Log in Register FAQ Memberlist Search pcHDTV Forum Index
pcHDTV Forum

pcHDTV Forum Index -> General pcHDTV topics -> HD-2000 vs. HD-3000 in a new 64 bit system
Post new topic  This topic is locked: you cannot edit posts or make replies. View previous topic :: View next topic 
HD-2000 vs. HD-3000 in a new 64 bit system
PostPosted: Thu Oct 05, 2006 10:57 am Reply with quote
Scott Larson
 
Joined: 15 Oct 2003
Posts: 713
Location: Portland, OR




I've gotten the HD-3000 to work fine in my new system but it looks like the HD-2000 is not going to cut it.

In my old system the two cards worked flawlessly for years side by side with the HD-2000 having slightly fewer reception problems than the HD-3000 but no significant problems with either. In the new system, it's the HD-3000 that's getting far better reception. dtvsignal on the HD-3000 gives me slightly lower numbers than it did in my previous system, but all of the HD-2000's numbers are lower and some are dramatically lower. One station that was in the 90's is in the md-60's now and the crummy reception of that station reflects that. Oddly another station that was also in the 90's has only dropped to the 80's and reception of that station is fine.

I did all this testing at the console without X running to exclude Nvidia's driver and any other video things that might be causing RFI.

I did some obvious things like swapping the two boards on the motherboard (this motherboard only has two PCI slots) but that didn't help. For fun I put an RFI barrier made from aluminum flashing around the cards and grounded it to the chasis but that didn't help either of course.

So I'm out of ideas. This is the only serious problem I'm having at this point. It looks like I may have to throw the HD-2000 away and buy an HD-5500 and hope it does better.
View user's profile Send private message
PostPosted: Thu Oct 12, 2006 9:54 am Reply with quote
Scott Larson
 
Joined: 15 Oct 2003
Posts: 713
Location: Portland, OR




So I replaced the HD-2000 with an HD-5500. Do you think it fixed everything?

Although I can't use dtvsignal to make numeric comparisions since I'm using the kernel drivers, it looks like it everything works again. I had the HD-3000 and HD-5500 recording the two channels that the HD-2000 couldn't receive well while I was displaying one of them with an 8xxx version of the Nvidia driver (the 7xxx ones don't compile on my system). Neither recording had any errors at all.

This isn't conclusive (I only had an hour to test) but it's a good sign.
View user's profile Send private message
PostPosted: Thu Oct 12, 2006 10:30 pm Reply with quote
Scott Larson
 
Joined: 15 Oct 2003
Posts: 713
Location: Portland, OR




I copied the lgdt330x.c from the pcHDTV driver over to the v4l source so dtvsignal would work with the PC-5500 and the numbers I'm getting from the PC-5500 are just outstanding. Here's what dtvsignal is reporting for the seven channels I care about:

Code:
PC-2000     PC-3000      PC-5500
67          86           93
82          87           90
62          81           87
68          71           82
78          85           91
94          94           94
75          82           92


Not only am I back to the quality I was getting before I swtiched motherboards, I'm getting even better reception. I recorded for three hours with both boards tonight with mplayer running the whole time and have not seen a single bad packet.

I think I'll keep the PC-2000 with the other motherboard (along with its CPU, memory and video board that I couldn't reuse) for backup. It did work very well for years in that system.
View user's profile Send private message
PostPosted: Fri Oct 13, 2006 12:50 am Reply with quote
xyzzy
 
Joined: 12 Feb 2006
Posts: 225




The signal strength numbers aren't directly comparable between different demodulators.
View user's profile Send private message
PostPosted: Fri Oct 13, 2006 8:50 am Reply with quote
Scott Larson
 
Joined: 15 Oct 2003
Posts: 713
Location: Portland, OR




Big number good. Small number bad. How do I weight them so they are comparable?
View user's profile Send private message
PostPosted: Fri Oct 13, 2006 11:39 am Reply with quote
nybbler
 
Joined: 06 Jul 2006
Posts: 125




Essentially you can't. With sophisticated equipment you could determine the card's noise floor, its resistance to adjacent-channel interference, tolerance of multipath, production of cross-modulation products, etc. But one number that will compare how good one is over the other probably isn't going to happen.

In your case it sounds like the HD-2000 is less resistant to noise generated within the system itself, perhaps on the 5V bus.
View user's profile Send private message
PostPosted: Fri Oct 13, 2006 5:50 pm Reply with quote
xyzzy
 
Joined: 12 Feb 2006
Posts: 225




First off, none of the demodulators measure signal strength at all. It's inherent in their design. The RF signal goes into the tuner, which does AGC (as well as mixing, filtering, etc.), and then presents an IF signal to the demodulator. The demod has no way to know what the level of the original signal is, since the tuner has already done AGC. If you look at the demod code, most demods just use SNR to report signal level.

Even if you look at SNR, the number the driver makes up is related to SNR, but no one really knows exactly how. If you had a plot of real SNR vs number from driver, it would be completely different from demodulator to demodulator.
View user's profile Send private message
PostPosted: Fri Oct 13, 2006 6:14 pm Reply with quote
Scott Larson
 
Joined: 15 Oct 2003
Posts: 713
Location: Portland, OR




xyzzy wrote:
First off, none of the demodulators measure signal strength at all. It's inherent in their design.

Who said anything about "signal strength"??? I sure didn't. I'm plenty aware that signal strength is meaningless when multipath can cause interference.

Quote:
Even if you look at SNR, the number the driver makes up is related to SNR, but no one really knows exactly how. If you had a plot of real SNR vs number from driver, it would be completely different from demodulator to demodulator.

So you're saying that since no one knows how to weight the numbers to make them comparable, they just can't be comparable. That's fine.

I've found that with the HD-2000 and HD-3000, I started getting bad packets whenever the number was below 70. It was very consistant between both cards. I don't know why.

I haven't gotten any bad packets from the HD-5500 but then none of the numbers are below 82.
View user's profile Send private message
PostPosted: Sat Oct 14, 2006 7:47 am Reply with quote
xyzzy
 
Joined: 12 Feb 2006
Posts: 225




Scott Larson wrote:
xyzzy wrote:
First off, none of the demodulators measure signal strength at all. It's inherent in their design.

Who said anything about "signal strength"??? I sure didn't. I'm plenty aware that signal strength is meaningless when multipath can cause interference.

You did. Go look at the code for dtvsignal and see what it (and thus you) were reporting.
View user's profile Send private message
HD-2000 vs. HD-3000 in a new 64 bit system
  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 1 of 1  

  
  
 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