When I purchased my HDTV, I noticed that trying to find what the actual resolution a set could display was like trying to pull hen's teeth. The manufacturers would say things like "HDTV compatible" or "over 900 lines of resolution", but the claims seemed bogus. I finally settled on a plasma screen because I *know* the number of pixels it has, but I was curious if there was a good place for information on what the various sets are supposed to display and how close they come to that theoretical best.
Actual Resolutions of TV Sets
-
LIVE: Tech Talk Podcast with Scott Wilkinson, Episode 8 Click here for details.
- Status
1 - 17 of 17 Posts
Several months ago Gary Merson wrote an article comparing resolution of several direct view HDTV monitors. The number of vertical lines that can be resolved across the screen varied from 800 to 1200. Maybe the article can still be found in Widescreen Review archives.
One thing to keep in mind, the best picture quality does not always tie to the maximum resolution number because many other factors apply.
One thing to keep in mind, the best picture quality does not always tie to the maximum resolution number because many other factors apply.
Joined
·
786 Posts
The question comes up a lot.
The answer really is no. No one has actually gone through measuring each separate monitor and published the information in one place.
Even reviews done on the sets almost never mention the actual resolutions, unless provided to the reviewer. The manufacturers don't even communicate the actual resolutions.
Granted resolution is a big part of the equation but, as noted in the above post, other factors are just as important.
Still, it is sad you can't get the information even from the companies selling the sets.
The answer really is no. No one has actually gone through measuring each separate monitor and published the information in one place.
Even reviews done on the sets almost never mention the actual resolutions, unless provided to the reviewer. The manufacturers don't even communicate the actual resolutions.
Granted resolution is a big part of the equation but, as noted in the above post, other factors are just as important.
Still, it is sad you can't get the information even from the companies selling the sets.
Joined
·
11,139 Posts
Like the pulling-hen's-teeth comparison. I'm also surprised, this far along in HDTV marketing, there are no unbiased sources of set resolutions. Think it might take publications that aren't advertiser supported to provide such data. Buying a set because images look good seems insufficient. Unless you're aware of the limitations imposed by shadow masks or grilles in direct-view-CRT sets, you'll discover your new set can't display about 40 percent of the details in a typical taped HDTV broadcast (say, less than 1400 lines of resolution, full 16X9 width). And if a broadcaster manages to crank out, say, 1700 lines or more during a live broadcast, you'll be missing 53 percent or more of the details (luminance or black-and-white resolution, anyway). Filters within sets lop off about 20 percent of received HDTV horizontal resolution just for starters. And MPEG-2 encoding varies the resolution according to motion and details within scenes.
The ability to deliver and view fine details is what HDTV is all about. Such high resolutions enable you to sit closer to the screen and fill a wider viewing angle for a better simulation of realism. Other image parameters such as contrast and color are important, of course, but without fine details, IMO, it's like constantly viewing impressionistic paintings. A director might want to provide such fuzzy impressions for a dream sequence or other creative reasons, but with limited-resolution receivers intricate details will never reach the eyes of HDTV viewers.
Would have selected a plasma set myself a year ago while shopping for HDTV, but what seemed to be the most desirable models, such as Panasonic's 60-inch, full-HDTV-resolution model, were only at trade shows, and still aren't available. -- John
------------------
STOP DVI/HDCP AND DFAST
[This message has been edited by John Mason (edited 05-28-2001).]
The ability to deliver and view fine details is what HDTV is all about. Such high resolutions enable you to sit closer to the screen and fill a wider viewing angle for a better simulation of realism. Other image parameters such as contrast and color are important, of course, but without fine details, IMO, it's like constantly viewing impressionistic paintings. A director might want to provide such fuzzy impressions for a dream sequence or other creative reasons, but with limited-resolution receivers intricate details will never reach the eyes of HDTV viewers.
Would have selected a plasma set myself a year ago while shopping for HDTV, but what seemed to be the most desirable models, such as Panasonic's 60-inch, full-HDTV-resolution model, were only at trade shows, and still aren't available. -- John
------------------
STOP DVI/HDCP AND DFAST
[This message has been edited by John Mason (edited 05-28-2001).]
Joined
·
3,650 Posts
I have the Sony KPxx-xbr300, and the DRC circuit upconverts NTSC 720x480i to 1440x960i. Since the set does 1080i native, I **assumed** that the set could display a 1440x1080i HDTV resolution... of course, just most folks, I've never seen actual resolutions documented...
Oh well...
-Tony
Oh well...
