Log in Register FAQ Memberlist Search pcHDTV Forum Index
pcHDTV Forum

pcHDTV Forum Index -> General pcHDTV topics -> Specifications for HD-3000? Goto page Previous  1, 2, 3, 4
Post new topic  This topic is locked: you cannot edit posts or make replies. View previous topic :: View next topic 
Re: Quality of Service
PostPosted: Wed Apr 12, 2006 12:04 am Reply with quote
inkling
 
Joined: 05 Feb 2004
Posts: 342




xyzzy wrote:
Maybe the signal strength is wrong? Could be the driver isn't reading it correctly.


What exactly happens that is wrong? I don't seem to have any problem reading signal strength without a lock.


Since I have to aim at a 100' antenna tower nearby, I'm willing to wager it's a double reflection from each side of the tower arriving just enough in phase to trigger a strong signal, but enough out of phase to give a bad stream. I do not often get good signal when pointed directly at the antenna farm because of all the buildings in my direct line of sight, so I bounce off that nearby tower instead.

If I point it off to the side of a direct beam to the nearby tower, it usually works better than looking directly at it. The rotor helps some, especially when I want to get back to a pre-set position that I know worked before. That doesn't always work but often it does.



No lock no clock, no clock no data.

Sure you can read it without lock, but it's worthless data and it will eventually desync the demodulators' i2c. That's been my experience, and also the experience of at least 3 people I know locally. The i2c failure used to occur over and over again, until I kicked that bad driver habit. I would have fixed it early last year, because even the pchdtv 1.5 and 1.6 V4L and 2.0 V4L2 drivers had the problem, but my attention was focused elsewhere. I didn't want to complain about it very much until I actually had a solution. I figured a year was long enough to wait for someone else to fix it, which never happened.

I imagine somewhere in the or51132 and or51211 datasheets, there is a big warning, in 36 point type, in blinking red letters, in all capitals, that says:
DO NOT READ THE SNR IF THERE IS NO LOCK.

Maybe we can ask Jack if that's true?


I plan on forking the v4l-dvb driver by completing the removal of all the non-HD2000 and non-HD3000 junk, fixing the interrupt routines to something near sanity, then Mr. Krufky can come read my code and see how it's supposed to be done. Just for grins, I'll probably make it work on kernel 2.4.x too.

Either way, it's a moot point. I will not be using any of the latest code from v4l-dvb EVER again. Their broken driver has already killed two of my pcHDTV cards because they're writing to registers willy-nilly.

Some of the cx23883 register addresses they're writing to aren't documented in any of my cx23883 datasheets and aren't even commented with any kind of explanation as to what they do. Other cx23883 register writes are flat-out wrong and in direct contradiction to the datasheets.

Code like that, like some of the old poorly designed hardware that used to be discrete TTL, is always suspect and gets excised if it's proven to do nothing for me. I've been doing things that way for the past 30 years I've been flipping these keys, using my xacto knife, the wirewrap tool, a roll of kynar and my soldering iron. No reason to change now. People complaining about noisy computers makes me laugh. Strap a person to a punch card machine for 4 hours a day and then an ASR-43 for another 4 hours and their hearing will be shot, just like mine.

My chopped down driver version hasn't killed any cards. Jack thinks he knows why the cards died, or at least how to fix it, and has sent me something to try. If it doesn't fix it, well tough, they're both out of warranty. I still have 3 working cards left. I'll use the dead HD3000 for a dedicated IR input, since the cx23883 chip still works fine, and use the HD2000 as framed wall-art, with a new Osram ultra-bright LED lighting it up.

No reason for me to waste my time trying to convince Mr. Krufky or anyone else what is right and what is wrong. That might infringe on my sarcastic message posting time, not to mention that I'm an old fart who needs to get the most out of whatever time I have left on this big blue globe. I fix it, use it and move along to something else interesting.

If Mr. Krufky doesn't like it, he is cordially invited to .... ... not use it.


It's just TV, after all. All the TV networks could go belly-up and all it would do is give me more free time to do something else.

