| By Date: | <-- --> |
| By Thread: | <-- --> |
Mac Michaels wrote:
On Monday 27 February 2006 12:59 pm, Jon M. Lamb wrote:I have the ATI HDTV Wonder which does not utilize the gpio the way your card does. Were there any cards that had more than one RF input when the structures for these drivers were started? It seems that the structures describing the cards may not be robust enough to handle cards that do not use the gpio. I think that wonderful work has been done making these drivers. I am by no means an expert on any aspect of them or on programming. I just do not understand why this issue ,not being able to select an RF input, keeps being pushed off as a missing feature of the v4l2 API.
Hello to all,
I posted this in a thread and got no response but it may not have been the proper place to ask. Anyway....
Michael Krufky wrote:
Please see the discussion in the list between Mac Michaels, Kirk Lapray and I about this very topic. We agree with you, but the API doesnt allow it in analog mode, unless you can change tuner input using GPIO...
I was reading the Video for Linux Two API Specifications (Is this the API referred to) trying to understand them. Under section 1.6.1 Tuners of the specification it says:
Video input devices can have one or more tuners demodulating a RF signal. Each tuner is associated with one or more video inputs, depending on the number of RF connectors on the tuner.
And in section 6.2.9 V4L2 in Linux 2.5.46 2002-10 it says:
6. The struct v4l2_tuner <http://v4l2spec.bytesex.org/spec-single/v4l2.html#V4L2-T UNER> |input| field was replaced by an |index| field, permitting devices with multiple tuners. The link between video inputs and tuners is now reversed, inputs point to the tuner they are on. The |std| substructure became a simple set (more about this below) and moved into struct v4l2_input <http://v4l2spec.bytesex.org/spec-single/v4l2.html#V4L2-I NPUT>. A |type| field was added.
To me this implies that the API spec does handle RF input selection by selecting a different video input. The driver should be able to define any number of inputs, RF or not, which could point to the same tuner. Am I misinterpreting the API Spec?
Jon
This is how I interpreted the spec when I implemented the driver for DViCO HDTV3 Gold-Q. This is one of the few cards that uses a Tuner with two RF input connectors used for analog TV. Either RF connector can be used for over-the-air or cable. The actual frequency tuned is under control of the application.
This card also handles digital TV using the same tuner. It works a bit differently. For digital signals one connector is assigned to VSB-8 for over the air and the other connector is assigned to QAM-256 for cable. (The connectors are actually labeled ANT and CABLE although either RF connector can be selected for VSB-8 or QAM-256.) This was done because the DVB API has no way to select a video source ... it is digital data not video. Neither spec had an RF input selector.
-- Mac
Jon
-- video4linux-list mailing list Unsubscribe mailto:video4linux-list-request (at) redhat.com?subject=unsubscribe https://www.redhat.com/mailman/listinfo/video4linux-list