Hmmm, inception?
So I just recently released a 'techlog', all about the Pattern Recognition Algorithm that I made during LD compo. *If you haven't seen the video, then you can see it a few posts down, or just click here. *
Strangely, I found the production of this video to be far more complex than the actual pattern recognition algorithm algorithm that I made, so I thought that I would detail all of the different challenges I faced, and how I overcame them:
:book: Scripting:
No sorry, I don't do impromptu. All of that was scripted, and whilst there's a negative stigma about using scripts, it definitely improved the quality of the presentation. This was actually the fastest part of the production, but I think that this was a big mistake. I rushed the scripting, and there are times that you can tell. When editing later, I often found myself wondering; what I was thinking... Times like those, I had to create text in the video, to explain the things that I failed to mention in enough detail.
There was some times that I branched off from the script, and this worked out pretty good. There are even a few instances that you can tell, just off the changed tone of my voice. Not sure if such a clear transition is good, but the impromptu parts made the video seem a bit more casual. I had initially intended to cut out the ramblings at the end of the video, but in many ways, it felt good to be freed of the script and just speak openly at the end of the video.

:loud_sound: Audio:
Audio was a real pain to pull together. I spent some time complaining about it, which certainly felt good. I had to record at the dead of night, otherwise family members were active, birds were chirping and cars were driving around at full throttle. I had to strip my room bare of all sound making devices; the clock was moved to the corridor, and I even had to remove my computer, since the fan was too loud to record over. Instead, all of the recordings were done into my phone, and transfered later. I initially used the phone's microphone, but I found that my head moved enough to create audible changes in the volume and quality. So instead, I used a headset, which worked better.
Then once I had the audio sample, I had to edit the sound in Audacity. This was strangely even more of a pain, not cos Audacity is bad, but cos I'm bad. I'm an aussie, and whilst I hate to blame my heritage for it, my voice sounds kinda... trash. It's way too coarse and certainly not the deep, stereotypical Australian voice. No matter how hard I tried, I couldn't get it sounding perfect. Most notably, audacity was unable to repair the coarseness of my voice, so I wasn't able to get the 'narrator' level of audio quality. I was able to make it sound a bit deeper by increasing the bass and reducing treble, but still far from David Attenborough. With the mic so close to my face, the recording picked up everything, including breathing and all sorts of other mouth sounds. These had to be individually repaired, and all up, the audio editing alone would have taken around 8 hours.

:paintbrush: Graphics:
I had no clue what I wanted to do for the graphics of the video, but I'd recently installed Krita and was amazed by the software, so I knew that would be the focus. I had initially planned to use more footage than graphics, since that would save SO much time, but the only footage I could really show was the code, which wasn't particularly engaging. I decided to go for a 'hand made' feel, with everything drawn, including text. Every time that text was to be displayed on the screen, I had to draw it. I don't have a drawing tablet so I was stuck with my mouse. It dawned on me halfway through to use my Laptop's touch screen, which greatly improved the quality of the drawings. I also discovered that there was different smoothing options available, which I used to further improve the quality of later graphics. Once I felt more confident, I even created some drawings, a few characters and fancy titles, just to spice it all up. :hot_pepper:
Small issue, but there was 174 graphics used in the video... So just importing them into my video software became a pain. I had a pretty neat idea though. For each 'set' of graphics, I used a different prefix, so all of the graphics related to a specific part of the video could be found in one place. Don't even get me started about the time spent making them, let's just not discuss that.

:video_camera: Video:
Okay, so as you've likely noticed, I'm stuck with freeware. I don't have any nice quality tools, just free stuff from the internet. And the video editor I selected felt free. It was pretty damn painful to do anything. Keyframes could only be linear, and things as simple as using the transparency channel of an image required a filter be applied. Even just transforming and layering graphics ended up being a pain. It doesn't seem you're meant to have 15+ layers of graphics, because at this point, the layers went up and out of the layers panel, covering everything else on the screen. So this phase of development was basically a battle against the software.
I wanted the graphics to look very professional, and I wasn't sure how to achieve this. I looked over so many different videos, but all of them used complex motion graphics, that I really couldn't replicate. I experimented with blurs, and different blending modes, which worked out surprisingly well. This was strangely simple to do, and the background throughout the video is just a long clip of me playing the game, put through an intense blur filter. The strobing colours of the enemies prevented the background from seeming static, which was a nice bonus.

:heavycheckmark: Overall:
Altogether, the video was well over a 60 hour investment, but it was fun, and I learnt some stuff. I wish that I could get some professional video editing software, but other than that, the freeware has performed surprisingly well. This is likely my last post for the season, so see yas next jam! It's been great, and I'm really looking forward to LD45. Cheers, JUSTCAMH. :wave: