View Full Version : I Could Use Some Advice

01 January 2013, 10:59 PM
This is going to be a bit of a long read but i would appreciate all the advice i could get. Ok so i am still very much a beginner at drawing (both traditionally and digitally) and after reading so many of the tips within this forum i'm wondering if anyone could tell me if i'm practicing the "right way" (as mentioned in some of the threads here) so I'm going to go through a step by step process of a couple of things i've drawn.

One of the things i practiced drawing were skulls. I first gathered about ten reference photos and a few sketches done by other artists. I then decided to use this image as my main reference:

I decided wanted to draw it in Painter rather than with pencil and paper. My main goal was to go through the motions of drawing a skull - to find out all the details and where to put them, the correct proportions etc. Copying the image exactly wasn't that important to me. I even changed some things on purpose. Here is my digital drawing:

I picked up on a few things while drawing it that would simplify the process of drawing a skull in the future. For example: When i drew the teeth on that image i drew the outline of each tooth one by one and i realized in the future it would be much easier and quicker to draw parallel lines with the spacing inbetween each line increasing until i got to the middle of the mouth kind of like this:

and then from there finish the rest of the outlines on the teeth.


01 January 2013, 11:01 PM

A short while after i had finished my digital drawing i decided to draw another skull (similar to the digital one) but this time trying to draw as much of it from memory as possible (without looking at any reference photo) only looking at the reference image when it was necessary. I also decided to change even more about the skull to try to learn more. This is my sketch (i don't have a scanner so i had to take a photo of it, sorry about the glare on the left eye socket):
A couple things I changed were the cheak bones and i made the eye sockets look angry. As i said i tried to draw as much of it from memory as i could and when i was almost done i realized the teeth were way off. they didn't have an overbite and the bottom teeth were too big and there should have been more in the front. i also noticed the left eye socket is bigger then the right.

So given all that im back to my main question: Would you say i'm practicing the "right way"? i just want to know if i'm on the right track to getting better.

EDIT: please excuse the pencil strokes on my sketch because i didn't have the proper tools to make is smooth

01 January 2013, 06:35 AM
When learning and practicing, you have to give a lot of thought to the true purpose of the exercise, and how it should benefit your overall artistic growth, so you know exactly what to focus on in your practices.

The reason for learning the structure and proportions of a skull, is so you can understand the underlying structure of the human head, which benefits you when you depict human heads. So when learning about the skull, you must also relate it to normal faces and heads, so you can see exactly how the bone structure affects how a human face/head looks, and you need to reference a wide range of different faces and heads, with different age groups (baby all the way to very old people), ethnicity, sex, attractive to ugly, thin to fat, and in various different lighting situations.

Memorizing details about the skull that don't even show up once covered up by fat and muscles and skin, is nearly useless. If you need to actually depict a skull in a piece of artwork, just reference skulls. Only memorize what's actually useful to majority of the types of images you would actually create as an artist.

01 January 2013, 05:23 PM
oh ok. i see what you mean. thanks for the response and the tips! i appreciate it. and my purpose for drawing the skulls was because i like dark art so i figured that's where i'd start. just to get to know the features of skulls so i can draw them much better in the future. i did hear somewhere that you should learn how to draw skulls to help you when drawing faces but they didn't go any deeper then that so i'm glad you pointed out that certain details of the skull would be almost be pointless to memorize when drawing a face. that probably just cleared up some confusion i would've had down the road lol

02 February 2013, 04:19 AM
No detail is pointless to memorize. Memorize everything. Draw that skull until you know every detail. Then you can better recreate it from memory.

02 February 2013, 04:31 AM
No detail is pointless to memorize. Memorize everything. Draw that skull until you know every detail. Then you can better recreate it from memory.

I think some things you absolutely must memorize, such as the basic standard proportions of the figure and facial features, or the main muscles that's visible to the eye even with a layer of fat and skin. But something like the skull, you only need to know the general structure for proportions instead of the micro details, because how often is the average artist going to be depicting an actual skull in his entire artistic career? Some artists never have to depict skulls even once in his entire lifetime as an artist. For esoteric and rare subject matters, I think memorizing isn't important, because the minute details can be gained from using proper references.

02 February 2013, 04:40 AM
Of course. But it never hurts to try and memorize as much as possible. When I am out and about I find myself looking at unique surface, taking pictures for reference of couse, and feeling the textures of them. Reference is amazing, and you should always have it. But sometimes you can get amazing ideas just from something that you might have memorized that jumps out at you. I know every artist is different. Just wanted to share my thoughts.

Edit: You are the award winner. I am not.

