Sherwood Amphos Air 2.0: Dive Data import Error

[Bryan Jacquot]

Hello,

I have a Sherwood Amphos Air 2.0 and am attempting to load data from our dive computers.  I'm using a Mac with OSX 12.4 with Subsurface 5.0.8.

When I attempt to download the dives using the FTDI mount point, the download begins, it progresses through multiple stages, and when the progress bar is near the end, I get the following error:

I've enabled logs, and they end with this each time:

[284.418129] INFO: Write: size=3, data=B10FFF
[294.972458] INFO: Read: size=0, data=
[294.972466] ERROR: Failed to receive the answer. [in src/oceanic_atom2.c:681 (oceanic_atom2_packet)]
[294.972495] INFO: Sleep: value=100
[302.746418] INFO: Purge: direction=1
[302.747353] INFO: Sleep: value=1
[302.759445] INFO: Write: size=3, data=B10FFF
[303.769107] INFO: Read: size=0, data=
[303.769109] ERROR: Failed to receive the answer. [in src/oceanic_atom2.c:681 (oceanic_atom2_packet)]
[303.769111] INFO: Sleep: value=100
[304.969832] INFO: Purge: direction=1
[304.970359] INFO: Sleep: value=2
[305.946909] INFO: Write: size=3, data=B10FFF
[307.540407] INFO: Read: size=0, data=
[307.540412] ERROR: Failed to receive the answer. [in src/oceanic_atom2.c:681 (oceanic_atom2_packet)]
[307.540433] INFO: Sleep: value=2
[307.642792] INFO: Write: size=4, data=6A05A500
[308.044079] INFO: Read: size=1, data=A5

The full log is attached.

Any suggestions?

Thanks in advance,

Bryan

[Jef Driesen]

On 12/07/2022 16:32, Bryan Jacquot wrote:
> Hello,
>
> I have a Sherwood Amphos Air 2.0 and am attempting to load data from our dive
> computers.  I'm using a Mac with OSX 12.4 with Subsurface 5.0.8.
>
> When I attempt to download the dives using the FTDI mount point, the download
> begins, it progresses through multiple stages, and when the progress bar is near
> the end, I get the following error:
>

The download fails because the Amphos Air 2.0 isn't supported yet. The protocol
is the same, which is why the downloading mostly works, but the remaining
details requires some changes in the code. The log file appears to be for a full
memory dump, so you already provided the data I need to take care of this.

I have already figured out most stuff. Can you try to import the attached xml
file into subsurface. Carefully check everything for errors (especially the
temperature and pressure graph), and report back. To avoid messing up your
logbook, I suggest you create a new disposable logbook for this purpose, because
not everything will be correct on the first attempt.

Jef

[Bryan Jacquot]

Thanks Jef.  Really appreciate your help.  I imported the XML file.  Here's what I observed.

• The lower left of the graph says "Unknown dive computer" (expected with XML import?)
• The dive numbers are reversed from chronological order, so the oldest dive has the ID of 12, with the most recent dive is 1.
• The start and end pressures are slightly different from what eDiverLog Lite shows.  I originally thought it was a difference in rounding, but there's something else going on.  I downloaded the same dives in that software on Windows, and here are a couple of examples of the differences I'm seeing when looking at the Equipment Tab in Subsurface for the start/end pressures:
• Dive #3 on June 22, 2022 at 10:31 AM, the Start Pressure in eDiverLog is 3020 and Subsurface matches with 3020; End pressure in eDiverLog is 1114 and Subsurface is 1130.
• Dive #2 on June 23, 2022 at 8:58 AM, the Start Pressure in eDiverLog is 2938 but Subsurface is 2930; End pressure in eDiverLog is 756 and Subsurface is 770.

Overall the shape of the pressure, temperature, and depth curves on the graph are accurate between Subsurface and eDiverLog.  As I closely examine the details, these are the data points I observed were off slightly.

Thanks again,

Bryan

[Jef Driesen]

On 14/07/2022 03:42, Bryan Jacquot wrote:
> Thanks Jef.  Really appreciate your help.  I imported the XML file.  Here's what
> I observed.
>

> * The lower left of the graph says "Unknown dive computer" (expected with XML
> import?)
> * The dive numbers are reversed from chronological order, so the oldest dive

> has the ID of 12, with the most recent dive is 1.

> * The start and end pressures are slightly different from what eDiverLog Lite

> shows.  I originally thought it was a difference in rounding, but there's
> something else going on.  I downloaded the same dives in that software on
> Windows, and here are a couple of examples of the differences I'm seeing
> when looking at the Equipment Tab in Subsurface for the start/end pressures:

> o Dive #3 on June 22, 2022 at 10:31 AM, the Start Pressure in eDiverLog is

> 3020 and Subsurface matches with 3020; End pressure in eDiverLog is 1114
> and Subsurface is 1130.

> o Dive #2 on June 23, 2022 at 8:58 AM, the Start Pressure in eDiverLog is

> 2938 but Subsurface is 2930; End pressure in eDiverLog is756and
> Subsurface is770.
>
> Overall the shape of the pressure, temperature, and depth curves on the graph
> are accurate between Subsurface and eDiverLog.  As I closely examine the
> details, these are the data points I observed were off slightly.

The first two are due to the import and not a direct download. For the third
one, can you send me your Diverlog Lite database? It's usually located in this
directory (might depend on the Windows version):

C:\Program Files\DiverLog\data\

Jef

[Bryan Jacquot]

I realized I dropped the group off the thread, adding back for historical reference.  Turns out Amphos Air 2.0 was not supported.  Jef kindly provided an XML of the file and made the necessary changes to add support in the next release.

Jef, when is the next release expected?

On Fri, Jul 15, 2022 at 5:24 PM Bryan Jacquot <br...@jacquot.me> wrote:

Fantastic, thank you.

On Fri, Jul 15, 2022 at 3:15 PM Jef Driesen <j...@libdivecomputer.org> wrote:

The begin/end pressure in subsurface match with the pressure samples in
ediverlog. Subsurface just looks at the first/last sample. I'm not sure where
ediverlog gets those slightly different values from.

The changes for the Amphos Air 2.0 have been pushed. It should become available
in the next subsurface release.

On 14/07/2022 16:01, Bryan Jacquot wrote:
> Sure. Here it is.  Thank you.