Fwd: direct import form dive computer

[Jef Driesen]

On 2017-03-14 00:43, Alessandro Volpi wrote:
> I have tried once again to obtain the SmartTec memory dump with 
> dctool.exe,
> on the WXP VM . VirtualBox was running on my  Ubuntu trusty laptop.
> 
> The command was THE RIGHT ONE, as suggested by Jef : "dctool.exe -v -l
> smart.log -f smart dump -o smart.bin" ; the failed trials described in 
> my
> previous message were carried on with the WRONG COMMAND :
> "dctool.exe -v -l smart.log -f smart.dump -o smart.bin"
> 
> The smart.log and smart.bin are now available in subdirectory
>  smarttec_import within my Dropbox folder :
>  
> https://www.dropbox.com/sh/dsg43qcbo013i1k/AAC_vCS3cZeZar5i3HngCNq3a?dl=0
> 
> I apologize for the inconvenience.

As I already suspected, downloading the Smart Tec works fine on Windows 
XP too. That pretty much confirms the problem is most likely located in 
the linux IrDA drivers, and not libdivecomputer or subsurface.

Can you repeat the same test, on the Ubuntu Trusty system where 
downloading the Galileo worked? You can use subsurface, or the linux 
version of the dctool command-line application:

http://libdivecomputer.org/builds/stable/linux/dctool

> Anyway it is still to be understood why does the subsurface memory dump
> work with the Galileo and not with the SmartTec; the irda interface is 
> a
> FIR and, AFAIK, a FIR device is characterized by high baud rate (16M in 
> my
> trusty system), but should also be able to work at lower data rate when 
> the
> connected device does not support such a high baud rate. The time 
> required
> for the SmartTec memory dump has been much longer, compared to the same
> operation carried on the Galileo DC

I don't think this is related to the IrDA data rate. I think the 
explanation is a lot simpler: The Smart Tec memory dump is almost twice 
as large as the Galileo one (259K vs 154K). So it's not unexpected it 
takes longer. I don't know how long the Galileo download took (the 
subsurface log doesn't include timestamps), but the Smart Tec download 
took 613 seconds. So if the transfer speed is the same, I expect the 
Galileo download took approximately 364 seconds. If that's indeed 
correct, then you have the answer for the longer download time.

(If you want to take the test, just download your galileo with the 
dctool on the same Windows VM, and check the last timestamp in the log.)

Jef