Hey everyone, I'll admit I'm pretty nervous about starting this. I've been a long-time lurker, but I think it's finally time for me to start here and really get serious about improving my art. I will do my best to post regularly now as I work (now that I said it hopefully I'll stick to it). So without further ado, here's some of my stuff:
Lately I've been trying to get better at drawing people (from reference right now) so I've been drawing a lot of portraits, mostly of pretty girls.
First off, here's a WIP of Belruel (fitting that my first post be PA fanart, right?). I believe I need to lower her breast level, and I should probably work on her armor more; add details and do better on the colors. I would say this is my best work of face and hair to date, if that helps to gauge skill level; I'd call myself a very early intermediate? For the background I actually just cut in the background Mike/Gabe drew in her comic in the last panel (meant as an homage to Mike). I intend to do the border in the style of a fancy old wooden portrait frame, and display the Wren standard at the bottom.
(This picture is fairly large so I'll spoiler it, unless people think it doesn't have to be)
Next is another portrait, this one inspired by a post on reddit. This is more of a sketch to practice hair and features, and also because Nordic warrior girls are rad. I'd like to possibly do more with her, finish the body and stuff.
Anyway, if you're reading this, thank you! I'm cautiously optimistic about starting this and I know with the help of everyone that comes by here I'll get much better.
Are you building up your figures with geometric shapes? Using guides? Or just winging it?
Keep at it!
This video is about 30 minutes but is worth watching:
For example I'll use some really old and shitty drawings of mine from like a million years ago.
Attempting to use colour and value in combination with boring single-thickness outlines:
Using contrasting values to define edges:
Hopefully this is a little progress towards a better style. One of the things I'm afraid of when I draw like this is the painting turning out too softly, with hardly any definition. I guess the style I'm really trying to shoot for is that of Marc Brunet (Cubebrush) and the way he works line work into his finished paintings as detailed in this video:
Anyway thanks everyone for your help so far, I'm already learning quite a bit.
Being able to compare what you're doing to a source will help you see how and where to push your values, which is pretty key to creating definition without using lines.
I did a GIS on "potraits with strong lighting" - check out the resuts. Do you see how the lighting brings out the models' features, and gives the photos more of a feeling or mood?
There's a reason why "use references" and "study your subject" is repeated here so often in every thread like a mantra. It really is the best way to improve.
I dunno. On it's own the painting definitely looks weird, but when compared to the reference it's pretty disappointing. But so it goes, onward and upward.
EDIT: fixed a couple more things, baby steps.
You probably didn't need me telling you to know that the drawing isn't right, so how do you fix it? Well, there are a lot of methods that people use to measure and correct observational work. Generally speaking the more novice you are the more rigid these techniques will have to be in order to help you. A good one is the 'plumb line', where you have some kind of vertical or horizontal line that crosses your drawing and the reference in the same spot, and then you compare how the features line up against this line in your reference and drawing to find discrepancies.
I placed this plumb line along the inner corner of the eye on both drawings to get them roughly in the same "spot". Immediately we can see that this plumb line reveals, among other things, that the mouth and jaw are much too far to the left in your drawing. In the reference, the contour of her cheek and jawline come to a nearly 45 degree angle, much more than in your drawing, and that's why her chin seems too square.
Many of the individual features are drawn inaccurately in addition to being slightly misplaced on the face, there's really too much to point out. The nose and lips are probably the worst offenders. Using measuring techniques like a plumb line can also help correct the shape of an individual feature, but a lot of it comes down to taking your time and being really particular about accuracy.
You can potentially use many plumb lines in the same drawing to give you many points of comparison to ensure accuracy, and they don't need to be vertical. Taken to its extreme with an exercise you may have heard about or possibly been assigned in school at some point is a 'grid drawing', which is really just a study like this done with many equidistant plumb lines arranged to cut the image up into a bunch of squares.
