I think you're holding yourself back as an artist by bogging yourself down with needless constraints and styles. I'm not sure of your end goal, and whether you wish to work in the industry, but I do assume you want to improve, as all artists should. But to improve, you have to do a good amount of work. I could be wrong, but it seems like all these constraints and "theories" are preventing you from actually doing.
I'd suggest looking at why you go into all the detail and theorizing; and see if it's because you love theorizing, and the concept behind art, or whether it's because you don't want to just sit down and take a risk putting all your energy and force into art that may well end up not being as good as you like. Art's a scary thing, in a lot of ways. And it does suck to put effort into something, only to have it not work out. But it's the only way to move forward and improve.
All I mean to say is, make sure you aren't deceiving yourself. If you truly want to improve, you'll have to invest the effort. Even when it's boring, and even when you fail.
I know it will be hard to believe, but it is this thread that makes me want to work in the industry and is helping me to get there faster.
Even when I post art that might not be good, I recognize that I made the effort to still post something consistently and learn from it. As opposed to starting with an idea (say: an AAA gun), maybe post 1 or 2 things about it, and then never post again in a long while either because something went really wrong or if it just ends up being too much work to do.
Basically, I'm getting into the habit of posting art while being able to show something for it.
Csn you post the scene that your flat-coloured lamppost sits in? As it is, it looks stumpy and short and lumpen and inelegant, at odds to the reference you keep posting. It might look fine in situ, no idea right now. If you don't actually have a scene to place it in and judge accordingly, I don't think it's possible to be so sure if the style choices you're making, other than feeding off simple Intrinsic gut feeling ... But from someone who has spent years putting together scenes, my gut feeling is that, yeah, in a scene it will be stumpy and short and lumpen and inelegant .
Annnnnnnnnnnnnnnnnnnnnd now it's time for a big update!!!! The first handpainted textures are up. There's a lot of detail here that may be hard to see.
I've added some cute patches of grass. This is my first drawing running very smoothly in Unreal Engine 4.
They don't use polygons. They're completely 2D against some 3D backdrops.
I've also painted some ground textures that can only be seen from the top.
What speaks against a more default approach of learning 3D and developing skills? Just find a concept or a real life model/Scene you like and try to rebuild it and learn modeling, mapping and texturing this way? I think you´d be way better off that way.
those grass tufts - are they just PaperSprite objects? If so, they are 3D. They're 2 triangles making a plane - you can see that in the wireframe - rendered along with everything else in 3D space, exactly the same as if you exported a plane from a modelling package. They might not have any depth, but that doesn't make them 'completely 2D'.
Its convenient for Unreal to refer to all of the Paper stuff as "2D" in order to differentiate that side of the editor, but if you're just going to drop them in as planes in a 3D scene, then 3D they are and they always will be. Again, its important to understand what you're dealing with and get your terms right, so you're not confusing others but most importantly so you're not confusing yourself.
Tbh I don't get, at all, how a 2d trace of a car is a 'prop'? Can you elaborate how you intend to use this?
Btw, unless I've missed something vital, shouldn't this whole thread be under the 2D subforum? The last 'complete' image was a screenshot that was painted over to get the final result.
Tbh I don't get, at all, how a 2d trace of a car is a 'prop'? Can you elaborate how you intend to use this?
Btw, unless I've missed something vital, shouldn't this whole thread be under the 2D subforum? The last 'complete' image was a screenshot that was painted over to get the final result.
The prop is from a 3D model that will be used in the gas station. I drew it over till I actually place it in the scene.
As for why this is not under 2D, I made this thread different as I want to focus on making an artstyle for a potential 3D game. There is actually very few paintover as all the assets are meant to originate in game. Only the depth of field and toon outlines were added in (but I could still render them in-game if I wanted to). This is an unedited real time screenshot of the final scene.
