[DogParkList] Re: RE: MacLoggerDX and Keyspan [Incident 020920-000000]

Don Agro dagro@dogparksoftware.com
Sun, 22 Sep 2002 11:40:53 -0400


Hi Scott,

On Sunday, September 22, 2002, at 10:36  AM, B. Scott Andersen wrote:

> Regarding Keyspan serial interface troubles with MacOS X:
>
> We need to find another program (or even two or more programs) that
> also have trouble with serial communications under MacOS X. If we
> can show it is not related to a single application then they might
> be more interested.

So far the only other one I have heard of is John Bastin's problem with 
WriteLog running in Virtual PC.

Both of these programs (MacLoggerDX and WriteLog) were trying to 
communicate with an Icom radio and failed to get a proper response from 
the Icom.
Simple IO as with a TNC and ZTerm seems to work fine on these machines. 
The Icom rigs are VERY unforgiving about RS-232 timing issues and will 
echo back a "0x7E 0x7E" flag sequence when they don't like what they 
are getting from the CI-V.

This is exactly what you, John Bastin, and Patrick Merat are seeing in 
your Debug Logs.

But how to explain all the other MacLoggerDX customers for whom the 
Icom presents no problem at all under OS X ?

Realistically, this doesn't seem to be the type of problem that Keyspan 
will be able to address. I am sure they don't have any Icoms lying 
around to test with.

Since ALL the Keyspan products are running here flawlessly on a variety 
of Macs running OS X, (and at 98% of my other customer's sites), I 
would suggest that the only way to track this down is with a clean 
install of OS X, and an install of ONLY the Keyspan driver.

I know that this is not a trivial exercise, but if you have a spare 
partition or Firewire hard drive, you can leave your working system 
untouched and experiment with this other disk/partition.

If this works, then the other 3rd party drivers can be added one at a 
time until the conflict is found. If not, it must be a hardware 
incompatibility.

At that point Keyspan might be interested in resolving the conflict.

> I might have a program that controls a telescope I could try (I don't
> know if it is out for MacOS X yet...) and I could also try setting up
> a regular serial line between the Mac and a PC or other UNIX 
> workstation.
> I'd then Kermit or XMODEM some files back and forth to see if there is
> a problem. [If I can remember how to do all this!] Other folks could 
> try
> this, too.

I think a clean install of OS X and the Keyspan driver will tell you a 
lot more in a shorter period of time.

> On the other front, perhaps Mr. Argo could ask Keyspan for some sample
> code they use to drive their own devices. There is a (very small) 
> chance
> that Keyspan is testing with something set (buffer sizes increased, 
> etc.)
> that MacLoggerDX has not set up. While I believe it unlikely that this
> could be fixed by something on "our" end, if it COULD be fixed on our
> end we are likely to get a fix much, much sooner!

I asked Keyspan for just that when I was porting MacDopplerPRO to OS X. 
They did not respond to the email.


73 Don Agro VE3VRW

D o g   P a r k   S o f t w a r e   L t d .

email: dagro@dogparksoftware.com
   www: http://www.dogparksoftware.com