In this Blender 2.5 video tutorial learn how to retopologize a high-resolution, sculpted creature head to create a low-resolution version that is suitable for rendering and/or a realtime envrionment.
You will also learn how to bake out a normal map from the high-resolution model to the low-resolution.
Yay first (i think)
thx johnothan this is great
This is just blender secksieness
BTW, this is why I’m glad I’m a citizen, is to support these great tuts.
Glad to be a citizen too. Thanks for the great tutorial Johnattan, you are a great teacher!
This looks really easy, and most importantly extremely useful. I’m definitely going to be doing this in some future projects.
Hey Jonathan,
Awesome tutorial, Retopo can be really daunting without some instruction before hand. Just wanted to point out that as far as I know the sun lamp works as well in GLSL mode. It makes it somewhat easier to get a good preview than setting up a point lamp system.
Keep up the awesome work!
simply awesome! thanks for the tut.
Thanks for the awesome tutorial! I’ve actually been looking for something to this effect in Blender 2.5 recently.
Keep up the good work, man! Your tutorials are the best!
Awesome tutorial.
Been working with the baking system lately and it, indeed, is quite buggy and crashes a lot currently.
Hey Johnathan, great model and tut.
Where you have hidden the operator panel from the tools panel – it is minimised to the small plus icon at the bottom corner of the 3dview. Clicking that will put it back in the tools panel.
Seems cool Jonathan, but also seems you didn´t take a long break after the training series was complete
how do you rotate the camera around the cursos?
By pressing “.” you can change the pivot point to around the 3D cursor.
-Jonathan
Looks great
I wonder what will be possible, once the render branch and ptex are merged into trunk
Thanks, it looks great !
Awesome! Going to be needing this handy feature very soon in the near future so thank you in advance for this…
your models are a treat to look at and learn from
In some cases (haven’t tried this with a head) I’ve been duplicating the hi-res object, lowering the multires, applying it and then bake. Saves a lot of retopo time, but may not work with all object types. I mainly use this method for solid, non-deforming objects. Nice tutorial as always.
Very nice , when I get to this level I gotta / must learn this .
Amazing the small amount of time used to render the normal map model. Thanks for the tutorial.
Awesome tutorial Jonathan, Topo has always been taboo or even an evil task. Thanks again my friend, this one
will be valuable to have. I have a question, if you don’t mind
http://durian.blender.org/development/three-4k-posters/
On here they are talking about un-touched 4k renders. Can you explain what they are talking about? the textures? or a High res render?
my renders are only 96 x 96 dpi…
Once again thanks for all of your time and knowledge
-Tungster
Renders which are not yet composited, I’d guess…
@ kram1032,
Ahhhh, renders left as is,
Many thanks my friend,
Peace
-Tungster
Very cool stuff. As always, I’m impressed. I also like the creature model you’ve done. I think it would be cool to see this thing fully colored, spec mapped, and all that good stuff.
Cool tut!
Nice you are in contact with the dev of surface sketching script,
please let us know when a working surface sketchign update is out!!!
Woah… no way! Very cool. It makes me wanna go out and find some complex head models and practice with this tool for topology exercises. Great tutorial!
Great tutorial as always Jonathan. Watching this made me want to make a bust
Can I request tutorials here? If so I would like to have a tutorial on how to make SSS textures for human models (with the epidermal and subdermal maps) because this is something I have looked for – for around three months now and haven’t found even one for any software
Great tutorial! So that’s where the Retopo feature from 2.49 went..
And, I find it really awesome that you plan all the times you will create a tutorial! (And I now know one you are about to create..)
i’d love to see a tutorial on something like this, i liked your old monster tutorial but it was like in 2.46 or something
great tutorial as usual
One extra thing you could do is:
-open the normal map in GIMP
-duplicate layer
-set the duplicated layer blending mode to overlay
-merge layers
-run GIMP’s normal map filter and set it to “normalize only”
This way the normal map will be stronger. Blender’s baked normal maps seem to be a tad weak/flat by default, so this extra step can help you push out the detail added by the normal map.
awesome job jonathan
Man, you are so good at retopology; and, I guess, topology in general. I can never figure out where to put poles or how to evenly distribute vertices.
Awesome, it helped me a lot on this project: http://niuarts.squarespace.com/blog-design/2010/9/24/testing-lights-and-shape-keys.html
Hey thanks alot for this, now I can get on with doing the clothing on my character (shrinkwrap wasn’t working out for me)
i think the in the influence pannel you need to turn up normal.
Hello, first I’d like to say awesome tutorial.
second. I’m using blender 2.56 and the polygons are viewable in the normal maps that i created.
I’ve tried deleting doubles, flipping normals adding smooth shading but nothing seems to work,
the only normal map setting that seem to work is world, but even that is leaving seems in the middle of the maps. Any help would be greatly appreciated.
Im a Blender user for about a month now and this is the single most incredible tutorial on topology I have watched EVER.Combined with the tutorial about sculpting,I believe that jonathan Williamson has given us the best info on Blender on the internet.Its like PRON for Blender users.
