Most Interesting Project You Worked on?

Pages: 12
Title^. What was the most interesting, and fulfilling project that you have worked on. I'm interested because it will give me some ideas on what I can work on in the future.
closed account (3hM2Nwbp)
I'd have to say that writing software for a custom PCM1 would definitely be the most interesting, although it ended in complete and utter failure. There's a lot more that I need to learn about engineering before reattempting it. :)

[1] http://en.wikipedia.org/wiki/Powertrain_control_module
closed account (z05DSL3A)
The control system for a custom made dedicated thermal transfer printer. That was a few years ago for work.

I'm currently gearing up for a home project of a 3D printer. I'm planning on doing everything myself (the whole mechatronics bit), lots of projects leading to one goal.

I'm pretty sure I can't do it, but I'm working on an OS written in a combination of C, assembly and Haskell (just 'cause). I've already accepted that I'm going to fail, but I don't really care, I'm doing it anyway.
Every project I've worked on. Sadly, my confidence usually crashes it (well lack of confidence rather).
the most interesting, and fulfilling project

Working on an RTOS was probably most interesting, I got to live on the other side of the fence, scheduling and queuing and waking all those user (and kernel) threads, servicing syscalls, reaching everywhere in hardware. It was also my first major corporate job, lots of new experiences. It was also not C++ (except where dealing with the clients and their C++ code)

Most fulfilling was probably when I wrote a high-speed open-road vehicle identification/classification system from scratch in pure C++/boost, replacing a stone-age RogueWave+ACE/TAO+proprietary embedded language setup that only worked with single-lane gated traffic. Hooked up a ton of new hardware, signed for a ton of new requirements, and team-led a bunch of new people getting them done, and then it all just worked. It felt very fulfilling to see the key customer, whose billion dollar revenue depends entirely on my code, make jokes and discuss philosophy at the end of acceptance testing, after the governor and the press left.
Made a prototype piece of kit for driving vehicles remotely (or just from the safety of sitting in a completely enclosed cabin).

Hemispherical lens on top of the vehicle dumping light to a CCD, and an image presented to the driver on a helmet mounted tiny little flip-down screen, where the image is in the direction his head is pointing, via an inertial sensor on the helmet, making the cabin he's sitting in essentially transparent to him. The image shown to the driver generated in real time from the huge hemispherical fisheye lens image; right section picked out based on the inertial sensor and then various processing to turn it from a distorted fisheye image into something sensible a human can use to drive with.

As added bonus, the entire image was recorded, so at any time one could review any part of the complete hemisphere; he couldn't just check over his shoulder - he could see what was over his shoulder ten minutes ago. Sometimes other people were very interested in having a good look at things after the drive.
Last edited on
Well damn, these projects sound really cool :( How does one manage to do these?
Just do it. I know my project will fail, but I'm doing it anyway.
Well, unfortunately I don't have any giant spherical lenses I can stick on my car :(
Think of something else you can do with it, then. You could get some sensors to detect the distance to the nearest object in front of a car as well as the speed at which the car is travelling and then use that information to tell if the car could stop in time to avoid a collision with that object. You could then warn the driver that they're going too fast or whatever.
Last edited on
Oh geeze, I've got a lot of failed 'experiments' lying around.

2 Newtonian physics simulators

3 Game engines (1 isometric 2D, 1 side scrolling 2d, and one 3D (one of the physics simulators was geared towards this))

1 failed attempt at making an OS in C++

And, several failed chat-bots, each of which more complete and functional than the last. I'm actually drafting designs for another attempt right now. The idea would be to have it scour BBS style forums to use for it's primary input in learning.

chrisname wrote:
(just 'cause)

Funny, that looks more like LISP syntax than haskell :P
Oh, don't worry, SchemeOS is just around the corner :P
How does one manage to do these?


That prototype was funded entirely by the company. At home my projects are more mundane. How much cheese can I fit inside a toasted sandwich, that kind of thing. :)
closed account (o1vk4iN6)
Just out of curiosity what company and what is it's purpose ? Sounds interesting and fun if you get to do that kind of stuff often.
chrisname wrote:
Oh, don't worry, SchemeOS is just around the corner


I think I've actually read about a real SchemeOS, or some OS written in a FP language.

... my roommate pointed out my chatbot will just become a gigantic troll. That's a risk I'm willing to take. For science. You monsters.
Last edited on
he idea would be to have it scour BBS style forums to use for it's primary input in learning.


I didn't notice this earlier, I now see why your roommate said that haha
Just out of curiosity what company and what is it's purpose ?


Mid-sized company of a multi-national defence megaCorp.
To be honest, seeing all of these responses makes me want to cry. I was going to say my most fulfilling program was when I designed Tetris from scratch, wrote it completely on paper, and inserted it into my compiler for a total of 180 lines...I feel like wasted talent -.-

On the plus side, I've strived to improve all of my high school programs and have already completed part of one and learned some new standards along the way. =)
I once attempted to write my own compiler for my own, custom-designed programming lanauge. The project fell threw however when I lost intrest. I learned a fair bit though.
Pages: 12