimog is the locking diff... back then we only had open and viscous. The locking diff gives it much better ability to climb steep hills. It finally can do most of what a Unimog should be able to. But locking impacts cornering, so it would be nice to be able to turn it off.
  10. The 914 is a very stable car in reality, one of mine was a 914-S with early k injection. It's center of gravity is very low and the turn-in is amazing, I never spun out in one and I can't say that for a few others. I always ran pirellis on mine and loved it's handling. The 914-6 was a little heavier in back but it wasn't disagreeably so, imagine this in black and you'll know what mine looked like..
    http://www.cannillomotorsports.com/pastinventory/9146/9146.htm Really nice example, I miss mine now.

    Alex Forbin

    P.S. Welcome back Justin :)
  11. So do they not feel like throttle changes / weight transfer are more magnified than a car that doesn't have the big difference in front/rear camber/toe curves?

    I would describe my Celica as stable in the sense that it's very difficult to get it to oversteer for long; when provoking oversteer, it tends to rotate a bit then hook back up. Not a good drift car (not that I care ;)). It's stable, but still, you feel the effects of weight transfer and suspension geometry changes more than most cars. Hard to describe, really. Given that 914's have almost identical camber and toe curves, I would've expected a similar feel, but much more nimble due to lower inertia and sightly rearward instead of slightly nose heavy weight bias.

    All that said, I do think the 914 is too easy to spin right now...
    • Like Like x 1
  12. KS95

    RACER Moderator

    I thought that too.
  13. The biggest thing is the low polar moment lets the car respond very quickly, more so than any of the other mid-engined cars that I've owned/own. There is throttle-off oversteer but it was never a problem for me, it's not twitchy or prone to snap-oversteer but it does recover very quickly and can throw an inexperienced driver off. Entering a corner at around 3k gives you enough room rpm-wise to balance it with the throttle and brake, if it does start to come around just give it a little throttle and it will straighten out. My cars were '72 models and had ARBs in front and back which caused them to be reportedly more stable than earlier years,
    I haven't driven any earlier models so I can't say for sure. I do remember that tire pressure would make a lot of difference in stability.

    Alex Forbin
    • Like Like x 1
  14. That actually sounds about like I expected, so I'm probably on the right track if I can get the snap oversteer tuned out. You say it's more nimble than any other mid engined sports car you've driven - have you driven a 1st gen MR2? The SAE real world inertia doc lists inertia values for the MR2 that are about 10% higher than what I calculated for the 914, so that would be a nice validation of my inertia calculator if so...

    Also out of curiosity, have you driven any other cars that had stuts front / semi-trailing arms rear, like an E30 3 series? Curious how they compared to the 914 with regards to lift throttle oversteer and how quickly they recover.
  15. No, on the MR2 and I didn't own the E30 so I didn't feel right about driving it too hard since it would have scared the owner. But I've done a full 360 in my Bertone X19, my 2.5L fieros are fairly predictable overall (especially the 88s) but at the limit the 84-87s can snap pretty hard if you brake hard and late without a little throttle too. The 3800 Fiero has bitten me twice, and it's too early to tell on the L67 Fiero (just finishing it up), and the 308s and the Esprit S4tt are not a problem.

    BTW I really wish they would put out another study for us. :)

    Alex Forbin