![]() |
| If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|||||||
|
|
Thread Tools | Display Modes |
|
#11
|
|||
|
|||
|
On Sat, 2 Nov 2013 09:52:01 -0000, "Brian Gaff"
wrote: Now this was a conversation with a sighted person about a week ago. He pointed out that many TVs like Sony in the analogue days had a vast dynamic range on the picture. The example he gave was of a spotlight momentarily dazzling the camera also dazzled the viewer on a good quality transmission, but he claims, no lcd seems to do more than give a kind of flat medium bright effect on such content, as if there is a clamp fitted to stop this. Is he right? I'd have thought that as long as the backlight was bright enough this effect should still be good. Brian The effect of of "dazzlingly bright" effects within a scene are simply "Simulated" within the much narrower dynamic range of the display (and transmission system). In fact, I've seen this effect put to good use by games software writers in the F1GP racing game with the Monaco GP racing circuit where your driver's eye PoV suffers the effect of coming back into daylight at the end of the tunnel stretch of the track. The only time I've genuinely been dazzled was when I was viewing a colour slide photo of the pit lane at a race track taken on a bright sunny day where I had to allow several seconds for my eyes to adapt to the very much darker scene area of the interiors of the pit garages before being able to pick out details (and, conversely, adjust for the high brightness of exterior scenes when moving my gaze away from the dark interiors). Photographic film (whether it be monochrome or colour - negative or positive) is capable of capturing a contrast ratio of some 4 to 6 orders of magnitude. Reflective phot prints made from such sources, otoh, rarely manage better than a 30 to 1 contrast ratio (about 1 and a half orders of magnitude) and that's for glossy prints - matt prints reduce this even further. I think TV displays can manage a slightly better range (255:1) just over 2 orders of magnitude (assuming 8 bits per primary colour, ignoring gamma correction effect). If the camera adjusts dynamically to optimise for the area of interest in the scene and the area of interest matches that of the viewers, the simulated effect can be quite effective. If it doesn't match viewers expectations, it can become seriously annoying. Fortunately when viewing well made professionally produced programme material (movies, documentaries and so on), most of us will accept that the part of the scene that is framed within the working brightness range of the camera is the part of the scene the director intended us to concentrate upon. When it comes to material such as holiday documentary home movies which have no underlying storyline, the mismatch between what the camera revealed and the viewer's interest will highlight the limited contrast range of the camera. In practice, display contrast ratios in excess of 300:1 tend to be superfluous to requirements due to typical viewing environmental requirements and there being no point in forcing the viewer to sit through a 20 or 30 minute dark adaptation process. Properly photographed scenes will apply all the necessary exposure control that the human eye would have applied when viewed directly, nicely limiting the range to comfortably fit within the constraints of the transmission and display system, leaving the viewer to experience a realistic simulation without their eyes having to make the extreme adjustments to the illumination levels of the original scenes. IOW, we don't normally require very high contrast ratios from the display device to nicely serve our needs when viewing cinematic content. -- Regards, J B Good |
|
#12
|
|||
|
|||
|
Ah, so if you make the blacks black the detail goes and the brights are not
so bright, and if you make it bright, then nothing is quite black so greay detail is lost. Both symptoms of low dynamic range then I suppose. Is this due to the back light leaking through the black opaque lcd pixels then? Brian -- From the Sofa of Brian Gaff Reply address is active "Stephen Wolstenholme" wrote in message ... On Sat, 2 Nov 2013 09:52:01 -0000, "Brian Gaff" wrote: Now this was a conversation with a sighted person about a week ago. He pointed out that many TVs like Sony in the analogue days had a vast dynamic range on the picture. The example he gave was of a spotlight momentarily dazzling the camera also dazzled the viewer on a good quality transmission, but he claims, no lcd seems to do more than give a kind of flat medium bright effect on such content, as if there is a clamp fitted to stop this. Is he right? I'd have thought that as long as the backlight was bright enough this effect should still be good. Brian The only problem I notice is at the dark end of the range. The contrast is too low to see details. The bright end seems much the same as it was with analog TV. Steve -- EasyNN-plus. Neural Networks plus. http://www.easynn.com SwingNN. Forecast with Neural Networks. http://www.swingnn.com JustNN. Just Neural Networks. http://www.justnn.com |
|
#13
|
|||
|
|||
|
Yes, but is this just fixed on TVs these days. many moons ago when colour
was just about to start, where I worked imported and converted a lot of tvs, and some of these in addition to saturation, had tint and gamma controls. Of courese this meant you could get some very unrealistic displays indeed, but on the other hand it would be loved no doubt by know twiddler types these days. Tint was good fun though, Brian -- From the Sofa of Brian Gaff Reply address is active "Bill Wright" wrote in message ... Stephen Wolstenholme wrote: The only problem I notice is at the dark end of the range. The contrast is too low to see details. The bright end seems much the same as it was with analog TV. Adjust the gamma? Bill |
|
#14
|
|||
|
|||
|
A few points, yes, the 625 system we used had negative going video, however
