Hi My bag full of useless adapters were indeed "class compliant" back in the day. Not so any more under the new approach to signed drivers.
Bob On Dec 19, 2010, at 3:09 PM, Chris Albertson wrote: > Yes. It can if done wrong. > > The "correct" way to use USB is to make your USB device "class > compliant" for some class. Then most OSes (and with recent releases > even MS Windows) will have built-in drivers for each class of USB > device. Never buy or design a USB device that requires the user to > instal device drivers. > > The worlds has enough poorly designed USB devices that only have > drivers for old versions of Windows. We don't need more of those > > Still. If I were designing a counter I'd just have to talk to a local > bus using SPI, "2 wire", I2C or whatever and then have a secon module > that converts "whatever" to USB, rs232, or just a pannel with nixi > tubes and toggle switches > > > On Sun, Dec 19, 2010 at 7:55 AM, Bob Camp <[email protected]> wrote: >> Hi >> >> USB ties you into the same silly PC driver two year life silly stuff. The >> bus has a lot of staying power, but keeping the stack up to date is a pain. >> Even for so called "standard" parts that interface to a "common" interface - >> neither one really turns out to be true. I have a big bucket full of serial >> adapters that were standard parts under XP. No drivers to install, just plug >> and play. Under the more modern stuff - no driver available. The gizmos are >> now Christmas ornaments. >> >> Bob >> >> >> On Dec 19, 2010, at 10:35 AM, jimlux wrote: >> >>> Gerhard Hoffmann wrote: >>>> Am 19.12.2010 05:52, schrieb Bob Camp: >>>>> The real thing you would learn about is writing code that runs an FPGA. >>>> Yes. A good starting point would be a Xilinx SP605 kit. >>>> It's about the $750 for a decent 5370 and includes >>>> many points from Bobs option list. >>>>> The other gotcha here is that the feature list can get pretty large: >>>>> >>>>> >>>>> 3) USB 2.0 interface >>>>> 4) RS-232 interface >>>>> 5) HPIB emulation of an HP box (w/o drivers) >>>>> 6) Ethernet interface >>>>> 1) Web server software >>>>> 1) Front pannel controls (W/o mechanic) >>>>> 2) Front pannel display (VGA output) >>>>> 3) Front interface connections (DUT's) >>>>> 4) Rear pannel standard interfaces and controls (w/o mechanics) >>>>> 2) Flash card storage >>>>> 3) USB stick storage >>>>> 4) SD card storage >>>>> 5) Battery backed RAM storage >>>>> 1) Some number of counter inputs ( some programmable SMA I/O) >>>>> 2) Some number of reference inputs >>>>> >>>>> 1) Battery power >>>>> 2) Auto 12 V power >>>>> 3) AC line power >>>> use as a PCIE card in a pc or mac >>> >>> >>> I hate cards that plug into a PC. The PC bus respin cycle is much too >>> short, and you have device drivers, etc. to worry about. At work, we use >>> rack mounted PCs to control a bunch of test equipment. Since we're doing >>> spacecraft stuff, the "design use life" of the rack is 2-3 years, but it >>> will be used beyond that, and often, will get reused for the next project. >>> >>> I've spent way too many hours hunting for another ISA bus machine, or >>> trying to resurrect NT4.0, because there's no device drivers for anything >>> newer. >>> >>> In my book, RS232 is fine for low end, Ethernet is even better, USB looks >>> pretty good and has some staying power. Folks implementing things on USB >>> tend to use simple conceptual models (e.g. emulated serial port). >>> >>> My only gripe about USB is that it's a very master/slave sort of thing. >>> >>> >>> _______________________________________________ >>> time-nuts mailing list -- [email protected] >>> To unsubscribe, go to >>> https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >>> and follow the instructions there. >> >> >> _______________________________________________ >> time-nuts mailing list -- [email protected] >> To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts >> and follow the instructions there. >> > > > > -- > ===== > Chris Albertson > Redondo Beach, California > > _______________________________________________ > time-nuts mailing list -- [email protected] > To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts > and follow the instructions there. _______________________________________________ time-nuts mailing list -- [email protected] To unsubscribe, go to https://www.febo.com/cgi-bin/mailman/listinfo/time-nuts and follow the instructions there.
