Afternoon 'Polycount Army'!
I have spent the last couple of weeks traveling around to a couple of Universities here in the UK breaking down some of my work and going over some tips and advice for getting into the industry. Well, I'm done with that now and I thought this stuff *might* be useful to share with you folks.
I am working on putting together a page (
http://www.scotthomer.co.uk/freestuff.php) that also works on mobile, however you can read/see the slides from the breakdown part of the talks here:
30(ish) page presentation:
http://goo.gl/CVdVAx
HD Turntable video:
https://vimeo.com/109541762
Mirror on Scribt:
http://goo.gl/XVkJHD
Let me know if anything doesn't work, or if there are any major issues with things like spelling and functionality and I'll do my best to fix them. Feel free to ask questions and I'll do my best to answer them here honestly.
I'm soon to be starting a new, dreamlike life at my dream studio with alot of my biggest influences. Polycount was a HUGE part of that journey, so..'thank you' and enjoy!
Scott.
[vv]109541762[/vv]
Replies
I've been putting the PDF to great use, it's helping me loads! I'll be sure to check out the pages I've not seen yet.
Thank you very much.
Add he's great at art :poly122:
Pages 15 and 16 are not loading for me though.
Glued.
"After completing the texturing and baking process,
the model is ready to be reconstructed. Using
the original high-poly model as a template I
reconstructed the low-poly model of the book by
hand."
Why did you reconstruct the low poly again AFTER you had already unwrapped and textured the original low poly? Isn't it a hassle to try and match up the new low polys' uv to the old one?
He means he put the parts back together once he had finished his exploded bake, that's how it came across during his talk on Monday.
Cheers for the talk as well Scott, learnt a tonne and had a great day.
I'm wondering about one part though. On your Baking theory page, on the bottom left pic of the UVs you say "A: Unfold mapped, all faces connected and welded together, vertices welded." Same text for C but the UV edges for A are green (which means unwelded verts), so should that say unwelded? Cheers
In my experience metalness maps have only ever been greyscale, Toolbag 2 only ever checks a single channel for the metalness map anyway. So what benefit are you getting from having a metalness map that has a full range of the RGB channels rather than one that's more efficient on a single channel?
https://drive.google.com/file/d/0By01RVm2KLyQU003WFZ2M3JwdXc/view
https://www.scribd.com/doc/243816021/ScottHomer-OrnateBookBreakdown-V2?secret_password=V2oHrdIOLlzrVoM4Cvgo
@Notes, exactly as Loktor said, at that point the model was still exploded, so I reconstructed it back snapping it all back together. I have clarified this in the PDF just to save any confusion in future.
@DonofDon - Hmm, not sure how that happened, I will fix the image, but yes, the images should be the same, only the fact that A uses 1 smoothing group and B uses split normals makes the 2 examples different. Sorry about the confusion! I have updated the image in the PDF which should now be online!
@Mazz423 - OMG, thank you for spotting that dude, I completely f#cked up on that. I originally used a specular map with Marmoset (which I accidentally plugged into my PDF) and then switched to metalness (which I didn't plug into my PDF) later on. I have updated the presentation with the correct textures for PBR this should be online now, what a mess. Apologies for the confusion and thank you for spotting it.
Thank you for all the kind words folks!
(Cheers for John Declan Paul and CJ Furguson for spotting my dumb mistakes..:D)
Does the cage need to be triangulated if I'm baking using my 3d modeling software (Maya) as well? Or is it Xnormal's specific workflow? And can someone care to explain why does it need to be triangulated? Thanks!
edit: well by "in engine" auto triangulation I mean the fbx or obj exporter auto-triangulation process.
So when getting to the baking, you triangulate both the LP mesh and the cage? Cause as far as I can see in Scott's work, only the cage is triangulated? Also, can someone please answer if this workflow is for Xnromal only, or should be implanted in any software?
Thanks!