05 May 2013, 01:59 AM
alright so i figured i'll post an update. i started reading Andrew Loomis' book 'Fun With A Pencil' and while it provided some insight, it started to get boring. i felt that it was more for the casual artist who just likes to sketch and doodle. so i moved onto the book 'Successful Drawing' by Andrew Loomis and this one has really opened my eyes and makes me think about things much differently then i ever did before. the book is really amazing. right now im working on perspective (which is also helping with proportion) using horizon and vanishing points. i'd post images of what i'm learning but there's not much to it because it's just boxes, lines, and planes. i'm even starting to learn beyond what the book is teaching. while doing the exercises in the book i've noticed that the vertical corner edges of buildings and houses will always be perpendicular to the horizon. and most of the time they will be perfect straight up and down. the only exception to that would be if your whole drawing was purposely angled/tilted but the side edges of the buildings would still be perpendicular to the horizon. is this correct?

05 May 2013, 02:17 AM
after thinking about more, the side edges of buildings wouldn't be perfectly perpendicular to the horizon if the image was at an obscure angle because the horizon is always straight across correct?

05 May 2013, 05:14 AM
The angle doesn't necessarily have to be "obscure"--it simply needs to not be straight on.

It's all about what angle you choose for the "virtual camera" in your scene--the one we, the viewer, are looking through. If it's tilting up, such as standing in the middle of a city and looking up at an angle--all the buildings would converge to a vanishing point up high. If you look down at an angle, such as standing on a tall building and looking down, all the buildings would converge to a vanishing point below. If you simply look in front of you straight, then all the buildings would be perpendicular to the horizon.

05 May 2013, 11:18 PM
ah, i see. thanks. this stuff gets so confusing sometimes. i do have another question, is the grid method a "must" for learning? i ask because personally, i can't stand doing it. it can get really boring and i find the grid distracting. for some reason, my eyes like seeing the whole picture that im trying to draw rather then small sections.

05 May 2013, 05:41 PM
Are you asking about the grid system that beginners often use when trying to achieve accurate proportions when copying an image?

You can set the grid to low opacity so it's very faint but still visible.

The grid isn't "necessary"--it's just a helpful guide for beginners who lack the ability to do accurate visual comparisons using only the notable landmarks in the image. If you can do accurate copies without it, then you don't need it. But if your copies always end up looking wrong even though you tried hard to be accurate (including using the horizontal flip trick to check your proportions to combat the brain bias), then you do need to use the grid system to help you get over the initial hump of learning to observe and analyze proportions. After you've done a few accurate copies using the grid system, you can try without it and see how you do.

05 May 2013, 07:17 PM
alright. yeah i have a problem with proportions when im not using a grid. i was hoping there was an alternative method to help with proportions. i'll start using the grid and really give it my all. thanks for all your advice your giving me. it's much appreciated

05 May 2013, 07:23 PM
i started again with the grid method to help me achieve better accuracy and i have a couple questions. 1) i want to draw things using a grid that will be benificial for me to know in the future when i eventually get good. stuff that i'll be drawing a lot of. should i draw landscapes, animals, foliage/plants/vegetation, architecture etc.? i basically wanna kill 2 birds with one stone here. learn accuracy while learning things i'll be drawing a lot of in the future once i go pro. and 2) right now i'd prefer to work digitally. i know my tablet and all my software like photoshop and painter very well so that's not an issue. i prefer to do it digitally because it's quicker and much easier to get rid of mistakes. plus i do not have a tilted desk to help me draw traditionally. and my question is if i were to practice mostly digitally, and develope some good skils, would those skills transfer over to traditional style using pencil and paper? for example: since my tablet is flat on my desk and i look at the screen when im drawing and not at the tablet and i ended up becoming quite good at getting proportions correct, would that skill transfer over to pencil and paper without giving me any trouble? i know you said in another thread that in some aspects it's almost better to start digitally as long as you know the programs and know how to use a tablet but im just wondering if i would suffer any set backs or if i would have to re learn anything if i were to transfer my knowledge from digital to traditional

05 May 2013, 07:54 PM
1) Sure, no reason why you can't.
2) Drawing skill is universal across all mediums. There are idiosyncrasies to how each medium handles, but the underlying motor skills are more or less very similar. It's like once you learned to ride a bicycle or drive a car, you can ride/drive all other variants. Even with variants that are very unique, once you learn the idiosyncrasies, the basic motor skills are still very similar. So no, you won't suffer any setbacks.

