"The algorithm is hard to predict" Me, sitting here watching a game devlog after seeing an engine simulation that was recommended to me

I’ve finished all of your videos and left wanting more. I’m surprised just why you’re still replying to comments since you should have gotten so famous your inbox would have filled with millions of comments from fans 😂. Your content is way better than a lot of other dev logs, you cramped so much knowledge in one video it felt like a roller coaster of information (not that I wish to imply your content is confusing). I felt like this channel is going to be big. Keep up he videos.

While I found the channel through the engine simulation video, I actually really liked the ray tracing video. Two things: 1. Videos on some stuff you only mentioned please (especially one about how you went about making your own programming language). 2. Really liked the quality of the videos and they are engaging. Genuinely was surprised to see this channel wasn’t as big as I thought because you couldn’t guess by looking at the video. Anyway excited to see where you go and it’s pretty cool to be here at kind-of-the-beginning (I think your channel’s already blowing up)!

looking forward to seeing these numbers quadruple by december!

Wow thats crazy. I expect that your channel boomed the last months. I hope your goals for the car ray trace videos are now reached. I really appreciate the effort it was amazing! Would be nice to see a part 2 and your thoughts to it :) Keep it up

Your channel is blowing up right now! I absolutely love your content! Your engine audio simulation video is the first recommended to me. The production value is fantastic and I was surprised to see that your first video was also awesome! Please keep it up! I love that you actually take some time to explain the concepts you’re working with. You’re definitely going places! Thank you for this awesome content!

Interesting video, a follow up could be nice. I find your perspective quite insightful, it's a thoughtful view from someone who puts care in his work first, but that doesn't want to let it "die alone" and I personally think that's the best way of doing it. Maybe the followup that makes the most sense right now could be in livestream form in a 2nd channel to make it easier to produce, given that you probably have some bigger priorities rn. (Then leave it up). Try to use the community tab to see what kind of feedback you get from it, if it reaches your audience. I see you've posted an update on the github thing but maybe you could test the waters on it too and see if there'd be interest for something like that (if you have time for it) I personally think that the Car rendering video was awesome, it was the 2nd video i watched from ur channel. maybe it did worse before in the short term because it wasn't that tied to the previous content enough and the topic was of less interest from the return viewers and too niche to find its algorithmic boost, but the numbers it has now prove to it having been a good choice of video. It now has the good thumbnail + good title + "thematical tie-in" with the biggest video in the channel (plus it's a great video) so it makes sense. Obviously much easier to see when it's already happened, it's quite telling to see how you saw it back then and a good reminder for new creators that these kind of bets can make sense in the future if there's a good context for them but that there can be an initial underwhelming feeling of "this was all for nothing" and that it's definitely not obvious at the moment.

You are almost famous now! Btw the raytracer inspired me to make my own and I find it to be one of the best videos on KZfaq!

The first video recommended to me by KZfaq from your channel was the rotary engine Sim. Then I checked your Channel out and watched some of your other engine Sim episodes. Now after seeing this video it seems the start of your engine sim series has led the algorithm towards car enthusiasts and it appears that has really helped your channel. Now I'm going to watch that last engine sim.

After the engine sim video, the first two I watched (and I watched them till the end) was the raytracer video and this one. This one had my interest because both me and my collegue got your engine sim video on our YT start page, so I started wondering about the algorithm.

This actually helped me plan for my next vid, thank you

Still here. I am not a creator, but one thing that can oddly impact is what other creators may also be doing at the same time, or even just what is happening in the world in general. Honestly not sure what I was watching when YT recommended your video, but, glad it did. Sometimes it does get it right.

Better thumbnails, browse popular channels and observe what choices they make. Great job amd good luck

Wow....amazing and insightful....thanks

I loved you ray tracer video also loved your ego

-Low retention rate but high amout of subscribers might indicate, that content is interesting to a lot of people, but the video is too long. So people check the interesting bits and drop out. Idea: you might try YT Shorts, these could work for asset creation videos, since they allow to show off quick and with lower effort. -Your first devlog video attracts better, since it states "c++ game from scratch", which more of people can't do, while "designing a game level" sounds very familiar and generic. Idea: you might counter with either: 1. Keep "c++ game from scratch" in each devlog title and add ": doing X thing" after. 2. Come up with interesting hook for each new video 3. Drop "devlog #1 , #2, #3" entirely to avoid "mid series video problem", and just add videos to playlist for continuity. Disclaimer: I am not a youtuber, nor any sort of marketing person, so take my words with grain of salt. A few of those :D

Would be interesting to see this again after your recent increase over the past few months despite the reduced quantity of uploads!

Awesome to see your analytics! You're doing much better than me and I'm sure you'll be monetised very soon.

Heh. This video goes to show that devs can tend to view human activities through a mechanical lens as well and misread what is going on. :) The raytracer was good. People just didn’t discover it yet.

Thank you kind sir

funny that raytracer video is 2nd most watched now.

Well it seems that after giving it more time the raytracing video is doing well comparatively speaking. Maybe with the finicky algorithm ruling everything sometimes things just need more time.

Bro you deserve 20x more views

TIP: Make an intro video or part 0. This way you can tell people what is to come and show the progress you've made. Viewers will be more likely to click through and sit through earlier 'lower quality' content.

Thanks for sharing! One thing you haven't mentioned is what you did to promote your videos? Did you simply upload and waited for the algorithm to suggest it to people or did you actively promote it through forums, Facebook, Reddit, etc?

Perhaps a way around getting people to watch videos that are later in a series when they haven't seen it before would be to leave the numbers out of the title / thumbnail and make the titles seem more standalone. It is a shame to get rid of the numbers for catagorical sake but I've seen this sort of titling work very well for certain channels.

More Flex/Video plz

The sad part is that no matter how awe-inspiring, ambitious or groundbreaking your work is, you may stay unnoticed for years until some of your videos randomly goes viral and then all of a sudden you start getting lots of views from all over the world. It's like winning a lottery, shame that clickbait thumbnails or dumb freaky shorties tend to give more chance of getting noticed than making quality content like yours.

Hi guys!

4:43 "age and gender" - male / female. you can see that this was made in 2021, not today.

"male: 100%" LOL

Kind of funny to see there are pretty much only males watching these videos.

As you're approaching 100k subs this video didn't age well 😆

lol 100% male viewers