Jump to content


Welcome to DBSTalk


Sign In 

Create Account
Welcome to DBSTalk. Our community covers all aspects of video delivery solutions including: Direct Broadcast Satellite (DBS), Cable Television, and Internet Protocol Television (IPTV). We also have forums to discuss popular television programs, home theater equipment, and internet streaming service providers. Members of our community include experts who can help you solve technical problems, industry professionals, company representatives, and novices who are here to learn.

Like most online communities you must register to view or post in our community. Sign-up is a free and simple process that requires minimal information. Be a part of our community by signing in or creating an account. The Digital Bit Stream starts here!
  • Reply to existing topics or start a discussion of your own
  • Subscribe to topics and forums and get email updates
  • Send private personal messages (PM) to other forum members
  • Customize your profile page and make new friends
 
Guest Message by DevFuse

Photo
- - - - -

HR24 vs HR34 better PQ?


  • Please log in to reply
91 replies to this topic

#51 OFFLINE   Rich

Rich

    Hall Of Fame

  • DBSTalk Club
  • 22,128 posts
  • LocationPiscataway, NJ
Joined: Feb 22, 2007

Posted 25 December 2012 - 09:30 AM

I wish my eyes were good enough to tell the difference :eek:


As long as we're happy with our PQ, what does it matter? Of course it's subjective, most things are. When Vizios were introduced they had what I thought was a pretty terrible picture. But if all you had to compare it to was an SD set, the PQ on the early Vizios (with an HD feed) would simply blow away the SD set's PQ.

Rich

...Ads Help To Support This Site...

#52 OFFLINE   Rich

Rich

    Hall Of Fame

  • DBSTalk Club
  • 22,128 posts
  • LocationPiscataway, NJ
Joined: Feb 22, 2007

Posted 25 December 2012 - 09:49 AM

You mean paragraph regarding the 500 series measuring better? I'm a calibrator (Google my "handle" and AVS Forum if you don't know who I am) and have measured the differences with my calibration equipment. The HR24-500's variance from a reference pattern generator (Accupel 4000) is < 1dEuv on grayscale. The gamma is also linear from 5-100% stimuli and matches gamma measurements from a Accupel 4000. I cannot say the same with the other sat boxes. Next time I am calibrating my displays I will save measurements of the HR24-500, H25, H24 and a HR24-100 for your viewing pleasure.

BTW, anyone can see the differences between the HR24-500 and other box simply by looking at the guide. The gray steps are brighter on other boxes compared to the HR24-500.... which is a gamma issue.


OK, I believe you. Gotta admit I've struggled with your posts, I just don't have the knowledge to understand them. Not your fault.

In one room I've got a 24-100 and a 24-500. What do you mean by "gray steps*"? I'd like to see exactly what you're trying to explain.

I have wondered about calibrating TV sets properly since I tried it myself with a BD calibration disk. Why would a BD disk be used as the reference for a set that's gonna play mostly 1080i or 720p content? I've done a lot of calibrations on meters and industrial equipment and we always used a reference point that matched the gases or materials that would be in the process. I really never understood using a BD disk for a TV set that wouldn't see all that much 1080p content.

Rich

#53 OFFLINE   P Smith

P Smith

    Mr. FixAnything

  • Registered
  • 20,227 posts
  • LocationMediterranean Sea
Joined: Jul 25, 2002

Posted 25 December 2012 - 10:17 AM

A better question would have been how he managed to measure that.

To begin with, while D-Nice has known qualifications, so why he is posting stuff like the above has people wondering - I am not sure what his motive is.

First, while everything is mastered in 8 bit, because of digital jitter (timing error) and other issues, errors can essentially reduce 8 bit down upwards of 40%. By having a path for 10 bit resolution, you have built in headroom.

To make it simple, think of having a water pipe that can deliver 8 gallons max a minute. If everything is working properly along the way, you get 8.0 gallons a minute at best. If you have a water pipe that can deliver 10.0 gallons per minute, there is no issue getting 8.0 gallons out even if there is variations in the stream.

As stated before, just because HDMI 1.3 can do Deep Color does not mean that Directv can do that.

If want a outside third party opinion, can call the people at Lumagen (maker of the Radiance Processors) and ask if you should set your processor out to the TV @ RGB 4:4:4 or YPbPr 4:2:2. They will tell you exactly what I told you. RGB 4:4:4 is 8 bit and YPbPr is 10 bit so you use YPbPr 4:2:2

Now, to measure a HR24-500 from 5-100% stimulus as D-Nice claims he did, he would need to be able to produce these known values directly out of the HR24-500 (or the other units he claims to have tested) and as we all know, there is not a 0-100 IRE Pattern Generator inside a HR24-500, thus making his claims impossible.

Also, the Accupel 4000 listed by D-Nice has a known problem with digital output which is why the Accupel 5000 was released.

Last I heard, D-Nice used a cheap meter that was profiled to a more expensive meter (though not a 1nm Spectroradiometer). Unfortunately, this has known limitations as well.

While I respect D-Nice's calibrations, his explanations here have me really wondering......

The thread become more and more interesting, at least for me, sort of tech geek. Unfortunately it's not my area of expertise and I'm all ears to listen and see all deepest technical details of the discussion. Perhaps in new dedicated thread ?

#54 OFFLINE   TomCat

TomCat

    Broadcast Engineer

  • Registered
  • 3,699 posts
Joined: Aug 31, 2002

Posted 25 December 2012 - 01:58 PM

The HR24-500 uses RGB colorspace while YPbPr is the standard and what virtually every TV expects to see.

RGB 4:4:4 is 8 bit and YPbPr 4:2:2 is normally 10 bit, but can be 12 bit.

As thus, the YPbPr would have more detail (2 more bit of data 10 v 8).

So in that case, the HR34-700 at 10 bit YPbPr should look better than 8 bit RGB in the HR24-500.

If the content you might be viewing were either of those, that might be somewhat true.

But the reality is that all HD content at the consumer level is 4:2:0 8 bit, which means that neither colorspace can provide any benefit beyond that, regardless of what they might be capable of. That would be equivalent to putting a mono source into a stereo system. What do you get? Dual-track mono. My car has a speedo that goes up to 120, too, but in traffic I'm kinda limited to about 75.

And "detail" is not really the issue here. Detail has a ceiling set by the pixel map. If your DVR is putting out 1080 to a 1080p TV, that is the top limit of detail.

There are a lot of things that can compromise that detail, such as 1440 imagers, bad focus, crummy lenses, interlace error, bad lighting even, but among them is neither the color scheme nor the bit level; neither is a factor in detail.

A "lower" color scheme (for example, 4:2:0 instead of 4:4:4 or 4:2:2) means that the accuracy of the chroma is dithered a little bit. Instead of each Pr or Pb coefficient representing each pixel as is the case in 4:4:4, in 4:2:2 each coefficient is a blend of one pixel and its next-door neighbor. In 4:2:0 one coefficient represents the value of four pixels, two on one scan line and the two directly below it. It can be considered that the resolution for chroma is then 1/4th what it is for luminance, but detail is really determined by luminance, because our foveal vision is much sharper than our color vision (which is why they can get away with this perceptual coding technique). So that does not really apply to detail, at least as perceived.

A shallower bit depth (8 bit vs. 10 bit, for instance) also has nothing really to do with detail. With 8-bit there are some 232 quantization levels while with 10-bit there are some 1006. Honestly, that is the only real difference; each digital word has two more bits of resolution.

What that means is that the quantization error in the digitization is less than 1/4th as inaccurate for 10 bit as it is for 8 bit, which means the tendency for visible artifacts is a little greater for 8 bit. 8 bit is fine for delivery while 10 and 12 bit really only makes a difference in production and acquisition, where generational losses are a factor.

Again, that does not manifest as a compromise of detail. What it usuallly manifests as is color banding or contouring. You can often see this in pastels or shots of the sky. And a 4:2:0 color scheme allows it to manifest a little more visibly than say 4:2:2 or 4:4:4 would, but only slightly more.

Regardless, for all intents and purposes, RGB vs YPbPr as a final color display method is essentially the same in the normal consumer application, perceptually speaking. True, some TVs handle one slightly better than the other, but just the other differences between TVs, screen types, and manufacturers are much more dramatic in how they affect PQ than the difference between RGB and YPbPr is, at least in this consumer application.
It's usually safe to talk honestly and openly with people because they typically are not really listening anyway.

#55 OFFLINE   TomCat

TomCat

    Broadcast Engineer

  • Registered
  • 3,699 posts
Joined: Aug 31, 2002

Posted 25 December 2012 - 02:25 PM

... comes does down to what HDTV you use to connect with.

Mostly, yes, that's about 99% of the game; once content leaves the digital domain (and it leaves it surprisingly early inside a HDTV, often in the HDMI receive chip which often also contains a DAC) that is when things start to stray from the PQ as delivered.


Put simply...one would think that since the HR34 has newer components inside...it has the potential to do the better video presentation,..

And no.

What really affects PQ before digitization is how well it is prepped, digitized, and encoded, and how much it is compressed is a part of that. That fixed PQ is what eventually hits all decoders equally.

As far as the decoder side (in our DVRs and STBs), all decoders, new and old, are pretty much made equal as far as PQ goes, because they are built to a standard that conforms them to do as close to the exact opposite as possible of what the encoder does, which is the goal. And since what they do is all math, there is only one way to perform that correctly, and they always do, always have, and always will do it exactly as correctly as they always have. There is no room for improvement once things are inside the digital domain.

If the oldest decoders are really based on performing a mathematical function, which is really all they do, and the oldest decoders come up with identical mathematical results to what the newest decoders do, neither can be more accurate than the other; once you get the math correct, there is no way to improve the decoder to get it "righter". And that result is exactly what defines what the coefficients for each pixel are, meaning they are the same everywhere. If your device is using an MPEG AVC decoder, new or old, it will provide the same exact PQ at decode because all of them do the math perfectly, and exactly the same.

There may be improvements in current draw and size and component cost, but the newest decoders really do exactly the same job that the oldest decoders out there do, and they do it pretty much precisely the same way they always have.

If they began to do it differently, the end product would be less fathful to the original encoding, and not more faithful. That translates to increased artifacts rather than diminished artifacts, and worse PQ rather than better PQ.

DACs, same story. There is really only one way to DAC a digital signal into analog, and it is by now such a mature process that there is only the tiniest invisible difference in the outcome between one manufacturer and another or an old DAC and a new one.
It's usually safe to talk honestly and openly with people because they typically are not really listening anyway.

#56 OFFLINE   Rich

Rich

    Hall Of Fame

  • DBSTalk Club
  • 22,128 posts
  • LocationPiscataway, NJ
Joined: Feb 22, 2007

Posted 25 December 2012 - 03:53 PM

You mean paragraph regarding the 500 series measuring better? I'm a calibrator (Google my "handle" and AVS Forum if you don't know who I am) and have measured the differences with my calibration equipment. The HR24-500's variance from a reference pattern generator (Accupel 4000) is < 1dEuv on grayscale. The gamma is also linear from 5-100% stimuli and matches gamma measurements from a Accupel 4000. I cannot say the same with the other sat boxes. Next time I am calibrating my displays I will save measurements of the HR24-500, H25, H24 and a HR24-100 for your viewing pleasure.

BTW, anyone can see the differences between the HR24-500 and other box simply by looking at the guide. The gray steps are brighter on other boxes compared to the HR24-500.... which is a gamma issue.


Just checked a 500's Guide against a 100's Guide and you certainly are right. There's a major difference in what you call "steps" (I would have called them rows). The 100 has a much more finished bevel between rows while the 500 has a straight gray line. If you stare at it long enough, you can just see a small bevel to the gray line.

So, now I'm confused. Which one is better? :nono2:

Rich

#57 OFFLINE   D-Nice

D-Nice

    New Member

  • Registered
  • 9 posts
Joined: Dec 25, 2008

Posted 25 December 2012 - 06:54 PM

A better question would have been how he managed to measure that.

To begin with, while D-Nice has known qualifications, so why he is posting stuff like the above has people wondering - I am not sure what his motive is.

First, while everything is mastered in 8 bit, because of digital jitter (timing error) and other issues, errors can essentially reduce 8 bit down upwards of 40%. By having a path for 10 bit resolution, you have built in headroom.

To make it simple, think of having a water pipe that can deliver 8 gallons max a minute. If everything is working properly along the way, you get 8.0 gallons a minute at best. If you have a water pipe that can deliver 10.0 gallons per minute, there is no issue getting 8.0 gallons out even if there is variations in the stream.

As stated before, just because HDMI 1.3 can do Deep Color does not mean that Directv can do that.

If want a outside third party opinion, can call the people at Lumagen (maker of the Radiance Processors) and ask if you should set your processor out to the TV @ RGB 4:4:4 or YPbPr 4:2:2. They will tell you exactly what I told you. RGB 4:4:4 is 8 bit and YPbPr is 10 bit so you use YPbPr 4:2:2

Now, to measure a HR24-500 from 5-100% stimulus as D-Nice claims he did, he would need to be able to produce these known values directly out of the HR24-500 (or the other units he claims to have tested) and as we all know, there is not a 0-100 IRE Pattern Generator inside a HR24-500, thus making his claims impossible.

Also, the Accupel 4000 listed by D-Nice has a known problem with digital output which is why the Accupel 5000 was released.

Last I heard, D-Nice used a cheap meter that was profiled to a more expensive meter (though not a 1nm Spectroradiometer). Unfortunately, this has known limitations as well.

While I respect D-Nice's calibrations, his explanations here have me really wondering......

Hmmmm, where to start.....

Let's see, how does one measure a DirecTV box??? I take it you don't know how to place test patterns on a USB stick or stream it over the network. Duh!

Bit depth..... which bit depth to send a display depends on the display itself. For instance 2012 Panasonics work best with a 4:4:4 8 bit signal. 2011 Panasonics are best with a 4:2:2 8 bit signal. Sony LEDs 4:4:4 8bit. Not one of them show a superior picture with 10bit and/or higher signal types which is no more than padding an 8 bit signal with zeros.

Accupel.... there is nothing wrong with the 4000 and its digital output. If you think it is, show me the data.

Meter I use.... I use a Klein-10 in the field and it is referenced to a Photo Research 5nm PR-655. Neither of these meters would be considered cheap. They are used my many in the industry, respected in the industry and are very accurate. Oh and no one needs a 1nm spectro in the field or to use on consumer grade TVs with their very course control set. You would know this if you knew anything about calibration.

Finally, you wondering.... I think you need to reread your posts in this thread and ask yourself why you chose to post such inaccurate information become combative when corrections are posted and also attempt to insult someone. Amateurish at best :)

Merry Christmas

#58 OFFLINE   D-Nice

D-Nice

    New Member

  • Registered
  • 9 posts
Joined: Dec 25, 2008

Posted 25 December 2012 - 07:03 PM

Just checked a 500's Guide against a 100's Guide and you certainly are right. There's a major difference in what you call "steps" (I would have called them rows). The 100 has a much more finished bevel between rows while the 500 has a straight gray line. If you stare at it long enough, you can just see a small bevel to the gray line.

I guess a better term to describe what the DirecTV engineers were attempting to emulate would have been gray ramp. On a calibrated display, the ramp, including the dividing lines, is brighter than on the HR24-500.

So, now I'm confused. Which one is better? :nono2:

Rich

The HR24-500 adheres more to the standards than the other models I listed OOTB. That does not mean the others models cannot be calibrated to the same standards.

#59 OFFLINE   SomeRandomIdiot

SomeRandomIdiot

    Godfather

  • Registered
  • 1,260 posts
Joined: Jan 06, 2009

Posted 25 December 2012 - 10:39 PM

Hmmmm, where to start.....

Let's see, how does one measure a DirecTV box??? I take it you don't know how to place test patterns on a USB stick or stream it over the network. Duh!


You have just inserted added more variables into the stream which have the ability to influence the outcome.

Bit depth..... which bit depth to send a display depends on the display itself. For instance 2012 Panasonics work best with a 4:4:4 8 bit signal. 2011 Panasonics are best with a 4:2:2 8 bit signal. Sony LEDs 4:4:4 8bit. Not one of them show a superior picture with 10bit and/or higher signal types which is no more than padding an 8 bit signal with zeros.



Clearly you know of calibration, not digital error aka digtial jitter and how it reduces a 16 bit CD down to 8-10 bits at time and reduces an 8 bit video down upwards of 40%.

Furthermore this has nothing to do with RGB output when the TV wants to see YCbPr over HDMI. If you have one of the few that can force the RGB input (which, again is 8 bit on the HR24-500) and the user knows to make sure that it is set for RGB in the user settings (which I would bet very few have that knowledge), then they can live with their 8 bit output from the HR24.

Accupel.... there is nothing wrong with the 4000 and its digital output. If you think it is, show me the data.


Might want to read the calibration forums about that - or speak to the people @ Accupel

Meter I use.... I use a Klein-10 in the field and it is referenced to a Photo Research 5nm PR-655. Neither of these meters would be considered cheap. They are used my many in the industry, respected in the industry and are very accurate. Oh and no one needs a 1nm spectro in the field or to use on consumer grade TVs with their very course control set. You would know this if you knew anything about calibration.


Interesting as the Photo Research units do not match the Minolta Spectroradiometers (CS2000/CS2000a) which are considered reference.

Furthermore, there are plenty of know issues with the Klein, also in the Calibrator's forums.

And while we are laying cards on the table, let's just admit the reason calibrators use the more inaccurate Klein unit is because it takes a measurement quicker than most Spectroradiometer and getting out of your house faster trumps using a more accurate meter.

Finally, you wondering.... I think you need to reread your posts in this thread and ask yourself why you chose to post such inaccurate information become combative when corrections are posted and also attempt to insult someone. Amateurish at best :)

Merry Christmas


Right back at ya.

Edited by SomeRandomIdiot, 25 December 2012 - 10:46 PM.


#60 OFFLINE   Kerry

Kerry

    Mentor

  • Registered
  • 50 posts
Joined: Jul 17, 2008

Posted 25 December 2012 - 11:10 PM

Both the hr 21 and hr 34 have better pix quailty than the hr24. The 24 smears low level detail,,,, like backgrounds,,,,,,, wood panneling in a house epsoide is one test I used and many others. It was the dvr. I sent the 24 with its better speed back and kept the 21 because of the better pix quailty. The 34 was also line the 21 with no difference with many video test done.

#61 OFFLINE   D-Nice

D-Nice

    New Member

  • Registered
  • 9 posts
Joined: Dec 25, 2008

Posted 25 December 2012 - 11:14 PM

You have just inserted added more variables into the stream which have the ability to influence the outcome.