05 May 2013, 02:33 AM
ah. thank you so much. i really liked the driving analogy you used. it makes perfect sense

06 June 2013, 11:24 PM
i thought i'd post another update

i'm in the process of drawing this wolf

here's my basic sketch

i recently realized that i was making a few mental errors when i drew. for some strange reason after i would draw just a line in my sketch i would just keep going without checking to see if it was correct first. which is why my drawings looked so distorted. i guess my thinking was "i'll fix it later if it's wrong" or something. needless to say i stopped doing that immediately! now i check right away if the line i just drew was correct and everything has been going much smoother and easier for me now. anyway, with this drawing i decided not to use a grid to sort of test/challenge myself. i did however use the ideology behind it by trying to break down the image and focus on small sections of it and draw them. i aslo put the picture on a layer so i could check every once in a while to see if everything lined up. no tracing though. i'm currently in the process of painting it. i'll post it as soon as i'm done. any thoughts or words of wisdom?

06 June 2013, 11:39 PM
oh and my sketch is smaller then the wolf in the image because of the way i have been working on it. and when i shrunk down the images so they would fit on the forum my drawing seems smaller. but my real sketch in my painter file is the exact same size.

06 June 2013, 01:15 AM
I don't really see anything alarming at this stage.

One piece of wisdom for you: Always keep in mind that painting and drawing aren't two totally separate processes. When you are painting, you are still drawing, because you still have to make sure all the shapes are correctly proportioned, with accurate distance, angles, curvatures, etc. The only difference is that you might be using bigger brushes that draws a bigger patch of value/color instead of a tiny brush that only draws a thin line. While painting, you are always constantly still making corrections to your drawing, so you never actually stop drawing--you're always making adjustments as you paint.

And don't treat backgrounds as an afterthought. They are just as important to the whole image in terms of composition. Too many people have the misguided mentality that the foreground main subject is all that matters, and they can just slap on any arbitrary background or get sloppy with them. A whole image has to work in every aspect--nothing is "less important"--it's ALL IMPORTANT.

06 June 2013, 11:19 PM
yeah i'm definitely going to put a background in there. i was thinking of just doing the one in the photo but maybe i'll put something else there. anyway, here's what i've done so far

i'm probably going to spend a few more hours getting the fur just right and then i'll move onto the background

06 June 2013, 06:12 AM
Don't just slap an arbitrary background behind a photo without knowing exactly what you wanted to express in the image. A painting is a cohesive whole, not subject cut from a complete scene and then with an arbitrary background swapped in. When you start designing the overall layout/composition of an image, you need to know what the whole thing is supposed to look like. Do thumbnail sketches and color studies that are rough versions showing the composition, the lighting, the color palette, etc, and once you have worked out all the potential problems in the image, you start working on the actual painting.

Another problem with less experience artists doing background swaps is that they have very weak understanding of perspective, lighting, and colors, and they often end up with the subject and the background mismatched, with contradicting lighting, incorrect perspective, wrong color casts, etc.

If this is just a technical exercise to practice your skills in copying an image accurately, then why bother swapping out the background?

Next time, spend the time thinking about what you want to express/convey first, then do sketches to work out the composition, lighting/color studies, etc before you even start the actual painting. (This is for original works where you came up with your own narrative and want to express emotions/moods and communicate ideas, or to portray original designs such as concept art).

For technical copy exercises, you only need to focus on getting things looking as accurate and as much like the original as possible.

For artistic interpretations (such as portraits, landscapes, still life, etc), you focus on the execution itself such as expressive brushwork, aesthetic sensibility in color choices and value management, or if you're after realism, push for the utmost realistic details and accuracy (realism from photos is a bit of a pointless artistic statement--it's better suited for technical exercises. For SERIOUS realism, you work from life only).

07 July 2013, 11:22 PM
You should listen to Lunatique, he usually charges for one on one mentoring ;)

I would like to make a comment about memorising details. I don't believe you need to memorise every detail in order to draw a subject really well. Noticing details in everyday life and references you capture, is different from committing them to memory, and probably more important. Studying detail and thinking about things like that in a way to understand the nature of what you are looking at, is what will help down the line.

Studying a reference and replicating it in a painting is all well and good in a mechanical sense, but you have to be mindful of the how and why of what you study.

11 November 2013, 04:13 AM
i haven't posted in a while but i did finish that wolf. here it is.

i have some more questions.
1. i understand that you can't draw what you don't know but when it comes time to draw something you've never drawn before how do you go about drawing it accurately? aside from studying/analyzing the subject (i have some understanding of that part), what's the approach? do you draw basic shapes like circles and stuff and then build off of them like andrew loomis describes in "Fun With A Pencil"? or is there some other method?