I don't understand what you said there ... you have a 3D model of the car already? if so, wouldn't it make more sense to post that, rather than a trace? If not, tbh I don't think these traces are going to get you very far. Maybe increasing your confidence in controlling your line work and keeping it clean ... Which is fair enough as part of a practice exercise I guess ... but at the moment you don't seem to be applying this to your own actual drawings, the ones you have posted anyway, which remain without any real quality of line. Outside of practise, these traces don't appear to contributing to your project at all, and posting them as updates is just confusing. Apply it, or leave it out, would be my advice to you here.
speaking of tracing ... Your right, the dof and toon outlines could and should be done in engine and very very easily too. Control the dof with a post process volume, and maybe do outlines (most quickly and easily) using the old school push'n'flip method ... Clone the object, push the polys out along the normals, flip the normals, apply a black material with no back faces. 10 seconds to do and still effective in 2016 for this sort of simple scene. More complicatedly, you would do it in a material. On the whole that would give you a much more polished and consistent result to your end image (your 'complete' image has glaring errors on both of those elements), as well as being more in line with what you say you're trying to do, so it begs the question why you didn't approach it originally. You'd also probably get more feedback having presented an actual 3D scene rather than a paint-over which you called "done".
I don't understand what you said there ... you have a 3D model of the car already? if so, wouldn't it make more sense to post that, rather than a trace? If not, tbh I don't think these traces are going to get you very far.
I changed the presentation so it now matches the ones The Iron Giant had. People have now seen what my 3D models look like, so I don't need to point them out anymore.
Clone the object, push the polys out along the normals, flip the normals, apply a black material with no back faces. 10 seconds to do and still effective in 2016 for this sort of simple scene.
lets be clear here, to iron out some of the confusion : a "prop" is an asset. Thats all anyone working in 3D will ever describe it as. A tangible object, that you can place, select, move. Your lamppost, that's a prop. A line drawing, thats not a prop. Thats a drawing. Because it looks very much like a vanilla trace, without any sort of modifications on your side, it barely qualifies as a concept, any more than the original photo that i suspect you traced over does.
Is it a style test? Possibly, but you'd want to work it up more with colours and shading and sense of how it will be rendered to be a decent style test, because line-work is only one part of this. You started doing some of this earlier, it looked like you were thinking about advancing the style, thinking about lighting and so on, and then the updates suddenly revert back to another basic trace. Not a prop, not something you can use in a scene, not a development of the style or a suggestion of a new technique. Suddenly, its going nowhere. All you've done is post an update of how you killed a few hours doing the same old thing you were doing at the start of the thread. Its advancing nothing but post count. Do you see what i mean? Do you believe that retreading the same ground is useful to you? I'm trying to be helpful with that question.
Claiming that showing 3D models - on a forum specifically designed for showing off the development of 3D models and receiving feedback about them - to be pointless ... well thats just silly in the extreme. Nothing more to say about that.
On your last comment about the outlines and DoF. Well, its up to you. Personally i think you're wrong, there are fundamental errors in what you've done which totally destroy the image, which would immediately be solved by doing it in realtime. Since i already mentioned this but you decided that you 'prefer' it, i'm not inclined whatsoever to point these out , it would be more beneficial at this point for you to look at it critically once more and try and see these errors for yourself. Think camera, think z depth.
(Also on that last point, well regardless, you're admitting choosing to favour what would be termed a complete Bullshot, with major core components of the image done after the event in photoshop . This makes the [UE4] tag in your title pretty much redundant)
I retracted my point on the outlines/depth of field. Originally, I wanted to treat my final image as a concept and draw in what I felt would make it perfect. But since I've already made the effort to generate everything in engine, I might as well go 100% with it or as you said, it just becomes a bullshot.
I never claimed showing 3D models was pointless. I wanted to say it was to help streamline how I present new art pieces. I want the focus to be more centered on what each final prop now looks like, instead of any intermediate or unfinished details.
You´d do yourself a favor showing less art, less test and whatsoever and instead showing plain 3D models, untextured and textured, UC maps and shader test and how your 3D work looks in a nice render engine. I feel like nobody has an idea what you actually do or want to do since there no real 3D work here.
I'm confused about how the 2D cars and guns relate to the above scene. I like the look of the scene, and it seems like it'd be more functional to design the scene, make a list of needed props/sprites, make those, and assemble the scene.
I'm confused about how the 2D cars and guns relate to the above scene. I like the look of the scene, and it seems like it'd be more functional to design the scene, make a list of needed props/sprites, make those, and assemble the scene.
I've got a story that a whole bunch of scenes are tied to.
The guns will make sense when I get to revealing the Canadian Army and the Devil Girl blowing stuff up.
You have complete control over your image with dynamic and baked lighting. Thats a basic core feature of every decent engine, and in ue4 Epic have worked hard to make it so. What specifically do you think you'd be missing by using it?
