Messages

Oh my gosh I loved this so much! Amazingly fun idea, great aesthetic and the gameplay really worked well (although maybe people would prefer to have autosaves before death bits, though it didn't bother me in a game this short).

Final score: 1296.

Thanks for this one!

Maybe I did! Who can tell what I say with that accent?

I tried using the navigator window but couldn't quite get used to it, so I went back to doing the slow way. I wish I could get used to it, I use preview windows constantly when doing animations, but somehow it just doesn't feel the same when painting scenes.

A great way to combat this is to zoom out regularly and see the big picture, the whole thumbnail again. When I detail I sometimes zoom out several times a minute, just to make sure new ideas aren't throwing the whole thing out. Even when colouring I sometimes use a layer set to "Saturation" blending mode filled with pure black, which strips all the colour out and allows me to check if my new colours aren't throwing the whole thumbnail off again. Obviously not a factor when detailing as you've chosen to do here, but figured it may be interesting advice for some.

However, I'm a bit stuck at the moment - I have tons of windows in my pic, but how should I colour the glass? I googled for tutorials, but they mainly show how to draw a window frame, not the glass...

One suggestion (and what I try to do) is to think in terms of the layers of light that happen on the glass. So if you paint them in order you have: The room behind the glass, and any curtains/blind that might be near enough to the window to be ssen, the glass itself and any colouring that dirty/tinted/stained glass might be adding on top of that room layer as well as any direct reflection of the light source on that surface, and then the surroundings that are reflected on the glass if the room behind the glass isn't brighter than the outside.

Here's a quick step by step that might work for you:



First I put the room in, which is in shadow. Next, because I'm painting without curtains, and because they're plain windows, I just very lightly put a bit of the white light of the sun over the top to show me where the glass is. I keep the bit off to the left unlit because the frame of the window is going to block the light off here. Finally (and the third and fourth panes of glass are the same here) I took a tiny bit of very transparent blue and just touch the lit area, showing the blue of the sky that we can assume with be reflected at this angle. I also boosted the contrast of the frame on the second window to try and show the depth of the structure a bit better.

This isn't some official method I've read or anything, just the way I usually do glass when I need to. If it helps you then that's great!

My first thought is that taking that boat shape out opens the image up really nicely, and probably acts as a decent, really clear south exit, and the part I had intended to be the exit now blocks my eye off from thinking that's an exit really nicely. I actually really like how much more open the comp feels like this, it's a really interesting contrast to what I tried to do, thanks!

I want to do some measuring and stuff of your other ideas, but so far this has already been quite informative and revelatory, thank you!

I never had to deal with this "doesn't look believable" shit when I was painting neon green skies.

Jokes aside, and as for c&c, I think it's always useful to get another opinion on stuff to break down what could be done differently!

I think a lot of it is build in experience, as you say. I don't think about every single cloud, that would be incredibly inefficient, but I've painted enough clouds by now that I know what I can usually get away with. Your hand and mind kinda go on automatic, like forming chords on guitar.

Making shapes without lines vs making shapes with outlines!

...sorry

One thing I noticed here and with other artists is that "lines first" artists tend to flavour forms with texture in the line phase, before the colour phase, which is kind of the opposite of what I'm used to (see here with the bumps on the tree trunk, stone details, tiny windows, etc). It's an interesting thing, because here it feels like you already textured your forms a fair bit, and spent a lot of careful time detailing them, and are now spending a lot of time carefully detailing them all over again, whereas the format suits me super well because I slap big rough areas around for my shapes first, then do the little details last. I'm lost whenever I try to draw any lines, let along careful ones like this, and it's interesting to compare.

I'd love more focused workshops. Even without the need to take it to a "finished piece", I think there's worth in just helping each other improve a single core aspect of drawing & sharing processes. I loved the value study portion of this workshop alone, and felt like it could have been its own thing, without the need to go "right, you spent a whole week tweaking this value study, now do a background from it" after.

Stuff like doing workshops on essential stuff like getting a solid silhouette/thumbnail, establishing an interesting colour scheme, drawing believable stonework, experimenting with details on a single object could all be unique activities that would be very valuable without having the whole fatigue set in, I think. At the moment there's so much to learn, which is awesome, and I really value all the knowledge being shared, but it also makes sense to learn in the little steps you mentioned.

First, and I think you agree, the placement of the character is less than ideal. This was something I was afraid of in my piece, which is why I included them early; the disadvantage being that I had a lot of emptyness in the layout.

I'm a bit wary of this approach, simply because all characters - whether player or not - have the potential to move through a scene over the course of a game and I think a piece should work without characters. So many screenshots of stuff I've worked on have made me cringe just because some journalist or programmer took a shot with the player character in a really weird spot and threw the balance off completely.

