Hauppauge quadHD problems

[NumberFive]

Hello,

I've spent this evening messing about with codecs as I've had occassional glitches since moving to a new machine.  I've now discovered that the glitches are in the recordings.    Confirmed using VLC I see the same glitches at the same points in time.

I'm 99.999% sure this is not a signal issue, as I have a good strong signal and in the last 11 years of my old HTPC using a Nova-T Dual, it would be absolutely exceptional to see a glitch in a stream.  In fact, I can't recall a case in recent history when that has happened. Sadly the new PC has no PCI slots for the old tuner.

I've updated to the latest drivers, it's Windows 10.  Not sure if there's much I can do but any tips appreciated. 

Failing that, recommendations for a rock solid tuner? 

Thanks for your help!

[NumberFive]

Just an update.  I do get a high number of MPEG-TS errors when using the Hauppauge quadHD when installed in an ASRock J5005-ITX.  I've moved the card into another PC, with same antenna feed, and thus far I've not seen any MPEG-TS errors.

I reckon the issue might be with the J5005-ITX, but I'll keep investigating.

I'll keep the forum updated with my findings.

[NumberFive]

A further update. 

Signal / Glitches

It appears that adding a very strong amplifier (20 dB) in front of the quadHD helps.  I'm a bit surprised by this, because I've had other quite cheap tuners on this feed without errors, and a Nova-T-500 has worked reliably for many years.  A few comments across the net do mention that these multi-tuner boards are more prone to weak signal, which seems odd.  I'd have assumed they have sufficient on board amplification.  Besides, the Nova-T-500 is dual tuner.  I'll keep an eye on this.

Remote

Anyway, a fresh new problem.  The IR remote on the quadHD is giving issues.  It works solidly as long as there are no records in process.  When idle (powered on, nothing happening) the remote is still responsive after many days.  When there are recordings in process, the remote works for a bit after the recording starts but will after a time fail to respond until IR.exe is restarted.  All that is required is to reboot IR.exe and it's good to go again.  This can happen anywhere between 1 minute or 20+ minutes.  It makes the remote fairly useless from a PVR perspective.  I'll open a ticket with Hauppauge support and see what happens.  Would be a shame if they can't fix, their remotes are very good and convenient to configure.

[NumberFive]

Just an update for the community.

Remote - There was indeed a driver problem. Hauppauge have now fixed this. You will also need to make a change to the registery to get the remote working effectively.

Signal issues - My quadHD continues to exhibit a lot of dropped frames / TS packets. It is very frustrating. I have a good signal, as evidenced by my other tuners (the ones the quadHD is intended to replace). I've tried amplifiers, attenuators, different PCs... the only common denominator is the quadHD. Disappointed. Since this experience, I've read a handful of Amazon reviews reporting the same.

[NumberFive]

Just adding a link to this post in another thread as it contains some information re picture pixilation using the Hauppauge quadHD.

https://forums.nextpvr.com/showthread.ph...#pid552925

[mvallevand]

With a quad tuner I don't think you can get around the engineering fact that it is a 4 way signal split.

Martin

[NumberFive]

With a quad tuner I don't think you can get around the engineering fact that it is a 4 way signal split.

Martin

I've seen this comment across the web quite a bit, but you 100% definitely can engineer for this.  It's kind of trivial as far as radio circuits go.  The first components after the antenna socket should be an amplifier and impedance matching circuits, then into the tuner ICs.  About 6 dB of amplification is needed, as each time you split the signal in two you lose 3 dB.  This amplification and matching stage would mostly consist of printed components on the PCB (i.e. PCB printed inductors, capacitors), just like you can see in masthead amplifiers, so minimal cost in terms of production.  Pennies/cents for a transistor or two.  

Some of the tuner ICs even have a buffered pass through for such daisy-chaining applications, but I don't know how the costs of those compare against the conventional approach above.

[mvallevand]

Are you saying the process is lossless?

Martin

[NumberFive]

Are you saying the process is lossless?

Martin

In most practical applications, yes.  The amplifier will introduce a little bit of noise, but as long as the noise figure is less than the SNR of the input signal, then this won't have an impact on the application.  Most TV amplifiers would have a noise figure < 3 dB.  I have a Hauppauge dualHD which reports that the SNR on a multiplex the quadHD definitely has an issue with as 32 dB.  That's a huge margin, there would be no issue amplifying and splitting that signal.  Frankly, that signal should be absolutely fine even if split passively.

I don't understand why this thought that splitting the signal is difficult on PC based DVB tuners exists across the web.  Yes, for fringe signals it's going to be a problem, but not for good quality signals.  Distribution amplifiers exist and are conceptually equivalent.  Maybe the PC tuners are passively split, but 32 dB SNR should be ample.

I think the Hauppauge quadHD has a design fault in that it appears to be dropping MPEG-TS packets, but I could be wrong. I don't have access to diagnostic tools to say for sure.  Hopefully it's post demodulator and can be fixed in a firmware / driver update.

[NumberFive]

Looks like the Hauppauge dualHD (USB) has the same issue. It's dropping a lot of packets, when another tuner running in parallel is dropping none.