-Tony
Joined
·
19,586 Posts
Well, that's kind of the point. Your Toshiba probably does not have 1600 lines of horizontal resolution. My Dwin FP system, which costs way more and has fewer compromises and has the same size CRTs won't do 1600 lines (on a bigger screen), so its highly unlikely that your Toshiba will. Its more likely in the 1200 to 1300 range.
Those numbers are most like the theortical maximum it could do if all the planets were in alignment, given the specs of the parts they used. But in the real world, it'll never happen.
One reason that no one has hard numbers, at least for projection systems, is that its not a black and white issue. If you go up and count the alternating lines you can see in a pattern, when do you stop counting them? If you can see any black between the white at all, is that good enough? Do you have to see a black line at least 1/2 the width of the surrounding white lines? Do they have to be completely clean, and if so how do you confirm that? Do you get up there with a micrometer and measure each one? But what if the focus attaintable isn't perfect so there is a little fuzz between them? Do you count that?
And so on and so on. Same with the resolution wedges such as on Avia. Where in there does the moire reach the point where its clearly not resolving enough to count?
------------------
Dean Roddey
The CIDLib C++ Frameworks
[email protected]
www.charmedquark.com
Those numbers are most like the theortical maximum it could do if all the planets were in alignment, given the specs of the parts they used. But in the real world, it'll never happen.
One reason that no one has hard numbers, at least for projection systems, is that its not a black and white issue. If you go up and count the alternating lines you can see in a pattern, when do you stop counting them? If you can see any black between the white at all, is that good enough? Do you have to see a black line at least 1/2 the width of the surrounding white lines? Do they have to be completely clean, and if so how do you confirm that? Do you get up there with a micrometer and measure each one? But what if the focus attaintable isn't perfect so there is a little fuzz between them? Do you count that?
And so on and so on. Same with the resolution wedges such as on Avia. Where in there does the moire reach the point where its clearly not resolving enough to count?
------------------
Dean Roddey
The CIDLib C++ Frameworks
[email protected]
www.charmedquark.com
Joined
·
6,723 Posts
Greetings
I put the Accupel Multiburst pattern on the screen for Pioneers and then Toshibas ... The 2 Pixel grouping that indicates the TV's ability to resolve at least 960 lines of h-resolution of out 1920 lines is far more crisp on Pioneer units like the Elite 510 than on a 56H80 unit for instance.
Since Pioneer only rates their 510 unit at about 1150 lines ... by default the TOshiba would be capable of much less given how the 2 pix group is shown.
In this case, I tend to believe the Pioneer number as being more reasonable and this would likely put the 56" Toshiba in the 1000 to 1100 line department. Certainly not 1600 lines.
Regards
------------------
Michael @ The Laser Video Experience
I put the Accupel Multiburst pattern on the screen for Pioneers and then Toshibas ... The 2 Pixel grouping that indicates the TV's ability to resolve at least 960 lines of h-resolution of out 1920 lines is far more crisp on Pioneer units like the Elite 510 than on a 56H80 unit for instance.
Since Pioneer only rates their 510 unit at about 1150 lines ... by default the TOshiba would be capable of much less given how the 2 pix group is shown.
In this case, I tend to believe the Pioneer number as being more reasonable and this would likely put the 56" Toshiba in the 1000 to 1100 line department. Certainly not 1600 lines.
Regards
------------------
Michael @ The Laser Video Experience
Joined
·
1,797 Posts
If you have an HTPC, and an illustration program such as Corel Draw! or Adobe Illustrator, you can do this yourself. You need to construct several different patterns with increasing resolution - such as 1400 alternating vertical black and white lines (700 of each), 1600, 1800, and so forth.
You can then view these images via your HTPC on your monitor, and estimate your horizontal resolution in pixels. You will be faced with the interpretation that Dean mentioned, but it still gives you some idea.
------------------
BOYCOTT DVI/HDCP & JVC!
You can then view these images via your HTPC on your monitor, and estimate your horizontal resolution in pixels. You will be faced with the interpretation that Dean mentioned, but it still gives you some idea.
------------------
BOYCOTT DVI/HDCP & JVC!
The more things change...the more they remain the same.
In the past, most of the TV manufacturers measured the Video Amp (measured in Mhz) and attributed 1 Mhz to 80 lines of Hort, resolution.
It appears that this pratcise still continues. I would bet that the Toshiba in question has a 20 mHz Video Amp and 20 times 80 equals 1600.
FACT: No 7" Crt with "normal" focusing (as opposed to magnetic) can do anywheres near 1600 lines of H. Res. The number should be somewhere between 900 and 1200.
A Sony G90 FPTV can do 1600 lines. It has 9" CRT's with the magnetic focus system and costs $35,000.
Lee
In the past, most of the TV manufacturers measured the Video Amp (measured in Mhz) and attributed 1 Mhz to 80 lines of Hort, resolution.