UE4 has a post process that ruins my cartoon. Its tonemapper automatically clamps values so they appear more "grey". You can see when I first posted the Lamp post that the colors are all wrong.
I've gone into the engine and removed the code that does it before, but then the lighting becomes useless (the engine is built around this feature).
Its not built around it. Set a post process volume for full control over everything it does.
The tonemapper can't be disabled in the post process volume. It's why I said it can only be removed through code. Playing through all the settings just to get a color is annoying so I rather choose to limit or not use any lighting at all for this.
Try and excuse the irritation here, but fuck me dude. Seriously, you are in no position right now to be worrying about slight tonal shifts, whether you can control them or not. That's SO far beyond the fundamemtals I can barely describe it. And let's be clear, you are still building up to the fundamentals. You are. And still a way off. Harsh? Maybe.
If you had the fundamentals nailed, it would tell you that vertex colours are not the way to go. Leave aside the engine for one moment. Imagine you don't have one, and your model is previewed purely in the modelling program. Can you not see the restrictions of manually shading each vert, the natural blocky look to the colouring, that you're hampered at every step by vertex placement and edge direction and the need to carefully author every single vert and edge to get the look you want? All to try and get around some debatable gamma issue [edit - totally irrelevant since post process means that whatever you produce will still be affected , whether it's dynamic/baked lighting or vert colours or a picture of a unicorn] and when even slightly experienced artists will be 99.9% sure you'll still get worse results? Give over.
I went back and used the real time lighting instead. I still had to deactivate the tonemapper that was giving me a lot of problems but at least now I don't have to use the vertex colors anymore to preserve my cartoon.
Look man, I dont want to beat you up. But, you should nail getting good at one thing.
If you want to do cartoons, and that art style. Pick one thing and go ham on it. Your really slicing up your workload and your going to learn at a snail's pace.
But, shit. do what makes you happy. Just my 2 cents.
Hm, really curious about that one. If it is just some line work, then I really don´t get it. If it is one of the 3D works Jordann was talking about, then I´d say, dude, show us more stuff like that. That´s cool.
Here's a tip; for the "outline" mesh, set the metalness to 1, and the base color to black. That'll make it solid black, the way you're probably hoping for.
ok ok , now we're talking... before you were just playing. But now!! Now we're talking. The only thing i would say is that it needs a bit more yellow in the sky. Just that. Some more yellow.
Other than that. It's perfect. Leave the perspective lines also. They add to the character of the concept art. Really good!
Keep in mind, my goal for this is to replicate 90's animated movies so a lot of my colors try to stay close to the source material. Of course, I could break a rule if it can make it look better (but I think 90's animators had incredible color sense that I find it hard to deviate from it).
The perspective lines idea sounds fun though. I'll definitely try it out in Unreal.
@DANR I've been waiting for someone to explode lol. Bravo, sir. xD
@Jordann Besides the harsh criticism that Danr's said, the message in it is very true. At the same time I don't understand why you don't just paint your light into your model instead? considering that you're going with unlit shading and vertex painting in your lighting it would be more ethical to just paint it into the texture itself? if you really want to achieve the cartoon look and art style... why don't you just produce hand painted assets? instead of tracing things and trying to achieve this overwhelming workload of different assets and scenes with all this technical mumbo jumbo. There are literally zero benefits for you to vertex paint in your lighting.
On a side note, all this commotion about your colours being desaturated, it's adjustable in post processing (like previously mentioned) and if you really don't believe in UE4's system calculations/colour gamma, blah blah.... just grade your scene with colour lookup tables from a colour grade pass that you adjust in Photoshop, Bam, done... exactly the way you want it with full control over your colour and no more needless fiddling with UE4 colour settings.
I would say the worst part of your attitude is that you're receiving a lot of good feedback from artists here on how you should go about some things and so on... but every critique has an excuse or an explanation and most of the time it's a technical restraint that hardly dignifies itself as a reason to not do what the artist suggested... Just take the critique and build with it... Honestly, in the nicest way possible I haven't seen much improvement here... just post after post of new unorthodox techniques to recreate something that already exists within stable fundamental pipelines like PBR / Unlit Painted Texturing.
Again, like previously mentioned by @Stranger definitely just do what makes you happy... but if you don't care about getting better and you just want to do what you want... then just tell us that you don't want help and all you want to do is show off your work.
