Johnny Destroy's sketchbook

[JohnnyDestroy]

I'm a senile old man who started out modding textures for games in the early -00's. As I grew more and more frustrated with my inability to edit the 3D files themselves, I began to download tutorials for various 3D-editing software geared for specific game engines. From there it was only a matter of time before I started to look around for the real deal, and since then I've worked in 3DS max, Maya and LightWave.

Being curious by nature, I've fooled around with most aspects of 3D. I have tried out animation, modelling, texturing and anything in between, which have sadly led to a vey incomplete demo reel as most projects have been abandoned halfway through once I've come across a new field I wanted to fiddle around with. I recently inherited a large sum of money and decided to quit my job and try to get into some art programs, but since my demo reel was much thinner that expected, I decided to try out Zbrush, hoping that it would be able to streamline my workflow. And did it ever!

I'll add my work here as I complete it. Please feel free to hand out critiscism and opinions, that is the reason I post my work after all. I'll start out with the very first thing I made in Zbrush a few weeks ago: The Moon Rocket from Tintin.







My goal with this project was to test the hard surface possibilities of Zbrush. I had recently bought a Tintin desk decoration and felt inspired. I wanted to stay true to the original design, but also add some details of my own. Hergé was the master of the school of clean lines, and thus his rocket design is very minimalistic. I began by following his design exactly, but quickly realised that it was going to be a very, very boring 3D model unless I added some details.

For example, the original design is completely devoid of gaps between the various parts of the rocket, the antenna is extremely rudementary and there is no engine housing to speak of. I found that the model became much more interesting once I added some minor details to these areas. I also added some lines and crevises to the pontoones not found on the original.

The model is at roughly 9 million polygons, and most of those are just there to make sure that the rounded surfaces are perfectly smooth. I also played around with multi pass rendering for the first time.

All in all, I'm pretty pleased with it.

[VISEone]

Nice job - you captured the style of Hergé but with your smooth changings you came up with a nice piece of something I would like to see on my desk. I was in a toystore at the atlantic coast in france and this guy had an incredible collection of TinTin and Hergé - he also had a rocket about 16 inches height - and I remember it was not nice to look at - even if it was worth a couple of 1000 euro - your's is much better and like I said before - you captured the spirit top notch!

[JohnnyDestroy]

For my next project I had two specific goals:

1. To find a workflow that's comfortable from start to finish.

2. To test the fibermesh functions.

I started with a rudimentary concept sketch of a landscape in Photoshop and then used that to identify the subtools I would need to create. For the landscaping, I created some low poly versions in maya and then exported them to Zbrush using the wonderful goZ function. For the tree, I used polysheres for the main body and then added several layers of fibermesh branches, and for the grass and leaves I used fibermesh. I also made good use of the HD geometry function. Without it, my system would have been brought to a halt (even though it's pretty powerful) as the complete tool weighs in at roughly 49 million polygons.

Once the subtools were done, I began polypainting them, using both manual painting techniques and textures. I also made several custom alphas for minor rocks, cracks and the stepping stones that covers the pathways. These alphas were great for quickly generating geometry and they were also used during polypainting. The result:







In conclusion, I'm pretty satisfied with this workflow. I find that making low poly models in maya suits me, since I'll get a clean topology from the get go, thus saving me from the hassle of a retopology. I'm also happy with the result, even though there are some things I would do differently in hindsight. The tree could use some more work, the fibermesh branches in particular; I never even bothered to paint them. The field of crops is also not all that it could be. But in the end, I acomplished what I set out to do, and I'd much rather move on to new projects than dwell on this for too long.

[JohnnyDestroy]

Nice job - you captured the style of Hergé but with your smooth changings you came up with a nice piece of something I would like to see on my desk. I was in a toystore at the atlantic coast in france and this guy had an incredible collection of TinTin and Hergé - he also had a rocket about 16 inches height - and I remember it was not nice to look at - even if it was worth a couple of 1000 euro - your's is much better and like I said before - you captured the spirit top notch!

Thank you for the kind words!

What you're describing is exactly what I found to be true; the rocket original rocket is far too stylized to really work as a 3D model. It looks great in the comics, and I'm a huge fan of Hergé, but the lack of details makes it dull to look at once you see it in all three dimensions. The biggest problem I had while making the model was to make sure I didn't add too much detail. At one point I had induvidual bolts and rivets for all the plates, but it made the model stray too far from the original. It was a balance act...

[MealeaYing]

I love every thing in here, very very nice work!



Cheers!
Mealea

[chalkman]

Very nice work, welcome! Look forward to seeing more of your work! Cheers, David

[EZRA52]

Great stuff Johnny. Excellent beginning here in Central. Looking forward to more of your work.

Ezra

[CeeKay]

Yep ~ Have to agree that you've got the subtle nuances of the rockets textures down there perfectly...