I suspect that some companies did use shall we say, creative means to get the effect I mentioned. I can recall for example modifying a black an white tv so it had dc clamping and a slightly non linear law at extreme high white, and this really did look impressive on things like top of the pops, but it needed very well regualated eht etc to not balloon when this happened! Those were fun times. I suspect similar 'enhancements' were used to make pictures look better on certain content on colour sets, which was probably why it was so hard on some to get the colour to look right all the time. I also remember the early Philips VCRs having an issue with vertical sine wave bars on saturated colours too, which was quite disconcerting at times. Brian -- From the Sofa of Brian Gaff Reply address is active "David Woolley" wrote in message ... On 02/11/13 09:52, Brian Gaff wrote: Now this was a conversation with a sighted person about a week ago. He pointed out that many TVs like Sony in the analogue days had a vast dynamic range on the picture. The example he gave was of a spotlight momentarily If carefully set up, an analogue TV, with a CRT display, is capable of going to near perfect black. In practice, they would either be set up so that maximum black was still slightly lit, or they would be set up so that they went black whilst the image was still grey. (Any emissive display can go to black, which was the big advantage of plasma displays.) dazzling the camera also dazzled the viewer on a good quality transmission, That's not possible as the transmitted signal only had a limited range between peak white and black level, and, except for night time scenes, one would expect the full range to be used. but he claims, no lcd seems to do more than give a kind of flat medium bright effect on such content, as if there is a clamp fitted to stop this. Dynamic range on LCDs is limited by the inability to to make the LCD cell completely opaque, i.e. it limits how black blacks can be and doesn't affect the white end limit. There are also issues to do with quantisation errors in the digital signal. Although the signal can represent pure black, the next level up is not infinitesimally lighter. Is he right? I'd have thought that as long as the backlight was bright enough this effect should still be good. Digital signals have a maximum possible whiteness, just like real world analogue ones. Also, one would generally want to use the full dynamic range of the LCD cells for all normal pictures, to minimise the consequences of poor contrast ratio due to blacks being grey. Note that the ability of real world analogue receivers to provide a high contrast ratio was affected by the quality of the DC restoration. Black and white sets generally had none, so low contrast night scenes would come out mid-grey, rather than almost black. Also, if I remember correctly, UHF analogue signals were actually transmitted with maximum white being zero modulation, so it was impossible to transmit a whiter than white colour, even if you temporarily exceeded the maximum transmitter power. (I believe one of the advantage of LED backlights is that they can be automatically turned down in night scenes, allowing the full range of the panel to continue to be used.) |
|
#15
|
|||
|
|||
|
"Brian Gaff" wrote in message
... Yes, but is this just fixed on TVs these days. many moons ago when colour was just about to start, where I worked imported and converted a lot of tvs, and some of these in addition to saturation, had tint and gamma controls. Of courese this meant you could get some very unrealistic displays indeed, but on the other hand it would be loved no doubt by know twiddler types these days. Tint was good fun though, Tint control was not normally necessary on European (PAL) sets because the PAL process automatically corrects for phase shifts in the colour signal during the broadcast chain (which manifest themselves as changes of hue) whereas the older NTSC system doesn't have this inbuilt automatic correction and so manual correction at the receiver may be necessary. The only exception to this is some early sets from Japan (eg Hitachi) which circumvented a patent by using an inferior decoder at the TV which didn't apply such good correction. I remember even in the early 70s (when I'd have been about 7 or 8) knowing about the advantage of PAL over NTSC (though not knowing the technicalities of how how it achieved this superiority - I wasn't that much of a child geek!) and hence being puzzled that my friend's parents' TV (a Hitachi) had a tint control whereas ours and other friends' sets didn't have one. https://en.wikipedia.org/wiki/PAL - see the paragraph "In the early 1970s some Japanese set manufacturers..." |
|
#16
|
|||
|
|||
|
On Sat, 2 Nov 2013 18:00:20 -0000, "Brian Gaff"
wrote: Yes, but is this just fixed on TVs these days. many moons ago when colour was just about to start, where I worked imported and converted a lot of tvs, and some of these in addition to saturation, had tint and gamma controls. Of courese this meant you could get some very unrealistic displays indeed, but on the other hand it would be loved no doubt by know twiddler types these days. Tint was good fun though, Brian 40 odd years ago I had an argument with a Canadian. His position was that UK colour televisions were crap because they didn't have the "hue" control he was used to back home. -- Graham. %Profound_observation% |
|
#17
|
|||
|
|||
|
On Sat, 2 Nov 2013 19:06:25 -0000, "NY" wrote:
