Portfolio critique for 3D enviro/prop artist?

3dcaster

Hi there ,
I am aspiring environement hard surface artist.
Self tought hardworker by night after hours.  it has been 2 years now that i am doing polygonal modelling with some previous CAD experience.
Trying to land my first job in game industry getting little or no response to my applications. 

I work mostely with Maya-Zbrush-Substance-Unity workflow and recently started incorporating fusion 360 for hard surface details.

https://www.artstation.com/3d_caster

I guess i have a lot of  contradictory ideas for my work and a lot of questions. Below are some of them:
1.Should I aim at bigger(more complex) pieces? or rather should i get more fidelity  from my smaller props? 
2.Should I concetrate more at the Unity part of my work?
3.I feel like i need to focus on hard surface at the same time I know that my portfolio lacks purely environmental pieces.   
4. I love to learn new software and i know learning Houdini is very popular now but i don't want to get burned out steaming in the wrong direction.  

To be frank I feel a bit stuck. I do not know what should i concetrate on in the first place.
The big questions is: Is my skill good enough for an entry level position as a
3D Environment Artist?
Have no mercy. and please let me know  how to improve my work.
thanks for your time,
Marek

Seeberg

Hello, hope all is well! 

Portfolio Feedback

• So for starters make sure you always hyperlink your website. Even in a very informal post. When apply or writing emails to employers having it be super easy with just a simple click is just a good habit to practice. 
• "Zombie hunter Sci-Fi motorcycle" could use some more topology. Getting that close to the prop for renders you definitely want it as smooth as possible. 
• "Engine of creation" needs better material definition, maybe add some wear and tear to it like engine burns. 
• "Lovebirds playing hooky" could use some more Ambient Occlusion. The car itself needs to be wwwaayy more rusty or deteriorated if a tree was able to grow out the engine block. Like that is a lot of time. Same with the K+M, that needs to be more worn down and rough instead of looking like fresh spray paint. 
• "Abandoned Military Warehouse" What makes it military? I would add a few small props, debris, and trash. Also take a look at the windows, they could use some depth to them. 
• "Arch Vis - a flat in Cracow" Shadows my dude, also in some of the render shots there is visible pixelation. 
• "Cafe Racer - Gauges" & "Honda Tank" could be combined with your "Used Work bench" when presenting a prop it's best practice to add a small scene around it or diorama, so I think you could combine those three into one render / portfolio piece. 

General Feedback

• Materials and Textures need more focus on generation, I would recommend adding some substance / material studies 
• For Environment Portfolios it is recommend to demonstrate 3 Environments. Exterior, Interior, and one with Foliage. 
• Focus on Texel Density, saw a few examples where your up close shots are border line blurry. 
• Focus on lighting, a few scenes could really use some improvements to the lighting. Also dont be afraid to add some ambient

Just my opinion so don't listen to me

• My last point, and you can take all of this with a grain of salt. If you are shooting for an artist / environment portfolio I would render out the environments and props in Unreal Engine rather than Unity. Unreal offers better rendering and control to help produce great looking assets. I use unreal for personnel projects and Unity at my 9-5 and from personnel experience UInreal is super easy to use to quickly pop out renders. Unity is really good to use if you are applying for a level designers job. But everyone has their own preference and both engines have the weaknesses and strengths.

Best of luck! 

Doxturtle

the workbench ends have some weird UV warping going on

Alemja

@Aeseeberg04 DId a great piece-by-piece breakdown so I will try to answer the questions you asked specifically. Also +1 to learning Unreal, I work in outsourcing and it's being used on 90% of our projects.

1.Should I aim at bigger(more complex) pieces? or rather should i get more fidelity  from my smaller props? 

If you do environments I would keep the scope small so you can work on better fidelity, atmosphere and story in your props and environments. Corners, small dioramas and the like are great for building environment set pieces while keeping the scope smaller. One of my favorite pieces to show environment artists is this piece by Scott Homer: https://www.artstation.com/artwork/KrVx

It was done years ago and yet I remember when it was put here on polycount, it's just a corner of a room that oozes atmosphere and story and goes the extra mile of making the diorama aspect interesting by making it look like it's falling apart in space. This is the kind of thing you should start to work towards in your pieces, those final 20% details (textures, lighting, presentation) that push things to the next level.