Actually I have not. Its ones and zeros. No video cards, no additional processing. As far as I know, DirecTV boxes do not color what they get from the NIC or external storage sources.

Clearly you know of calibration, not digital error aka digtial jitter and how it reduces a 16 bit CD down to 8-10 bits at time and reduces an 8 bit video down upwards of 40%.

No offense but i know far more about this stuff than you do. Its a requirement for what i do.

Furthermore this has nothing to do with RGB output when the TV wants to see YCbPr over HDMI. If you have one of the few that can force the RGB input (which, again is 8 bit on the HR24-500) and the user knows to make sure that it is set for RGB in the user settings (which I would bet very few have that knowledge), then they can live with their 8 bit output from the HR24.

A HDTV wants to receive a YCbPr signal? Really??? List one display manufacturer that explicitly requires their equipment receive a YCbPr signal.

All DirecTV boxes send out an 8 bit signal. That is true for those sending 4:4:4 and RGB. Get an analyzer and see for yourself.

Almost all displays are capable of recognizing RGB and YCbPr signal types without the end user having to do anything.

Might want to read the calibration forums about that - or speak to the people @ Accupel

Accupel is a Chromapure product and has been for a few years now. I know the owner, Tom, quite well and he is always in contact with the original owner and programmer of Accupel pattern generators, Greg. Neither have ever stated that any Accupel has issues with their digital outputs

Interesting as the Photo Research units do not match the Minolta Spectroradiometers (CS2000/CS2000a) which are considered reference.

You know nothing of this field so I suggest you read more on what meters are defined as reference. Photo Research is the defacto brand use to calibrate displays in Hollywood and broadcast studios. There isn't one studio that would allow you to touch any of their mastering displays without you having a Photo Research meter. Not one.

Furthermore, there are plenty of know issues with the Klein, also in the Calibrator's forums.

From one vendor? Sure. I don't use that vendor's tables

And while we are laying cards on the table, let's just admit the reason calibrators use the more inaccurate Klein unit is because it takes a measurement quicker than most Spectroradiometer and getting out of your house faster trumps using a more accurate meter.

Wrong. I, as well as many other calibrators, do not put quantity over quality. Guaranteed to fail. The largest difference between my Klein and the PR-655 its tables were created from is <1 dEuv on grayscale and <1.2 dE2000 on color for plasma, ccfl LCD and LED LCD. Translation for normal folks... visually imperceivable.

How many Kleins have you used? Better yet, how many spectros have you used?

Edited by D-Nice, 25 December 2012 - 11:33 PM.


#62 OFFLINE   hdtvfan0001

hdtvfan0001

    Hall Of Fame

  • DBSTalk Club
  • 32,343 posts
Joined: Jul 28, 2004

Posted 26 December 2012 - 07:35 AM

While the debate over who's test equipment is better and why is intriguing...going back to the OP....I suspect the goal was to get viewer feedback in contrast to an analysis-paralysis dissertation.

As others have pointed out...there are multiple factors in play for the final results.

Having actually viewed both the HR34 and HR24 on the exact same pair of 55" and 60" displays for extended periods of time...my simple response would be that the HR34 appears to present a very slight improved image, using the most sophisticated and complex visual technology in existence - a highly picky set of eyes. ;)
DBSTalk CHAT ROOM MODERATOR
DirecTV Customer Since 1996

#63 OFFLINE   D-Nice

D-Nice

    New Member

  • Registered
  • 9 posts
Joined: Dec 25, 2008

Posted 26 December 2012 - 09:15 AM

Having actually viewed both the HR34 and HR24 on the exact same pair of 55" and 60" displays for extended periods of time...my simple response would be that the HR34 appears to present a very slight improved image, using the most sophisticated and complex visual technology in existence - a highly picky set of eyes. ;)

Can't debate that :)

#64 OFFLINE   P Smith

P Smith

    Mr. FixAnything

  • Registered
  • 20,227 posts
  • LocationMediterranean Sea
Joined: Jul 25, 2002

Posted 26 December 2012 - 10:11 AM

Can't debate that :)

While

Both the hr 21 and hr 34 have better pix quailty than the hr24. The 24 smears low level detail,,,, like backgrounds,,,,,,, wood panneling in a house epsoide is one test I used and many others. It was the dvr. I sent the 24 with its better speed back and kept the 21 because of the better pix quailty. The 34 was also line the 21 with no difference with many video test done.

what you'll address to the observation ?

#65 OFFLINE   Laxguy

Laxguy

    Never say 'never'.

  • DBSTalk Club
  • 14,146 posts
  • LocationWinters, CA, between Napa and Sacramento
Joined: Dec 02, 2010

Posted 26 December 2012 - 11:13 AM

Both the hr 21 and hr 34 have better pix quailty than the hr24. The 24 smears low level detail,,,, like backgrounds,,,,,,, wood panneling in a house epsoide is one test I used and many others. It was the dvr. I sent the 24 with its better speed back and kept the 21 because of the better pix quailty. The 34 was also line the 21 with no difference with many video test done.


Would that not depend hugely on how your monitor handles things??
"Laxguy" means a guy who loves lacrosse.

#66 OFFLINE   Rich

Rich

    Hall Of Fame

  • DBSTalk Club
  • 22,128 posts
  • LocationPiscataway, NJ
Joined: Feb 22, 2007

Posted 26 December 2012 - 12:25 PM

Would that not depend hugely on how your monitor handles things??


You'd think so. Kinda hard to tell who knows what they're doing and who's trying to baffle the other guy with BS. As for the 21s having a better PQ than a 24...that's gotta be a poorly tuned TV.

I think I'll stick with the Mark I Eyeball. I think having to have a new TV calibrated is kinda criminal. People don't buy TVs with plans to calibrate them. They buy them to watch programming.

Rich

#67 OFFLINE   TomCat

TomCat

    Broadcast Engineer

  • Registered
  • 3,699 posts
Joined: Aug 31, 2002

Posted 28 December 2012 - 08:19 PM

Mine's bigger than yours.

But seriously, the PQ as decoded by DTV DVRs is all the same. There can be some differences in how things are displayed from that point forward; one would expect that they all would handle grey scales and gamma about the same, but maybe not, we seem to have some anecdotal evidence here that they really don't. It's a little difficult for any of those reports to be all that faithfully empirical, as none of them appear to be in double-blind testing with only one element different per test, most are not "blind" at all and have many components different which may influence their credibility. Personally, I think gamma adjustments can make a lot of difference in how much I "like" or "dislike" a picture (I wish my current TV's had more control over that like my older HDTV's did).