I would say the worst part of your attitude is that you're receiving a lot of good feedback from artists here on how you should go about some things and so on... but every critique has an excuse or an explanation and most of the time it's a technical restraint that hardly dignifies itself as a reason to not do what the artist suggested... Just take the critique and build with it...
Again, like previously mentioned by @Stranger definitely just do what makes you happy... but if you don't care about getting better and you just want to do what you want... then just tell us that you don't want help and all you want to do is show off your work.
It's not that I'm making excuses or explanation on purpose. I want to recognize all the critiques. It's just very hard to communicate these things.
I also want to make better art. It's why I'm on Polycount and I like posting here. I realize people look at this thread and may get confused but I'm hoping to fix this.
Keep in mind, my goal for this is to replicate 90's animated movies so a lot of my colors try to stay close to the source material. Of course, I could break a rule if it can make it look better (but I think 90's animators had incredible color sense that I find it hard to deviate from it).
The perspective lines idea sounds fun though. I'll definitely try it out in Unreal.
Oh my ... Before you go running off on another quest to find something else to distract you from learning how to make half decent art ... Jordan, he was taking the piss.
what he was saying is that your 'concept art' is at a level where it's almost impossible to give decent feedback. It doesn't serve as a concept because it doesn't seem to resemble anything that exists in the real world, or give a tangible sense of the fantastic either. As a 'colour' guide, it's just shapeless blobs where we can't tell what the various colours are supposed to be, so nothing to say there. As a lighting guide, well everything is in shadow, so it's all flat. The only thing to comment on there is that your cast shadows are massively off across the whole image. In general, you presented a nothingness that still managed to be littered with basic fundamental errors, including the guide lines that you still insist on scribbling everywhere but do not appear to be actually using. Breaking it down into proper feedback would send someone mental, so basically, all that was available to post was sarcasm. Absolutely not the right thing to do, but at this point in the thread, it was actually understandable.
my advice to you now : stop trying to pre-visualise this sort of thing in 2d. Your skills aren't there and the chronic errors you are making are going to confuse you. Don't rely on people to point out the errors to you - as you've seen, the inclination to do this is fast diminishing. Instead, mock it up in ue4. Use primitives and a ground plane, and a light source. It'll take you 2 minutes to do. Set your camera, and screenshot it. Compare against what you've drawn, and hopefully you'll see that actually using the 3D tools you have at hand will give you much more consistent and useful results for pre-viz work, as well as a base to build on with actual assets. However, we don't want to see that shit, it's only useful for you, as one of the many steps you need to take.
For the thread itself, my suggestion would be : go back to your tank and work on that. You have direct reference for this sort of thing in the film you've posted shots of. Build it up and texture it so the style matches or gets to a point where you're happy with it. Use the techniques you've picked up on so far (such as the outlining), see how it fits with the lighting and whether you need to do more there, such as using a toon lighting ramp. Balance the forms and the details with the rendering style so it's entirely readable as a tank, but hits the cartoony look you're after (as discussed further down the thread, about the Cadillac). Try it in a small scene with your lamppost, so you can judge relative scales, and also more interesting lighting such as the source being from the lamp itself. Do it all, ALL, in engine. Get something cool looking, something that people will be happy to give feedback on, and progress from there
danr said: Oh my ... Before you go running off on another quest to find something else to distract you from learning how to make half decent art ... Jordan, he was taking the piss.
what he was saying is that your 'concept art' is at a level where it's almost impossible to give decent feedback. It doesn't serve as a concept because it doesn't seem to resemble anything that exists in the real world, or give a tangible sense of the fantastic either. As a 'colour' guide, it's just shapeless blobs where we can't tell what the various colours are supposed to be, so nothing to say there. As a lighting guide, well everything is in shadow, so it's all flat. The only thing to comment on there is that your cast shadows are massively off across the whole image. In general, you presented a nothingness that still managed to be littered with basic fundamental errors, including the guide lines that you still insist on scribbling everywhere but do not appear to be actually using. Breaking it down into proper feedback would send someone mental, so basically, all that was available to post was sarcasm. Absolutely not the right thing to do, but at this point in the thread, it was actually understandable.
Man... how was I suppose to know it was sarcasm? Especially when I just had Kid.In.The.Dark say "JordanN, if you don't take artists all feedback you're not learning?"