I admire your work and the way you present each tutorial is very pleasing and easy to asimilate and learn.
Salutations from Bulgaria to you sir.
I had no idea there was such a fantastic surface snap tool… I knew about the other snap options for the most part… but this…
I now know how to do the complex retop I’ve been avoiding!
Keep up the awesome tips.
Can you do all of your animations for a movie with the lower poly version, and then link the rig and animations to the high poly when you’re ready for the final render?
Jeremy
Yes and no… There are a few problems with doing that. First and foremost being that a 3 million polygon highpoly mesh isn’t going to be the most cooperative thing for animations. The second issue is that the reason we bake normals is to allow most of the detail quality of the highpoly mesh, on a lower-poly mesh. However, if you watch Jonathan’s facial rigging and animation tutorials, you’ll see a method that could -theoretically- allow you to do what you described. You would use the lowpoly mesh as a Mesh Deform modifier, and the highpoly mesh would be it’s area of affection…
I say theoretically because Blender, your computer, etc., may not be able to handle that kind of processing power.
Generally with animation, we want to have the lowest adequate polycount… with still renders, highpoly meshes are ok, provided you only want to render once… (Higher counts equate to longer render times. Professional art studios can set up renders that take multiple days to render a single frame, though this is done rarely, and usually faster methods, like lowering polycount, lighting accuracy, etc., are employed, as they aren’t noticeably different)
That said, I’m not Jonathan Williamson, but those are my instincts as moderately experienced artist. I hope that helps.
You’re absolutely right. I should have expanded on my previous comment. You can use the multiresolution modifier to sculpt in some nice extra details but in order to use it for animation you’ll want to bake that normal detail out so as to be able to remove or disable the multiresolution during animation. Using the “Bake from Multires” feature should make that quite simple and allow you to keep just a single mesh rather than retopologizing.
-Jonathan
That’s a great explanation thank you!
When do you know to use a base mesh with multi res and bake multi res, versus making a multi res and retopologizing it?
It seems like an extra step to retopologize when you’ve already created the low poly mesh at the beginning of the sculpt.
Jeremy
when i try and unwrap my head one side is always much bigger than the other in the UV image editor can anyone help?
Jonathan i appreciate your help by showing me this tutorial though what should I do if my original sculpt is not showing up in the render??
Is it disabled at render time? Check the camera icon next to the name in the outliner. You can also press CTRL + ALT + H to re enable it.
Thanks I thought i almost lost it
Jonathan, trying to wrap my head around normal mapping … do I understand correctly?
each pixel projected on a face of a lower poly model stores a normal value, allowing fine details to show insofar as the face isn’t that foreshortened. The more foreshortened the face, (ie turned away from us) the less detail that is shown until finally the silhouette of the lower poly model is revealed as we look at faces “edge on”
does this sound right?
what kind of specs do you have on your computer that you’re able to go up so hih on the multiresolution modifier. Usually for me the highest i can go is 4 before my computer begins to lag.
Hey Jonathan
when i am in edit mode with my sculpted face there is an underlying ‘wire-frame’ and all the topology videos I’ve watched there is no vertices or anything in edit mode and so is like stopping me from doing topology please reply i need your help bad.
Hi there,
When sculpting topology isn’t so much of an issue, so long as you have evenly distributed meshes with adequate subdivisions to create the surface. After the sculpting is done you’ll generally retopologize your model, at which point the topology is crucial. This tutorial ought to clear a bit of it up for you: http://cgcookie.com/blender/2010/08/30/retopology-and-normal/
Cheers,
Jonathan
what do you mean?
this tut will be helpful when can get started correctly all topology i have seen is in edit mode were there is no vertices and edges and such on the mesh i feel like i’m missing something important that no one is telling me. i’m following everything to the letter is there a option i don’t have enabled. I’m extra confused. when other people do it they have no vertices around there mesh and then they make a mesh with dots in all the faces? helpp
hey jonathan, big fan of your tuts, and even a bigger fan of the workflow you’re using here! suuuuper thankful for the education we’re receiving at such a low cost.
one slight note on this video- I think you sell yourself short in the render by having such balanced lighting. Show off the normal detail with a bit more dynamic lighting.
oooh and also, about last few seconds where you compare renders of the hi-res and low-res versions: I have a hunch that the main visual difference between the two renders is the quality of the ambient occlusion. The models would look much more similar if AO was turned off for the render. The better-looking option would be, of course, to bake the AO from the hi-res to the low-res, and apply it to the diffuse. ( I understand that AO wasn’t really the focus of this tut, but don’t let it take away from the accuracy of your normal bake after all that time spent on the re-topo )
awsome turorial
Thanks Joel!
thx jonathan for all your gr8 tutorials! They are like a daily soap to me.
But at the moment it feels like i never get the topology thing right, damn xD
greetings from germany
p.s. i love this community <3
This might help you out: http://cgcookie.com/blender/cgc-series/learning-mesh-topology-collection/
thx for reply =)i figured out i’ve never activated the snap element to ‘faces’, it was still set to vertex so it felt kinda weird. I guess now it works as it should^^