[mythtv-users] Questionable capture quality from a PVR-250

[Jarod C. Wilson]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Wednesday 04 February 2004 10:37, brentfisher at shaw.ca wrote:
> I know that this has been touched on before but I still don't have a sense
> of where the problem lies based on the inconclusive discussions to date.
>
> The video that I capture is on my PVR-250 is of questionably quality;
> slightly fuzzy and washed out.  My symptoms are similar to those that Cory
> described a few days ago, refer to:
>
> http://www.gossamer-threads.com/archive/MythTV_C2/Users_F11/PVR-250_slightl
>y_fuzzy_picture_P104263/
>
> I have been chalking this up to the fact that MPEG-2 is a lossy compression
> - I'm taking an analog signal, sampling it into a compressed digital format
> - some degradation is expected.  But increasing the sampling rate should
> then impove it, right?  This doesn't seem to be the case.

I think a bigger issue may be that you're capturing an interlaced signal, 
deinterlacing it (presumably) with MythTV, then spitting it out your video 
card, which then re-interlaces the picture for the TV. This is the primary 
reason the PVR-350's output is superior -- you never have to deint/reint.

> Then I read some of the conversations on this mailing list and there are
> those claiming that their systems are outputting video that is in some
> cases superior to what the tuner on their TV is producing.  Jarod says that
> it should at least be on par.

Correct. And I'm viewing on a 47" HDTV. ;-)

I believe some of the major factors are signal quality and grounding. 
Televisions are a bit more fault-tolerant of dirty signals and grounding 
problems than computer TV capture cards. I have everything properly grounded 
and an amplified signal splitter feeding my capture cards.

Also, upon further VERY close examination, I do see the PVR-350's output as 
being crisper than that of my GF4 (even when using VGA-Component, thus no 
reinterlacing). The difference I see isn't huge, even when using SVid out on 
a GF4, but I have a very clean signal going into my capture cards, so that 
may help. And the issue is NOT capture side (for me), because I can watch the 
exact same recordings on the GF4 and 350, with different visual results.


> The tuner in my TV produces a noticably crisper picture compared to what I
> get out of my MythTV box.

For me, the GF4 playback is nearly indistinguishable from what comes off my 
cable boxes. Perhaps a touch softer/less sharp, but also reduced in noise. 
The 350's output is indistinguishable (I haven't really compared noise versus 
the cable box though).

> I don't suspect that the problem lies with my TV-out as DVD output from
> xine is quite good.

Keep in mind that DVD output is non-interlaced, so less conversion steps, 
fewer places for things to go awry.

> I have also downloaded some TV show episodes from a
> newsgroup (which are presumably captured using similar means as I am using)
> and they seem considerably higher in quality (also played back using xine).

There's a chance those episodes were already deinterlaced when saved, using a 
higher-quality deinterlacer than the one internal to MythTV, like lavcodec, 
kernel deint, dscaler, etc.

> I have played around with the dnr_* settings and the bitrates but have been
> unable to affect the capture in any noticable way.
>
> Questions:
>
> 1. Is it actually possible to get captured video with a PVR-250 that
> rivals/exceeds the output from my TV's tuner?

A resounding YES. But I'm not certain exactly how. If you can post up a small 
clip of what you say looks like crap, I can take a look at it relative to 
what I'm getting. I still need to get together some sample clips myself, per 
a request from Cory -- I can share those with you as well.

I just had another thought... Is your card a rev1 or rev2? I see Cory has a 
rev1, so I'm wondering if things are better w/the rev2 (which is what I have, 
along with an M179 and a 350).

> 2. Can the location of the card on the motherboard actually make a
> difference?  I've read some claims to that effect.

Doubtful.

> 3. Is it possible that my card just isn't very good?

Always a possibility...

> 4. Does anyone have any suggestions as to how I can improve my captures?

- -Signal amplifier (I use a relatively cheap 4-way amplified splitter from RS)
- -Double-check grounding (usually outside by where cable comes into the 
premises)

> odds & sods:
>
> card:
> PVR-250 w/ remote (not M179)
>
> I'm using:
> ivtv: 0.1.9
> Firmware: 1.8.22024

Have you tried other firmware versions? Some folks have had considerably 
different results with different firmware versions.


> lspci says:
> 01:08.0 Multimedia video controller: Internext Compression Inc iTVC15
> MPEG-2 Encoder (rev 01) Subsystem: Hauppauge computer works Inc.: Unknown
> device 4001 Flags: bus master, medium devsel, latency 64, IRQ 5
>         Memory at f4000000 (32-bit, prefetchable) [size=64M]
>         Capabilities: [44] Power Management version 2

Hmm... From the look of that output, you DO have a rev1 PVR-250, just like 
Cory. I don't know if that really matters, but I'm suspicious... ;-p


- -- 
Jarod C. Wilson, RHCE

Got a question? Read this first...
     http://catb.org/~esr/faqs/smart-questions.html
MythTV, Fedora Core & ATrpms documentation:
     http://wilsonet.com/mythtv/
MythTV Searchable Mailing List Archive
     http://www.gossamer-threads.com/archive/MythTV_C2/
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFAIXiRln8CX+oYdJURAl2yAJ90v1o9vhLedss3UE7W/Y3LtftfqACg2s32
b9ALa5eltD3LT7no4o4s7dE=
=3SBX
-----END PGP SIGNATURE-----