-ink
View user's profile Send private message
antenna tweaking ; signal strength ; S/N
PostPosted: Wed Apr 12, 2006 1:01 pm Reply with quote
dieter
 
Joined: 20 Jan 2005
Posts: 43
Location: US




scott> if you're not down with that, you should use a Windows machine
scott> with an HDTV card

It is not obvious how running virus server would improve reception/capture.

scott> [ info on working system ]

Thank you.

> Most PC-3000 cards aren't having problems although it's very easy to cause
> problems if you aren't careful about antennas and where you put them.

Even tweaking the antenna to the point where analog stations come in quite
good (no snow, multipath reduced to the point that you have to stare at the
screen awhile looking for ghosts to catch a very faint one) I still lose
lock in approx 0.02% of samples, in all digital stations, both VHF and UHF.

dieter>> Just because you do not find signal strength to be useful
dieter>> doesn't mean it isn't useful to others.

inkling> Are you going to be manually moving your antenna around
inkling> during capture to find the optimal signal path?

Depends on what you mean by "capture". If I'm trying to record
a program, no. If I'm experimenting with antenna setups, then
yes. If someone is trying to setup an antenna to DX a weak station,
knowing the signal strength is very important. If someone is in a
strong signal area, signal strength is likely to be sufficient and
multipath is likely to be far more important. From what I've read,
multipath is pretty much always a major problem for digital.

dieter>> We need to KNOW what the abilities are.

inkling> Why? You can't collect the data to determine what the limits are
inkling> for yourself? Do you have to have the teacher draw the answer on
inkling> the blackboard?

Perhaps you enjoy reverse-engineering products that are supposed to
be open source, but I consider it absurdly inefficient. The
documentation should be available on the web, and the pchdtv site
should have a documentation page with links to it.

inkling> Lock is the clock. The data is only valid when the clock is valid.
inkling> If you read the data when the clock is invalid, you get invalid data.

Thank you.

inkling> Can you guarantee how long write() will take? Of course not.

Guarantee? Does Linux have real-time guarantees for *anything* ?

xyzzy> Maybe the signal strength is wrong? Could be the driver isn't reading
xyzzy> it correctly.

The signal strength as reported by azap bounces around all over the place
on my system. There is a thread on this with some data. I added a check
of the return code from ioctl and it returns 0, so it isn't because the
ioctl is failing.

I tried taking the median of a large number of readings and that worked
somewhat better.

Taking the average (arithmetic mean) of a large number of values gives results
that appear to be at least somewhat meaningful (better than the median).
Add attenuation and the average signal reading goes down slightly. Disconnect
the antenna and the average signal reading goes way down.

I wonder where the signal reading is taken? At the RF input, or after some
processing? Perhaps after some AGC?

inkling> I know someone mentioned changing azap to print the values as percentages
inkling> instead of the 4 digit hexadecimals for strength and SNR.

Percentage? I've been assuming that SNR stands for Signal to Noise Ratio,
and those are ususally in dB, not percentage.

I've been wondering if there is a formula for converting the 4 digit hex SNR
into dB? Assuming it isn't already in dB? (just doing 10 log10(hex) would
be far too easy :-) And nothing in DTV is easy. :-( )
View user's profile Send private message
Re: Ringing the bell
PostPosted: Thu Apr 13, 2006 12:41 am Reply with quote
xyzzy
 
Joined: 12 Feb 2006
Posts: 225




inkling wrote:

Did Mike Krufky send you out here to annoy us after I hurt his widdle feewings when I told him the cx88 driver sucked? I don't recall you piping up with any of your flawed misconceptions before that.


It's pretty clear who the one with hurt feelings is. Funny how I'm such a fool, yet all my patches are accepted and yours aren't? I bet that keeps you up at night. Probably just as well, as you seem to be willing to do hours of research to try to find some way to insult me. For a bitter old man, you sure act childish.

It's clear you have nothing to offer, so consider yourself ignored. I have no more interest in arguing with someone on the Internet than in running in the special olympics.
View user's profile Send private message
Specifications for HD-3000?
  pcHDTV Forum Index -> General pcHDTV topics
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
All times are GMT - 7 Hours  
Page 4 of 4  
Goto page Previous  1, 2, 3, 4
  
  
 Post new topic  This topic is locked: you cannot edit posts or make replies.  


Powered by phpBB © 2001-2003 phpBB Group
Theme created by Vjacheslav Trushkin