2.Should I concetrate more at the Unity part of my work?

Environment artists need to have more engine knowledge than character artists because they often get involved in level building. Unity is fine but as recommended above, you should start looking into Unreal too.

3.I feel like i need to focus on hard surface at the same time I know that my portfolio lacks purely environmental pieces.   

Why not both? you can easily make hard surface pieces that are part of a small environment, 2 birds with one stone. Also as an extra work, start including more breakdowns of your work, if you can make things in efficient, smart ways that will make you an extremely valuable artist.

4. I love to learn new software and i know learning Houdini is very popular now but i don't want to get burned out steaming in the wrong direction. 

This is a understandable fear. Houdini is still new and while there is a definite advantage to getting on the bandwagon early, you're still in the phase where you're still figuring out how to put the wagon on the horse. Meaning you don't quite have the fundamental skills yet so focus on making good, solid pieces that can get you jobs and then once you get that job you can start learning Houdini. It will still be around and who knows, you might be able to learn it on a job. I've seen a lot of people stretch themselves thin by learning software, not skills.

The big questions is: Is my skill good enough for an entry level position as a 3D Environment Artist?

I have to be honest, no... not quite yet. I think you probably have about another year or 2 of really focused dedicated practice before you will be good, sooner if you really go all in and push yourself. I would start making threads for your projects here on polycount, find some discords if you can and reach out to environment artists you admire to see if they can give you feedback. You have some skills, you're just in the sucky stage where you're getting close but you aren't quite there yet. I think you can do it and hope to see you post more stuff around here!

3dcaster

Thank you @Alemja and @Aeseeberg04 for your time and proffesional cryt. 

After reading the answers I came up with the idea for a hard surface diorama scene.
Please let me know if you think it is a good step for me at this time.

Story:
Gererator unit of the big ship has been detached from the main body as a result of impact. It is now drifting in to the space. Since gerator is still working behind the floodgate the drifting piece is still enlightened inside by the computers and other light sources. Bits and pieces are driftng around like in "Gravity" movie.
Directly behind the focal point of the wrecked ship would be a planet (framing effect) and  far away light source (sun like) that gives the "hairlight" effect on the ship and may cause lens flare on the camera. 
very rought sketch below (sorry... I am not to good with drawing)

Benefits:
- Practicing lighting (realtime?) in engine. I decided to give Unreal a try. 
- Both Hard surface and environment pieces. (two birds with one stone) 
- good setup for story telling
- a lot of objects to be duplicated. It would take propably about 10-15 unique pieces needed.

Cons/week points:
- I do not know much about re-using textures with decals.
This piece needs decals and propably I will end up with unique texture sets.
- Do not have many textures needed for this piece in my library so I will need to look for it or make new substance materials.

Letting me know if this would be a good practice at this stage would be crytical.
If it is a good idea. What do you suggest:
- Making a new thread on polycount with update from time to time?
- postig an update in current thread from time to time?
- Making a piece from start to finish and than making a short article like this?
https://3dcaster.weebly.com/process.html

 I do not want to spam the forum.

Thanks once again for the feedback. I think it helped me to clear my mind a bit. 
I am very appreciated.

Marek

3dcaster

Thank you @Alemja and @Aeseeberg04 for your time and proffesional cryt. 

After reading the answers I came up with the idea for a hard surface diorama scene.
Please let me know if you think it is a good step for me at this time.

Story:
Gererator unit of the big ship has been detached from the main body as a result of impact. It is now drifting in to the space. Since gerator is still working behind the floodgate the drifting piece is still enlightened inside by the computers and other light sources. Bits and pieces are driftng around like in "Gravity" movie.
Directly behind the focal point of the wrecked ship would be a planet (framing effect) and  far away light source (sun like) that gives the "hairlight" effect on the ship and may cause lens flare on the camera. 
very rought sketch below (sorry... I am not to good with drawing)

Benefits:
- Practicing lighting (realtime?) in engine. I decided to give Unreal a try. 
- Both Hard surface and environment pieces. (two birds with one stone) 
- good setup for story telling
- a lot of objects to be duplicated. It would take propably about 10-15 unique pieces needed.