That said, the differrences between the PQ of any top two brands of HDTV, say a Sharp Aquos and a Sony XBR, for instance, will probably make a much more significant difference in how things appear than any technical difference between any of the available HD DVR+ models out there.

There is also probably more difference between the outcomes of calibrations done by two top calibrators on the identical TV, since that is more an artistic or impressionistic black art than it is a scientific practice. Still valid; just not as absolute as they would like us to think it is. It's not that different than tuning a piano; if you do that precisely by frequency matching sound waves on a scope it will probably sound awful, but if you do it by ear with decades of skill and experience behind you, it can sound astonishingly perfect.

And if quibbling about those tiny differences between DVR models is the biggest issue in your life, then you have it pretty good compared to most.

Edited by TomCat, 28 December 2012 - 08:32 PM.

It's usually safe to talk honestly and openly with people because they typically are not really listening anyway.

#68 OFFLINE   Hoosier205

Hoosier205

    Hall Of Fame

  • Registered
  • 6,612 posts
Joined: Sep 03, 2007

Posted 28 December 2012 - 09:24 PM

The fact that someone tried to call out D-Nice brought us the most humorous DBStalk moment of 2012! :) He's an industry god. Haha
DTV = Digital Television

#69 OFFLINE   TomCat

TomCat

    Broadcast Engineer

  • Registered
  • 3,699 posts
Joined: Aug 31, 2002

Posted 29 December 2012 - 06:36 PM

...while everything is mastered in 8 bit, because of digital jitter (timing error) and other issues, errors can essentially reduce 8 bit down upwards of 40%. By having a path for 10 bit resolution, you have built in headroom...

I think this might be a misunderstanding of how bit depth and timing errors might manifest as deterioration to PQ. Skip to the last four paragraphs for what is and what is not possible, and you are welcome to read and pick apart the rest (the indented part in between) if you are still wondering how we can come to this conclusion. No guns to the head here, so easy on the hatin'. If an issue is complicated, sometimes verbosity is required.

the largest 8-bit word is 11111111, or 255 decimal. This means that there are 256 possible values to assign quantization levels to (in practice, about 232). The largest 10-bit word is 1111111111, or 1023 decimal, and this yields 1024 possible quantization levels (in practice, about 1006). If you multiply 232 (for Pr) times 232 (for Pb) that yields over 53,000 shades of available colors or chroma for 8-bit video. If you multiply 1006 by 1006 that yields over a million available shades of color for 10-bit video.

The largest digital word in 8-bit video (which is what all consumer HD is limited to) even when processed at 10 bits is 0011111111, again, 255 in decimal. Yes, there are two more bits of resolution in 10-bit, but if the source is 8-bit, those extra two bits are truncated, or zeroed out, meaning that the end result is exactly the same as if processed at 8-bit.

The limitation of maximum number of values or quantization levels is still 256 (232), the available shades of color for RGB or for YPrPB is still about 53,000, and the number of quantization levels and color values (assuming sample rate [number of pixels in the pixel map] is unchanged) is what fixes and defines how accurate the video is in relation to its source, and the pixel map itself is what defines the amount of possible detail. And that, in a nutshell, is exactly how video digitization works.

In digital delivery, the ones and zeroes representing these coefficients are usually modulated as low/high voltages, low representing zero and high representing 1. This can be seen as a series of square-wave pulses in pulse-code modulation. Degradation to the bit stream due to frequency response losses manifests as a rounding of corners on those pulses.

Timing errors manifest as the pulses arriving a bit early or a bit late compared to a reference or to adjacent pulses, which smears the pulses' rise times and fall times.

So after transport through a hostile environment, the representation on a scope can be smeared with the pulses being rounded. That is analog degradation to the carrying medium, which is unavoidable.

But an MPEG decoder or a DAC can still identify each rounded pulse as a one, and each absence of a pulse as a zero, which means it can extract the original coefficients perfectly, making rounding degradation of the pulses meaningless. It can also know that even if offset in time, at what time the pulse is supposed to occur; it reclocks them, resetting them back to where they were originally in reference to each other, which takes jitter out of the equation completely. What we are left with is a perfect representation of the coefficients as transmitted, even if there is error correction used to supplement some of the potentially missing numbers.

All of the information can still be extracted perfectly, which is why we use digital in the first place. In analog, the information is married to, even part of, the carrying medium, and degradation to the medium also degrades the message. In digital, the message is turned into a mathematical construct, essentially a number, or a stream of binary numbers, which divorces it from the medium. If the medium is degraded (up to a point) the message still survives 100% intact.

All of this is according to how much carrying medium degradation there is. If there is little degradation, it is easy to extract the information and recreate the coefficients perfectly. If there is a lot of degradation, error correction can fill in the blanks intelligently so that the numbers are still extracted perfectly. If there is too much degradation, the signal is muted and the screen goes black. It's all, perfectly, or none. There is no inbetween.

That is what is sometimes referred to as the digital cliff; at any one point in time an MPEG decoder can either extract the 187 digital words in a packet, or replace corrupted values with redundant copies using error correction even if the carrying medium is severely compromised, or it can extract nothing if the packets are compromised too much. It is all or nothing; you either end up with a perfect picture or a blank screen. The decoder either has enough (all) of the sent information to recreate the other 99% that was not encoded (discarded in compression) or it doesn't have enough information that it can make sense of and so can't make intelligent guesses about how to reconstruct the other 99%, and so does nothing (mutes to black).

It might not appear that way if the stream is pixellated due to reception, but at any one instant in time each macroblock is either painted onto the screen perfectly or not, and when not, the previous macroblock remains there frozen until eventually updated (or until a time-out mutes the entire screen), which is why you might briefly see a mosaic effect. Each part of the picture is still perfect, but older macroblocks mixed with newer updated macroblocks destroy the stitching illusion and what you see over all then does not look as it really should, which is considered a perceptual artifact (this is different from pixellation due to overcompression, which obviously can present an imperfect picture, although still faithful to what was encoded and compressed).

The point is that while the carrying medium can partially or gradually degrade, while inside the digital domain, the information doesn't, and can't; at any one particular instant in time you either have it all extracted and reclocked perfectly faithful to what was sent, or you have nothing. There is no visible artifact that can degrade 8-bit video by "40%", and having 10-bit processing would not in any way provide "headroom" to an 8-bit signal, whether it was degraded (which it can't be) or not.

