[Coco] DriveWire problem
bryan at clingman.org
bryan at clingman.org
Tue Feb 16 18:10:49 EST 2016
Hi,
It could be because the ftdi drivers in certain cases inject some warning text into the serial data stream when they detect a counterfeit device.
http://www.eevblog.com/forum/microcontrollers/ftdi-gate-2-0/
--bryan
On February 16, 2016 3:02:30 PM PST, Steven Hirsch <snhirsch at gmail.com> wrote:
>On Sun, 14 Feb 2016, Aaron Wolfe wrote:
>
>> On Sun, Feb 14, 2016 at 5:07 PM, Steven Hirsch <snhirsch at gmail.com>
>wrote:
>>
>>>
>>> The DW GUI is showing pairs of messages:
>>>
>>> Sun Feb 14 2016 16:53:47.362 WARN DWProtocolHandler dwproto-0-9
>>> Timed out reading from CoCo in OP_WRITE
>>>
>>> Sun Feb 14 2016 16:53:47.367 WARN DWProtocolHandler dwproto-0-9
>>> OP_WRITE took 286ms.
>>>
>
>> This means the server is waiting for data from the CoCo and not
>getting it
>> within the expected time period. Some operations like writing
>involve a
>> few back and forth exchanges between server and coco. If either side
>takes
>> longer than ~250ms iirc then the other side is supposed to assume
>failure,
>> which is what the server is warning about. It really shouldn't
>happen.
>> So... first thing I'd check is the serial cable itself for anything
>loose,
>> etc. Some people have had this kind of result when the cable was
>just too
>> long. OP_WRITE is one of few times the coco actually sends a lot of
>bytes
>> in a row over the wire so it can reveal a more general problem.
>
>Give the man a prize! I bought a new name-brand FTDI cable, swapped
>out
>my $5 no-name flea market adapter and life is now good. I copy data to
>
>the DW volume to my heart's content. The original was probably a
>counterfeit FTDI (as so many cheapie adapters tend to be) and likely
>succumbed to electro-migration.
>
>Thanks, Aaaron.
>
>
>
>--
>
>--
>Coco mailing list
>Coco at maltedmedia.com
>https://pairlist5.pair.net/mailman/listinfo/coco
--
Sent from my Android phone with K-9 Mail. Please excuse my brevity.
More information about the Coco
mailing list