Cons/week points:
- I do not know much about re-using textures with decals.
This piece needs decals and propably I will end up with unique texture sets.
- Do not have many textures needed for this piece in my library so I will need to look for it or make new substance materials.

Letting me know if this would be a good practice at this stage would be crytical.
If it is a good idea. What do you suggest:
- Making a new thread on polycount with update from time to time?
- postig an update in current thread from time to time?
- Making a piece from start to finish and than making a short article like this?
https://3dcaster.weebly.com/process.html

 I do not want to spam the forum.

Thanks once again for the feedback. I think it helped me to clear my mind a bit. 
I am very appreciated.

Marek

Alemja

Glad to be helpful! Your new project sounds like a solid plan, be sure to grab a lot of reference and put together a mood/reference board. This will help people when they give you crits so they can push you in the right direction.

Most people tend to make a thread per-project here, so that would be the best idea. If you do consistent, meaningful updates, say every 1-2 weeks you will get some attention and people will help you out. Don't be discouraged if people don't respond at first, sometimes it takes a while to get noticed. However if you give as much information as possible in the beginning people will be more likely to help, otherwise they won't know what you're making.

Getting feedback as you are working on the project is the most important thing you can do and will allow you to grow the quickest. If you only show pieces when they are finished you are more likely to keep making the same mistakes, the goal is to get as many eyes on your work as possible to catch things early. As long as it looks like you are trying to implement feedback and making progress, people won't think you're spamming.

Seeberg

if you plan on reusing textures definitely look into Trim Texture techniques, here is a link to get you started!

https://www.gdcvault.com/play/1022323/The-Ultimate-Trim-Texturing-Techniques

3dcaster

Thnaks guys.

@Aeseeberg04
Ultimate trim technique looks super cool. I will deffo try to utilize it for the purpose of my scene. 
I have a question though. Have you ever used it with the more photorealistic textures?
I Morten mentioned that because of the stylized style of the "Sunset overdrive" it was not an issue. I will definietely try to go for more realistic look. 

Thank once more. this talk was super useful!

Zi0

IMO the best thing is to start with props and become very efficient in high and low poly modeling, UVing, baking and texturing, when are able to produce high quality assets it will be easier to make a scene full of quality models but then again this is also depending on what kind of scene you are creating and in what style but since you mentioned you want to go hard-surface I think this approach will help. Find yourself a nice concept and try to ace it! . Fusion 360 is pretty neet but afaik most studios  use subDs and Zbrush for the high poly models so I would suggest to become good in making high poly subd models in  (for instance) maya.

Also I noticed that you are from Krakow and I saw this on skillshot today:
https://www.skillshot.pl/jobs/11511-junior-level-artist-mesher-at-flying-wild-hog

Seeberg

@3dcaster I have really only just started to use them for stylized stuff, however this artists seems to have done it with PBR materials. Def check out his portfolio, 

https://www.artstation.com/artwork/86vOw

AlecMoody

Looking at your work you should slow down and focus on quality with smaller props. When I look at prop art I would much rather see something smaller where the artist took extreme care than something larger that is simplified or sloppy.

I would suggest you focus on how things fit together, how things are constructed, and getting the surface texture/detail just right. At an old job we used to talk about texture sheets as having a love per pixel ratio- Same idea here but focus on the high res art.

In your case, I would skip the low and bakes for now, and work on just making great high res art. After you have that down you can move onto learning to bake well and build quality materials and lighting.

Some suggestions for first projects:

A small leather bag. Don't use marvelous for this, the whole point is to learn:

https://cdn.shopify.com/s/files/1/0043/9252/products/the_railcar_duffle-1.jpg

A big industrial globe valve:

https://www.indiantradebird.com/itb-admin/members/176/images/0_0877a5df.jpg

3dcaster

Hello,

Thank you for the guidance. 
I guess making a new scene is going to be a good idea for later. 
First i will start with a simple prop and try to ace it. 
enxample below. 
http://nutool.pl/536-thickbox_default/wiertarko-wkretarka-12v-npk12l.jpg

Than I will make exaples of from the reference. 
The valve seams pretty easy once the pipes are in place but the bag looks waaay out of my comfort zone. 