Ultimately your eye for accuracy has a lot to do with how much you practice observation, and as you practice more you will need fewer crutches to keep your accuracy in check.
Speaking very very briefly about the quality of the painting and rendering, which could be an entire post unto itself:
Use the biggest brush that you can get away with for a particular detail, don't make scratchy little marks like you've done on the lips, eyelashes and eyebrows. Think of those areas in terms of masses and shapes rather than individual hairs or cracks in the skin. I always tend to advise people to simplify their studies so that they can isolate their problems and not have to spread their attention between too many new concepts simultaneously, so I would suggest that you try doing more studies like this in grayscale rather than color, until you feel like you've got a solid command over rendering when just working with value.
Adding to what @Scosglen has said, I'd suggest taking a small piece of that reference and just work from that to see what you can do with it. It will also be less overwhelming, and it would allow you to really nail the observation on a smaller level and work more with what you are actually seeing instead of symbols that your brain has developed over time and piecing them together like a Mr. Potato Head. Think of this as just one square if you cut up the reference into a grid.
You know that thumb thing artists do when drawing from life? They're doing that to see how things line up like what Scos did with the vertical line. I'd try doing that first with just your pen or pencil, then check your work with a vertical line on the ref second.
Just the painting:
Up next: fix the nose shape, brighten some of the area around her left (our right) eye, and fix other mistakes (why can I see these when uploading but not in Photoshop, even when flipping the canvas?). Then hopefully be able to finally move on to the hair and maybe body before starting another project.
It sounds like you're already aware of some of the problem areas that need more attention. I might also encourage you in general when doing studies like this to at least rough in some loose indication of the entire mass of the head, hair and neck at the beginning. In this case, her darker hair serves as a nice value contrast to help us make better judgments about if our values and colors are accurate than relying solely on the gray canvas color. A person's hairline and skull shape can significantly affect their appearance as well, so it helps to work with everything in place rather than polishing up a floating disembodied mask and trusting that it all falls into place when you get to the other parts of the head.
I dashed out a quick paint-over to demonstrate the way I might make some tweaks. I'm mostly just eyeballing where parts seem off and making quick approximations closer to the ref, so this is definitely not meant to be airtight.
The biggest structural changes I made were adjustments to the nose and mouth (third frame of the gif). The nose was looking a bit elfish, so I straightened the bridge out and did some minor remodeling around the nostrils. I brought the mouth up closer to the nose a smidge, extended the corner of the mouth on the left, and did some softer modeling on the lips. You have to be very judicious with the planes and hard edges on lips in order to avoid creating a "wax lips" look.
I also went around and punched up or down any values that seemed off as I saw them, probably most notable around the eyes. A little bit darker around the lids and nostrils, a little bit lighter around the lit side of the mouth (anywhere that bright light off to the right is hitting) and the left cheekbone. Some very minor specular hotspots on the nose and forehead. Most of these should be fairly evident and self explanatory in the gif.
At any rate, hope this is helpful, carry on.
Quoting for embellishment. The rest of the image would act as a frame of reference for what you are currently working on and help your image come together much more smoothly. Are you just hiding your sketching layer or do you really going in and try to fully render one section by itself?
Oh, and by the way, you're already showing improvement from your first post. Keep at it, dude.
Unfortunately, after that I had a pretty awful combination of art block and classes being particularly brutal as we get closer to midterms. However, I've finally been pulled out of this slump thanks to Noah Bradley's "#21 Days", which you can read about here. I'm really excited to get into this and see how much I can learn and improve coming out of the next three weeks. For day one tonight I worked based on Week One of Noah's Art Camp 2, and did a fairly quick (a little over an hour) master study of a Sargent piece. I'll probably do some more of these for the next few days to really work on my fundamentals. I worked on the piece upside-down and using the same adjustments as in the first part of the video.