A very well thought out balancing act there Johnny (from another Herge fan).

[JohnnyDestroy]

Thank you all, I'm very glad my work was well recieved.

I'm currently planning to make some character models, and I have a few concept sketches ready. Originaly I had decided to do old stoney face himself, Judge Dredd, but the level of realism I want for that model requires a completely new base mesh with good topology. I have a decent human base mesh from maya, but the topology is all screwed up on that one, so I'll have to retopologize it first. I'm not fond of the idea of using someone elses base mesh, and even if I would decide to go that route it seems all but impossible to find good base meshes with a topology suited for shoes and clothing rather than muscle anatomy, so I'd stil have to retopologize.

As I don't want to slip into my old habits of puttign projects on halt while I delve in to technical issues, I think I'm gonna have to put the Dredd project on the back burner for now. I can work on my own base mesh in between projects, and focus on putting out more finished work for now.

So with that in mind, I think I'll choose something more cartoonish that's not as dependent on a perfect topology. It would also allow me to try out the dynamesh functions of Zbrush.

[Jerick]

Welcome to the awesome club, Johnny! 3D artists are like some of the coolest people on earth and a large portion of them congregate on this site. Your rocket is refreshingly old school. To me, the 1950's era vision of the future is the most captivating future ideal, as strange as it sounds. Very stylish.

[JohnnyDestroy]

I've chosen my next project, and threw together a quick concept sketch last night. I've had this idea of an old-shool officer of some fictional nation, almost Preussian in nature. I figured it would be a suitable object for a more stylized and "cartoonish" project, so I fired up photoshop and started plotting the project.

The pose I have in mind isn't very complicated, so I think I'll try to model it directly, ignoring the t-pose phase of the model. This is something I've never attempted before, so it's going to be interesting to see if I can handle it. Also, since I have no plans to animate the model, it looks like a prime candidate for trying out the dynamesh function. As soon as my son poops and falls asleep, I'll try to find the time to get this project rolling. Here's the concept sketch:

[JohnnyDestroy]

I haven't had much time to model this week, my four months old son demands a lot of attention, but I did start on the head of my Officer.

As planned, I tried out the dynamesh function, and I was pretty pleased. It quickly became obvious that I would still need to retopologize the head, but I still liked the work flow. It allowed me to quickly produce the basic shape and feel of the head, and I was able to try a lot of different ideas without being constrained by topology. Using dynamesh, I felt that I didn't "paint my self into a corner" quite as often as usual, and I'll be using it again.

That being said, I sure do hate to retopologize models. It's a pain, and worse of all is the fact that it's obvious that I need to do it more often, as my feel for edge flow is not up to par. I had a lot of problems with the retopology on this one; the normals got flipped (causing problems during the projection), I had a ton of dud points and about 75% of all my points became locked in place about halfway through, meaning I could no longer move them. There are a few areas with weird topology on the model still, but again, I didn't want to get bogged down as usual, so I decided to let them be and rely on practise making perfect in the future.

All in all, I don't like the retopologizing features in Zbrush. There's not enough visual information, and it's very, veeeery easy to make a mistake, without noticing, that will later render the model nearly useless. If anything, this ordeal made me look around for a dedicated topology software, and I'm almost certain I'll end up buying TopoGun.

Once the head was retopologized I had to export it to maya, clean it up, import back to Zbrush and project the details (mostly polished edges and things like that) from the original dynamesh. This proved difficult until I noticed that the retopologizing had flipped the normals of the model and then turned on double sided faces.
And at last, today I had some time to sit down and start the aging process of my Ofiicer's head. Images below:

[voodoodad]

Johnny, I love the work on the head. It looks great! One very minor thing I noticed though. You juxtaposed the "three quarters" and "side" labels in your last upload.

[JohnnyDestroy]

Johnny, I love the work on the head. It looks great! One very minor thing I noticed though. You juxtaposed the "three quarters" and "side" labels in your last upload.

Oops! You're right, I completely missed that.

Anyway, here's where I'm at now.



I think I'm pretty much finished with the head at this point, except for hair which will be added later with the use of fibers. Once again, I'm trying not to get bogged down, it's so easy to just keep going and going on a model, it never feels quite done. But I'll leave the head as it is for now at least, and start working on the coat.

[JohnnyDestroy]

I just finished the basic shape of the coat. I discarded my original idea of completely bypassing the T-pose, it's too much work getting down all the mirrored basic features, like cuffs and things like that. Instead I decided to model out the basic shape of the coat in a T-pose, along with all the features that are going to be mirrored anyway. The model is pretty clutter free, I only added the parts that will affect the topology, and now I'm going to retopologize the whole thing before I start adding details and pose it.

The arms are too long, but I'll fix that during the posing. Right now I just focused on getting the model ready for retopology.