[sane-devel] scanning negatives using HP4850 and genesys backend
7xsuazlhhtls at gmail.com
7xsuazlhhtls at gmail.com
Sat Aug 1 14:51:25 BST 2026
Thanks for the additional info.
I don't have a machine running Windows so USB captures from that
platform aren't possible at the current time.
Jon
On 7/31/26 12:18 PM, Ralph Little wrote:
> Hi,
> There are other backends that support transparency but the genesys
> chips are pretty much a world of their own. On the upside, the GL646
> is pretty well documented.
> I believe the main ways in which the transparency support would be
> invoked would be via the source and options related to capture type. I
> don't see any specific image capture type-related options in the
> backend but some would have to be added I suspect (e.g. for selecting
> between positive and negative, IR etc).
>
> The main challenge with genesys-based devices is that every model is
> different to every other in implementation. The Genesys chips are very
> flexible and offer all sorts of driving options and they need to be
> set up for the particular hardware that they are matched with. The
> main way to tackle this task is to observe Windows driver behaviour
> through capturing traffic and looking at the register values that are
> set. It's a little more involved but that is the main gist.
>
> I have had some dealings myself with looking at a GL646 device before
> although it was a fairly superficial look at motor stepping tables but
> I may be able to look at it in the future sometime.
> One thing you *can* do (if you have the inclination) is to make some
> USB captures on Windows and post them to an issue at our GitLab issue
> tracker (https://gitlab.com/sane-project/backends/-/work_items), if
> you don't intend to do the implementation yourself. That way someone
> could pick up the issue at some point. Ideally a developer needs the
> hardware at hand, but a rough stab could be made by looking at the
> traffic.
>
> Cheers,
> Ralph
>
> On 2026-07-31 10:01, 7xsuazlhhtls at gmail.com wrote:
>>
>> Ralph,
>>
>> Thanks for the reply.
>>
>> I looked at the Genesys Datasheet for the chip (GL646USB) that is
>> used in the HP4850. I see the register that appears to control the
>> slide scanning lamp.
>>
>> Is there another backend that has support for a similar feature?
>> Possibly I could use that as a template to better understand how
>> support for this feature might be implemented in the genesys backend.
>>
>> Thanks again,
>>
>> Jon
>>
>>
>> On 7/31/26 11:21 AM, Ralph Little wrote:
>>> Hi,
>>> According to our Supported Scanners list
>>> (http://www.sane-project.org/sane-supported-devices.html) the
>>> genesys backend doesn't yet support the slide scanning feature of
>>> this device. It requires additional code to drive the extra hardware
>>> involved.
>>> We don't have an active maintainer for the genesys backend at the
>>> moment unfortunately, so I don't know when someone would be able to
>>> get to this task in the future. :(
>>>
>>> Regards,
>>> Ralph
>>>
>>> On 2026-07-30 20:58, 7xsuazlhhtls at gmail.com wrote:
>>>>
>>>> |I recently acquired an HP Scanjet 4850 scanner. The document
>>>> scanning works with no problems. This particular scanner has a
>>>> negative/slide scanning component built in to the lid of the
>>>> device. This includes a separate lamp to illuminate the film for
>>>> scanning. The lamp in the lid comes when I run a hardware self
>>>> test procedure but I have not discovered any way to have the lamp
>>>> come on when scanning using sane.
>>>>
>>>> Is this capability supported?
>>>>
>>>> info about my setup ---
>>>>
>>>> HP Scanjet 4850
>>>>
>>>> $ scanimage -V
>>>> scanimage (sane-backends) 1.4.0; backend version 1.4.0
>>>>
>>>> genesys backend
>>>>
>>>> Fedora Linux 44 (Workstation Edition)
>>>> Linux 7.1.5-200.fc44.x86_64|
>>>>
>>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/sane-devel/attachments/20260801/55279916/attachment.htm>
More information about the sane-devel
mailing list