> On Dec 10, 2021, at 12:57 PM, Guy Sotomayor via cctalk 
> <cctalk@classiccmp.org> wrote:
> 
> All valid points.  My frustration has been where I see projects that use a 
> RPi for something that a simple HW circuit/CPLD/FPGA could have done it 
> simpler and more efficiently.
> 
> I've lost count of the FPGA boards that I have.  I also typically don't use 
> eval boards for actual projects other than testing a few "flows".  Everything 
> gets done with a custom board because I typically need other components and 
> it gets too messy if I'm just using an "off the shelf" eval board (and more 
> than likely the eval board doesn't have enough I/Os).

I've done some pretty large VHDL designs (though not structured in the most 
conventional manner).  It's definitely one option.  It's a good one for large 
designs, or ones with severe performance requirements.

FPGA (VHDL or Verilog) work requires learning not just a new language but a way 
of looking at design that's unfamiliar to software people.  It's interesting 
and rewarding but not a simple thing to do.  And even "simple CPLD" projects 
aren't necessarily simpler done that way, at least not for people like me.

To pick an example: I have a rudimentary software-defined radio I built around 
1998 (using Harris SDR chips) which uses a Lattice CPLD to interface to an EPP 
mode parallel printer interface.  That's conceptually a simple task but it 
turned out to be surprisingly hairy.  At the time it was the best answer; if I 
were to attempt a similar project today I would most likely pick a Raspberry 
Pico to do the interfacing.

The Pico is in fact a good example to point to.  While a Pi is a full Linux 
system, a Pico is a deep-embedded device, tiny and dirt cheap even by Arduino 
standards.  I've seen low end FPGA boards but I don't know if you can beat $4 
as the price tag.  And add to that the fact that FPGA development software 
often only runs on *gag* Windows, while embedded software boards such as RPico 
offer open IDE systems that run on real host operating systems.

> I should also note that the Beagle Bone MFM emulator isn't actually fast 
> enough.  It works OK if you only have one drive but it's not fast enough to 
> handle the drive select signal when you have more than one.

I haven't run into that, it's rather puzzing why that would be so given that 
MFM does one drive at a time.  In any case, the original comment was about 
RC11, which is massively slower than MFM disks.

        paul

Reply via email to