In binary math, and therefore in digital video, there is no "up to 40%" of a "one" or "part" of a "zero", there can only be one of two states of being for each bit of the information itself: one, or zero. And the only way those numbers can be changed from one to zero or back is if a mathematical process is performed on them, which does not happen by accident in nature during transport. It happens only on purpose at decode and conversion to analog (assuming conversion happens in the DVR, which it only does for component and composite) and if the pixel map is rescaled, but it always happens in virtually the same way, providing virtually the same result, and completely irregardless of limited deterioration of the carrying medium during transport.

As D-Nice says, the only differences are due to differences in chipsets. The standard for what is accomplished is complicated and rigid, but how they get to that finish line, how they do that job, is up to the chipset designer. That can cause small, nearly imperceptible differences for test patterns, and virtually invisible differences for garden-variety video.

One other place where math is done is if there is YPrPb to RGB conversion, such as in the HR24-500. Since that is a matrix equation and resultant values are based on mixing percentages of other values, there is room for tiny amounts of quantization rounding error to creep in, the amount depending upon how sophisticated the conversion might be (we have to assume it is not sophisticated to keep costs down). That would imply that true YPrPb processing (no conversion needed) would be more accurate, although not necessarily "better". There would be less error if it were 10-bit video, but it isn't.

Edited by TomCat, 29 December 2012 - 07:25 PM.

It's usually safe to talk honestly and openly with people because they typically are not really listening anyway.

#70 OFFLINE   Hoosier205

Hoosier205

    Hall Of Fame

  • Registered
  • 6,612 posts
Joined: Sep 03, 2007

Posted 29 December 2012 - 07:21 PM

I think this might be a misunderstanding of how bit depth and timing errors might manifest as deterioration to PQ. Skip to the last four paragraphs for what is and what is not possible, and you are welcome to read and pick apart the rest (the indented part in between) if you are still wondering how we can come to this conclusion. No guns to the head here, so easy on the hatin'. If an issue is complicated, sometimes verbosity is required.

the largest 8-bit word is 11111111, or 255 decimal. This means that there are 256 possible values to assign quantization levels to (in practice, about 232). The largest 10-bit word is 1111111111, or 1023 decimal, and this yields 1024 possible quantization levels (in practice, about 1006). If you multiply 232 (for Pr) times 232 (for Pb) that yields over 53,000 shades of available colors or chroma for 8-bit video. If you multiply 1006 by 1006 that yields over a million available shades of color for 10-bit video.

The largest digital word in 8-bit video (which is what all consumer HD is limited to) even when processed at 10 bits is 0011111111, again, 255 in decimal. Yes, there are two more bits of resolution in 10-bit, but if the source is 8-bit, those extra two bits are truncated, or zeroed out, meaning that the end result is exactly the same as if processed at 8-bit.

The limitation of maximum number of values or quantization levels is still 256 (232), the available shades of color for RGB or for YPrPB is still about 53,000, and the number of quantization levels and color values (assuming sample rate [number of pixels in the pixel map] is unchanged) is what fixes and defines how accurate the video is in relation to its source, and the pixel map itself is what defines the amount of possible detail. And that, in a nutshell, is exactly how video digitization works.

In digital delivery, the ones and zeroes representing these coefficients are usually modulated as low/high voltages, low representing zero and high representing 1. This can be seen as a series of square-wave pulses in pulse-code modulation. Degradation to the bit stream due to frequency response losses manifests as a rounding of corners on those pulses.

Timing errors manifest as the pulses arriving a bit early or a bit late compared to a reference or to adjacent pulses, which smears the pulses' rise times and fall times.

So after transport through a hostile environment, the representation on a scope can be smeared with the pulses being rounded. That is analog degradation to the carrying medium, which is unavoidable.

But an MPEG decoder or a DAC can still identify each rounded pulse as a one, and each absence of a pulse as a zero, which means it can extract the original coefficients perfectly, making rounding degradation of the pulses meaningless. It can also know that even if offset in time, at what time the pulse is supposed to occur; it reclocks them, resetting them back to where they were originally in reference to each other, which takes jitter out of the equation completely. What we are left with is a perfect representation of the coefficients as transmitted, even if there is error correction used to supplement some of the potentially missing numbers.

All of the information can still be extracted perfectly, which is why we use digital in the first place. In analog, the information is married to, even part of, the carrying medium, and degradation to the medium also degrades the message. In digital, the message is turned into a mathematical construct, essentially a number, or a stream of binary numbers, which divorces it from the medium. If the medium is degraded (up to a point) the message still survives 100% intact.

All of this is according to how much carrying medium degradation there is. If there is little degradation, it is easy to extract the information and recreate the coefficients perfectly. If there is a lot of degradation, error correction can fill in the blanks intelligently so that the numbers are still extracted perfectly. If there is too much degradation, the signal is muted and the screen goes black. It's all, perfectly, or none. There is no inbetween.

That is what is sometimes referred to as the digital cliff; at any one point in time an MPEG decoder can either extract the 187 digital words in a packet, or replace corrupted values with redundant copies using error correction even if the carrying medium is severely compromised, or it can extract nothing if the packets are compromised too much. It is all or nothing; you either end up with a perfect picture or a blank screen. The decoder either has enough (all) of the sent information to recreate the other 99% that was not encoded (discarded in compression) or it doesn't have enough information that it can make sense of and so can't make intelligent guesses about how to reconstruct the other 99%, and so does nothing (mutes to black).

It might not appear that way if the stream is pixellated due to reception, but at any one instant in time each macroblock is either painted onto the screen perfectly or not, and when not, the previous macroblock remains there frozen until eventually updated (or until a time-out mutes the entire screen), which is why you might briefly see a mosaic effect. Each part of the picture is still perfect, but older macroblocks mixed with newer updated macroblocks destroy the stitching illusion and what you see over all then does not look as it really should, which is considered a perceptual artifact (this is different from pixellation due to overcompression, which obviously can present an imperfect picture, although still faithful to what was encoded and compressed).The point is that while the carrying medium can partially or gradually degrade, while inside the digital domain, the information doesn't, and can't; at any one particular instant in time you either have it all extracted and reclocked perfectly faithful to what was sent, or you have nothing. There is no visible artifact that can degrade 8-bit video by "40%", and having 10-bit processing would not in any way provide "headroom" to an 8-bit signal, whether it was degraded (which it can't be) or not.

