If anyone is interested in testing whether your display properly deinterlaces 1080i sources, HQV has just made Blu-ray and HD DVD versions of the benchmark tests available. The disc includes tests for:

HD Noise Reduction
HD Video Resolution (motion adaptive processing)
Jaggies (diagonal processing)
HD Film Resolution (inverse telecine processing)


The Jaggies tests are the same as those available on the SD version, but the other tests are new. The disc is easy to navigate and it provides clear instructions and examples that allow you to determine whether the display passes a test or not.

It's a great diagnostic tool for those who want to know the quality of their HD processing for video and film sources.

http://www.hqv.com/benchmark.cfm

My Blu-Ray/HD-DVD versions arrived today. The only really interesting pattern on these discs is the film-mode resolution test. HD-DVE, for instance, is all-video, so you cannot determine whether your video processing is giving full resolution for film sources. No other test disc that I know of right now has this test, making them both worthwhile just for that. Otherwise these are *very* bare and basic test discs.

Mark

How does the film-mode resolution test work? I ask this because a question arose in another forum regarding Toshiba's RPs. Apparently, they fail the HQV HD 3:2 pulldown test. However, under testing done by The Perfect Vision, the Toshiba "picked up the 3:2 pulldown reliably, but not so quickly." TPV used a test not yet available to the public. The question is, does the HQV benchmark measure a slice of time that could possibly occur before the Toshiba locks into the 3:2 cadence?

It's a pan of stadium seating. If it fails to lock on you'll get moire patterns.

Sounds then like it's long enough to pick up a delayed lock in. Thanks.

Tom, have you checked the disc? ;) The stadium pan is only on the SD disc. On the HD disc there is a moving resolution test pattern - if resolution is lost you will not see the single pixel lines in the pattern. The pattern runs for quite a few seconds and if the Tosh cannot lock during that time it's implementation is poor.

Mark

Yes, I own it. You are mistaken. If you stay on the film detail test, after the pattern you describe runs for a while the stadium pan I refer to appears.

Tom, have you checked the disc? ;) The stadium pan is only on the SD disc. On the HD disc there is a moving resolution test pattern - if resolution is lost you will not see the single pixel lines in the pattern. The pattern runs for quite a few seconds and if the Tosh cannot lock during that time it's implementation is poor.

Mark

Tom, my apologies, I didn't let the test run far enough.

Mark

Does this test disc have value for a CRT RPTV since the deinterlacing test would have no value? I have orderred the disc assuming the other tests are meaningful.

Thanks.

Ruby,

The disc is primarily intended to test your video processing and in particular your deinterlacing. This disc will be of limited use if you don't use any video processing in your display chain...

Mark

Ruby,

Does this test disc have value for a CRT RPTV since the deinterlacing test would have no value?

Is your CRT RPTV high definition or standard definition? You didn't say. Your assumption that it would have no value appears to indicate you have an SDTV. In that case you would be right.

Sorry, I should have said it is HDTV. Basically, my question relates to the fact that CRTs are 1080i devices and don't have to create a 1080p image.

The noise reduction and jaggies test would certainly be useful. The deinterlacing tests would only be useful if you are deinterlacing 1080i into 720p.

CRTs are not like fixed pixel devices. They can display a variety of resolutions, within limits. I expect your set will accept 480i, 480p, 540p and 1080i signals, at least. It may also accept a 720p signal but convert it to one or more other formats. Your TV may also accept computer resolutions and formats.

The HQV Benchmark discs would still be of value to you, both the SD and HD versions. It's a test and evaluation program. The SD version would help you determine whether a 480i signal from a source component looks better using the TV's processor or the one in the source (such as a progressive DVD player). The HD versions will work in a similar manner for additional signal formats from a BD or HD DVD player. These players allow you to determine what output signal to send to the display.

Best regards and beautiful pictures,
G. Alan Brown, President
CinemaQuest, Inc.

"Advancing the art and science of electronic imaging"

Thank you for the helpful replies. I thought it would be af value but I didn't see anything in the documentation online that confirmed that. In any case, I also have a 720p DLP RPTV that I knew it would be useful for. I just wanted to know if I should even bother with the CRT.

Most importantly, I intend to use the disc to evaluate potential replacement displays for
the CRT RPTV.

Thanks again.

Greetings

Tried 1080p 60 on the PS3 ... and a Runco 1080p plasma. Ugly deinterlacing/processing. The Runco takes a 1080p native signal ... but the one from the PS3 was ugly ...

Would be interested in confirmation from others on this.

Toshiba XA2 feeding 1080p 60 into the same Runco fared much better. No resolution loss ...

Regards

According to this disc, the Gennum processor in the JVC RS1 and the Anthem AVM50 fails the HD Film Resolution test. Strange.

The Lumagen ProHD passes film resolution but fails video resolution tests.

Mark

Is the HD HQV benchmark the only one currently available for testing 1080i film and video sources?

To follow up on my earlier post, Kris Deering has confirmed that Gennum processors fail this test, but they do nonetheless correctly perform inverse telecine on 1080i film sources. Thus, this appears to be a false positive.

