[David Wallis]

Quote:

Digital Video Essentials Quote:mcfoo There is a blurb in the First Look section of the July/August SGHT on it. It will first be available on D-theater, JVC's hi-def tape. Standard def DVD will follow...in other words, same old stuff from Joe Kane... http://www.avsforum.com/avs-vb/showt...hreadid=144493

Not sure how many people new about this ! So just thought I would put it here with a link to the original thread I found it in.

I have not yet got my SGHT mag but its great news that we can get a test tape for setting up the whole chain from 30k to projector.



DavidW

 

[Dean Roddey]

This tape should be a good test case for the robustness of tapes. Since it will be constantly FF'd, rewound, paused, and started and whatnot over particular patterns, it will get more abuse than your average tape over a similar period of time.


I'll keep my Accupel HD pattern generator myself, but for folks who have these tape decks, it will be a nice thing to have these patterns available to them. They won't be terribly convenient to get to (because a calibration requires a lot of back and forth movement between patterns), but it'll be considerably better than nothing.

 

[Abq-Pete]

The most important feature of this for me is the ability to calibrate to a specific output device and the entire signal chain. Using pattern generators has always had me wondering what I am really calibrating to. The fact that the pattern generator is outputting a perfect signal does not imply that any of my HD boxes are doing the same.


What I'd really like is for the Accupel to output 8vsb that my boxes can decode.


Just my $0.02.


Regards, Peter

 

[gravy]

If anyone has information about the technical specs of the images used for test patterns, please send it to me. I am considering authoring a D-VHS calibration tape.


I have the tools, the equipment and the know-how.


Come forth, or forever hold your peace!


-g

 

[Den]

Dean - agreed. But the Accupel needs circles, vertical resolution wedge and a moving pattern for scaler assessment - I really miss those.


Peter - so true. Sencore has one I believe but mortals can't afford.


Gravy - I believe everything you need is in the SMPTE papers.

 

[Dean Roddey]

Quote:

The fact that the pattern generator is outputting a perfect signal does not imply that any of my HD boxes are doing the same

I dunno. Other than the very early DTC-100 (prior to the subsequent software updates), has anyone got any real proof that any of the boxes fail to meet the signal specifications?

 

[Abq-Pete]

I have various DTC-100s, Panasonic TU-DST50, Panasonic TU-HDS20, and Hipix. I can tell you that there are marked differences in the picture quality of all of them, specifically in the areas of brightness and color cast. Some equipment has to go through transcoding which further alters the picture.


As such, I would like to correct for these differences and get an accurate picture. I can't do that with the Accupel as it stands.


I spoke with the director of my local PBS station. He is willing to broadcast some test signals for me as soon as he acquires a file server to store the information. If he does, then I'll record that on the Hipix and D-VHS and do a more accurate setup.


Come to think of it, it would be nice if the D-Theater Test Tape was NOT D-Theater encoded. It makes it only useful for the component out of the JVC (which I do have) and not my other D-VHS systems.


Regards, Peter

 

[Dean Roddey]

Quote:

I have various DTC-100s, Panasonic TU-DST50, Panasonic TU-HDS20, and Hipix. I can tell you that there are marked differences in the picture quality of all of them, specifically in the areas of brightness and color cast. Some equipment has to go through transcoding which further alters the picture

Maybe. But unless you've watched the same material at the same time, via multiple boxes, switched back and forth on the same well calibrated display, I'm not sure how you can be sure that they have significant differences.


I've calibrated so far 5 systems with the Accupel, all different HD boxes and displays, and the results when displaying real content from the boxes look quite correct. If there was any differences between them in terms of their being incorrect in their manipulation of the signal, it wasn't terribly apparent. Maybe I'd see small differences if I used this tape, but then again I'm not going to lug a VCR around to do calibrations with.


Also, if these boxes are non-trivially incorrect in their processing of the signal, there's no guarantee that what you get after a calibration with the tape is going to be 'correct' or acceptable anyway. So you are probably just going to end up 'salting to taste' by looking at real world content, which I can do just as well with the Accupel.

 

[Den]

I'm of Peter's mind on this Dean. Common sense tells me the differences are there. Only with end-to-end calibration can we ensure the optimal setup. Examples:


Which has the more accurate (or less objectionable) RGB matrix decoder, the STB or the display? That governs whether we run YPbPr or RGBHV, which in turn determines whether the display's Color saturation control is available to us.