In binary math, and therefore in digital video, there is no "up to 40%" of a "one" or "part" of a "zero", there can only be one of two states of being for each bit of the information itself: one, or zero. And the only way those numbers can be changed from one to zero or back is if a mathematical process is performed on them, which does not happen by accident in nature during transport. It happens only on purpose at decode and conversion to analog (assuming conversion happens in the DVR, which it only does for component and composite) and if the pixel map is rescaled, but it always happens in virtually the same way, providing virtually the same result, and completely irregardless of limited deterioration of the carrying medium during transport.

As No Dice says, the only differences are due to differences in chipsets. The standard for what is accomplished is complicated and rigid, but how they get to that finish line, how they do that job, is up to the chipset designer. That can cause small, nearly imperceptible differences for test patterns, and virtually invisible differences for garden-variety video.

One other place where math is done is if there is YPrPb to RGB conversion, such as in the HR24-500. Since that is a matrix equation and resultant values are based on mixing percentages of other values, there is room for tiny amounts of quantization rounding error to creep in, the amount depending upon how sophisticated the conversion might be (we have to assume it is not sophisticated to keep costs down). That would imply that true YPrPb processing (no conversion needed) would be more accurate, although not necessarily "better". There would be less error if it were 10-bit video, but it isn't.


Show off :)
DTV = Digital Television

#71 OFFLINE   Laxguy

Laxguy

    Never say 'never'.

  • DBSTalk Club
  • 14,146 posts
  • LocationWinters, CA, between Napa and Sacramento
Joined: Dec 02, 2010

Posted 29 December 2012 - 07:52 PM

Heh.

And thank you.

I've spent a lot of time explaining that the number of pixels a camera can handle is just one factor in how good the resulting images can be. Also, that I can make a 5,000 x 5,000 pixel image out of a 300 x 300 image, but it'd look like crap unless it was printed very large and you viewed it from a half mile away..... and so forth.
"Laxguy" means a guy who loves lacrosse.

#72 OFFLINE   P Smith

P Smith

    Mr. FixAnything

  • Registered
  • 20,227 posts
  • LocationMediterranean Sea
Joined: Jul 25, 2002

Posted 29 December 2012 - 09:01 PM

Packet's first byte eg sync byte [47h] for MPES stream is still a part of 188 bytes packet's length. If it will not equal 47h then data will be discarded up to next sync byte, next packet.
MPEG video/audio decoder will process not the packets, but ES where packet's header and other system bytes stripped down.

#73 OFFLINE   SomeRandomIdiot

SomeRandomIdiot

    Godfather

  • Registered
  • 1,260 posts
Joined: Jan 06, 2009

Posted 31 December 2012 - 09:29 PM

I think this might be a misunderstanding of how bit depth and timing errors might manifest as deterioration to PQ. Skip to the last four paragraphs for what is and what is not possible, and you are welcome to read and pick apart the rest (the indented part in between) if you are still wondering how we can come to this conclusion. No guns to the head here, so easy on the hatin'. If an issue is complicated, sometimes verbosity is required.

the largest 8-bit word is 11111111, or 255 decimal. This means that there are 256 possible values to assign quantization levels to (in practice, about 232). The largest 10-bit word is 1111111111, or 1023 decimal, and this yields 1024 possible quantization levels (in practice, about 1006). If you multiply 232 (for Pr) times 232 (for Pb) that yields over 53,000 shades of available colors or chroma for 8-bit video. If you multiply 1006 by 1006 that yields over a million available shades of color for 10-bit video.

The largest digital word in 8-bit video (which is what all consumer HD is limited to) even when processed at 10 bits is 0011111111, again, 255 in decimal. Yes, there are two more bits of resolution in 10-bit, but if the source is 8-bit, those extra two bits are truncated, or zeroed out, meaning that the end result is exactly the same as if processed at 8-bit.

The limitation of maximum number of values or quantization levels is still 256 (232), the available shades of color for RGB or for YPrPB is still about 53,000, and the number of quantization levels and color values (assuming sample rate [number of pixels in the pixel map] is unchanged) is what fixes and defines how accurate the video is in relation to its source, and the pixel map itself is what defines the amount of possible detail. And that, in a nutshell, is exactly how video digitization works.

In digital delivery, the ones and zeroes representing these coefficients are usually modulated as low/high voltages, low representing zero and high representing 1. This can be seen as a series of square-wave pulses in pulse-code modulation. Degradation to the bit stream due to frequency response losses manifests as a rounding of corners on those pulses.

Timing errors manifest as the pulses arriving a bit early or a bit late compared to a reference or to adjacent pulses, which smears the pulses' rise times and fall times.

So after transport through a hostile environment, the representation on a scope can be smeared with the pulses being rounded. That is analog degradation to the carrying medium, which is unavoidable.

But an MPEG decoder or a DAC can still identify each rounded pulse as a one, and each absence of a pulse as a zero, which means it can extract the original coefficients perfectly, making rounding degradation of the pulses meaningless. It can also know that even if offset in time, at what time the pulse is supposed to occur; it reclocks them, resetting them back to where they were originally in reference to each other, which takes jitter out of the equation completely. What we are left with is a perfect representation of the coefficients as transmitted, even if there is error correction used to supplement some of the potentially missing numbers.

All of the information can still be extracted perfectly, which is why we use digital in the first place. In analog, the information is married to, even part of, the carrying medium, and degradation to the medium also degrades the message. In digital, the message is turned into a mathematical construct, essentially a number, or a stream of binary numbers, which divorces it from the medium. If the medium is degraded (up to a point) the message still survives 100% intact.

All of this is according to how much carrying medium degradation there is. If there is little degradation, it is easy to extract the information and recreate the coefficients perfectly. If there is a lot of degradation, error correction can fill in the blanks intelligently so that the numbers are still extracted perfectly. If there is too much degradation, the signal is muted and the screen goes black. It's all, perfectly, or none. There is no inbetween.

