[mythtv-users] Tuner cards that do HD?

Geoff Mishkin gmishkin at acs.bu.edu
Mon Apr 16 17:54:57 UTC 2007

Rod Smith wrote:
> On Monday 16 April 2007 13:00, Geoff Mishkin wrote:
>> Rod Smith wrote:
>>> On Monday 16 April 2007 02:02, Geoff Mishkin wrote:
>>>> I use the pcHDTV HD-3000 which I highly recommend.
>>> I couldn't disagree more, particularly for digital cable use. I've got
>>> two pcHDTV 3000s, and neither can reliably tune a digital cable (QAM)
>>> signal.> My anecdotal evidence is better than your anecdotal evidence :-)
>> No, this does have all the signs of a problem with interacting with
>> other hardware.  But, I've had my HD-3000 in two different computers,
>> and connected to four different cable systems plus OTA systems in two
>> different regions.  No problems.  So the OP shouldn't count it out just
>> yet.
> It's not just me who's had problems; lots of others have reported difficulties 
> here, on the pcHDTV forums, and elsewhere. Note that this is a QAM issue (for 
> cable TV reception); AFAIK, OTA users are unaffected by this issue. My point 
> is this: Even though the device works fine for some people, it's not worth 
> the risk for the average buyer; other devices do *NOT* have this problem, and 
> the only real advantage of the pcHDTV 3000 over them is its analog tuner, 
> which is a minor advantage, IMHO. I'll also say that I don't know if the 
> current model (the pcHDTV 5500) suffers from this problem.
I wasn't saying it's just you, that's why I started my message with "my
anecdotal evidence..."  I'm not sure either of us know exactly what the
risk is.  If you do, I'd be glad to hear so that I can make more
appropriate recommendations to my friends.  But keep in mind that the
people who it works fine for generally won't spend much time hanging
around the spam-infested pcHDTV forums.

If you could link to a specific post or posts, I'd actually be
interested to see what specific problems people are having.  It's hard
to tell the serious problems from the trivial ones.  For instance, a
couple of years ago when I was getting my card set up, I'd see problems
where people were simply missing critical steps in the setup process.

Another advantage of the pcHDTV cards (or similar) is that it's one less
box on your shelf, only less thing to plug in, fewer wires, etc.
>>> For the OP's purposes, too, the pcHDTV 3000 (and I believe the 5500) use
>>> framegrabber analog tuners, which increases the CPU load when recording
>>> analog signals. IMHO, it's better to go with two tuner cards, one analog
>>> and one digital (or more if you want more than that).
>> My CPU load when recording analog is between 5% and 7%, which hardly
>> seems exceptional.  It still records fine even if the rest of the system
>> is under load.  The biggest disadvantage in my opinion is the fact that
>> recordings are larger.  But, disk space is cheap, and there's always
>> transcoding.
> The exact CPU load will vary depending on the CPU speed and encoding format. 
> Given your "recordings are larger" comment, I suspect you're using RTJpeg, 
> which uses less CPU than MPEG-4 but produces bigger files. Certainly a 5-7% 
> CPU load isn't likely to cause serious problems, but if somebody's using a 
> less capable CPU than yours and/or wants to encode using more CPU-intensive 
> options, this could become an issue. Also, no matter what the CPU load, it 
> will be lower if a hardware-encoding card is swapped in. This might or might 
> not be important for any given configuration, but I think it's important to 
> mention it so that the reader can judge the matter him/herself.
RTJpeg is about a gig an hour for me.  That's nothing compared to how
big HD recordings are, so I don't worry about it much.  Also, I pity the
person who has a worse CPU than me.  Mine was low-end when I got it
almost five years ago.  I guess it's important enough to mention.
> FWIW, on my system with a 3.06GHz Celeron-D 347, framegrabber cards such as 
> the pcHDTV 3000's analog tuner consume about 20-40% CPU time when recording, 
> depending on the encoding options used. (The true range might be greater than 
> that; I haven't tried every single option.) This is a big enough load to be 
> important, particularly when using two tuners and/or when playing back HD 
> content (which consumes 70-100% of CPU time, depending on the content and 
> settings) while recording analog content.
I really hope that 20-40% is with MPEG-4 and not RTJpeg.

             --Geoff Mishkin <gmishkin at bu.edu>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mythtv.org/pipermail/mythtv-users/attachments/20070416/c8f52902/attachment.htm 

More information about the mythtv-users mailing list