It appears that this pratcise still continues. I would bet that the Toshiba in question has a 20 mHz Video Amp and 20 times 80 equals 1600.
FACT: No 7" Crt with "normal" focusing (as opposed to magnetic) can do anywheres near 1600 lines of H. Res. The number should be somewhere between 900 and 1200.
A Sony G90 FPTV can do 1600 lines. It has 9" CRT's with the magnetic focus system and costs $35,000.
Lee
Joined
·
19,586 Posts
"I put the Accupel Multiburst pattern on the screen for Pioneers and then Toshibas ... The 2 Pixel grouping that indicates the TV's ability to resolve at least 960 lines of h-resolution of out 1920 lines is far more crisp on Pioneer units like the Elite 510 than on a 56H80 unit for instance."
I told Greg that, even though its not actually required to do a calibration, that he would probably be better from a marketing perspective if he'd done a resolution wedge instead of a multi-burst pattern. Really, everyone wants to know what their resolution is (roughly anyway), because its part of the big 'mine is bigger' thing that drives so much of this thing of ours. By providing that measurement, he'd probably sell a lot more HDGs. The multi-burst basically tells you very roughly where you fall, but the problem is that almost all HD capable sets will fall between the 2 and 1 pixel bursts, and interpreting amounts of fuzz in them really is too inaccurate to even bother. A wedge, though still subject to interpretation, would have given a lot better measurement.
------------------
Dean Roddey
The CIDLib C++ Frameworks
[email protected]
www.charmedquark.com
I told Greg that, even though its not actually required to do a calibration, that he would probably be better from a marketing perspective if he'd done a resolution wedge instead of a multi-burst pattern. Really, everyone wants to know what their resolution is (roughly anyway), because its part of the big 'mine is bigger' thing that drives so much of this thing of ours. By providing that measurement, he'd probably sell a lot more HDGs. The multi-burst basically tells you very roughly where you fall, but the problem is that almost all HD capable sets will fall between the 2 and 1 pixel bursts, and interpreting amounts of fuzz in them really is too inaccurate to even bother. A wedge, though still subject to interpretation, would have given a lot better measurement.
------------------
Dean Roddey
The CIDLib C++ Frameworks
[email protected]
www.charmedquark.com
Joined
·
11,139 Posts
Nice to have at least three members here that have actually fed high-resolution test patterns into HDTVs and tried to read them. Many years ago I did this to rank NTSC sets, so I'm familiar with the indistinct point Dean talks about in wedge patterns (black and white wedge lines no longer distinctly separated). With stationary NTSC wedge patterns, we used to get several opinions and found a consensus wasn't too difficult.
Can someone elaborate a bit on what the problem is with Accupel's multibursts? You'd think, if you had enough multiburst resolution variations (like the DIY patterns dkeller_NC describes), that it would be fairly easy to tell when lines in one grouping are no longer resolveable. What is the one- and two-pixel pattern, and why would it only show a resolution of 960 lines?
Intrigued by Michael's Accupel tests of popular RPTVs since, over a year ago, I was impressed by Toshiba's 1600-line claim. Still not clear what he measured, though. (Wish ISF folks with cameras and pattern generators would snap screens and publish results online after they've fine tuned all the models they handle.) Anyway, his results seem to confirm some remarkable Toshiba test results forum-member Abdul Jalib provided a while back at hometheaterforum.com. He'd contacted member LB here, who expanded the list of model specs he'd provided in his 12/5 post for the thread listed by JohnDG above. The low Toshiba number may confirm Michael's tests. -- John
------------------
STOP DVI/HDCP AND DFAST
[This message has been edited by John Mason (edited 05-29-2001).]
Can someone elaborate a bit on what the problem is with Accupel's multibursts? You'd think, if you had enough multiburst resolution variations (like the DIY patterns dkeller_NC describes), that it would be fairly easy to tell when lines in one grouping are no longer resolveable. What is the one- and two-pixel pattern, and why would it only show a resolution of 960 lines?
Intrigued by Michael's Accupel tests of popular RPTVs since, over a year ago, I was impressed by Toshiba's 1600-line claim. Still not clear what he measured, though. (Wish ISF folks with cameras and pattern generators would snap screens and publish results online after they've fine tuned all the models they handle.) Anyway, his results seem to confirm some remarkable Toshiba test results forum-member Abdul Jalib provided a while back at hometheaterforum.com. He'd contacted member LB here, who expanded the list of model specs he'd provided in his 12/5 post for the thread listed by JohnDG above. The low Toshiba number may confirm Michael's tests. -- John
------------------
STOP DVI/HDCP AND DFAST
[This message has been edited by John Mason (edited 05-29-2001).]
