El 15/06/17 a las 19:53, Jim Unroe via chirp_devel escribió:
I discovered something interesting while testing the driver with the sleeps removed. While using my main programming cable with (counterfeit) Prolific chip, the driver uploads just fine with no delays in the driver. However, when using my FTDI chip based programming cable, it fails immediately after the 1st block is sent.
Hi developers, I found a behavior like that at home, I have this options as interfaces at home:
* Real Motherboard Serial via a DB9 back port. * Rather OLD all-in-one USB dongle (USRobotics) that has LPT/ethernet/Audio/Serial adapters (DB9) this one needs a extra feed of +9 to +12 V dc * Cheap chinese CH340 USB-Serial adapter (DB9) * Cheap chinese ARK316? USB-Serial adapter (DB9) * Arduino USB-TLL serial adapter (0.1" headers)
Followed by a homebrewed RS232 to TTL levels using a MAX232 like Chip that can go as high as 2 Mbps (External power feed, not powered from the Serial line)
The real Serial at the MoBo and the old all-in-one adapters work fine no matter what combination I shoot to them.
But all the rest of the adapters have it's problems keeping up with the speed... some simply don't work with some radios, other trough errors due mainly to out of sync problems at some point after the download has started, etc.
I found this behaviour programming mainly the Serie/Family 60 and 60G from the Kenwoods...
So different serial adapters (drivers) has different performance and some of them fails... I has Ubuntu Linux & Windows here, and this affects both OS.
Just my 6 cents...