So if you run Max units of mm, then your system unit scale must be metres for that workflow to work...?
Ie, Max just multiplies all values by 1000 to transform from metres to display mm values.
That is pretty limiting because it means whatever you use you are modelling in the range of ~ 0-10 max system units to get 0-10 metres in Racer... that is really small and all the spinners are approaching useless when you get down to offsets like 1mm or something (ie, 0.001m)
Is it difficult to add a scale variable on the export panel?
Actually using 3DS Max properly results in this exporter not working correctly. I can see you maybe using metres for a track, but for a car it really isn't playing into the scene scale.
Things like viewport culling start to be a problem (ie, you have to pull it right back)
Using metre system and mm Max units means the default chamfer step size is not even working at the sizes you'd want to control a chamfer.
Ie, 3mm or 2mm chamfer gives no difference in the viewport at all. That is a fine chamfer I'll admit, but the fact the 3mm chamfer is 50% bigger and Max doesn't even see it is not ideal!
Building a car in metres is just not a good idea it seems... even though the Max units are mm, it's really just multiplied every number you see by 1000, not given you 1000x more accuracy
(a fundamental flaw in my view, but this is the way Max works and so we have to work with it I guess)
So I still stand by my idea that working in cm is best for cars, then going via ASE > modeller, doing the /100 scaling and working from there. You can keep meshes separate which smooths workflow and non-linear workflow possibilties, and when you export via ASE it connects them to one DOF too.
For tracks then metres is more acceptable as a unit I suppose, but still, if you are building small props I'd prefer cm (currently make them elsewhere and import/scale as appropriate for export)
I still love this plugin because it's fast for tracks, but it's limited for cars still on my recent findings with this.
Yes I could merge my meshes right before export, and scale it down, but then it's no faster than going via ase/modeller.
Surely just adding a scale variable would be super easy and allow authors to be flexible with the tool, rather than forcing them to follow specific and possibly not ideal workflows?
Cheers
Dave