I also think it's a bit much to add another stage to the workshop where we design characters, mostly because I think designing, iterating and drawing a character is a separate, very different set of skills and ideas and worth exploring in a completely separate workshop. I'd love a character workshop or something - naturally, once everyone has finished their piece and we've all recovered from such intense drawing/feedback/critiquing over the last month.

That also means people who don't care much for background art or don't have the confidence to work on a whole scene yet can join in, too. Maybe yourself or someone else here would be interested in running it (or maybe a few people together, seems to work well!)

Those are my thoughts!

I feel like I've entered a forum of masonry enthusiasts suddenly.

Interesting point, though, and again, I know basically nothing about building steps, so thanks for your insight!

That's an interesting point, and shows that I didn't really research cobblestones enough before drawing. I've edged it to keep it a bit more "real" looking.

And with the detailing done, and the due date just around the corner anyway, I'm submitting this as my finished piece. I'll probably share some ideas on what I could have changed/done better in a few days once I've got fresh eyes, but for now feel that trying this method was quite educational, and am interested to see what other people feel I could have done better. Thanks to loominous for hosting & JudasFM for the theme and everyone who entered, shared their process, gave feedback and helped keep me motivated to work on the piece. Look forward to seeing all your finished works. It's been fun and educational!

[Misj':]

Misj': New colours look way better. I'm curious to see how strong your final thumbnail reads, it's still kinda hard to gauge while you're doing your detailed values last. It's an approach I've never really seen before.

With regards to science vs romance, I think painting is a series of decisions and then lots of working those decisions in. The more theory you have to inform your decisions the easier it is to achieve the ideas you have in your head, I think. Eric recently linked me to an AMA with a background artist from Song of the Sea and it was interesting to read how they specifically put many tangents into the backgrounds for the film in order to get the look they wanted. Shows that knowing a theory doesn't necessarily homogenize your art, it can even help diversify it by showing you the pattern your work follows and give an understanding of how to break that pattern, I think.

As for my piece, I'm nearly done. Been using some reference photos to allow my designs of stuff to be better informed by reality (I stole this term from Ilyich, it's a good one, thanks il-chy ^_^) seeing as I'm not really going for a "cartoony" look. Included them here to give an example of how they influenced my designs. I've got some more cleaning up to do tonight before I call it finished, but I'm pretty close.



As for guiding the eye with colours and stuff, saturation is probably as good for this as hue. I have orange foliage over a yellow sky, but because the difference in how vivid they are is so big the tree stands out. If I bumped the saturation of my sign up to the same as the dark wall behind it then it'd be the exact same colour, but because it's less saturated it stands out from that wall. I think the real thing to keep in mind is that the eye is caught by all forms of strong contrast, no matter how you achieve it.

[edit]

This is something that I found often with people using the value-first approach. And faking colors can be great, and relying solely on value and saturation will often create a unified look and feel.

...

The difficulty is, that each approach makes people afraid to loose their 'starting point'. Value-first-artists tend to be afraid to loose the feel of their values when they go bold with hues, line-artist are afraid to loose line-details when adding shadows or values that may obscure them. People relying on a color-first approach are afraid to use saturation because it will mess up their predefined colors.

I actually very, very rarely go with a values first approach, I start with all three and work as I go (I didn't for the sake of joining in this workshop) because I like to plan out my hues & saturations very carefully at the start. Means I always get a tight palette that gives a unified look. Look at James Gurney's thoughts on gamut mapping for an example of the sort of idea kept in mind when I start working.

Therefore, rather than being afraid to use saturation, I incorporate it into the planning of the piece, despite picking my colours first. That's not to say I don't make big changes along the way (obviously, as you can see by my progress in this piece ).

I've done some more refining and tweaking. More tidying up. As I saw Daniel say earlier, rendering really is the biggest part of working on a scene, putting in all the details and trying to get it all consistent. You can see down the bottom I've made some decisions about shapes that I put in just to put some nice forms in that area, here they become setting appropriate objects without losing too much of the form. I do this quite a lot, put down a design without thinking about what it will actually be and then decide on the actual object once I have the form looking kinda how I want.



edit And I spent another hour working on detail, texture and stuff. Ilyich reminded me that my ground was a big boring flat plane, so I've sketched in some paving to break it up some and gone over it with a multiply layer and a dark colour to drop the value down a little bit and push more focus onto the harbor master's building. Darkened the sign to make it pop more against that big, bright wall, and just various texture painting.