Again, not crazy accurate but already learning quite a bit, and I know that training myself in how I see while drawing is something I've really needed to work on. I'm a little worried about making the jump back from drawing to painting, but that's something for Future-Phill to take care of. I will admit, even with it's problems I'm pleased with how this first day's drawing turned out, and I'm especially pleased to see that I have actually improved since beginning this sketchbook. Onwards and upwards I suppose.
This one was actually not a master study, but a study of a pretty cool painting I found online. I think it served its purpose of providing some nice abstract shapes though, but if anyone has some good examples of turn-of-the-century realist painters I'd love to discover some new ones. I tried Googling for some tonight but couldn't find many I wanted to emulate, and those that I did didn't work out well after the image adjustments I did for these. Anyway, more learning to be done here, but I did see some improvement in my placement before measuring with this one.
This one was pretty quick/sloppy. Had to get it done pretty quickly which, yes, really defeats the purpose. Still saw some good things happening while drawing and measuring though, better than before I started these studies. Still, wish I would have been able to give this one the time it deserves.
I had fun with this one! I've been experimenting with different brushes throughout these studies and I liked how this one felt. I was a lot looser than yesterday drawing this, and it felt like it went pretty fast (didn't have to go back and fix a lot of big measurement problems as often).
Your work is continuing to improve. Great! I dunno about you, but for my studies, I make the canvas the same size as the source image - it makes it easier for me to see if something isn't placed or aligned correctly.
Will you start doing studies with more value range soon?
Another fun one. Thanks for commenting Ollie, keeping the canvas the same size is a great idea; usually judging the horizontal distances are one of the harder parts of these, so that should help! And yes, I was hoping to start working on value soon, probably with grayscale first then teaching myself better color (if for no other reason than because it's getting harder to find good references that work after the adjustment). These have definitely been interesting exercises though, now I NEED to practice some more pencil studies from life to see how what I've learned from these translates. I think if not, I'll need to work on my focus more.
I think this will be the last of these studies, now onto grayscale and then hopefully color (thanks Ollie for the suggestion!). But these have been good, I think my ability to see and draw has improved, and I've refined some Photoshop stuff. Beginning tomorrow I'll have more time and so can start doing the grayscale studies and be less rushed through the studies.
Gearing up to start doing more intense grayscale stuff, so I did some landscape thumbnails. A Bierstadt and two Morans. Interesting studies, trying to not go too dark or light, so I'm focusing on creating enough distinction of light without crazy contrast. The darks are looking too dark though, time to fix it in the coming days.
Wellp. I started off wanting to do a good grayscale landscape study but after starting and deleting 3 or 4 times I gave up; I just couldn't seem to get anything right today, and my brushes weren't helping at all. I'm thinking about looking around for some different brushes, and I know everyone says you don't need any certain brushes to paint well but it really helps. Do most people have days where you just can't get things right no matter how hard you try? I definitely do. Anyway. Decided to try doing a study in my sketchbook since digital wasn't working out, and was disappointed in this as well. The likeness isn't great, and I feel like it took far too long to do this (between one and two hours) for what it is. Obviously I shouldn't work on my traditional skills and my digital skills at the same time and I'm definitely going to stick to photoshop for the vast majority of the 21 Days, but there's some stuff in this one that are kind of unforgiveable and just sloppy or lazy. And that raises the question, "why didn't I see and fix these while I was drawing it?" Well for one, it's easy to see the flaws afterwards, and for another I wanted to go to sleep, but really I don't know. I feel like I can do far better than what I've posted in this sketchbook but between procrastination and just plain apathy I haven't made much I'm proud of lately. And now I'm kind of pouring my heart out, sorry to anyone that wasn't ready for it to get heavy up in here. But I guess I'm just frustrated. What I really need is better discipline, and I'm hoping that sticking to these 21 Days will help build that.
