Developers December 16, 2020

The Problem With Trying to Crunch To Success

Beekey Cheung @ProfessorBeekums

The Cyberpunk 2077 launch has been both a success and a disaster at the same time. The disaster portion is the huge number of bugs, especially on consoles. What I find interesting is that while they tried to not resort to crunch at first, the studio decided to have the development team crunch the last 2-3 months before release.

The problem is that software development is not an assembly line. More hours does not always mean more productivity. Sometimes, more hours can actually lead to less productivity. I once saw a 5 person game development team attempt to crunch for 3 months. At the tail end, the entire team spent several days trying to fix a bug that ended up being an "A" being capitalized in a configuration file. These weren't stupid devs. They were tired devs.

I started developing my own beliefs about crunch when I spent two 14-hour days working on a single bug ridden feature. Later on, a well rested version of me rewrote the entire thing bug free in 2 hours. It wasn't that those 28 hours were a valuable learning experience either. I saw precursors of the 2 hour solution in the code I deleted. I was so tired that I must have forgotten what I was intending and went down a different (and awful) path.

I definitely work long hours as an entrepreneur, but I literally own everything now. My product is also my baby. There's a love of the work that helps energize me. There's a difference between that and an employee working for a salary and 0.01% (or less) of equity. Those employees can't be expected to work long hours and still be productive.

  1. 2

    I kind of view it as an addictive cycle. You get hooked on the "deadline rush" and you find yourself needing it to get off the couch. Then you crunch and you're exhausted and burnt out, so your productivity crashes and you find yourself behind on another deadline, which means crunch and burnout, and the cycle continues, because you can't stop, things have to get done.

    I'm not gonna say I'm past it, I still get stuck in that self-destructive cycle, but I can see it for what it is.

    I'll also say that working at a company, there's also a malaise that can creep in where you're just sort of keeping up with the rest of the pack and not really maintaining the level of production you see working on your own stuff. A lot of people solve this problem by not really shipping much of anything, and that's.... in my opinion that's almost worse. If I'm sitting in this chair for a reasonable work week I want something to show for it. Sure, I can put 40 in, clock out, and be well rested, but that's the single biggest chunk of my life spent for that week. If I didn't get anything out the door after all that, I'd be pretty bummed, rested or not.

    1. 1

      I'll also say that working at a company, there's also a malaise that can creep in where you're just sort of keeping up with the rest of the pack and not really maintaining the level of production you see working on your own stuff.

      This is really on point. I hit this really badly with my first two startup attempts. I don't think I ever really found a great solution. My current situation is a bit of a hack. I built a tool to help my productivity so it helps with my day job. Doing my day job normally doubles up as product research because I learn what my own pain points are.

  2. 2

    The problem is that software development is not an assembly line. More hours does not always mean more productivity. Sometimes, more hours can actually lead to less productivity.

    So true. See also: more people. Very early in my career, I learnt that adding more people to a late project will make it later.

    1. 1

      Very very true. My best story about this is a project that was expected to take 3 developers 3 months. One week before the deadline, they added 17 developers to the project. This was done while "The Mythical Man Month" was literally on someone's desk. One developer even raised doubts about it during the pump everyone up meeting (I am ashamed it was not me). That developer was chastised for trying to bring down morale.

      Fast forward 5 months... the project finally launched as a buggy mess.

      1. 1

        Whaaaaaaasat? 17 devs?😂

  3. 2

    "I definitely work long hours as an entrepreneur, but I literally own everything now."

    I think when we are working on our babies, it's harder to notice when you are tired. I hate rest days, when I take them I miss coding, I really do, but I've learned that it's better long term to not push it too much even if you feel motivated. I've had the exact same experience as you, trying to fix a bug for days, then when I'm exhausted and frustrated taking a break and coming back to find it was the simplest thing. It would have been so much more productive to notice when I'm not making progress and stop for the day.

    1. 2

      I agree with you in many ways. But I also find that the "work" as a solo founder is more varied. Networking and marketing doesn't always feel like work as a developer, but it is every much as important as writing code. I find myself enjoying it at times to. So while I try not to work too hard on writing code all the time, I find that spending a few more hours doing the non-coding work is somewhat invigorating.

      1. 1

        I know what you mean, I'm my own graphical designer and I do see creating icons for my program, thumbnails for my videos and so on as not work, relaxing even. I should do more networking and marketing like you, those are definitely the things I'm worst at because I have almost zero experience, but maybe I would like them.

Recommended Posts