2. lunatique, i remember you telling me that i should learn to draw things like photos with or almost with photorealistic accuracy before moving on to things like perspective and stuff but im still having trouble with accuracy. even though the wolf came out looking good, it takes me a long time just to get my outline/basic sketch accurate. i've tried applying basic shapes and then building off of them, i've tried pushing myself to using the principles of the grid (by breaking the subject down into simpler forms) without actually using a grid but that doesn't work either. the thing that works the best is the grid but it doesn't feel like im learning anything from using it and it feels more like a crutch then a tool for learning. like it helps with whatever im drawing at the time but when i move onto something else i'll try to not use a grid but i have the same problems with accuracy all over again and im not really seeing too much improvement. maybe im just not drawing enough. im not drawing 5 or more hours a day, every single day. some days i'll draw for a few hours and then there are other days that i don't draw at all because there's other stuff that i need to take care of. any thoughts?

11 November 2013, 02:05 AM
1) If you want to depict what you see convincingly, even if you have no understanding of the subject, then you simply depict what you see, as if the image is nothing but abstract patches of shapes, values, colors, and edges, and you simply copy them as well as you can. This is part of traditional art training--to depict what you see and not what you know. Of course, if you combine that with insights about the subject, you can achieve even more, because you can then decipher what it is you're looking at (such as understanding why a blue sky leads to blue color cast in shadows, or the reason behind atmospheric haze, or the anatomical structure of animals, etc.). Having the observation/analysis skill of depicting what you see exactly is a technical skill that can be trained with time and patience, while having understanding of the scientific reasons behind why things appear the way they do, takes time to learn and memorize. Ideally, a proficient artist should have both qualities.

2) Exactly how many training exercises have you done? Most beginners have no idea how much practice it takes to become a proficient artist. It takes hundreds of drawings and paintings, so you need to put whatever efforts you've made in that context.

You need to know that people don't improve at a steady rate. They go through up's and down's and a period of plateau, before they reach the next steady climb, and then it become erratic again. This is the same for every single human endeavor. There might even be times when it feels like you're regressing--that is perfectly normal too. The key is to be tenacious and not give up, and to have discipline and keep on learning and training, and you'll eventually hit your next steady climb up.

When you use a grid to learn, it's main purpose is to help you be able to make visual comparisons with an easier set of reference anchors, as well as help you develop synaptic connections necessary for critical visual analysis. Once you have done at least one excellent copy using a grid, you can move on and learn to be accurate without it. Without a grid, the visual reference anchors are a lot more challenging because you have to extrapolate them from the image itself, using it's main landmarks as comparison reference points, and if you mark those wrong in your copy, you'll be using unreliable reference points, thus make one mistakes after another. They key is to continue scrutinizing those reference points as you work, to make sure that as you add more to the image, they still hold up against more landmarks. There isn't as much to consider as most people think--it's really just:

Amount of curve
Relative proportions (both micro and macro levels)

And then for tones, you have:

Edge quality (from sharp to soft)
Amount of gradation (gradation values and distance)

That's pretty much all there is. So scrutinize your image using that list and analyze every single detail, as well as the overall proportions (including the negative spaces).

BTW, the wolf looks pretty good. It's left hind leg gets too thin at the bottom though, and there are some other minor issues, but for a beginner this is very good.

11 November 2013, 11:49 PM
thanks for the insight and the critique on the wolf! the problem i was having with the wolf's back legs was that in the photo it actually looks like it has 3 legs instead of 2 and i couldn't decipher which of the 2 things (behind the right leg) was the left leg so i just tried to draw it as i saw it in the photo. i found your advice on analyzing what i'm drawing very reassuring because that's exactly how i'm practicing when i draw anything. finding reference points and how everything else relates to these points. analyzing not just one, but several reference photos of a particular subject/object and finding "truths" that apply to all every single one. basically, common denominators so to speak. like the wolf for example, as i was drawing it i was making mental notes of basically what is common to all wolves. the shape of the head (almost like a stop sign), the ears are of medium size that are flat at the top rather then pointed. the structure/posture of the wolf. the snout area which is about medium sized compared to other animals in the canine family. and how the fur flows. it all flows away from the face toward the butt/tail and straight down on the legs. so basically if the wolf was standing on 2 legs and it's nose pointed to the sky the fur would be pretty streamline going from the nose to the feet. shortly after i made my previous post i had a big "aha!" moment when it came to breaking an image down to simpler shapes. i've been attempting to draw a woman who is sitting and i was in the process of drawing her arm (which was bent at the elbow at about 130 degrees) and it dawned on me that instead of seeing a full arm fully detailed and textured i saw two diagnal lines that meet to form an angle of about 130 degrees. almost as if the image was a silhouette of a woman. so i think im finally starting to understand what i've been reading in andrew loomis' books about building off of shapes and basically using a stick figure as a skeletal structure of a person and building off of it.