On the concept art, I wanted to keep the thread's tradition (having at least one drawn art next to 3D ones). But I'm starting to realize though I'll post less of them (as I've done away with most props/character drawings now).
Okay I'm gonna be honest and harsh and say that your going to look back on this project, hopefully sooner than later and say "wtf was I thinking." You have far more to learn. This is not practicing or making you a better artist what so ever and when you create these projects trying to do something crazy like this your really just stunting yourself and putting yourself at a complete stand still. I think the best place for you if your not the type to want to follow good tutorials is the Monthly Art Challenge thread, participate and don't try to do anything crazy, first just prove to yourself that you can follow a concept 1:1 all the way through the pipeline. Good luck man I'm really just trying to help you out.
Okay I'm gonna be honest and harsh and say that your going to look back on this project, hopefully sooner than later and say "wtf was I thinking." You have far more to learn. This is not practicing or making you a better artist what so ever and when you create these projects trying to do something crazy like this your really just stunting yourself and putting yourself at a complete stand still. I think the best place for you if your not the type to want to follow good tutorials is the Monthly Art Challenge thread, participate and don't try to do anything crazy, first just prove to yourself that you can follow a concept 1:1 all the way through the pipeline. Good luck man I'm really just trying to help you out. I
INDEED! Sorry, but this is the harsh truth! I like that you have ambitions to work to the end. But this is not looking like it's really moving YOU forward!
Replies
I'd suggest looking at why you go into all the detail and theorizing; and see if it's because you love theorizing, and the concept behind art, or whether it's because you don't want to just sit down and take a risk putting all your energy and force into art that may well end up not being as good as you like.
Art's a scary thing, in a lot of ways. And it does suck to put effort into something, only to have it not work out. But it's the only way to move forward and improve.
All I mean to say is, make sure you aren't deceiving yourself. If you truly want to improve, you'll have to invest the effort. Even when it's boring, and even when you fail.
Even when I post art that might not be good, I recognize that I made the effort to still post something consistently and learn from it. As opposed to starting with an idea (say: an AAA gun), maybe post 1 or 2 things about it, and then never post again in a long while either because something went really wrong or if it just ends up being too much work to do.
Basically, I'm getting into the habit of posting art while being able to show something for it.
Csn you post the scene that your flat-coloured lamppost sits in? As it is, it looks stumpy and short and lumpen and inelegant, at odds to the reference you keep posting. It might look fine in situ, no idea right now. If you don't actually have a scene to place it in and judge accordingly, I don't think it's possible to be so sure if the style choices you're making, other than feeding off simple Intrinsic gut feeling ... But from someone who has spent years putting together scenes, my gut feeling is that, yeah, in a scene it will be stumpy and short and lumpen and inelegant .
I've reduced the scope so now they're only "mini dioramas" or "islands" instead of huge environments.
The first handpainted textures are up.
There's a lot of detail here that may be hard to see.
I've added some cute patches of grass. This is my first drawing running very smoothly in Unreal Engine 4.
They don't use polygons. They're completely 2D against some 3D backdrops.
I've also painted some ground textures that can only be seen from the top.
This thread was always focused on cartoons, but some props I've shown here may show up again later.
Its convenient for Unreal to refer to all of the Paper stuff as "2D" in order to differentiate that side of the editor, but if you're just going to drop them in as planes in a 3D scene, then 3D they are and they always will be. Again, its important to understand what you're dealing with and get your terms right, so you're not confusing others but most importantly so you're not confusing yourself.
Now, the next step..................... IS TO MAKE MORE!!!!!!!!!
I am creating a small gas station. Some ideas I've been pitching,
Btw, unless I've missed something vital, shouldn't this whole thread be under the 2D subforum? The last 'complete' image was a screenshot that was painted over to get the final result.
As for why this is not under 2D, I made this thread different as I want to focus on making an artstyle for a potential 3D game. There is actually very few paintover as all the assets are meant to originate in game. Only the depth of field and toon outlines were added in (but I could still render them in-game if I wanted to).
