Hi all,

I am disappointed with glossy lcd screens, LD LED lcds, motionflow type 120hz motion enhancements that give me 3D headache. I am not too keen on flicker (don't want plasma) and I love static lcd images, the detail, the pop, etc. I read that Samsung wants to sell 240hz lcds in 2012? I only hope the screen will be matte,, no LD LED zones, black color rendition as good as on sony professional broadcast lcd monitor, or some kind of kuro lcd. What remains to be fixed is motion handling, so since I don't believe 240hz is that much of an improvement can they make over 2,000 Hz lcd( yes two thousand Hz) or higher, whatever it takes for perfect motion? What I want is an LCD picture quality as good in motion as it is when static (no cartoon walk, no gloss) I've allowed myself to start this thread in order to finally get the answers about LCD motion handling. Thank you very much in advance for any constructive info.

http://www.watch.impress.co.jp/av/docs/20070111/dg79_05.jpg (http://www.avsforum.com/avs-vb/showthread.php?p=7714953#post7714953)


If you want all those things, then you are an OLED man, dude!

Hang tight, the OLED Saviour is approaching with speed, his mighty footsteps rocking the land, shaking and breaking those nasty LCDs and PDPs!

With perfect blacks his mighty attacks
will roll PDP off its tracks!

And super speed to take the lead
on LCD's market-feed!

Thin is in and OLED wins
nobody thicker could be slicker!

No blur and no fur
on this mighty beast!

All your frames are his trivial games
he juggles them with ease in a microsecond sneeze
while the LCD sleaze does wheeze and PDP's chromatic freeze!

OLED TVs: Technology Advancements Thread (http://www.avsforum.com/avs-vb/showthread.php?p=7714953#post7714953)

http://www.watch.impress.co.jp/av/docs/20070111/dg79_05.jpg (http://www.avsforum.com/avs-vb/showthread.php?p=7714953#post7714953)


If you want all those things, then you are an OLED man, dude!

Hang tight, the OLED Saviour is approaching with speed, his mighty footsteps rocking the land, shaking and breaking those nasty LCDs and PDPs!

With perfect blacks his mighty attacks
will roll PDP off its tracks!

And super speed to take the lead
on LCD's market-feed!

Thin is in and OLED wins
nobody thicker could be slicker!

No blur and no fur
on this mighty beast!

All your frames are his trivial games
he juggles them with ease in a microsecond sneeze
while the LCD sleaze does wheeze and PDP's chromatic freeze!

OLED TVs: Technology Advancements Thread (http://www.avsforum.com/avs-vb/showthread.php?p=7714953#post7714953)

I don't know about approaching with speed. But I wish it would hurry up.

By the time it gets here in sizes appropriate for HT, LCD and plasma will be much much better

Think about it,
LCD will be in great shape, and the 10G Kuro with absolute black and almost as thin as OLED will probably be out.

Sounds to me that the Op has no choice but HD CRT TUBE Plasma then last LCD, in perspective to flat screen and the tube as best PQ. then oled (hopefully) soon will mass market but dont expect a 50" for 2k as soon as 2015

get a plasma its your best call

Sounds to me that the Op has no choice but HD CRT TUBE Plasma then last LCD, in perspective to flat screen and the tube as best PQ. then oled (hopefully) soon will mass market but dont expect a 50" for 2k as soon as 2015

get a plasma its your best callExactly. I'll just add that by 2015 it's going to be a different flat panel world out there. LCD and PDP might be on their way out then while OLED will probably share the fate of SED.

Why would you need 2kHz refresh when the source is still 30Hz or less? What am i missing?

Why would you need 2kHz refresh when the source is still 30Hz or less? What am i missing?

i agree.
2400hz?? my goodness. maybe he meant 240hz?

even with 240hz lcd still wont dice, they need instant on / off like plasma and its not possible because of the way the technology is. its just not ment to be Natrually fast response, add ons dont do do the job 120hz interpolation and adding a frame is not the right way.

lcd in general is a failed technology for TV viewing and compared to plasma and the only reason it still exisit is because

of idiots who belive lcd is better

and low cost to make lcd more revenues and back to idiots...

a very very small group that have really no control over lighting READ

maybe glass windows from over 5ft in glass or glass house, thats a diffrent story

but even then why not put blids or the tv behind the glass?

and pretty soon you might scratch that off as most companies are going glossy screen to get higher PQ ' more blacks' etc etc

also Led is another issue, it will cause Uneven wear like plasma so thats another thing lcd trolls wont be able to use....

yes i said it correct LED lcd will have uneven wear ie burn in ~ burn in can mean its wear is diffrent then the other area of tv , black bars side bars etc etc..

of idiots who belive lcd is better

unbelivable

Without getting into a argument, give me a good reason(s) why lcd is better tv expereince for watching movies and games and regular programing then a plasma would be :)

unbelivable

I never said plasma wasn't a better experience.