im not sure what you mean by "training exercises". what im doing is drawing random things (analyzing as i draw of course) so i can keep building up my mental catalog of things i know how to draw. and i always have a photo that i work on to challenge myself on a bigger scale. right now i have a photo of a woman that im trying to draw.

what i was previously lacking was setting goals for myself. my personality is more go with the flow then planning things out but since i desperately want to become a great artist i recently decided to set goals even if i have to force myself to do them.

right now my day to day goals are:
1. draw at least 3 random objects. right now i've been on a weapons kick so im drawing different types of cutlery (knives, hatchets, tomahawks, machetes etc.)
2. work on a much more elaborate piece from a photo
3. learn about at least 1 art style/movement and at least one architectural style/movent

yesterday i learned about romanticism and surrealism. salvador dali's, max ernst's, and H.R. Giger's (i suppose he can be considered a surrealist) works were some of my favorites in surrealism. romanticism was a pleasant surprise. i thought it was going to be basically "chick art". but it's more about creating an overall mood (doesn't matter what the mood is, it could be anything from peaceful/pleasant to terror/morbid) with your artwork. that's what i ultimately want to do with my art. create a mood. make people feel the art. i loved that style. Thomas Cole's works were amazing and inspiring.

a long term goal of mine is to eventually draw an entire piece of artwork straight from my head

sorry for the long response. i'll keep posting updates every now and then on my progress and posting my random sketches if anybody feels like critiquing or commenting

11 November 2013, 11:59 PM
also, i'm glad you pointed out how a blue sky gives a shadow a bluish hue. i never even thought of that. i was wondering why in the photo of the wolf that the shadow had a purple hue. i ultimately drew it the way i saw it in the photo but i couldn't figure out why i was seeing purple in the shadow

11 November 2013, 05:05 AM
By training exercises, I mean specific exercises designed to push you to strengthen particular weaknesses, or learn specific lessons. The kind of assignments I give my students in the workshop (the one I teach right here at CGSociety, as linked in my signature) are exactly like that. I designed them to shove the students out of their comfort zones and do things they've never done before, in order to learn lessons in the most effective way possible. I drill them on very specific topics like how to effectively manage values, how to create a convincingly lit scene from scratch and make it look real, how to assign believable colors onto a B/W image and make it look natural and creative, how to create interesting composition with just basic shapes, how to elicit specific emotional responses in narrative images with effective composition, how to create visual narratives that have intellectual and emotional resonance, how to impart brushwork that are visually compelling and also effective in describing surface properties, how to depict expressive characters with clearly readable expressions and body language, how to stylize effectively according to the artist's intent, etc.

If you are serious about your artistic development, then you need a training regiment that is very effective. You need to know exactly what type of exercises to do and what types to avoid so you don't waste your precious time on ineffective or even harmful methods that will create bad habits and cause long-term damage to your artistic development.

The reason I created the workshop was because aspiring artists don't know how to learn/practice/train properly, thus many of them waste years of their lives and don't really improve significantly. With an effective training strategy, someone can go from a total beginner to an advanced artist in just a short few years, but without that effective strategy, many end up spending decades or their entire lives and still are running in circles drawing/painting like a beginner/intermediate artist.

12 December 2013, 06:40 PM
thanks. i'd take your class in a heartbeat but i just can't afford it right now. i was wondering, i see a lot of artists, in various tutorials, use oil brushes and other artistic brushes in Painter when coloring their artwork and then they smooth everything out with blenders but that seems like unnecessary steps when all they need to do is use a soft airbrush. that's what i always use for my details and my question is, is there something im missing? aside from using those brushes for a certain texture, why bother using a rough brush if you're just gonna blend/smooth it out anyway?

12 December 2013, 03:43 AM
If you're talking about extremely smooth blending, then you're right, there's no need to use textured brushes and then blend the hell out of all the expressive brush marks and make them look like soft airbrush. The whole point of using distinct brush marks to create expressive brushwork, not to blend everything to death and snuff all spontaneity and life out of the brushwork. There's a time and place for different brushwork styles--some projects require very smooth and detailed work, while some require more painterly approaches. A good artist should be able to do both.

CGTalk Moderation
12 December 2013, 03:43 AM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.