This is an unedited real time screenshot of the final scene.
speaking of tracing ... Your right, the dof and toon outlines could and should be done in engine and very very easily too. Control the dof with a post process volume, and maybe do outlines (most quickly and easily) using the old school push'n'flip method ... Clone the object, push the polys out along the normals, flip the normals, apply a black material with no back faces. 10 seconds to do and still effective in 2016 for this sort of simple scene. More complicatedly, you would do it in a material. On the whole that would give you a much more polished and consistent result to your end image (your 'complete' image has glaring errors on both of those elements), as well as being more in line with what you say you're trying to do, so it begs the question why you didn't approach it originally. You'd also probably get more feedback having presented an actual 3D scene rather than a paint-over which you called "done".
I'll look into this.
Is it a style test? Possibly, but you'd want to work it up more with colours and shading and sense of how it will be rendered to be a decent style test, because line-work is only one part of this. You started doing some of this earlier, it looked like you were thinking about advancing the style, thinking about lighting and so on, and then the updates suddenly revert back to another basic trace. Not a prop, not something you can use in a scene, not a development of the style or a suggestion of a new technique. Suddenly, its going nowhere. All you've done is post an update of how you killed a few hours doing the same old thing you were doing at the start of the thread. Its advancing nothing but post count. Do you see what i mean? Do you believe that retreading the same ground is useful to you? I'm trying to be helpful with that question.
Claiming that showing 3D models - on a forum specifically designed for showing off the development of 3D models and receiving feedback about them - to be pointless ... well thats just silly in the extreme. Nothing more to say about that.
On your last comment about the outlines and DoF. Well, its up to you. Personally i think you're wrong, there are fundamental errors in what you've done which totally destroy the image, which would immediately be solved by doing it in realtime. Since i already mentioned this but you decided that you 'prefer' it, i'm not inclined whatsoever to point these out , it would be more beneficial at this point for you to look at it critically once more and try and see these errors for yourself. Think camera, think z depth.
(Also on that last point, well regardless, you're admitting choosing to favour what would be termed a complete Bullshot, with major core components of the image done after the event in photoshop . This makes the [UE4] tag in your title pretty much redundant)
I never claimed showing 3D models was pointless. I wanted to say it was to help streamline how I present new art pieces. I want the focus to be more centered on what each final prop now looks like, instead of any intermediate or unfinished details.
I realize it takes a bit of time for me to absorb new information/crits before I can demonstrate it.
I'm basically trying to let this thread slowly correct itself, instead of abandoning it.
The guns will make sense when I get to revealing the Canadian Army and the Devil Girl blowing stuff up.
As a first test, I used my hand drawing from page 2 to give me an idea.
You can see when I first posted the Lamp post that the colors are all wrong.
I've gone into the engine and removed the code that does it before, but then the lighting becomes useless (the engine is built around this feature).
Playing through all the settings just to get a color is annoying so I rather choose to limit or not use any lighting at all for this.
By making it more "grey," do you mean less saturated, or less contrast, or both? Both of those are controllable in the post process volume.
The colors look more colorful in that first lamp shot, if anything, so I can't see what you mean.
https://answers.unrealengine.com/questions/44088/gammatone-washing-out-image-how-to-fix-1.html
It's a feature that's more understandable for realism but for cartoons, it not something I want.
If you had the fundamentals nailed, it would tell you that vertex colours are not the way to go. Leave aside the engine for one moment. Imagine you don't have one, and your model is previewed purely in the modelling program. Can you not see the restrictions of manually shading each vert, the natural blocky look to the colouring, that you're hampered at every step by vertex placement and edge direction and the need to carefully author every single vert and edge to get the look you want? All to try and get around some debatable gamma issue [edit - totally irrelevant since post process means that whatever you produce will still be affected , whether it's dynamic/baked lighting or vert colours or a picture of a unicorn] and when even slightly experienced artists will be 99.9% sure you'll still get worse results? Give over.
If you want to do cartoons, and that art style. Pick one thing and go ham on it. Your really slicing up your workload and your going to learn at a snail's pace.
But, shit. do what makes you happy. Just my 2 cents.
But, if he did model it. good stuff.
Not everything is shown/complete, this is just to visualize the lighting.
Now we're talking. The only thing i would say is that it needs a bit more yellow in the sky. Just that. Some more yellow.
Other than that. It's perfect. Leave the perspective lines also.
They add to the character of the concept art. Really good!
The perspective lines idea sounds fun though. I'll definitely try it out in Unreal.
