[vdr] Re: VDSB solved, need fix...

["C.Y.M" <syphir@syphir.sytes.net>]

Vladimir Shved wrote:
> I think I figured out VDSB problem after playing with multicard setup, I
> think its diseqc switch, at least on my system. Most people have this kind
> of setup:
>
> (Card#1)-to-(diseqc#1)-to-(lnb#1..#4)-to-(diseqc#2)-to-(Card#2)
>
> When Card#1 sends command to switch to port X, i.e. lets say port 3 and
> Card#2 records from port 4, I think the diseqc command goes through LNB and
> reaches diseqc#2 and causing to set diseqc#1 and #2 at port 3. VDR at Card#2
> either loses signal or gets wrong signal, gets confused and VDSB.
>
> This is not caused by drivers, I've tested it. The problem is VDR does not
> handle diseqc parallel operation very well(I think that's the right way to
> describe it?) At first I had two cards in the same machine, then I've put
> both cards in different machines and the result was the same: both diseqc
> switches would switch at the same time. There is no way diseqc commands
> would go through power cords, power supply filters very much all noise and
> stuff. Sometimes the second diseqc#2 would not react but I think its because
> of low signal or some sort of noise would block diseqc command? When I
> configured my channels on both cards to same satellite, the problem
> disappeared. I think all VDSB fixes were messing around with diseqc signal,
> either make it very low so it would not reach other switches or very fast,
> that second switch would not be able to react to commands. 
> There are ways to solve it, one is to setup some sort of diseqc signal
> blocker from diseqc to LNB, similar to this one:
> http://www.soontai.com.tw/DB22K.html
> (I don't think the one we want exists, the one above would not work.)
> Second is to get rid of diseqc switches, which is not practical, or is it?
> What would be the alternative?
> Third, would work if it would be possible to control each diseqc switch
> individually without interfering with other diseqc switches, I think diseqc
> v2.0 protocol supports this but I'm not sure. Need to add diseqc v2.0
> support to VDR, where each card would have its own diseqc.conf file. I think
> this the most practical solution at this time and according to info here:
> http://www.satlex.de/en/diseqc_technicians-page_13.html
> it should be possible.
> There might be other solutions to this that I'm not aware of, anyone?
>
> Right now VDR handles only diseqc v1.2 commands? Diseqc experts any
> comments, does it make any sense?
VDR only requires diseqc v1.2 cmds to function.  I think the problem 
lies with the broken v1.2 backward compatability of diseqc v2.0 
switches.  What would adding v2.0 commands to vdr do?