BTW, the reason the Lumagens fail the video resolution test (and the diagonal processing test also) is that the SiI-504 chip in them is now a little dated and just doesn't do video deinterlacing very well. Lumagen added their own proprietary software to provide 1080i deinterlacing for film sources.

The upcoming Radiance, based on Gennum processing, should do both quite nicely.

To follow up on my earlier post, Kris Deering has confirmed that Gennum processors fail this test, but they do nonetheless correctly perform inverse telecine on 1080i film sources. Thus, this appears to be a false positive.


Why should it fail this test, which is after all essentially just a film sequence, and yet work correctly with other material? How did Kris confirm this?

Mark

BTW, the reason the Lumagens fail the video resolution test (and the diagonal processing test also) is that the SiI-504 chip in them is now a little dated and just doesn't do video deinterlacing very well. Lumagen added their own proprietary software to provide 1080i deinterlacing for film sources.The 504 shouldn't even come into play with an HD source as it only processes SD-resolution signals. I thought that Lumagen implemented their own HD video deinterlacing in an FPGA.

- Dale Adams

Why should it fail this test, which is after all essentially just a film sequence, and yet work correctly with other material? How did Kris confirm thisWhile I haven't seen this specific test, if it's anything like the SD SMPTE 133 test patterns I've seen (such as on the NTSC DVE disc) then the problem is most likely the alternating black/white horizontal line section of the test. This can look a lot like combing to many algorithms and may be triggering the Gennum processor to treat this as video rather than film. While there may be a few sources which have content like this, in my opinion it's an artificial test and does not represent the vast majority of real-world video and certainly doesn't represent natural image material.

- Dale Adams

The test pattern looks very much like this one, but also has a rotating bar in it (eg like the HQV jaggies test bar)

http://bjr.birjournals.org/cgi/content/full/77/913/52/F1

Mark

The 504 shouldn't even come into play with an HD source as it only processes SD-resolution signals. I thought that Lumagen implemented their own HD video deinterlacing in an FPGA.Yes, that's correct. I'm sorry if I implied otherwise. What I should have said is that the Sil-504 doesn't do very well with SD video sources and doesn't deinterlace HD sources at all.

I had also forgotten that they added motion adaptive deinterlacing for HD video sources. Given that, it's a little disappointing that it fails the HD video resolution test, but Greg Rogers review of the HDQ certainly left the impression that the film side of HD processing was a little better implemented than the video side.

While I haven't seen this specific test, if it's anything like the SD SMPTE 133 test patterns I've seen (such as on the NTSC DVE disc) then the problem is most likely the alternating black/white horizontal line section of the test. This can look a lot like combing to many algorithms and may be triggering the Gennum processor to treat this as video rather than film. While there may be a few sources which have content like this, in my opinion it's an artificial test and does not represent the vast majority of real-world video and certainly doesn't represent natural image materialThankfully, a standium film clip is also provided after the pattern as a real world test example.

If your video processor passes neither the pattern nor stadium clip, then you know you have a problem. If it clearly passes the latter (i.e. no moire whatsoever), but not the former, I suppose you could attribute to an algorithm-specific issue with that pattern, as you state.

If you are a video processor solutions provider, it seems to me that it would be in your best interest to make sure that future products pass both the pattern and the stadium clip tests, now that a number of home theater magazines are using this to benchmark VP products.

To follow up on my earlier post, Kris Deering has confirmed that Gennum processors fail this test, but they do nonetheless correctly perform inverse telecine on 1080i film sources. Thus, this appears to be a false positive.

BTW, the reason the Lumagens fail the video resolution test (and the diagonal processing test also) is that the SiI-504 chip in them is now a little dated and just doesn't do video deinterlacing very well. Lumagen added their own proprietary software to provide 1080i deinterlacing for film sources.

The upcoming Radiance, based on Gennum processing, should do both quite nicely.

Interesting. I read several posts on a different thread ( http://www.avsforum.com/avs-vb/printthread.php?t=832465 ) about people experiencing moiré artefacts in the HD DVD MI3 scene were there's a panning shot of the Vatican wall. They suggested that even when the XA2 was set to 1080p output the moiré was still present. The JVC RS1/Gennum does not exhibit any moiré in this scene with a 1080i signal. I assume this scene is a similar to the stadium test but it appears the HQV processor fails to resolve the issue 1080i>1080p properly in the Tosh.

Dazzer

So I order the HQV Blu-ray + DVD pack from them July 3rd and still haven't received the discs yet. Not even an email indicating that they have shipped. I sent three emails to two different person and not a single reply for them. WFT?:mad: They already charge my credit card. Can't find a telephone number to call.

Anyone have similar experience. All I really want to know if the disc have left the building.

I'll order my set (DVD+BR) in a couple of weeks, wedding in planning = hobbies suffer.

So I order the HQV Blu-ray + DVD pack from them July 3rd and still haven't received the discs yet. Not even an email indicating that they have shipped. I sent three emails to two different person and not a single reply for them. WFT?:mad: They already charge my credit card. Can't find a telephone number to call.

Anyone have similar experience. All I really want to know if the disc have left the building.

Finally got a reply today!!! They had some kind of inventory error that affected order fulfillment. SiliconOptix will issue me a refund and send me the discs at not charge.