On Monday 23 August 2004 08:41 am, Dave McGuire wrote: > > Hmm...does a 208-pin FPGA count as wide-pitch? How about a 128-pin > > PQFP > > (which is what the MIPS CPU comes in)? How about a BGA? > > No, I'm talking about something like SOIC. Man, those pins are > just *not that small*. Let's be realistic here.
I am being realistic. My kit was a home-brew computer, something on the order of complexity of a Commodore 64. But to implement the custom video logic, I was looking for an FPGA-based solution. The smallest FPGA I could find that still fit my needs has 208 pins. > So you're advocating using crap tools to assemble circuitry? If No. I'm advocating this: when you're trying to get into the kit-building business, you must design your circuit around who your customers are going to be. I already have a full-time job; I don't need to spend the remaining waking hours of my already copious time answering technical support calls on how they need to, after spending $150 to $200 on a kit, invest another $150 to $200 in a good soldering iron. Like I said, the overwhelming majority of the customers who are interested in the Kestrel NEVER built an electronic circuit before. Never! They don't even have breadboards. > you're really talking to people who have never picked up a soldering > iron before, then I fear for our profession. Most of these people > will give up in frustration with fried components and lifted pads. David, with a comment like that, I must question where you've been all these years. The homebrew kit industry all but died along with the introduction of SMT -- it's not a coincidence as to why. > You know, the philosophy of using the right tool for the job is not > obsolete. You seem to have this idea in my head that I'm ass-backwards. Please stop. I've explained no less than three times now that these decisions are based purely on a BUSINESS-level decision-making process. Maybe not as bluntly as that, but I was hoping that you might put 2 and 2 together by now. I apologize if I seem frustrated, but I am. I hate repeating myself. If I don't design my kits around the needs of my customers, nobody will buy them. Ergo, I'm essentially out of business. It doesn't take a rocket scientist to figure this out. > Don't fear SMT. SMT is good. :-) Again, I don't fear SMT. Those $22 superscalar MIPS processors are awfully appealing to me, and as long as I build for myself exclusively, the idea of using $100 FPGA chips isn't that bad to me (seeing as how it'll probably replace at least that much cost in combined board space and discrete components anyhow). -- Samuel A. Falvo II