@Zi0 - thanks for the tip! 

Thanks all for your help. 
I will deffinietely go back to this thread in future to reflect of my work.

have a good one!

Zi0

In that case I am going to subscribe to this thread. Good luck you can do it!

3dcaster

Hi. 

I could not post earlier because I was actually doing an art test in Unreal, for the Skillshot post you recomended me @Zi0. thank you for that !
Below is my progress on http://nutool.pl/536-thickbox_default/wiertarko-wkretarka-12v-npk12l.jpg.
I also bought "ultimate hard surface tutorial" by Simon Fuch. 
https://gumroad.com/l/spEkf 
There are some pretty amazing techniques there. I started making the prop below using them. 

Question. 
What do you think about quality of the sculpt? should i aim for the spotless edges? to be honest some of them seams to me a bit jagged as per .jpg below.

I want to know before i will proceed to the main bodypart and spend a lot of time on it:) 
Thanks!

Zi0

Hi,

I'm glad to see that you applied for the job at Flying wild Hog I hope your art test goes well and you will get the job  

Are you trying to sculpt that prop? I never did the tutorial you mentioned so I don't know what techniques they use in it but I would go for the Zbrush boolean workflow. That way you can make these parts very fast and the result will be clean and not jagged up which will show on the normal map I'm afraid. I quickly made one of the parts using this method:

In case you don't know this workflow here is the video about it: https://youtu.be/Fg6QPJ7XirE
I use this method a lot for hard parts that would take a lot of time using subDs.

If you want to nail this prop you also need to look at proportions:

The rounded shapes on the prop are a bit more oval then in your model.

Don't get demotivated and never give up!

And let me know what the outcome of your art test will be

Good Luck!

defragger

don't sculpt this! It is a classic modelling job. Use the right tools for the job. You can use the dynamesh method mentioned before but on this drill prop the front parts are round so you can model one piece 45° or 30° like a pizza piece and clone it round.

3dcaster

Hi All .
I wanted to post immiedietly however i had an ongoing Fusion 360 /MAYA/ Substance Painter paid project going (first renders dirty renders below)

Below is my take on the Boolean operation workflow in 4r8 and it is amazing to work with. 
Is that the quality I am after?

Please let me know your thoughts.
Thank you All  

@defragger -  thanks for the suggestion. I know that i  can model it but I am trying to get much better in Zbrush Dynamesh/ boolean / Panel loops and I feel that this drill will be good practice. 




First renders below. of the Fusion 360 / Maya / Substance painter workflow project.

Zi0

Hi,

If that strange texture on the drill part is only a material and the geometry is smooth than its good, could use a deformation->polish

3dcaster

Hi again:) I worked on quality a bit. Model is still far from being finished but i wanted to ask about the bevels. Please see the jpg below. 
The technique i am usin is pretty similar to the one described in your video @Zi0 but there is a twist. 
In Maya I am separating UV with hard edges and spliting UV islands using script. Than in Zbrush I can " Auto group with UV"  and finally "Mask borders" This gives me ability to polish edges separately form the rest of the model. Sometimes I can unmask edges that i do not want to be polished at all because they will be polish later on when I will boolean new geometry. 

Please let me know your thoughts! and as always thank you for your help! It really made me try harder!

Zi0

The chuck should be a bit larger to match the reference more. Personally I model things in subD and use zbrush for hard pieces to save time so I'm curious to see how you are going to create this only using zbrush.

3dcaster

Hello again. 
So i went ahed with the sculpting. 
I am not sure if it was all that time efficient because my Zbrush is seriously instable.
In the end I have a finished sculpt/high poly below. There are still some normal map details that i decided to paint directly in Substance Painter.
Is it clean enough? Can I start low poly modeling? for the record, at the beguining i used Maya for the general outline of the shape. 

Thanks and have a good day!