Additions:
∞
∞
Deletions:
∞
∞
Edited on 2005-02-08 18:56:14 by RichardBerg [move servers]
Additions:
∞∞
Deletions:
∞
∞
Edited on 2004-12-09 02:33:48 by RichardBerg
Additions:
∞Better, though not great. Neither of these images' deficits would be noticeable at 30fps, of course.
Tuner performance
Deletions:
∞Tuner sensitivity
Edited on 2004-12-09 02:27:05 by RichardBerg
Additions:
All of the above was done with the card set to highest quality, which comes to about 12 megabits per second of mpeg2. (Output files are directly parseable, BTW; no "transport stream" nonsense.) I capped a few hours of our favorite news channel at medium quality (~7.5Mbps) as well. Obviously talking heads won't show much degradation, so here's a sports clip:
Not bad for being able to fit nearly twice as many shows on your PVR. Unfortunately they switched their tape loops at the same time I switched quality settings, so I don't have direct comparisons for these images. Suffice to say I didn't see any mpeg2 flaws in the 12mbps clips, although I'm not as attuned to codec flaws as some people (morello on Ars, various wizzes on Doom9).
Above is the absolute worst still image from several hours of medium quality footage -- the first frame of a scene change during a trailer for some action movie. (Click for the context as Huff -- things aren't really as bad as they appear.) By comparison, the first frame of an explosion @ 12mbps:
∞(~3.7MB)
Nothing...as it should be. That isn't to say the separation is perfect -- blow up those pics and you'll see some rainbows -- but at least it isn't sensitive to stray noise. Nice, because if there's one thing that sucks in the desktop video world, it's when two enthusiasts (who are rare to begin with) can't duplicate each other's work.
Deletions:
All of the above was done with the card set to highest quality, which comes to about 12 megabits per second of mpeg2. (Output files are directly parseable, BTW; no "transport stream" nonsense.) Here are similar clips taken at a medium quality level (~7.5Mbps):
Oldest known version of this page was edited on 2004-12-03 03:58:49 by RichardBerg []
Page view:
Review: Hauppauge WinTV-PVR150 Media Center Edition
Thanks to Intrex∞ for letting me borrow the first batch of MCE goodies. Oh wait, that was 6 weeks ago. Hmm, guess I'm still a slacker despite this whole real-job thing.
The PVR-150, despite its name-of-lesser-cardinality and cut-rate pricing, is actually the replacement for the PVR-250. I've had it confirmed that Hauppauge has been making such 'Roslyn' boards exclusively for at least a couple months. Why? Well, the overarching magic of the computer industry has historically been tighter circuits giving rise to better deals, and sure enough we have here an example of Conexant's integrated "Blackbird" CX25843 reference designs∞ becoming much cheaper to manufacture than the PVR-250's hybrid chipset (mpeg2 from Philips, audio from Micronas).
How does it look? Let me paste in its entry from the WeatherMapRoundup:
∞(~6.1MB)
Clearly a step or two above the BT cards, though not the Compro's equal. Detail isn't resolved quite as well, as you can see when picking out the mountains (both the good ones and the misrendered ones) against a shadow. It doesn't help that levels are more comparable to the clipping-fest given by stock BT cards (note Huff's filesize agreement); not much you can do in post, and worse, I don't know of any PVR programs (for which such a card will certainly be used) that let you tweak the input amps or luma coring or anything of the sort. Finally, there is moderate comb artifacting -- in the still you can see that solid colors near hard edges aren't uniformly saturated, and sure enough if you download the video clip you'll see dots crawl. This is nothing AviSynth can't cure, however.
Clipping issues aside, the overall balance isn't bad -- you might even say better than the Compro, which tended towards oversaturation by default. Flesh tones look realistic, even if her perm doesn't.
With the wide range of colors outdoors, however, it's clear that the overall gamut isn't as expressive as it could be.
Pure CG looks quite good when the fonts are big. A tiny bit of trouble with y/c separation is evident, but I have a hard time calling this picture appreciably noisier than similar ones on the Compro.
With sports video, camera lenses are being strained to keep things sharp at changing distances, while in the digital domain you're losing a lot of vertical resolution to high-motion interlacing (and some horizontal rez to yet more computer glitches, apparently). Moral? Every bit of remaining rez helps. Unfortunately, the Hauppauge looks decidedly second-rate compared with the example from the Compro review.
Software
The PVR-150MCE comes with the bare minimum: 4 cryptically named driver folders and an applet called AMCap that's little more than the MS DirectShow SDK sample run through a compiler. Luckily, the drivers appear to be excellent, supporting almost the complete DX9 hardware capture specs with no crashes on 2K, vanilla XP, or MCE. I got trial versions of BeyondTV, SageTV, and others to work without difficulty despite the myriad capture hardware plugged into my testbed system.Compressor quality
All of the above was done with the card set to highest quality, which comes to about 12 megabits per second of mpeg2. (Output files are directly parseable, BTW; no "transport stream" nonsense.) Here are similar clips taken at a medium quality level (~7.5Mbps):
[pics]
Tuner sensitivity
In the WeatherMapRoundup we explored how rainbows could be greatly exacerbated when combining a poor tuner with electro-magnetic interference. The PVR-150 is known to exhibit minor rainbowing, so what happens when a cheapo cable is sprawled across my toy-laden room instead of quality RG6?
[pics]
Back to VidCapSamples