Website Rebuild, 24/7 Bench Operation, & Upgrades (Patreon)

Hey everyone,

We just stacked-up about 6 videos, so we're finally at full production after an extended 3-week period of internal operations improvement. I loved the time away from production we had to just make things even better, and the team is feeling fresh and excited to get going on the upcoming X3D parts. Not to get too deep into the philosophy of managing a company or anything, but one of the things we've decided as a team that we like is periods of 2-3 weeks where we decide "we're just going to lose money for a few weeks and stop production, but fix everything that's bothering us." This has been incredibly motivating -- I'd rather just burn cash for a couple of weeks than have in the back of my mind that "we really should update that test bench" or "these components really should be put back together." It helps reverse entropy of the lab in a way, and it's the same for the team. If a squeaky door has been bothering Mike (who also works facilities for us), then it's a good time to just fix that and move on. One of the things that makes this team great is that everyone here tends to have projects in the back of their minds, and so it's nice to just clear those all out and make some mental room. It feels freeing and helps get everyone out from that feeling of being overwhelmed by ever-building lists of things to do.

Website

I have intentionally kept this quiet for a while. Several years ago, I mentioned in an off-hand comment that I really wanted to update the website. I got my start there. We are finally rebuilding it (and have been for a few months) and are nearing a state where it's nearly ready.

I debated sharing this part, but I feel like some of you can relate to this feeling: Sentimentally, the website is part of my business that my dad was familiar with -- he didn't get to see the growth of the YouTube channel, and so I think of the website as sort of a continuation of what he knew of my work. Working on it reminds me of how I'd update him on every single change I made back when I was making functionally $0 per year. When working at the old house 15 years ago, I'd go downstairs to show him the new articles and the pagination features and whatever else I hacked together. The YouTube channel was secondary. We owe everything to the website, and it's bothered me that I had to stop updating it a couple of years ago. It got to a point where it was totally unscalable. I built it in a way that I understood, but could probably teach no one else because everything was broken somewhere if clicked wrong.

But, in terms of web development, I always was a 'hack' in that regard, never a pro. This time, we're working with someone in the industry (to be revealed at a later date!) to build a new website. It's mostly back-end changes. The front-end is incredibly simple: Show articles, show charts. That's about it. The back-end is designed in a way where I can finally teach others how to help me update it, which is the only way it's getting regularly updated. I'm excited for that. I know a lot of you -- including Patrick and I on the team -- prefer articles for long-term reference or lookback as it's quicker to judge charts, like if making a recommendation for a friend. The truth is that articles don't pay for a team like we have. The website couldn't even pay for me back before we had YouTube success - I was living off of money I'd saved from a previous job and taking side jobs to keep it going. But I think we have now built enough YouTube success to restart a website, even if we run it at a slight loss just as a project that satisfies the sentimentality for me and the pragmatism for you all/readers. Likely though, we'll find a way to make enough off of it where we can expand it, because obviously some profit would drive further development, and that's a good thing.

I think our plan will be to publish articles on a slight delay from videos. Transparently, this is for two reasons: (1) Just time. The staff writing the articles will be busy on the video project until it goes live, and converting the script to an article will take probably half a day of work for each one. We'll likely do that half-day of work after the video is published. It's just logistics. (2) Frankly and transparently, and I hope this isn't thrown back in my face later, the videos will make us significantly more money, and we have very real expenses these days, so we do want to make sure people have a chance to see the video first. Remember that we don't sell fully sponsored videos, we don't sell multiple direct ads per video (just one), and so we keep our possible bias down and we limit reliance on those advertisers. That does mean though that we have to maximize AdSense (which is the least conflict stream) and Patreon or GN store sales, because those are the more independent streams that allow us to snub advertisers if they suck.

Besides, some content is best consumed as a video -- like the EVGA backstory. Some is best as text though, and we've always known that because that's how I started.

Anyway, I don't have a timeline yet. I will say as much as it'll be in the first half of this year, at this point.

'Secret' Projects

Patrick spent the whole time working on a project we can't talk about yet, but I think you all will really like it. The most I can say right now is that it involves a hand-built computer part and it actually works. It's one of those "we're not going to make ROI, but who cares, it's cool" projects. 

On my end, I mostly spent time focusing on getting some better operating procedures in place for test benches, writing manuals and guides for the team to help me manage stuff that used to only be my responsibility (for like 15 years), and updating some hardware. One of the weeks was for the rest of the team to have a vacation while I stayed to just think through how to generally make things flow even better here (I'll also do a short personal break, but I stagger mine to be after the team's typically as the time to work alone when they're out is highly beneficial for tasks that require abstract thinking). The things I worked on included getting bench SOPs written in a way that enables 24/7 operation without having to change staff hours. For our reviews we're working on right now, the timeline is so tight that we actually need them running (just mathematically speaking) about 20-24 hours a day to get all the parts into the dataset for X3D. To achieve this, we're basically rotating testing based on schedule. Without saying who is who, we have one team member who prefers to start work at about 9AM (I give the team flexibility to choose their own start time range), so he checks work from overnight and starts the next test. Another team member starts mid afternoon, which gives us some overlap for a hand-off and then extension to the early night, then I cover the overnight hours. This allows us to cram-in all those CPUs or GPUs you see on charts in our reviews where we might have only had a week or two to run the tests.

Logistically, it's pretty cool. I love that we've managed to turn flexible start times into being something uniquely productive for our type of work.

Mike's 'secret project' was installing rails in the ceiling of the studio. It's super cool -- he has a behind-the-scenes video he's filming with Vitalii to show you all. Even with drop ceilings, we managed to get rails mounted (safely to the girders) to hold our studio lights, which gets tripods off the ground. It's hard to explain how much that matters unless you're in a studio environment, but imagine having a small forest in your workspace. That's what it was -- tripod forest. It's one of those projects that just feels fun to do, and sometimes that's good for us.

That's it for the update! Thanks, everyone. And thank you again for supporting us.