Production update 2023-10-06 (Patreon)

Hello! I have some good news, a bit of bad news, and more good news.

Let's start with the bad news: to the surprise of no one, things are going a bit slower than I had set up in my production schedule. Please send your shocked scream emojis.

Good news: the main reason for this is because the project is getting better and more refined. This video will probably end up being my most influential work yet for the speedrun community of the game I'm covering. Sure, some stuff I've made becomes a resource for newcomers or to explain tricks and glitches, but I've never done anything that really advanced the level of knowledge in a game. What whiteted, Onslatt and I are working on, with data collected by Seanjohn, Ace and Irie, is gonna further the knowledge of Frigate completion odds by a huge amount, and that's exciting.

Data collection

Whiteted's romhack is finished, and we've started the data collection. Seanjohn has already completed and logged 40 runs at 1:06 pace or better. I drove out to Perfect Ace's to lend him my own Everdrive so he can do runs in the coming weeks. Irie will soon join in the data collection as well.

With all this data, we're gonna be able to form a proper distribution of how long a hostage actually takes in practice to escape for each escape point, which will ultimately be a lot more precise to determine the real odds of success, compared to a theoretical number.

This data collection also helped unearth a small issue with the theoretically calculated times, and Seanjohn figured out a better way to get these figures: he eliminated every guard in the level, save for the 6 hostages, then sent them on their way one by one, unloading them as fast as possible, and he tracked their escape times. Most of them were right on par with the figures we already had, but some had discrepancies (which were already question marked so this confirms the suspicions). Therefore, these measurements are what we're gonna use as our theoretical escape times.

Manim

In this video, there's gonna be a lot of numbers, tables, graphs and so on. Therefore, I figured it would be a good time to try using manim! If you don't know, manim is an open source Mathematical ANIMation tool originally developed by Grant Sanderson of 3Blue1Brown. If you've ever seen his videos, he is next level. 

Now, what I'm gonna do is not gonna look that good, of course, but when I manage to get past the learning curve, I think manim is going to help me take visuals to the next level whenever I need to get a bit mathematical. For example, here's a very simple table with a few values and a highlighted column.

The two lines at the bottom define an animation that beautifully draws a table that looks like this: 

The animation is automatic! I have no easy way to embed it into this post, so you'll have to take my word for it.

Storyboarding

However, adding yet another tool like this is making what used to be a very streamlined process (write script, record script, edit from A to Z) is becoming a lot more complex. About 2/3 of the script is written and the rest needs the experimental data; I rewrote a significant portion of what I had already written (and some of it already recorded...); I gather footage with BizHawk, but I often need lua scripts to track additional data; editing is a combination of Photoshop, Premiere, After Effects and now manim; there's gonna be maths to be done programmatically as well, and I have yet to figure out if I'll be able to use manim directly or if it's gonna have to be somewhere else. Thankfully, Onslatt should be able to help me out with that part.

All this to say: it's gotten complicated to know exactly what I need to do on each platform. So, I decided to do a kind of descriptive storyboard of the second section of my video, the one that's all about the numbers and the probabilities. This will help me make a very precise and detailed list of each element that I need to make in After Effects, and especially, everything that I will need to create in manim. I'm not quite done with this part yet, but I completed most of it today.

So I have in pink the stuff I need to make in After effects, and in blue the stuff I need to make in manim. Once everything is done I'll make a detailed list of all of these elements so I can keep track of it all and not animate anything I won't use in the edit.

Progress

So, with all that said, how is the actual progress? Well, like I said, slower than originally planned but still going along. My original deadline of October 22nd was moved - the sponsor dropped all of their October deals so I moved on with a new one with November as the target instead. When exactly, I don't know yet. Anyway, here's how the progress looks:

A lot of my work was done outside of these three categories though, so that's why it doesn't reflect in here. Learning manim is going to take me a lot of work and time.

Pictures

Let's end this with a few pictures of what I've edited so far!

Perfect Ace is an extremely dominant player. In green are his world records in GoldenEye and Perfect Dark.

Explaining strafing/diagonal movement, and the control styles.

Showing the path taken and the hostages released. This is the first half of the run.

Explaining the pipe warp, or in general, how Bond is able to phase through obstacles if there's a tiny gap and enough lag.

That's about it for now. In other news, there are two new songs I'm thinking of adding for this video, but I'll hold off on saying what they are until that plan is more solid. Right now, I actually don't have my piano because my brother needed a MIDI keyboard for a side gig he got himself into and I should be getting it back in a couple of weeks. The new release date is unknown but it'll be in November. I'll talk more about it once I finish storyboarding and once I can reschedule the rest of the production now that I have a better idea of the entire scope of the project.

See you in a week or so!