"Brian Gaff" wrote in message ... Yes, but is this just fixed on TVs these days. many moons ago when colour was just about to start, where I worked imported and converted a lot of tvs, and some of these in addition to saturation, had tint and gamma controls. Of courese this meant you could get some very unrealistic displays indeed, but on the other hand it would be loved no doubt by know twiddler types these days. Tint was good fun though, Tint control was not normally necessary on European (PAL) sets because the PAL process automatically corrects for phase shifts in the colour signal during the broadcast chain (which manifest themselves as changes of hue) whereas the older NTSC system doesn't have this inbuilt automatic correction and so manual correction at the receiver may be necessary. The only exception to this is some early sets from Japan (eg Hitachi) which circumvented a patent by using an inferior decoder at the TV which didn't apply such good correction. I remember even in the early 70s (when I'd have been about 7 or 8) knowing about the advantage of PAL over NTSC (though not knowing the technicalities of how how it achieved this superiority - I wasn't that much of a child geek!) and hence being puzzled that my friend's parents' TV (a Hitachi) had a tint control whereas ours and other friends' sets didn't have one. https://en.wikipedia.org/wiki/PAL - see the paragraph "In the early 1970s some Japanese set manufacturers..." "Simple PAL" also did away with the need for an expensive glass chroma delay line in the decoder. I only saw (and worked on) one model like this, the "Granada Colourette" or to give it its real name the "Kuba Porta-Color" (actually made by General Electrics). Horrible thing! The control in question may well have been labeled "Tint" by the manufacturer, but in my opinion it was far more akin to the "Hue" control on an NTSC set. True "Tint" control just lets the viewer adjust the basic black and white picture slightly usually through the red/blue axis for personal preference of a cool vs.warm picture, and to correct grayscale errors as the tube ages, but tough if green/magenta adjustment is needed. -- Graham. %Profound_observation% |
|
#18
|
|||
|
|||
|
On Sat, 02 Nov 2013 11:06:51 +0000, David Woolley
wrote: On 02/11/13 09:52, Brian Gaff wrote: Now this was a conversation with a sighted person about a week ago. He pointed out that many TVs like Sony in the analogue days had a vast dynamic range on the picture. The example he gave was of a spotlight momentarily If carefully set up, an analogue TV, with a CRT display, is capable of going to near perfect black. In practice, they would either be set up so that maximum black was still slightly lit, or they would be set up so that they went black whilst the image was still grey. (Any emissive display can go to black, which was the big advantage of plasma displays.) dazzling the camera also dazzled the viewer on a good quality transmission, That's not possible as the transmitted signal only had a limited range between peak white and black level, and, except for night time scenes, one would expect the full range to be used. but he claims, no lcd seems to do more than give a kind of flat medium bright effect on such content, as if there is a clamp fitted to stop this. Dynamic range on LCDs is limited by the inability to to make the LCD cell completely opaque, i.e. it limits how black blacks can be and doesn't affect the white end limit. There are also issues to do with quantisation errors in the digital signal. Although the signal can represent pure black, the next level up is not infinitesimally lighter. Is he right? I'd have thought that as long as the backlight was bright enough this effect should still be good. Digital signals have a maximum possible whiteness, just like real world analogue ones. Also, one would generally want to use the full dynamic range of the LCD cells for all normal pictures, to minimise the consequences of poor contrast ratio due to blacks being grey. Note that the ability of real world analogue receivers to provide a high contrast ratio was affected by the quality of the DC restoration. Black and white sets generally had none, so low contrast night scenes would come out mid-grey, rather than almost black. Also, if I remember correctly, UHF analogue signals were actually transmitted with maximum white being zero modulation, so it was impossible to transmit a whiter than white colour, even if you temporarily exceeded the maximum transmitter power. (I believe one of the advantage of LED backlights is that they can be automatically turned down in night scenes, allowing the full range of the panel to continue to be used.) It might be helpful to Brian to pint out that outdoor LED display screens have come a long way, and can produce adequate resolution at the intended viewing distances, and enough brightness to give direct sunlight a run for its money. Eidophor Projectors anyone? -- Graham. %Profound_observation% |
|
#19
|
|||
|
|||
|
On 02/11/2013 17:56, Brian Gaff wrote:
Ah, so if you make the blacks black the detail goes and the brights are not so bright, and if you make it bright, then nothing is quite black so greay detail is lost. Both symptoms of low dynamic range then I suppose. Is this due to the back light leaking through the black opaque lcd pixels then? Brian indeed. that's why some sets have an array of leds behind the screen and dim some of them in an attempt to portray blacks as black. that's why oled is promising - with those, black really is black as you just turn off the led. -- Gareth. That fly.... Is your magic wand. |
|
#20
|
|||
|
|||
|
NY wrote:
The only exception to this is some early sets from Japan (eg Hitachi) which circumvented a patent by using an inferior decoder at the TV which didn't apply such good correction. The early PAL Japanese sets had the hue control to circumvent the PAL patent. Every other line was stored in a delay line, but instead of, as per the PAL spec, being used to combine and average out the previous and present lines, turning a phase error into a less noticeable saturation change, it was simply used to repeat the line again. So any phase errors would have ended up with the wrong hue, (just like NTSC) hence the hue control to manually adjust for 'best results'! -- Mark Please replace invalid and invalid with gmx and net to reply. |
| Thread Tools | |
| Display Modes | |
|
|
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| dynamic range? | les | UK home cinema | 1 | August 31st 08 10:50 AM |
| dynamic contrast ratio | Terry Smith | High definition TV | 3 | July 5th 06 10:56 PM |
| Dynamic Multipath | Bob Miller | High definition TV | 26 | January 2nd 06 06:06 PM |
| Dynamic range of DLP? | Pete Fraser | High definition TV | 14 | December 5th 03 04:31 AM |