There's no real secret to this part, I don't think. If I planned everything pretty well then the texturing bit is the relaxing bit at the end where I can just put on some nice music and paint away with a tiny brush, making nice patterns and stuff. It's more reliant on being patient than problem solving. Sometimes for work (where I spent 1 day on most backgrounds, as opposed to 1 month like here, because I usually have 70-80 to do per game) I'll use photo textures instead of painting texture in, just to speed the rather painstaking process up. The end result is never quite as quaint or handpainted, but it's a decent way to make the most time consuming part go a lot quicker.

I'm right on track to be done by the 6th. Still have to work that boat into the scene much better and go round and tweak everything more, but it's shaping up how I want it and the road to the end piece seems pretty clear at this point.

At this stage I'm mostly just rendering detail.

Could you elaborate a bit on that? I'd be interested in how you are working here - what kind of brushes you use (hard/soft edge, opacity,...) and your use of layers.

Yeah! Here's a look at my "workhorse" brushes, that I use for 99% of everything:



With all three of these I have the tablet's pressure sensitivity to affect both opacity and size. I also use the mouse quite a lot, for things that need more precision. I also always modify the size and opacity of the brushes manually constantly with the [ and ] keys (size) and the 1 to 0 keys (opacity).

The hard round brush is great for clear edge definition and precision work. I usually work with it quite small, for cleaning up messes, but it's also reliable for filling in large areas of flat colour and things. The best thing about it is that it never "spills" outside your edges. The worst thing is that it takes ages to blend and look natural, you end up with a lot of flat strokes you need to blend out if you're using it for shading and colouring.

The oil medium wet flow brush is one I've only been using a few months, I switched from the chalk 36 pixels to this because I used to have to set the chalk brush's "scatter" to 2% manually to get the edges to wander how I wanted, and this does it automatically. This is great for both dabbing down in a series of pokes to make textures form and for making more "natural" looking strokes than the hard round brush. I use this for the majority of the work in a piece like this where I want a less smooth, clean look. It can be a little hard to use at lower resolutions (320x200, basically) because it works better the bigger it is, but it's still probably my most favorite brush at the moment.

The soft round brush is one that I use at low opacity, and really big. I drop the brush down to 2% opacity and make it much bigger than the target surface, and that allows me to let the light bounce out and this makes stuff like light sources or areas of light colour being hit directly with a strong light really glow. Often I'll use this with an overlay layer (for making something lighter) or a multiply layer (for making something darker). I flatten those down as soon as I've put the colour onto them, so I don't get confused.

Other than that, I mostly just keep it to a single layer that I paint on, fixing up things just by painting over the mistakes, really. I always have my grid of thirds and my perspective grid on separate layers so I can refer back to them, to fix up the inevitable mistakes, and sometimes I'll keep my earlier version on a separate layer to refer back to it and make sure the changes I've made are ones that look better to me. I also convert back to greyscale now and then just to check and make sure my hues and saturations aren't throwing out my values. Because my early versions are so messy, it doesn't really help me to use many layers, and using a single layer for the painting itself means I never get lost and paint on the incorrect layer.

Let me know if I missed anything!

Very sorry to see you bowing out of the activity, I've really enjoyed following your progress and your piece as it is now is really shaping up nicely.

Another update from me, with another hour's painting or so. At this stage I'm mostly just rendering detail. I've still got a post about colour ideas coming, I promise!

Selmiak: Yeah, I'm using an Intuos at the moment, though I hope to some day get myself a Cintiq. I like the Intuos a lot, though, it's served me pretty well over the last year or so, and it's functionally identical to the Bamboo I had without one or two annoyances of shape etc. (less accidental pressing of buttons and stuff)

Misj': Yeah, wild variances in hue are the hardest for me to make look "good" so I mostly like to trick the eye into seeing different hues rather than actually using them. Part of cutting my teeth on pixel art, really! I wanna write a bit about this (still haven't gotten around to it ) but it's definitely an intentional thing. I find I get the "tightest" designs when I rely more on value & saturation contrast. Also means I can save new hues for little detail objects (though in honesty I rarely do this haha) Really interesting to see the different approaches.

Here's my progress - at this point I'm mostly just adding detail and cleaning up the loads of mess I made in the planning phase. This is the slowest part, but also the most enjoyable because it's less about trying to work out how to solve the problems in my image and more about refining all the groundwork. I'm enjoying cleaning up this piece, and making steady progress.

I retired my tablet from that era because the cable was getting worn. Sorry to see someone else having a similar issue!

Also it makes me sad when people try to put games/art into boxes. Just let stuff stand on it's own you know?

The intent is not really compartmentalization, more the grouping of like pieces in order to analyze and compare different approaches to the medium, and also differences within similar takes on an approach.