That is what is sometimes referred to as the digital cliff; at any one point in time an MPEG decoder can either extract the 187 digital words in a packet, or replace corrupted values with redundant copies using error correction even if the carrying medium is severely compromised, or it can extract nothing if the packets are compromised too much. It is all or nothing; you either end up with a perfect picture or a blank screen. The decoder either has enough (all) of the sent information to recreate the other 99% that was not encoded (discarded in compression) or it doesn't have enough information that it can make sense of and so can't make intelligent guesses about how to reconstruct the other 99%, and so does nothing (mutes to black).

It might not appear that way if the stream is pixellated due to reception, but at any one instant in time each macroblock is either painted onto the screen perfectly or not, and when not, the previous macroblock remains there frozen until eventually updated (or until a time-out mutes the entire screen), which is why you might briefly see a mosaic effect. Each part of the picture is still perfect, but older macroblocks mixed with newer updated macroblocks destroy the stitching illusion and what you see over all then does not look as it really should, which is considered a perceptual artifact (this is different from pixellation due to overcompression, which obviously can present an imperfect picture, although still faithful to what was encoded and compressed).

The point is that while the carrying medium can partially or gradually degrade, while inside the digital domain, the information doesn't, and can't; at any one particular instant in time you either have it all extracted and reclocked perfectly faithful to what was sent, or you have nothing. There is no visible artifact that can degrade 8-bit video by "40%", and having 10-bit processing would not in any way provide "headroom" to an 8-bit signal, whether it was degraded (which it can't be) or not.

In binary math, and therefore in digital video, there is no "up to 40%" of a "one" or "part" of a "zero", there can only be one of two states of being for each bit of the information itself: one, or zero. And the only way those numbers can be changed from one to zero or back is if a mathematical process is performed on them, which does not happen by accident in nature during transport. It happens only on purpose at decode and conversion to analog (assuming conversion happens in the DVR, which it only does for component and composite) and if the pixel map is rescaled, but it always happens in virtually the same way, providing virtually the same result, and completely irregardless of limited deterioration of the carrying medium during transport.

As D-Nice says, the only differences are due to differences in chipsets. The standard for what is accomplished is complicated and rigid, but how they get to that finish line, how they do that job, is up to the chipset designer. That can cause small, nearly imperceptible differences for test patterns, and virtually invisible differences for garden-variety video.

One other place where math is done is if there is YPrPb to RGB conversion, such as in the HR24-500. Since that is a matrix equation and resultant values are based on mixing percentages of other values, there is room for tiny amounts of quantization rounding error to creep in, the amount depending upon how sophisticated the conversion might be (we have to assume it is not sophisticated to keep costs down). That would imply that true YPrPb processing (no conversion needed) would be more accurate, although not necessarily "better". There would be less error if it were 10-bit video, but it isn't.


Not hating at all.

Digital is a wonderful medium when it works as expected.

Sony the world in the early 80s that 16bit 44.1k sampled digital audio was perfect - just like D-Nice in his statement that its only 1s and 0s. Those who used their ears instead of theory knew something was wrong.

10 years later it was finally measured that digital jitter was an issue in digital. Now digital jitter is accepted worldwide and measured.

Turned out that those 16bit 1s and 0s flying down a digital path worked fine - until the timing burped.

Same thing as an Interstate. You can have cars bumper to bumper with literally no space between them all travelling at 100 mph. However, the first car to fall to 99.999999 mph and you have a tremendous pile up.

And when that happens in the digital path, the 1s and 0s come crashing into each other. Turned out the 16 bit audio was reduced to 8-10bits (a reduction of 40%-50%) and the quality suffered greatly.

No difference in video. But as circuits are now made to try and reduce the issue as much as possible (and because of cheap circuits, you never prevent it).

So yes, giving a 10bit path gives some headroom.

However, as it turns out, this is a mute point, as testing over the weekend revealed that DirecTV boxes cannot even output YCbCr 4:2:2 10bit YCbCr. The output is YCbCr 4:4:4 8 bit, which one would think that an expert such as D-Nice should have known (despite arguing for 8bit, 10bit and higher), except he does not have the equipment to measure that.

Revealed a valuable lesson.

A building inspector might not be the best general contractor.
A Movie Reviewer might not be the Producer.
A School Teacher that can grade standardized test does not need to know the subject matter.

Lots I don't know. Some I do know.

However, I am at least smart enough to learn when I don't know.

Others might be wise to do the same (not looking at you).

I'm pretty confident in the earlier conversation what is right - and quite frankly, will not recommend D-Nice in the future to people that asked me about calibration, which I did prior (see my first response).

Unfortunately, he isn't smart enough to understand what he does not know about behind the scenes and his equipment limitations.

PS - the only way for Digital extraction to be perfect as you state above is with Error Correction. And there is no error correction in the HDMI chain.

#74 OFFLINE   Kerry

Kerry

    Mentor

  • Registered
  • 50 posts
Joined: Jul 17, 2008

Posted 31 December 2012 - 09:41 PM

You'd think so. Kinda hard to tell who knows what they're doing and who's trying to baffle the other guy with BS. As for the 21s having a better PQ than a 24...that's gotta be a poorly tuned TV.

I think I'll stick with the Mark I Eyeball. I think having to have a new TV calibrated is kinda criminal. People don't buy TVs with plans to calibrate them. They buy them to watch programming.

Rich


You guys asnd girs sure do yak alot. It dont matter with the test equment disk and bla bla, what matters in this simple test is what you can see with the eye ball, or rather in this case cant see. It would take me all of 2 min to show you on your display the sofness and loss of detail on the 24 vs the 34 and 21. Its harder to see something thats not there vs something that is.ps if you call a xbr65hx929 a bad display O well. That kinda sums it up.
Proced govner LOL!

#75 OFFLINE   Hoosier205

Hoosier205

    Hall Of Fame

  • Registered
  • 6,612 posts
Joined: Sep 03, 2007

Posted 31 December 2012 - 10:25 PM

You'd think so. Kinda hard to tell who knows what they're doing and who's trying to baffle the other guy with BS. As for the 21s having a better PQ than a 24...that's gotta be a poorly tuned TV.

I think I'll stick with the Mark I Eyeball. I think having to have a new TV calibrated is kinda criminal. People don't buy TVs with plans to calibrate them. They buy them to watch programming.

Rich


Having a new TV calibrated is kinda criminal? How so? The point of calibration to see content as it was intended to be seen. If you spend good money on a new display, why would you not want it too look as good as it can? I just don't see the logic in being opposed to calibration.
DTV = Digital Television




Protected By... spam firewall...And...