Fail
OK, it’s time to admit defeat. 3D Graphics Thingy is not going to happen. It’s been six months since I worked on it. Heck, I even let my web hosting account expire due to neglect.
So what happened? I ran hard into the memory interface wall. Getting a decent DRAM controller working proved to be far, far more difficult than I’d expected, even with the assistance of Xilinx wizards and prebuilt controller packages. And since getting a working memory interface is a precondition to actually doing any of the 3D stuff, well, that sure put a damper on things.
A second reason for failure is that I found working with FPGAs to be abstract and unsatisfying, and the tool software to be a nightmare. When I built BMOW, I was constantly wiring things, debugging with the oscilloscope, buying new chips, soldering switches, and generally being hands-on. In contrast, 3DGT development ended up being nothing but writing Verilog in a text editor, and wondering why the Xilinx synthesis tools never did what I expected them to. The FPGA hardware itself just sat, untouched.
So what’s next? Since last summer, I haven’t done any electronics work at all, except building a light saber from a string of Christmas lights and a flourescent tube cover. I’ve gotten pretty involved in remote control vehicles, primarily RC planes, which give a few excuses to solder and build simple circuits. I have half an idea to use an Arduino with my Slow Stick somehow, to collect acceleration data in flight, or automate aerial photography or something.
Maybe I’ll come back to the CPU design thing again at some point. I still have a 68008 and some other parts I bought last year that I never got to use, so those are still waiting for me. For all those who contacted me asking if they could build something like BMOW or 3DGT, or asking for advice, send me a note and let me know how your projects are progressing now.
Happy hacking wishes to you all!
Read 8 comments and join the conversation8 Comments so far
Leave a reply. For customer support issues, please use the Customer Support link instead of writing comments.
I was somewhat concerned when the web site went down after the long period of no updates, but I’m glad to hear that it was just a loss of enthusiasm rather than another medical crisis.
My adviser loves to tell people “you’ve got to smell the solder!” (although luckily he doesn’t mean it literally) and I know that’s certainly true for me, which is probably one of the reasons he is my adviser. Although I enjoy both hardware and software, building hardware has the nice benefit of positive feedback when I can show it to someone and say “I designed/built this myself”. It’s harder to do that with software and HDL designs since their physical embodiment is so much more ethereal.
Hi,
sorry to hear about you abandoning the project. I read all your posts last autumn when I was on a work assignment in western China and consequently had lots of free time in the evenings and tinkered with a Xilinx board I brought with me in my luggage (which I later upgraded to an an Altera DE2 board from the local ebay equivalent). I did couple of things with the boards (http://www.youtube.com/user/elpuri2#p/u/2/7UVuGzfeoWo (yes, a pong :)),
http://www.youtube.com/user/elpuri2#p/u/4/vMiwFCXo80k) and was pushing towards a Commodore Amigaish chipset with synthesized MC68k, but the thing kind of faded when I returned home and found out that I still had a life.
I never experienced that non-hands on effect, but I started really low and tried to always get tangible results fast. I would attribute most of your frustration to the Xilinx software and not to the process of working with FPGAs (which is a bit sad really).
Well anyway, I really enjoyed your 3DGT article about memory bandwidth. It really made me realize how it almost always boils down to memory bandwidth these days when logic is fast and memory is lagging behind.
I hope you manage to find the spark someday and we’ll get to read all about your adventures. Word of advice for your future endeavours: go Altera. The DE2 board is pretty nice and using the development software is pure pleasure compared to the Xilinx WebPack.
Is the Altera software really a lot better? I downloaded their tools and tried them (without hardware of course), but it didn’t really seem different from the Xilinx tools.
As for memory bandwidth, I think the Spartan-3A board probably could do the job, but getting a working memory controller is a big task. If I were designing custom hardware for 3DGT, I would use many memory modules in parallel, and use a slow but wide memory interface. That would be easier to get working, and provide similar bandwidth.
Hello!
I’ve had days like yours. I really do understand your problems here.
And of course when I could not find your site, I did infer that what happened was exactly what did.
Keep trying.
I’d like to say thank you for the well documented project that BMOW is; it is incredibly instructive to review the old posts.
I hope that renewed interest in 3DGT or another project will prompt further postings.
“If we hear, we forget; if we see, we remember; if we do, we understand.”
Don’t worry, I’m sure I’ll be doing something more. Just not sure what or when. 🙂
Ran into this and thought you might be interested: http://www.youtube.com/watch?v=jDzPfAbHLFI
That’s pretty cool– hadn’t seen that before. Crucially, he used regular SRAM, although at 50MHz. Can you even get SRAM with a sub-20ns access time? I’m also pretty amazed that somebody would create such a complex project, which I’m sure took months of effort, but have no public documentation or details about it other than a You Tube video. I can’t imagine putting in that kind of effort on a project, and not wanting to share it with others.