WIP Landscape, need some help

[Phlakes]

Well, I've started work on another project that will never get past pre-production, or maybe half a simple engine if I'm lucky, and I need this landscape thing. I made a quick paint over a while ago and I've been putting it off until recently, but naturally I'm stuck. Here it is-



Mine in the top left, reference to the right and dithering ref in the bottom left.

Now, this is one of many, many variations, mostly different palettes, and I think this one looks the best. But first, I've always sucked at dithering. It's been that one thing that I've never been able to get right. But I think I have the technique down now, I'm just not experienced enough to apply it well. Looking at the one mountain, I know it has basically no depth, and I'm pretty sure it's because I just made a straight gradient. So I'd like any kind of help from anyone more experienced, so hopefully I can get this down sometime. And obviously the sky was a failure. I'll get back to that later.

Also, this is my first big project in a REALLY long time, so I'm out of practice with larger-scale shading and textures.

[Juniorps]

Look, I see some pictures in this style or model of mountains and sky for example, the more distant mountains are not as detailed, and I think you should do the same because maybe the image is drawn very detailed with all the rocks, sometimes things Simple is better.

[ErekT]

Yeah, lose the dithering. It's not that it's badly done, it looks ok to me. But the dither gives your surfaces a sense of detail you wouldn't be able to make out at that distance, unless they were covered in grey-ish brush.

[st0ven]

So the major problem i have when looking at your rendering using various references, is that youve used one reference to judge color, and the other to inspire technique, but looking at your interpretation of both it feels like youve entirely abandoned composition, and i say this because when looking at it, i cant make heads or tails of what is happening in the fore/mid ground at all. Your current progress on the sky in the backgorund is simple and nice and using all the right colors from your reference (and will look much smoother with some quality dithering), but in the foreground i dont know if that dark blob is supposed to be a mountain top or a cloud piece or what.

What i recommend, is that you not worry at all about dithering or detail until you use just a solid hard brush, and scribble out your composition in a way that it can be coherently understood. Know that your more contrasting values are going to bring things closer to the foreground, and that the light source is coming from the left side of the screen, either at, or just below the horizon line of  your sky (this is why you dont see a lot of hard, bright outlines on the clouds. presumably the sun is already beyond the horizon). I roughed in a mountain in the fg and one in the bg, and you can clearly see that the darker one looks much closer thanks to the contrast. Right now whatever you have peering up through your clouds is too dark vs the clouds that are occluding it, making it look like its actually closer to the fg than it should be, and i think thats creating a lot of confusion to the eye.



Again i dont know what your original intention is with the image, ive read the post a few times over and i dont really understand whats happening even with the brief description. But i did start to doodle something on my own and i did it to illustrate that its probably MORE important to be doing a rough composition first before you start worrying about dithering and smoothness. Youll have worked out most of the hard stuff in this phase (like values and color), and that makes the blending very much procedural if you wait until thats all figured out beforehand.

[Phlakes]

...That is extremely helpful. I think part of the problem is that I was trying to stay too close to the ref, which doesn't translate well into pixels considering how the fog works and how it's made of mostly solid colors. Anyway, I'll be back with an edit, and I'm sure I still won't be there yet.

[Phlakes]

Alright, got back on track. I really played off st0ven's edit here, because after screwing around with the original I realized it was beyond saving. Plus, it's good to have guidance like that.

Just to clarify, I only used it as a reference, although a very close one.



I'm very happy with the left half of the big mountain, but I'm thinking I'll need another color for the right half.

[pistachio]

IMO, your latest update makes the mountain look too smooth as opposed to st0ven's edit, which gave it a fair deal of ruggedness and texture, making it look rather realistic. Even if you squint this is still obvious, while yours almost looks like a mountain of sand. You might need some dark bits mixed into the light and, occasionally, vice-versa. These would suggest imperfections.

[Phlakes]

It's supposed to be a bit far away so I don't want to make the details too big, but it's definitely better.

[EyeCraft]

Good improvement! One (imo) very important thing you've lost from st0ven's edit to yours is soft edges. All your edges are quite well defined, whereas st0ven's version has a whole bunch of delicious ambiguous forms all melding into eachother, an effect you would expect with heavy fog/clouds.

There's somewhat more contrast in st0ven's as well, due to do darker midtones creeping a bit more into the middle horizontal third of the composition, as well as the bold shadows of the biggest mountain pushing up infront, as well as cutting sharp edges in the form (which contrast wonderfully with the soft edges of the background).

Bleg kind of rambling... I just woke up  

[Phlakes]

Well, one thing I've been taking into consideration is that there'll be a big title at the top. I want to leave most of the detail and contrast near the bottom to keep the title readable, so I moved some of it down further than it should be in regular pixel art.

Also, bit of an update-



I do wish I had kept some of those sharper angles and fuller colors, but it's a bit too late to be getting back to that. I barely have the time to work on this at all, definitely not to go back and rework the foundation of it.

But next time. This is good experience, at least I'm not stuck where I was in the OP.