Trace Systems HID maker - anyone used it?

[dbotkin]

Has anyone used the Trace Systems HID Maker software? It's supposed to generate both PC-side and PIC-side source for PIC USB applicaitons. What caught my interest is its claim to procude CCS compatible C source for the 16C745, which would save me countless hours of work.

I haven't yet decided what processor to use for an upcoming project. The Cypress chips are lower cost but lack ADC and PWM, both of which I need. I can do both in firmware, but would prefer not to.

If you've used HID Maker I'd love to hear about your experiences, regardless of which processor or compiler. Of course I'm especially interested if you've used it to produce C source to use with CCS C.

[Guest in Japan]

Wow! What a neat tool!!!! Please buy and use it and let the group know how it worked! Just kidding - pretty much.

[MikeW]

I think that if one has to purchase a vendor ID, that costs several thousand dollars !!

I could be wrong, but make sure before you buy the $399 software

[dbotkin]

Yes, a USB vendor ID will cost $1500 if and when we go to production. I'm aware of that. However, the more I look, the more I think I can use the CCS USB demo code and adapt it to my needs. All I'll be doing is sending short blocks of data to and from the PIC on a pretty regular basis -- think low bandwidth streaming data in both directions, like 9-10 bytes in each direction a few dozen times per second. Well within the limits imposed by low-speed USB and the HID class drivers, I think.

I've just never done anything at all with USB before, so I'm taking a step back, ordering USB Complete and a 16C745/JW, and starting from scratch. It's new territory for me, as it is completely different from the stuff I know quite well. TCP/IP, serial, parallel, Morse code, motor control, servos, ultrasonic rangefinding, even RF and Manchester coding seem pretty straightforward compared to USB. In the past I've tended to approach it as a matter of firmware being something you write to make the hadware work; in this project, it's more like the hardware is just a physical container to make the firmware work. I wrote code to replace hardware, but USB is something for which there isn't a corresponding hardware concept from which to work. It's a pretty big paradigm shift for me.

There's also going to be a pretty heavy host-side component to this project with a good deal of the functionality provided by VB or VC++ code on the PC side, which I have so far avoided in every project. I'm not particularly looking forward to it, but we'll muddle through.

[rwyoung]

If you aren't dead-set on using the 16C745 you could use any other PIC and run the USB side with an FTDI FT232 chip. USB on one side and serial port with handshake lines on the other side. They also have a dual serial port versions and parallel port versions. Pretty simple and for low volume stuff I think it is a pretty cost effective choice. Pay a little more for the chip but pay less (money and brain-sweat) in software development.
_________________
Rob Young
The Screw-Up Fairy may just visit you but he has crashed on my couch for the last month!

[dbotkin]

Yep, looked at that already. In fact, up until recently the plan was to develop with the FTDI chip with a PIC and consider moving to a Cypress or PIC only solution later on, if we needed to worry about volume production. Having looked a little deeper, though, it almost triples the cost of the hardware, and also complicates things a little on the host side. It looks like if we do it right with a HID class device we will need no host drivers or anything, only the single application executable; this would be a BIG plus with only one piece of application layer software to maintain, and no install issues to worry about. Since the CCS drivers are there, we decided to go ahead and do it right the first time.

Incidentally, I have a brand new, never-used FTDI DLP-USB245M module if anyone wants one... I just paid Mouser $25 for it, would like to see $20 or better. It's USB on one side, 8-bit parallel on the other side.