Links and Notes - May 3rd 2021

Super meat boy post follow up

My post about playing Super Meat Boy with my son picked up a little bit of attention yesterday. From Team Meat, the creators of Super Meat Boy!

Not gonna lie. I tweeted out the article and went to sleep. Woke up, put the tea to brew, and checked out Twitter while waiting and this put a smile on my face. The post itself was a fun one to write. In fact, it was supposed to be part of my links and notes for the day but the idea kept getting bigger because I couldn't stop thinking about it. Initially it was supposed to be a small write up about how my son would feel frustrated when he first started the game but then became absorbed by the challenge. The only lesson there was supposed to be about resilience and the joy he found in conquering the challenges he set for himself. But then the idea grew so much that I skipped the links and notes entirely and poured my words into that post instead.

This is the second time this has happened. Where I skipped the links and notes in favour of a more complete post. I find this intriguing given that I have 3 essays (at least) that I want to write and so far I haven't been able to complete them at all. Whereas when it came to these topics, I managed to complete something decent in the span of a little over an hour. How?

I think the answer lies in this random tweet I saw:

It's a quote from the article that's linked in the tweet that is itself a quote from the Simpson's Writer. This was supposed to be a separate link in the links and notes, but it connected well into this section. Anyway. The interviewer asks the question

How much time and attention did you spend on these scripts?

To which, they receive the answer:

All of my time and all of my attention. It’s the only way I know how to write, darn it. But I do have a trick that makes things easier for me. Since writing is very hard and rewriting is comparatively easy and rather fun, I always write my scripts all the way through as fast as I can...

Then the next day, when I get up, the script’s been written. It’s lousy, but it’s a script. The hard part is done. It’s like a crappy little elf has snuck into my office and badly done all my work for me, and then left with a tip of his crappy hat. All I have to do from that point on is fix it. So I’ve taken a very hard job, writing, and turned it into an easy one, rewriting, overnight.

That actually feels about right. Even my links and notes, the reason it gets done is because I hammer through it. And I think the reason I don't get the longer essays done in a hurry is because I don't try to hammer through it first. I still think it's impractical for someone who writes only as a hobby to hammer through an entire long essay in one go. But large sections? Absolutely. I look forward to trying it out.

How to draw anything

This is a great video from the Proko channel by Antonio Stappaerts is a great look at how the fundamentals should be built upon to allow us to draw anything from our imagination. These are not novel ideas but they are certainly well put together and all from a single video. Generally, to get these ideas you'd need to put together a few different sources but Antonio does a great job in not only breaking it down but also in ordering them in importance. His points?

  1. Structuralization - Know the basic shapes: cones, spheres, cylinders, boxes. Learn how to draw them in perspective and intuitively in space together
  2. Manipulation - Take the basic shapes, and manipulate them. Twist, bend, squish, stretch. All of it. But more than that. Take organic shapes and turn them into contours and manipulate those too.
  3. Observation - Look at any subject and forget trying to understand the details like anatomy and outer characteristics. Observe to break things down into manipulated shapes. Mannequinize the things you see
  4. Education - Now begin understanding the inside and outside of the topics you are trying to draw. Educate yourself. Learn anatomy. Learn details
  5. Imitation - Start learning from peers. At this point you should be at the level where you can analyze other artists to identify their tools such as line styles and shape usage.
  6. Imagination - Put it all together. Go forth and create your own unique work


Antonio makes a great point that artists learn basics and then try to jump straight to complex topics. That's me. I've missed a bunch of middle ground that would help me adopt complex concepts without struggling so much. I think there was a great quote towards the end of the video:

If you cannot even make a simple shape [such as a cylinder] look interesting, then why bother making complex designs.

This was in response to Antonio asking a professional designer for advice on becoming a better designer to which she responded by asking if he'd ever tried to make shapes like boxes and cylinders look interesting.

The final advice given? (paraphrased a little)

The only way to get better is to have discipline and a sense of structure in your training. Whether that's a consistency with what you do, or just having a regular template of the things that you want to learn. Art is a challenging endeavor so a sense of structure in your artistic training is definitely necessary.

Google AI relighting

Early this morning I saw this tweet go out

This is wild. The AI can take a photo, extract the subject, understand the lighting and then replace the subject in a different background with matching lighting. For the record, this is incredibly difficult for a human to pull off. It's a complex topic and even in modern day animation, CGI components in movies can look out of place because of a lighting mismatch. It's something that directors and VFX artists will invest large numbers of hours into getting right.

And this AI does it to photos automatically. I'm sure there are some inaccuracies but this is huge on its own. This is what it looks like:


If the final photo still looks out of place that's because the camera focus is incorrect. The background is completely in focus while the foreground is also in focus. There's already a lot of research in this area and I'm sure it's only a matter of time before the two are combined. You can see a video demo of the research below:

It even has applications of this tool in videos, even if support remains basic. You can also read the research here

Not gonna lie. The implications of this are huge. With deep fakes and other deep learning based software shenanigans, the potential for abuse via faked media grows every day. At this point, I think it's safe to accept, if you've been ignoring it all this while, that you cannot trust what you have not witnessed further hand.

On a more positive note, I'd love to see how all this integrates with real film/photography pipelines someday. If a computer can do 90% of the work I wonder how much it can cut down on the grindy work required by VFX artists.

In other news

  • Bill Stoppard is at it again with a great video on Power Stops. His single subject videos never fail to delight.

This blog doesn't have a comment box. But I'd love to hear any thoughts y'all might have. Send them to [email protected]

Previous links and notes (May 1st)

Next links and notes (May 4th)

Posted on May 03 2021 by Adnan Issadeen