The "What Are You Working On?" Thread

My cousin who owns multiple Z cars says DATSUN stands for "Don't Ask To See UNderneath" Cause basically most of it is probably gone...

Fun car though especially with a small block chevy under the hood...

View attachment 285494
I'm a bit reluctant to search for owners' feedback on the handling because of things like half the car being gone from rust and no one actually owning original shocks anymore. :laugh:

I'll just trust the data and my instincts.
 
A few months back I came across an RF2 version of my local circuit, Atlantic Motorsport Park in Shubenacadie, NS. I've been in touch with the author and he kindly gave me permission to convert/continue his work and provided me with all source files. The circuit is built from survey data and is quite accurate to reality in terms of layout/elevation/camber, etc. The graphics are quite dated though. I have also found Lidar data for the circuit, but I'm not sure how to work with it yet.

Anyways, my main question is this- What's the best way to get the circuit into AC? Do a standard conversion from RF2, or since I have the Blender files would it be better to export the model and go from there? This is my first venture into 3D work and modding and I've been doing lots of research but it's all a lot to take in and process.


upload_2019-1-10_12-45-26.png
 
Last edited:
A few months back I came across an RF2 version of my local circuit, Atlantic Motorsport Park in Shubenacadie, NS. I've been in touch with the author and he kindly gave me permission to convert/continue his work and provided me with all source files. The circuit is built from survey data and is quite accurate to reality in terms of layout/elevation/camber, etc. The graphics are quite dated though. I have also found Lidar data for the circuit, but I'm not sure how to work with it yet.

Anyways, my main question is this- What's the best way to get the circuit into AC? Do a standard conversion from RF2, or since I have the Blender files would it be better to export the model and go from there? This is my first venture into 3D work and modding and I've been doing lots of research but it's all a lot to take in and process.

I have no idea how to properly implement a track of this complexity in AC, but I've done a bit of track and track prop work before. Cars are one thing but managing a scene this big is a nice rabbit hole you got yourself into. Good luck. :roflmao:
 
Haha, it certainly is a huge challenge but I'm up for attempting it, and I want to do it properly. This place is near and dear to my heart as I've spent huge portions of my life at this circuit. It also has a very interesting history, being as far as we can tell, the only club/volunteer owned and operated circuit in North America. It opened in '74 and that year Gilles Villeneuve set the outright track record, which still stands today, at 59.xxx in a Formula Atlantic Car. Other greats such as Keke Rosberg, Rahal and others have raced here over the years.
 
A few months back I came across an RF2 version of my local circuit, Atlantic Motorsport Park in Shubenacadie, NS. I've been in touch with the author and he kindly gave me permission to convert/continue his work and provided me with all source files. The circuit is built from survey data and is quite accurate to reality in terms of layout/elevation/camber, etc. The graphics are quite dated though. I have also found Lidar data for the circuit, but I'm not sure how to work with it yet.

Anyways, my main question is this- What's the best way to get the circuit into AC? Do a standard conversion from RF2, or since I have the Blender files would it be better to export the model and go from there? This is my first venture into 3D work and modding and I've been doing lots of research but it's all a lot to take in and process.

View attachment 285495
that one? search for ac reboot on google and you'll get it
it's already been converted
Bjk7m8X.png
 
Bunch of bad geometry as evident from the screwy reflections/bumps, but I THINK this is a more realistic topo for the front end. Not anywhere close to where I'd want it though.

slowlygoinginsane.JPG


I'm mostly working off pictures of glass headlight covers to get an idea for how it should flow. Connecting it to the bumper, mainly because of the curve on the side, is proving a bit challenging. Should I just model them separately?

17.jpg


I'm sure the experienced modellers here are scratching their head, but I've never modelled anything this organic. Strictly props and nice, easy to understand angles and forms up to now.

I think I'll take a break for now and see what people have to say.

EDIT: What kind of limit to good reflections is there working in polygons? I never really looked closely at real-time cars in games. To my understanding, you can't really get NURBS-like reflections with polygons no matter how dense or clean the mesh, so should "fixing reflections" be something done near the end, without too much worry about it? Or should I always keep reflections in mind when modeling the base mesh?
 
Last edited:
Bunch of bad geometry as evident from the screwy reflections/bumps, but I THINK this is a more realistic topo for the front end. Not anywhere close to where I'd want it though.

View attachment 285597

I'm mostly working off pictures of glass headlight covers to get an idea for how it should flow. Connecting it to the bumper, mainly because of the curve on the side, is proving a bit challenging. Should I just model them separately?

View attachment 285598

I'm sure the experienced modellers here are scratching their head, but I've never modelled anything this organic. Strictly props and nice, easy to understand angles and forms up to now.

I think I'll take a break for now and see what people have to say.

EDIT: What kind of limit to good reflections is there working in polygons? I never really looked closely at real-time cars in games. To my understanding, you can't really get NURBS-like reflections with polygons no matter how dense or clean the mesh, so should "fixing reflections" be something done near the end, without too much worry about it? Or should I always keep reflections in mind when modeling the base mesh?
hard to plan mesh without having to do it in front of me :)

upload_2019-1-11_9-32-18.png


or plan b avoid that angle at all since these light cover are an abomination :D
upload_2019-1-11_9-33-35.png


you can also look into smoothing group for your model, to see better how it works by surface angle
example id=1/2 (just an example) this way it breaks reflection
upload_2019-1-11_9-38-1.png

as for my arrow, the line seems wrong, coming from the angle of the windshield and following the side panel, yours goes down too much, hence your difficulties to find the spot for the light :)
 
hard to plan mesh without having to do it in front of me :)

View attachment 285623

or plan b avoid that angle at all since these light cover are an abomination :D
View attachment 285624

you can also look into smoothing group for your model, to see better how it works by surface angle
example id=1/2 (just an example) this way it breaks reflection
View attachment 285625
as for my arrow, the line seems wrong, coming from the angle of the windshield and following the side panel, yours goes down too much, hence your difficulties to find the spot for the light :)
That's it. I was thinking of taking a picture and drawing over it, in 2D with a tablet. I could probably map out anything that way, traditional artist background and all. Your picture is very clear. Thanks!

