[nanok]

no, it seems okay, when it sees DIGITAL_CAMERA it decides it must be a pentax, as far as i could see from the code. the k20d is the first which reports something else (something with K20D instead of just DIGITAL_CAMERA), as far as i can tell.

do you still get that behaviour? have you tried using the source code linked above by ramiro instead of the debian package?

[nanok]

ramiro, i will test the gx10 and report back. we have one in the house .

[Ramiro]

Nanok, the debian package is built from that same code.
It's actually a bug in pkremote, it should not show the 14MP and **** quality settings if the camera is a K10D.
I've updated the readme today, and will hunt that bug asap.

I'm eagerly awaiting your report on the GX10.

[nanok]

hi ramiro, i saw your reply only after i wrote my message, we were writing at the same time it
seems. sorry for the uninformed oppinions

gx10:

- taking a picture works
- focus works
- focus point selection works (but see bellow)
- changing sharpness, saturation, contrast works, if "natural" is selected, not if "bright" is selected.
- changing mode while dial is in "user" mode works
- Av and Tv modes work as expected, green mode seems to work (never used it before, so i might miss something)
- green button in M mode works

now the "bad" news:
- i can select 14mp and ****, as reported for the k10d (so the "bug" is somewhere in code which is common to the two, but i guess that was quite obvious already)
- i can select pef, along with dng and jpeg (gx10 cna only do dng and jpeg)

note: i haven't tried to use the obviously impossible settings, not sure what the reaction (i can
try, along with an strace, if that helps)

some random thoughts (not gx10 related)

- when i select a focus point, it glows green for a moment, and it goes back to red (this happens
with both gx10 and k20d). i never know what point is currently selected, except when focus is
confirmed. if this information cannot be retrieved from the camera, maybe a dirty trick would be to a) remember what the user has previously set, and display that b) read the exif of the last picture and perhaps crosscheck a) with that (not sure how usefull that would be though).

-the focus confirmation is not always displayed, eventhough the camera sounds like it is happy with the focus achieved (no more hunting, quick double check and lock). i am not sure why this could be, it will need further testing (i might be wrong and the camera might simply be refusing to focus, without "hunting")

- in av/tv, the shutter speed/aperture controls, respectively, can be changed, but will go back to
the previous setting imediately (for obvious reasons). maybe it would be nicer, from an usability
standpoint, to grey them out or something. same for sharpness,contrast,saturation in "bright" mode.

overall, i would say it is more than usable with the gx10, it works very nicely, no major complaints really. so as far as i am concerned, i have to usable cameras for tethered shooting, one samsung and one pentax, thanks to pkremote, great!

i might have missed some things, please let me know if i can test further (feel free to "order" specific tests)

thanks again for the excellent work, and your patience.

[Ramiro]

Hi nanok,
those are very good news. The bugs were expected, as the program handles the GX10 as if it were a K10D.
We'll go on working to improve the program. Thank you very much for your tests an suggestions.

PS: The polarizer turning function is high on our priority list

[glasbak]

Well,
I stumbled on this thread while I was searching for a solution to get pkremote working on my Debian Etch system. I downloaded pkremote a few months ago to adapt it to a command line version, but it looks that it is already done (thanks Ramiro).

But pkremote refuses to work on my debian etch system.

I found out that the program searches for : /sys/class/scsi_generic
That was not available on my system, but I solved that with : modprobe sg

But both '/sys/class/scsi_generic/%s/device/vendor' and
'/sys/class/scsi_generic/%s/device/model' stay empty,

The kernel log with dmesg shows this when I plug in my K10D :

Code:

usb 4-1: new high speed USB device using ehci_hcd and address 5
usb 4-1: configuration #1 chosen from 1 choice
scsi3 : SCSI emulation for USB Mass Storage devices
usb-storage: device found at 5
usb-storage: waiting for device to settle before scanning
  Vendor:           Model:                   Rev:     
  Type:   Direct-Access                      ANSI SCSI revision: 00
sd 3:0:0:0: Attached scsi removable disk sda
sd 3:0:0:0: Attached scsi generic sg0 type 0
usb-storage: device scan complete

Notice that the Vendor and Device fields are empty, if I put a memorystick in the same USB slot, I get values there.

'usbview' does show the K10D, so something is recognized.

Anybody knows what it takes to get this working with Debian Etch ?

I already increased '/sys/module/usb_storage/parameters/delay_use' to 15, but that did not help.