Joined
·
6,723 Posts
Greetings
The multiburst pattern via the Accupel shows a series of black and white lines. It starts out with a 5 pixel group ... then 4, then 3, then 2, then 1.
IF the TV can display a 1 pixel group ... it can display 1920 pixels aka lines. 1 pixel of white next to one pixel of black and so on.
Two pixels of white and then black and then white and then black by default is 1920/2 = 960 lines.
The Toshiba 34" tube set for instance ... cannot display the 2 pixel group either. Therefore it does less than 960 lines, but it does display the 3 pixel group (1920/3) so it does more than 640 lines. The TOshiba spec for the set is in the range of 800 lines so that is about right.
What I have found on the ... 40X81 Toshiba units is that the 2 pixel group is only faintly visible ... while much more visible on a 65" TOshiba.
The big thing here is ... these rptv sets all seem to do 1 million pixels so that is fine. Now can one tell if they are watching an image with 1.1 million or 1.2 million pixels? I doubt the difference is that obvious.
Regards
------------------
Michael @ The Laser Video Experience
The multiburst pattern via the Accupel shows a series of black and white lines. It starts out with a 5 pixel group ... then 4, then 3, then 2, then 1.
IF the TV can display a 1 pixel group ... it can display 1920 pixels aka lines. 1 pixel of white next to one pixel of black and so on.
Two pixels of white and then black and then white and then black by default is 1920/2 = 960 lines.
The Toshiba 34" tube set for instance ... cannot display the 2 pixel group either. Therefore it does less than 960 lines, but it does display the 3 pixel group (1920/3) so it does more than 640 lines. The TOshiba spec for the set is in the range of 800 lines so that is about right.
What I have found on the ... 40X81 Toshiba units is that the 2 pixel group is only faintly visible ... while much more visible on a 65" TOshiba.
The big thing here is ... these rptv sets all seem to do 1 million pixels so that is fine. Now can one tell if they are watching an image with 1.1 million or 1.2 million pixels? I doubt the difference is that obvious.
Regards
------------------
Michael @ The Laser Video Experience
At this time there are no Direct tube TV's in the consumer market that are actually capable of displaying the full glory of HDTV. For that matter there are very few RPTV that can perform this as well.
The CE industry has not yet figured out how to manufacture a HD direct set without compromising brightness and even more important, keeping the cost down.
As for RPTV. For the most part, if it does not have 9 inch guns then they too come up short in displaying the full resolution of HD.
This may be part of the reason why so many manufacturers are hesitant to give out the true numbers of resolution to their HD capable sets.
The CE industry has not yet figured out how to manufacture a HD direct set without compromising brightness and even more important, keeping the cost down.
As for RPTV. For the most part, if it does not have 9 inch guns then they too come up short in displaying the full resolution of HD.
This may be part of the reason why so many manufacturers are hesitant to give out the true numbers of resolution to their HD capable sets.
Good suggestion for getting a rough feel for the resolution by displaying a test image off of a PC. I've used the NTSC images off of video essentials DVD and my monitor looks like its displaying at the 525 horizontal lines the NTSC standard has. (This was the same test pattern that made me cringe when I played it on my old CRT) I've been thinking about getting a second HD set, but really don't like RP. It's good to know that the HD CRT sets are only around 800 lines. Maybe I'll keep dreaming that the stock market will go up and I can get another plasma screen. (They keep getting cheaper. Do I really need a car? or a second kidney?) Many thanks for the enlightening conversation.
Quote:
What is the res of your plasma screen? 1024X768? By what I have seen, you will know the actual resolution, but you will know that is it far less than the HDTV specs of 1280x720P or 1920x1080i. Of course, if I could afford one, I'd sure like a plama display. http://www.avsforum.com/ubb/biggrin.gif
| Originally posted by ResMan: Maybe I'll keep dreaming that the stock market will go up and I can get another plasma screen. |
I've got the pioneer 505-HD plasma. It's a 16:9 aspect ratio and it is 1280x768 pixels. I kept waiting for a plasma that would display 1920x1080, but I decided I wanted to see HD this century. I figure the resolution is at least one of the HD formats, and it's over 1 Mpixels, so it meets my definition of high definition. It really looks great (as long as I ignore that one intermittent green sub-pixel over there on the left.) It does a good job with black levels also, so I don't get that funny "darker blacks become lighter grays" thing that I saw on earlier plasmas. The built in scaler / line quadrupler is nice also. Sorry to go on a bit, but I still get excited when I talk about it. (It's so cool.)
1 - 17 of 17 Posts
- Status
Join the discussion