I was thinking of just re-modeling around the wheel arches, because all I'm doing is cutting out loops and slowly going towards nothing again. :roflmao: Would cut down on the unneeded loops.

Remodeled it based on your picture. This is more like it. Should approach things carefully from now on, until I get some more intuition. No need to go overboard with 20 loops outright...

benobro.JPG


UPDATE:

Thoughts? I think this is the cleanest I've done so far.

lowpolyrear.JPG
 
Last edited:
To get the headlight shape correct search ebay for hires pictures of the head lights and model them separately and then shape the car so those fit. Same tip goes for rear lights, mirrors, dashes etc.
 
That's it. I was thinking of taking a picture and drawing over it, in 2D with a tablet. I could probably map out anything that way, traditional artist background and all. Your picture is very clear. Thanks!

I was thinking of just re-modeling around the wheel arches, because all I'm doing is cutting out loops and slowly going towards nothing again. :roflmao: Would cut down on the unneeded loops.

Remodeled it based on your picture. This is more like it. Should approach things carefully from now on, until I get some more intuition. No need to go overboard with 20 loops outright...

View attachment 285631

UPDATE:

Thoughts? I think this is the cleanest I've done so far.

View attachment 285648

what i'm doing prior to starting mesh, is making splines from the blueprint. All the lines you see in side view, you will have them in top view etc. Positionning takes a bit of time fine tuning the x,y,z.
But then it makes a 3D blueprint i can refer too.
Also splines can have bezier corner, and nicely smoothed shapes, so i know it's a good reference for positionning my mesh afterwards.

your front looks much cleaner now, easier mesh to play with and adjust, now you can match to your blueprint to get the nose right ;)
 
@Ghoults
@Ben O'Bro

Yep, my idea was to do them separately and then get the flow right. Hence why I'm doing the base mesh of the car first, so I can graft the details into it later, instead of working the mesh off the details. I think you know what I mean.

I don't feel too good adding much more definition to the base mesh, because I don't know how the headlights, front and rear, will flow. Should I do those details now and then fit the front and rear quarter around them?

Physics also advanced. Gonna need a lot of reference and reading to really get it right, given such an old car, but everything's pretty reasonable. I think I'm gonna go for a 1972 240ZL because of the similarities between USDM and JDM.

Alright, this is shaping up a bit. I can get behind this. Not a fan of all the **** going on over the front fenders, but it'll probably be necessary in the final geometry, anyway. I think now would be a time to start modeling the lights separately.

110119_1637.JPG
 
Last edited:
So I've been experimenting with my AMP conversion project. I was able to use the MAS utility to extract the .RFCMP. I loaded the .SCN into 3dSimed and everything was as it should be. I exported to Autodesk .FBX and it automatically exported a texture folder with it. I imported the .FBX into KSeditor and when it loaded up everything looked strange and was pitch black and pure white and didn't resemble the original circuit. Where have I gone wrong?
 
So I've been experimenting with my AMP conversion project. I was able to use the MAS utility to extract the .RFCMP. I loaded the .SCN into 3dSimed and everything was as it should be. I exported to Autodesk .FBX and it automatically exported a texture folder with it. I imported the .FBX into KSeditor and when it loaded up everything looked strange and was pitch black and pure white and didn't resemble the original circuit. Where have I gone wrong?
benobro 24/24 assistance how may i help you today

upload_2019-1-11_16-44-39.png


persistence is the ini file that comes with the fbx, when you export from edit:3dSimEd, you have the option to create that ini file

it will not keep the object properties (such as transparency, or rendering on/off), but will keep material settings and dds maps
 
Last edited:
Thanks Ben and John.

So this is my export/import process. I load the .fbx in KSeditor, load persistence from the .ini file 3DSimEd provides and the final image is what I get in the editor.
upload_2019-1-11_12-4-41.png


upload_2019-1-11_12-9-37.png


upload_2019-1-11_12-12-30.png
 
Thanks Ben and John.

So this is my export/import process. I load the .fbx in KSeditor, load persistence from the .ini file 3DSimEd provides and the final image is what I get in the editor.
the editor is set to black & white initially. goto illumination tab and fix it. selecting natural will allow you to see your textures with shaders without to much effects.

you don't need vertex RGB in your list as Ac doesn't support it by default (actually it loads a new object for those vertex without vertex rgb afaiaa ) and turn off copy texture maps as you wont need copies.

nothing is wrong. also once you have set the colourscape and anything else as you like it. goto layout tab and press "save current", next time it loads as you set it and with better resolution.
 
Last edited:

Latest News

What's needed for simracing in 2024?

  • More games, period

  • Better graphics/visuals

  • Advanced physics and handling

  • More cars and tracks

  • AI improvements

  • AI engineering

  • Cross-platform play

  • New game Modes

  • Other, post your idea


Results are only viewable after voting.
Back
Top