Sure, Why Not: Another Post (and a video of sorts) (Patreon)

A little guy video I never released

So first, here's a video:

https://www.youtube.com/watch?v=LNxiqS5G2hg

This is a Little Guy I shot back in 2024. It's mostly finished, but is missing a few (unnecessary) on screen graphics, and has one section near the end where I forgot to delete a few retakes, but for the most part this is what I had planned on releasing.

I have distinct memories of showing this to people and getting reactions, so maybe you're all going to say "dude you posted this last year," but I can only find one uploaded copy of the video and it has exactly one view, by me. If it turns out I'm giving you the same booby prize I gave you before... well, sorry about that. Besides, it's on my mind anyway.

This video is kind of emblematic of the issues that made me put the Little Guys series on pause - in a nutshell, I saw Jon Bringus' shtick and wanted to duplicate the formula, but since I didn't want to actually rip him off, I couldn't just run slightly-too-new videogames on everything, especially since that's just not really my vibe to begin with. Without that however, there wasn't really a charming and relatable angle for a general audience to click with, so I kept getting devices (I have a ton of them now) but not finding any way to show off the stuff that intrigued me without a ton of frontloading that kinda takes the fun out of it.

Case in point: I thought it was really interesting that that nicotine-ad computer was so thoroughly locked down for what it was, but I think a lot of people have just never broken into an appliance-ized computer or OS before and so that angle didn't really land the way I would have liked. And with this video, well... it's a lot tougher.

The machine itself is a luggable Pentium 133 - not something I would make a video about, generally speaking. Then it's married to a rack of Motorola 68000-based computers through a wacky PCI>VME bridge, which is intriguing if you're used to thinking about that sort of thing but, again, very hard to make poppy and relatable enough for a general audience. Really, there's nothing remarkable about this from an EE / comp-sci perspective, it's just that you don't usually see this sort of thing. Unfortunately that means that whole section of the video has the tenor of a railfan going "oh my god a Mountain Railways EMR-2200 with red livery?????? no way!" Like, if you know you know, but if you don't know, you really, really don't know and no amount of explanation will fix it.

Most people don't want to watch me point at chips and talk about datasheets, I think, and it's made worse in this case by me not actually knowing what I'm talking about. it's interesting until you've spent 20 minutes not learning anything, plus I spent too damn long on the disassembly. It's just a real mess.

So this is the point in the post where I was gonna say that this device, as a topic, is basically dead weight and that I'm gonna dump this bad video on you and just move on - but with a year and change of separation, I find myself looking at it again and thinking "I could fix this."

I tried at the time, and it seemed impossible so I just shelved it, but I think if I were to refocus on two or three of the intriguing elements (say, the independent 68000 based computers in the simplest terms possible, and the E1 packet tracer software, and the funky neon backlight) and lose all the exploratory stuff where I went down a bunch of dead ends and drew a bunch of incorrect conclusions, it'd actually be watchable.

Maybe I'll make that my next project after I get the TEMPEST video released. In the meantime, if you'd like to watch the prototype of a potential future video, the link's up there, and if you want to give me pointers on what you think would enhance it, particularly for a general audience, I'm all ears. No Wrong Answers, just ones I might not use :p

Regarding the current video, I did want to mention that editing seems to be going well, and that the foot-pedal mark system did in fact make a tremendous difference and will probably speed up future projects considerably. Hopefully this contributes a little to getting me out of the rut I've been in.

Also, despite the frustration evident in my last post, it's very nice to be back to editing from a network drive. Predictably, it's been working flawlessly and I don't have to think about USB drive issues anymore. Of course, I've immediately run into space issues again - I only have 1.5TB available on the drive after offloading all my footage so far, but it's looking like I might need to go back and reshoot the intro section of this video for a really silly, preventable reason, in which case I'll probably not have enough space to offload that footage, putting me sorta-kinda back where I was.

My plan is to delete the first day's work off the NAS while leaving it on the original SSD I shot it on, so I have it in case I want to go back to it for some reason, which... is workable, but a bummer that I don't have storage parity between my server and my raw shooting media. Oh well, there are worse problems to have.

About the silly, preventable problem though: Apparently sometime after arriving at the studio, I got a big goddamn scratch on my arm.

I was aware of this, but thought it "wasn't that bad." And, I mean, it's just a little scab and a red mark, but god, it's really prominent in a number of shots, especially from the second camera angle. The scratch stands out like... hm. Well. I guess it stands out like a sore thumb. This is the closest I've ever gotten to literalizing that figure of speech.

Point is, I had hoped it wasn't visible, and it's not like it's oozing blood or anything - hell, maybe it looks unremarkable to you, but trust me, it's really noticeable in motion, and it's bad enough that it feels unprofessional and offputting to me. And my instinct was to go reshoot the whole intro to fix this, but that's a day's work, it'll eat up space on two SSDs, it requires me to completely disassemble and reassemble both of the PCs I'm showing off for a second time, and the result could be worse in a number of ways. So instead, I'm looking at ways to fix it.

This is the same frame, but with the injury removed using the Patch Replacer in Davinci. This is basically a Photoshop clone brush except you can animate it per-frame. And at first this seems like a slam dunk - "just put a motion tracker on it," you might say, "and it'll do all the work for you." But it turns out the FX tracker isn't really capable of handling objects rotating, which arms do, a lot. The resulting motion from a tracker is very artificial compared to what I get from just setting keyframes manually, so it looks like I can fix this, I just have to hand-keyframe the whole thing.

This, over and over for 25 minutes of footage, nearly frame by frame. I can do basic motion interpolation but I have to recheck after every single move to make sure it didn't wander off my arm. And I can't really get rid of the red scrape mark due to the geometry limitations of the patch replacer. Annnnnd I can't decide where the threshold is - it would double the effort to fix every frame, so I'm trying to stick to only the shots where it's super noticeable, but that's sometimes hard to judge. Also, is that going to make people notice it even more when they start realizing that it's visible in some shots but not others? I'm not sure, but it seems worth it to try this anyway.

My impression is that this is the sort of thing professional video editors have to do on a routine basis, so I'm trying to get used to it instead of going straight to a reshoot as soon as I find a visual flaw. Maybe if I get better at this it'll make it easier to deal with e.g. realizing there was a stain on my shirt for half a shoot. I have the feeling that it's gonna be tedious no matter how much practice I have, but I should still feel confident that I can do it if I need to.

Anyway, just thought you'd find that little production detail interesting. Thanks for sticking around, hope you enjoy the video!