Simultaneous Recordings

[Quaraxkad]

Until recently I've been having trouble getting a strong enough signal for consistently clear recordings, and I finally bought a decent antenna. I now have a Winegard HD7694P on the roof at maybe about 18' high, a 50' cable going inside to a cheap RCA 4-port splitter/amp with 10db gain from Lowes. Two of those ports are connected to two Hauppauge HVR-2250 tuners in my HTPC and a third directly to my TV, the fourth is unused. Using the signal strength monitor tool from Hauppauge I get a full strength 30db signal from all of my important stations, and the signal strength indicator on the TV (Samsung HLR5667W) shows most channels at or just below full strength. I can watch any channel and have a perfectly clear picture either through NextPVR or directly on the TV. But when I have multiple recordings scheduled simultaneously the resulting videos are very often corrupt and unwatchable. For example there are 4 simultaneous recordings every weeknight and morning. A single recording from one tuner with no others at the same time will come out clear. Does the HVR-2250 lose some signal strength when both tuners are being used at the same time? Or is it the amp/splitter? It's been my understanding that any one tuner or device should have no effect on the others. I'm not quite sure how to diagnose this issue, what should I do/how should I test this?

[Graham]

From my own experience ... and perhaps not relevant to you ...

Recordings when the signal strength was less than brilliant were affected by weather ... especially wind blowing nearby trees and affecting the signal to the antenna.

Recent attempts to record to a folder on a network device resulted in choppy recordings with the choppiness being increasingly bad as the number of simultaneous recordings increased. The problem was with the device and not the network.

I can't speak for the 2250 ... but I have used a variety of dual and quad tuners and have never seen a problem caused by loss of signal inside the device.

[martint123]

I can't see that using a second tuner will change what the first is getting.
As Graham suggests, I would look at disc writing throughput.
I would start checking around the possibilities of a very badly fragmented drive and lots of seeking with 4 simultaneous writes.

[johnsonx42]

what sort of drive are you recording to?

[Quaraxkad]

My recording drive is a brand new 3TB Toshiba installed locally in the HTPC (not over a network share). I ran ATTO benchmark on it, and the performance is pretty good:



The drive is used for nothing other than recordings. Although it IS under even more stress than just recording, because of my postprocess.bat file running VideoReDo on them. Four new recordings while running videoredo on 4 finished recordings is quite stressful! That may no longer be necessary now that I have better reception, so I will disable that for now and see how future recordings end up. I'd be surprised if the hard drive performance has an affect on the recording quality though.

[Graham]

... the performance is pretty good ...

I agree. I would expect a local drive to be able to cope easily with four (and more) HD recordings ... mine does.

It may not be the transfer rate that might be tripping you up ... The drive may be having to do a lot of seeking if your are post processing and that may cause occasional momentary peaks in the queue length and that may cause data to be dropped ... but I would be surprised if that were happening ... it's the twenty-first century and disk drives are very big and very fast.

[johnsonx42]

...because of my postprocess.bat file running VideoReDo on them. Four new recordings while running videoredo on 4 finished recordings is quite stressful! That may no longer be necessary now that I have better reception, so I will disable that for now and see how future recordings end up. I'd be surprised if the hard drive performance has an affect on the recording quality though.

As Graham points out, you may be losing recording data due to the load on the drive, or possibly CPU load. I suspect you'll find that eliminating VideoReDo solves the problem. If you want to keep your VideoReDo process going forward, you can write your postprocessing.bat so that it only runs one VideoReDo at a time and waits on the others until previous runs are done; I limit my system to 3 comskips at once, just so the CPU use doesn't ever get high enough to make the fan get loud (I used to have a software encoding analog tuner, so I was saving CPU cycles for that too).

[stustunz]

Next you will say you have comskip running also😃
What CPU you got and how much ram?

[Quaraxkad]

Well, I have postprocessing.bat disabled and the recordings from last night and this morning came out great! I originally set up VideoRedo to run a Quickstream Fix (which is not CPU intensive) on every recording because my OTA reception wasn't great and a lot of the recordings weren't watchable otherwise. I don't use comskip or anything else. Obviously up to 4 instances of Videoredo running at once is hard on the drive, but I'm still surprised that it would actually affect the recording quality. But now that my reception problems are fixed with the new antenna, it looks like I don't need any postprocessing at all.

[johnsonx42]

Obviously up to 4 instances of Videoredo running at once is hard on the drive, but I'm still surprised that it would actually affect the recording quality.

it doesn't exactly affect the recording quality directly, rather all the seeking back and forth causes the write queue to grow longer and longer until data gets dropped. it's like taking a shotgun to your transport streams, and they don't like that!