Edward, known as TRZ, is a 22-year old professional designer from Moscow, Russia. He is a former YouTuber, who ended up working with YouTubers.

His professional path began in 2020 with the creation of thumbnails for fellow YouTubers. Over time, he tried himself in various niches, from poster design to UI/UX and even NFT design, but eventually settled on thumbnail design.

He was featured previously on Unlayered #22. TRZ is available for hire.

Twitter (X) x.com/tearz_dsgn
YTJobs ytjobs.co/@Tearzzz
Behance behance.net/tearzzz

Featured design

Overview

This isn't the first time I've worked with Raden. I know him personally, and we get along well. My previous work on Project Zomboid videos were very well-received by the audience and by Raden himself.

Other thumbnails for Raden

This time, Valera (Raden) received a developer key for Subnautica 2 a week before its official release. He wasn’t able to obtain 3D models or asets directly from the game for this project, so I had to work with raw screenshots, posing, and his own photos.

Initial idea

We were brainstorming ideas for really long time, eventually came up to one where he could holding & showing a crab while being in danger of big sea thing behind him.

I already had an idea in my mind of what it would look like, but I still needed references, so I found these.

References

Editing

Valera made a whole archive for me with all the assets I needed: his photo, a reference image of the character's suit, a screenshot of the crab he'll be holding, and a screenshot of the large scary monster that should be behind him.

Starting assets

For like 90% of the thumbnails I create, the process starts with posing, which I then run through AI. I used to use Magic Poser, but now I use posemy.art (thanks to Dill for the tip).

This is the pose I ended up with. A great starting point to work from.

Posing

First things first, after posing, we should give our 3D dummy a bit of flesh tone. Just like in the previous Unlayered post, I’m using AI for this.

I used ChatGPT to write the prompt and Nano Banana 2 to generate the image. There were a lot of results, but this one is the most decent. Even so, there are still many aspects here that need fixing.

AI generation

From that point on, I began the editing process. I already had a clear vision of what I wanted to see, so all that was left was to execute it.

Editing process. *SD - Stable Diffusion

Unlayered

Workstation

Hardware

  • Main Monitor - Samsung Odyssey G4 G40B S25BG400EI

  • Second Monitor - AOC I2480SX

  • Mouse - Logitech G PRO X Superlight

  • Keyboard - Attack Shark X68HE

  • Headphones - Razer Kraken Kitty Pro

  • Graphic Tablet - XP-PEN Star 03 V2

  • Microphone - Samson C01U Pro

  • Chair - ZONE51 Arena Pro

  • Cat on the right - Michael Jordan

Software & Tools

  • Photoshop Beta

  • Nano Banana 2

  • ChatGPT (Prompting)

  • Stable Diffusion

  • Pinterest (Ideation)

What's the best part of making thumbnails?

For me personally, I'd say it's ideation. At the same time is the hardest. When you done with idea, messy part is over, only execution and flow state remains.

Best thumbnail you've ever made?

Best thumbnail you've ever seen?

If you found this edition of Unlayered helpful, please consider sharing it with someone who might benefit from this workflow too! 💙

Reply

Avatar

or to participate

Keep Reading