Crankshaft's Sketchbook

I think I've figured out why people are having trouble reading your image. Despite of the correct perspective, scale, and lighting, the camera angle you have chosen makes it hard to see the floor. Showing the floor will help us understand how the structural elements relate to each other.

This also means you will have to redo the whole thing. But I think it will be worth it.

Thanks Atto! That's a great tip and I'm going to apply it to my sketches. I didn't know that about making details thinner in line weight.

Thanks Yian! I've made the floor a bit bigger. I don't think I can re do it as its too hard. I'm relieved to find out that you think the scale, perspective and lighting are good so far. I added two vehicles to add some human element and to give more scale and reinforce the floor. I've added a fence also. Here's a rough comp. I'm been busy with other drawing goals so the engine room is really sitting on the back burner. Tell me what you think.

The perspective and value is only good in a technical sense. Overall it is still difficult to tell what is going on and this can only be fixed if you start with a better drawing with a better camera angle.

And just a minor thing: anamorphic lens flare is always horizontal across the screen. It does not follow the perspective line.

I see you also added the Chinese character back. Once again, it looks amateurish and hilarious in an inappropriate way. I strongly recommend you to reconsider. Chinese simply doesn't work like that.

Thanks for the prompt advice Yian! I wish you were here earlier then I could of avoided all these stupid mistakes from the beginning! Unfortunately I work 10 hours a day as a car mechanic so it took me 3-4 months just to get the line drawing down so redoing the floor isn't feasible...I'm a perfectionist and I want to change it but I really want to just render it and move on to other things...I really feel stupid for spending so much time on it

Considering the flaw I've done some work to make it more readable. I've cropped the top because there was a lot of dead space and my eyes were too fixated on only it. I've made the floor bigger. Is the floor hard to see because it's messy at the moment? Those dots are visitors on the ramp and there are two vehicles to give more scale. The horizon line is very high up in this picture (above the frame). At the beginning I spent a lot of time choosing the right camera angle and I choose this angle for the very reason so that we could see the floor.

I think the floor reads better now because now my eyes are fixated on the glowing orb which is closer to the floor and people. Is it decent enough? I'm not expecting a perfect read, just something decent.

How would changing the camera angle help? If we look down we would see the floor but vice versa (looking up) the floor would be squished.

Thanks for the tip about the flare, I've changed it. I'll get rid of the Chinese letters.

I apologize for the long reply but I really appreciate your help!

After reading the recent posts here and on the other forum I now find it hard to judge this picture. It still reads OK to me but I have witnessed every stage of its development here so I already know what the space should look like and as such I can't really say how a viewer would react having only seen the final image without any a priori knowledge. I guess this is a problem every artist has with every picture they paint. The artist is painting things that evoke the idea that's already in their head whereas the new viewer is expecting more explanation. It's at this point that you leave behind design and technical skill and move into the area of visual communication and perhaps that's where your next progression and improvement will come from.

I think with this picture your progress has stalled a little by focussing for so long on greyscale. By doing that you only have one tool to play with- Light vs Dark- and you seem to have pushed the contrast right to the limit pointing the focus on the orb and flare while pushing the rest into darkness. Perhaps trying out some quick studies in colour could give you ways to explore more options to control the viewer's gaze by playing with Warm vs Cool hues and varying saturation to communicate the more distant forms and space without having to take the focus away from the bright foreground areas.

Also on the issue of camera angles, this is maybe where 3D models would help. Not as an underdrawing but as a way to quickly explore different angles to find the most dynamic and explanatory view. Quite a lot of your pictures have a high, godlike viewpoint looking down at an angle from a distance. This does allow you to see more of the surrounding space and context but can sometimes, especially with closed off interiors, neutralise all the hard perspective work that you put in resulting in a kind of view like those old, isometric video games. A lower angle from closer in would give a bit more drama and connect you more to the figures and allow them to tell a bit of the story. The downside is the increased complexity of the image and the difficulty of representing all of that space in a single, more cropped picture. I guess the decision you have to make is what type of image you want to create: one based on design like a three dimensional blueprint explaining the environment or something more cinematic showing some kind of action taking place. Both kinds of image are valid and both often appear in concept art but the decision is always yours to make.
And I'm not sure if any of that helps. Like I said, the image is still understandable to me, already knowing your intentions and ambitions as I do, so perhaps you just need to come at it with fresh eyes. You've already shown that you can manipulate and rotate objects in space so I'm sure you'll eventually get to the picture that you want.

Edit: Also I got the latest issue of IFX the other day (issue 127) and there's a tutorial by Eliott Lilly that reminded me of you. Lots of high tech pipes and cables.

Thanks Valence for taking the time to post insightful advice! Rather than post another long reply I'll say that I actually learned a lot from your advice and will apply it to the painting. I've learned that once the technical fundamentals are down the design/self expression can never be wrong and is solely up to you.

Quick design sketch based off a loose reference. The idea was already in my head and the image was carved out of imagination. The picture is only there to show where I got the idea from (Final Fantasy 7, Mako reactor). More updates to come!

That reference photo is a great example of what I was talking about regarding greyscale and colour. When you convert it to a black and white image it is still readable but is nowhere near as clear.



If you look at point A you can see areas of similar value suddenly become separated into different planes when differentiated into warm yellow and cool blue.
Also at point B the pipes suddenly pop out much more into three dimensional space when separated by saturation from the surrounding blue light.



Greyscale is a fantastic way to perceive the necessary values and should never be overlooked but always remember that some problems can also be solved by merely introducing colour.

And that's another promising sketch.

Updates! Some gesture and anatomy studies. I've come up with a personal approach to my anatomy studies. I need to keep things simple and not over think things. I'm currently following Micheal Hamptons book and looking thru Kevin Chen's blog. Both great because they're very relatable to how I draw machines. I'm going to focus on the overall pose and not worry about finer details like the hands or individual limbs since I'll probably never get detailed in drawing characters in the first place.

My general approach

-Go thru both sources and simultaneously apply theory learned thru gesture studies or longer studies
-Gesture studies are fun and loose which forces me to focus more on the pose rather accuracy, it also gives me an idea of some landmarks and some proportions
-I'll do them in pen to force myself to draw smarter
-Draw thru the body just like a machine to better understand 3D space
-Block things in using the envelope method and major shapes like cylinders and boxes, this helps me feel the weight and counterbalance as necessary, carve out smaller forms later or leave them undeveloped for now
-Blocky forms are easier to get the gesture and to carve out smaller forms just like how I draw machines
-Treat the body like another machine I usually draw and not a figure

Thanks Valence! I see what you mean now! That' a great tip! However I understand it now in theory but I may struggle a bit to apply it in practice but I'm definitely going to try it out!

That first page of sketches is head and shoulders above the previous set you posted Crank, and I'm guessing the next few were drawn quicker. Though there are some proportion issues to them they are really nice as gesture studies.
Its great to see the improvement in both those and the new mechanical piece you've posted and reading the discussions your having is giving me some real insight into a lot of the issues I experience when working in colour - so thanks to you and everyone who has posted replies to your queries.

Stay curious!