George

[nanok]

do you have hal running? (not sure if it was default on etch)

[glasbak]

do you have hal running? (not sure if it was default on etch)

Nope, now I have it running, but it makes no difference.
It probably will boil down to making my own kernel (or module), that seems the preferred way how to solve linux problems.

George

[nanok]

i doubt if that has anything to do with it. are you using the stock kernel now?

try to just connect the camera and see what pkremote says, with sg module and hal running, and see how it behaves

[glasbak]

I doubt if that has anything to do with it. are you using the stock kernel now?

try to just connect the camera and see what pkremote says, with sg module and hal running, and see how it behaves

I am running the stock Etch 2.6.18-5-686 kernel.
Hal present or not does not make any difference.
pkremote says nothing, because it wants the right strings in the Vendor and Model fields.
These are empty, so pkremote can't do anything.
Can you, or somebody else look what 'dmesg | tail' shows when the K10D is plugged in ?

George

[nanok]

i only have a samsung gx10 (similar to the k10d), the k20d identifies itself differently,

[814469.152669] usb 3-1: new high speed USB device using ehci_hcd and address 6
[814469.300667] usb 3-1: configuration #1 chosen from 1 choice
[814469.304681] scsi8 : SCSI emulation for USB Mass Storage devices
[814469.304681] usb-storage: device found at 6
[814469.304687] usb-storage: waiting for device to settle before scanning
[814469.304681] usb 3-1: New USB device found, idVendor=0839, idProduct=005e
[814469.304681] usb 3-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[814469.304681] usb 3-1: Product: GX10
[814469.304681] usb 3-1: Manufacturer: SAMSUNG TECHWIN Co.

nanok@spartakus:~$ cat /sys/class/scsi_generic/*/device/vendor
ATA
TOSHIBA
SAMSUNG
nanok@spartakus:~$ cat /sys/class/scsi_generic/*/device/model
WDC WD1600BEVS-6
MK3006GAL
DIGITAL_CAMERA

(the last string in each case is what applies for the gx10)

note that the dmesg was without sg module, uppon inserting the module:

[814474.306349] scsi 8:0:0:0: Direct-Access SAMSUNG DIGITAL_CAMERA 1.30 PQ: 0 ANSI: 0

hope this helps, let me know if i can assist you any further.

i am running unstable, 2.6.26-1-amd64 stock kernel, pkremote compiled from source (i couldn't find a precompiled amd64 binary, iirc)

edit: i just realized, i meant udev, not hal. my appologies. try starting udev

[glasbak]

After a 'apt-get dist-upgrade' I got a new kernel.

Code:

[  112.462089] usb 4-1: new high speed USB device using ehci_hcd and address 2
[  112.594971] usb 4-1: configuration #1 chosen from 1 choice
[  112.596011] usb 4-1: New USB device found, idVendor=0a17, idProduct=006e
[  112.596011] usb 4-1: New USB device strings: Mfr=1, Product=2, SerialNumber=0
[  112.596011] usb 4-1: Product: K10D
[  112.596011] usb 4-1: Manufacturer: PENTAX Corporation
[  112.816683] Initializing USB Mass Storage driver...
[  112.821844] scsi0 : SCSI emulation for USB Mass Storage devices
[  112.823544] usbcore: registered new interface driver usb-storage
[  112.823631] USB Mass Storage support registered.
[  112.831931] usb-storage: device found at 2
[  112.831967] usb-storage: waiting for device to settle before scanning
[  117.828539] usb-storage: device scan complete
[  117.830218] scsi scan: INQUIRY result too short (5), using 36
[  117.830308] scsi 0:0:0:0: Direct-Access     PENTAX   DIGITAL_CAMERA   1.30 PQ: 0 ANSI: 0
[  118.025714] Driver 'sd' needs updating - please use bus_type methods
[  118.032912] sd 0:0:0:0: [sda] Attached SCSI removable disk
[ 1207.692342] sd 0:0:0:0: Attached scsi generic sg0 type 0

And this 'INQUIRY result too short (5), using 36' is probably the new part that does the trick.
Just made my first picture with pkremote and my k10d

george

[nanok]

great to hear it's working now. maybe ramiro can use the info if there is indeed a requirement for a newer kernel, to add it to the readme (though it might be a bug in the sg module of that particular debian kernel you were using)

[ahopkins]

Ramiro - Just want to say thank-you for taking the time to put this together. It is refreshing to see the convergence of FOSS with photography whenever it happens.

[and]

Just discoverd this, very cool stuff. great work. I wonder if the k7 will have remote capabilities even tho pentax themselves are not planning to release any remote software for it...