When set to output everthing at an HD scan rate, does the STB correctly remove black level setup from SD content? That governs whether we go with a compromise black level, ask the user to adjust Brightness when changing channels (!) or just forget it and go with another STB make/model lacking the bug.


When set to output everything at an HD rate, does the STB correctly detect and transform SD (BT.601) colorimetry to HD (BT.709)?


Not to mention just being able to assess STB scaler and resolution performance with reference signals, etc.

 

[Den]

Quote:

Maybe I'd see small differences if I used this tape, but then again I'm not going to lug a VCR around to do calibrations with.

No need to lug the VCR. I meant an 8vsb generator of course. If the customer has the JVC deck, of course you bring your DVE tape to optimize that path and let it all hang out.

 

[Dean Roddey]

I hear what you are saying. I guess I'm just saying that, given the setups I've done so far, the bit of difference wouldn't have been remotely worth the difference in the Accupel and one that has an 8-VSB modulator. If you've got a deck, and the tape is available, then of course its a no-brainer to get it, even if you do have a pattern generator, since its only a few more bucks for another source of possibly calibration information.


But as far as the Accupel vs a much more expensive modulator capable one, I wouldn't worry too much about it, because I just don't think the results would come out all that much different most of the time.

 

[jamoka]

If you guys need a calibration tape, why don't you download some test patterns with a pc, then dump them to D-VHS over firewire. That would be free, available now, and allow you to make your own custom set of patterns.


You can even rip the patterns from a DVD calibration disc, convert to .ts, and dump to tape in the same way. D-VHS test patterns are easy.


Joe

 

[trbarry]

If you search there is even a zip of free non-copyrighted HDTV test patterns that has been posted here from time to time.


- Tom

 

[David Wallis]

Well I am not sure what you guys all have setup, but my source settings (brightness, contrast etc) for my 30k are different than my DTC100 and they are different again for my Samsung SIR-T150. It was setup with a Sencore VP300 (not RF) but final adjustments where made for the ever so slight differences we saw after it was aligned and gray scale tracking was done. Definitely noticiable to my eyes even though they were supposed to be the same according to the VP300.


I certainly would not see a proper ISF calibration not using a test pattern from source to screen. This is surely the best method by taking into account the equipment chain. A test generator would get you in the ball park but then my settings on the G90 would all be the same from all sources which did not look quite correct IMO.

Anyhow we will see exactly once I get the DVE D-Theater tape.


DavidW

 

[Dean Roddey]

Quote:

A test generator would get you in the ball park

To be fair, it will get you exactly correct. The issue is that you have devices that are not implementing the standard well, or that have less than stellar processing (which is kind of scary for something that's supposed to be the end all and be all of sources right now.) So how large your 'ballpark' is depends on how badly your source devices vary from correct.


Also, if your source device isn't doing its job well, having the 8VSB based pattern generator wouldn't help much. Most of what you'll watch from the tape deck is pre-recorded and wouldn't go through the demodulation process anyway, right? So the fact that you've hooked up an 8-VSB based pattern generator doesn't mean its going to agree, even if you directly recorded what it puts out and then turned around and played them back from the deck, because it would take two different paths through the source.


So which is correct? Do you calibrate it for when you are watching the content coming back through the STB or directly out from the deck or through the deck and through the STB? I think you'll end up doing some combination of these when you do the 169time thing to get recordability of DTV stuff, right? If you calibrate with the tape, how do you know that when you are watching it with a path all the way in from the modulated signal that its going to be correct?


I'm nit picking of course, but that's what these forums are for, right?

 

[David Wallis]

Dean their can only be one answer in reality and that is you calibrate to make your display look its best. If that means slight allowences for equipment that migh be off a little then is it not better to have it balanced for what you see not what you might see if the device was perfect.

Thats the problem with a test generator over a test tape or disc. You are expecting your chain to reproduce perfectly at a standard and we know very well that that is stuff of fantasy land.

Its like not adjusting the input sound levels to suit different sources. It should not happen because everything should be perfectly built to the ninth degree but its not.

Balance what you see on the screen, so that it take's in all your equipment , because after all if your source is D-Theater tapes you hardly care that it would look perfect if you where watching the test generator !

Not even my G90 is perfect.



DavidW