I've been waiting for someone to explode lol. Bravo, sir. xD
@Jordann
Besides the harsh criticism that Danr's said, the message in it is very true. At the same time I don't understand why you don't just paint your light into your model instead? considering that you're going with unlit shading and vertex painting in your lighting it would be more ethical to just paint it into the texture itself? if you really want to achieve the cartoon look and art style... why don't you just produce hand painted assets? instead of tracing things and trying to achieve this overwhelming workload of different assets and scenes with all this technical mumbo jumbo.
There are literally zero benefits for you to vertex paint in your lighting.
On a side note, all this commotion about your colours being desaturated, it's adjustable in post processing (like previously mentioned) and if you really don't believe in UE4's system calculations/colour gamma, blah blah.... just grade your scene with colour lookup tables from a colour grade pass that you adjust in Photoshop, Bam, done... exactly the way you want it with full control over your colour and no more needless fiddling with UE4 colour settings.
I would say the worst part of your attitude is that you're receiving a lot of good feedback from artists here on how you should go about some things and so on... but every critique has an excuse or an explanation and most of the time it's a technical restraint that hardly dignifies itself as a reason to not do what the artist suggested... Just take the critique and build with it... Honestly, in the nicest way possible I haven't seen much improvement here... just post after post of new unorthodox techniques to recreate something that already exists within stable fundamental pipelines like PBR / Unlit Painted Texturing.
Again, like previously mentioned by @Stranger definitely just do what makes you happy... but if you don't care about getting better and you just want to do what you want... then just tell us that you don't want help and all you want to do is show off your work.
Either way, good luck with this.
Kid.
I also want to make better art. It's why I'm on Polycount and I like posting here. I realize people look at this thread and may get confused but I'm hoping to fix this.
what he was saying is that your 'concept art' is at a level where it's almost impossible to give decent feedback. It doesn't serve as a concept because it doesn't seem to resemble anything that exists in the real world, or give a tangible sense of the fantastic either. As a 'colour' guide, it's just shapeless blobs where we can't tell what the various colours are supposed to be, so nothing to say there. As a lighting guide, well everything is in shadow, so it's all flat. The only thing to comment on there is that your cast shadows are massively off across the whole image. In general, you presented a nothingness that still managed to be littered with basic fundamental errors, including the guide lines that you still insist on scribbling everywhere but do not appear to be actually using. Breaking it down into proper feedback would send someone mental, so basically, all that was available to post was sarcasm. Absolutely not the right thing to do, but at this point in the thread, it was actually understandable.
my advice to you now : stop trying to pre-visualise this sort of thing in 2d. Your skills aren't there and the chronic errors you are making are going to confuse you. Don't rely on people to point out the errors to you - as you've seen, the inclination to do this is fast diminishing. Instead, mock it up in ue4. Use primitives and a ground plane, and a light source. It'll take you 2 minutes to do. Set your camera, and screenshot it. Compare against what you've drawn, and hopefully you'll see that actually using the 3D tools you have at hand will give you much more consistent and useful results for pre-viz work, as well as a base to build on with actual assets. However, we don't want to see that shit, it's only useful for you, as one of the many steps you need to take.
For the thread itself, my suggestion would be : go back to your tank and work on that. You have direct reference for this sort of thing in the film you've posted shots of. Build it up and texture it so the style matches or gets to a point where you're happy with it. Use the techniques you've picked up on so far (such as the outlining), see how it fits with the lighting and whether you need to do more there, such as using a toon lighting ramp. Balance the forms and the details with the rendering style so it's entirely readable as a tank, but hits the cartoony look you're after (as discussed further down the thread, about the Cadillac). Try it in a small scene with your lamppost, so you can judge relative scales, and also more interesting lighting such as the source being from the lamp itself. Do it all, ALL, in engine. Get something cool looking, something that people will be happy to give feedback on, and progress from there
On the concept art, I wanted to keep the thread's tradition (having at least one drawn art next to 3D ones). But I'm starting to realize though I'll post less of them (as I've done away with most props/character drawings now).
You have far more to learn. This is not practicing or making you a better artist what so ever and when you create these projects trying to do something crazy like this your really just stunting yourself and putting yourself at a complete stand still. I think the best place for you if your not the type to want to follow good tutorials is the Monthly Art Challenge thread, participate and don't try to do anything crazy, first just prove to yourself that you can follow a concept 1:1 all the way through the pipeline. Good luck man I'm really just trying to help you out.