You are making good improvement, keep up the diligence, and you'll continue to see improvement. Your focus is in the right place, which always generates frustration. I've been working at some fundamentals, and man have I been making some shitty drawing for the last two weeks. It does feel bad when you put down a drawing and immediately go "Oh fuck, thats not right at all goddamnit" but it will continue happening especially as you hammer away at these basic assignments. Its alright, keep pushing forward. Slow down in the middle of the drawing when things are seeming off, walk away for a few minutes, come back and just look at the reference for a while.
Your instinct to be more accurate is your best asset at this stage, embrace it, and the frustration, and keep working through it. I also try and reduce distractions when in this particular mode of drawing, simple music or complete silence, if you can achieve that.
A better day, and I switched up my brushes/tools a little bit as well. Started this one off with very rough block in of the big color spots, then went back through and smoothed out the lines and paid a little closer attention to more detailed light and dark patches. I think I softened/render it too much; something to fix tomorrow. I took your advice Iruka and took a little break and that was definitely good. Anyway, better than yesterday, will be better tomorrow.
I think I got some better values, a little better study than yesterday. I still missed some important values in the sky, for example, after looking at it, but a little progress nonetheless. Not a huge fan of the pastel brush leaving white "canvas" dots all over, but it does to get that rougher look I want for these. After looking at it fresh off of a break you can tell the far right (sky and foreground) is pretty borked; too sloppy, I will work on that.
Well I missed last night's study due to moving back home for spring break, but I'll make up for it now that I have a lot more free time to practice. Fairly pleased with how this one turned out, but most notably it was probably the most relaxing practice piece I've done since the second day's study; it was really therapeutic to work through this one and I think things have started to fall into place a little more. I'm seeing better values and better accuracy within this one, even for being a short (hour-ish) project, than in a lot of what I've been doing in the past. Also, thank you everyone for all your support. It's really nice to have this community that gives such active feedback, and you're always constructive and I would say "tough but fair", which is really helpful for me to make sure I stay focused and take what I'm doing seriously.
Well this one felt even nicer than the last. It may be that I'm home and a lot more relaxed while doing these; no homework, tests, etc. to worry about that I "should" be doing. One thing in particular I noticed while painting this was that even at the roughing-out blocking stage I saw a much stronger similarity between the canvas and the ref. Also, I think the "use the biggest brush possible" advice has finally started to actually click in that I'm starting to figure out how to effectively use the bigger brushes and still get decent detail. Let's see what improvement tomorrow brings.
Rushed through this one a little more than I should have, but it still took a little too long. The accuracy is a little questionable; there's a lot of little details going on in the foreground that this kind of study doesn't mesh super well with, but I still got some good practice from them and even got to work with reflection, which was nice. Still seeing little improvements here and there, especially in the process rather than in the finished product which is interesting. I'm tempted to go back to one of the thumbnails I did a few posts up and work through them like I've started to be able to do, and this might end up being necessary as I start to slowly run out of master paintings that I want to tackle.
(from yesterday, didn't get to posting it)
and Day 15:
I kind of spun my wheels on these, they were both a little rushed to get my drawing for the day done. It's interesting going back and forth between palettes with different levels of contrast and seeing how many basic areas of value the artist used, I find that it links back a little to my first week's studies in seeing the big abstract shapes.
Fairly satisfied with how this one turned out, it felt pretty good to draw and I got more into it than the last few which was nice. I chose a painting that I thought would be a little more difficult than the rest, and I'll probably continue ramping up the difficulty as we finish out the last several days. But, I like how this study looks; it felt much more natural to paint, I did very little measuring at all, and it was enjoyable to work through.
This one was really quick, probably a little less than an hour.
ok but seriously
It looks to me that you're getting a pretty good handle on values in greyscale. Do some work in colors soon! It might feel scary and overwhelming, especially if you don't allow yourself to use the eyedropper tool, but if your current progress is any indication, you'll find that challenge easy to overcome.
Oh...and by the way, did you ever finish that illustration of the woman with the crow in your first post?