Well i wrote "idiots who think lcd is better ~ then plasma, then you replied saying "unbeliable" and i have a feeling is because you dont fellow my statement correct? if so then my question still remains why do you think of my statement and or question about lcd ? because i clearly wrote on my post for tv viewing expereince

gus, FWIW your posts are tiresome. Hell I actually agree with your general stance on things, but your responses to every single thread read the same way, and you're not adding anything useful to the discussion. The complete lack of punctuation, spelling, and grammar don't help either (I apologize in advance if there's some explanation for that).

This thread was asking about how super high refresh rates could help LCDs work better. You don't really know anything about how LCDs work, or what's in the pipeline, or how manufacturers might take advantage of these things to comment intelligently on it. Repeating the same old generalities doesn't add anything to the conversation. Your posts in the thread about the professional Sony LCD monitors were the same - you sounded like the guy who says "but it's impossible for them to play" during a barbershop discussion about whether Wilt could take on Shaq.

As for LCD buyers being idiots, well the fact is what people like is a personal preference thing. If Joe Blow likes the way an LCD looks and it fits his needs, who are you to call him an idiot? That's like saying that someone who likes mushrooms on their pizza is an idiot. Different strokes for different folks.

jeff

jeff yeah i did get off topic and i guess your post makes sense sorry for my miss spelling :) although i just wanted to explain the issues with lcd because i just wanted to let the op know all the facts, and how lcd cannot fix its motion speed even with 240hz and lol that led to the rest ....

Why Not

There are myriad reasons why not, firstly, motion artifacts. Secondly, it would take a beast of a processor to create 80 interpolated frames(30 fps * 80 = 2,400hz), I would be suprised if it was even possible to do this in real time with today's technology. Lastly, even if you could get all that working, response times are no where near fast enough to display 2,400hz content, considering how you need a bare minimum of 4 milliseconds to display 240hz content, you would need to increase speeds from current top of the line LCDs 10 times in order to accomplish this.

No worries, I just think boards work best when people stay on topic and stick to what they know best. In my case, that means not talking too much about LCD technology :D Don't take me too seriously though.

But I do believe that faster screen refresh rates and backlight tricks can improve certain aspects of LCD performance, hence the OPs question. If I'm not mistaken the new 950 series Samsung LED backlight models do some sort of back light flickering that apparently really improves the motion blur. Or at least I think that's what the CNET review said.

I also believe there was news about Sony and Samsung coming out with 240 Hz displays? More techno-trickery is coming.

jeff

even with 240hz lcd still wont dice, they need instant on / off like plasma and its not possible because of the way the technology is. its just not ment to be Natrually fast response, add ons dont do do the job 120hz interpolation and adding a frame is not the right way.

lcd in general is a failed technology for TV viewing and compared to plasma and the only reason it still exisit is because

of idiots who belive lcd is better

and low cost to make lcd more revenues and back to idiots...

a very very small group that have really no control over lighting READ

maybe glass windows from over 5ft in glass or glass house, thats a diffrent story

but even then why not put blids or the tv behind the glass?

and pretty soon you might scratch that off as most companies are going glossy screen to get higher PQ ' more blacks' etc etc

also Led is another issue, it will cause Uneven wear like plasma so thats another thing lcd trolls wont be able to use....

yes i said it correct LED lcd will have uneven wear ie burn in ~ burn in can mean its wear is diffrent then the other area of tv , black bars side bars etc etc..

Wow! You have a real fundamental misunderstanding of many aspects of display technology.

LCD is failed technolgy? For better or worse, it's dominating the market share and shows no signs of slowing.

Segmented LED backlights will not 'burn-in' like any phosphor based display (CRT, Plasma, OLED) can. (LEDs do age and fail, but not like phosphors.)

LCDs can be instant-on; ever heard of ferro-electric liquid crystal modes? Probably not... These are bi-stable, like a DLP. There is a huge ammount of R&D going into LC materials research; they will get better.

Plasma and LCDs both have their place. You're obviously a plasma fan-boy, but please stop posting mis-information you don't understand.

even if possible with the hz the speed is the issue with lcd, fix the speed of the ms and theirs no need for more hz ...

There are myriad reasons why not, firstly, motion artifacts. Secondly, it would take a beast of a processor to create 80 interpolated frames(30 fps * 80 = 2,400hz), I would be suprised if it was even possible to do this in real time with today's technology. Lastly, even if you could get all that working, response times are no where near fast enough to display 2,400hz content, considering how you need a bare minimum of 4 milliseconds to display 240hz content, you would need to increase speeds from current top of the line LCDs 10 times in order to accomplish this.


it is a failed technology if you were to asked your self what provides a better tv expereince, but as jeff said everyone has diffrent taste so YMMV
its taking over again part of it is missinformed people for the most part and lower cost as i said earlier,

the led based system will have uneven wear and burn in is considerd part uneven wear but ok lets retract burn in., pendragoon help with this explination.

if lcd were instant on wouldnt that solve motion blur? why not use this on cosumer displays? (iv heard of the high end sony bvpm or whatever) those dont have a problem compared to normal lcd in terms of motion blurr


how is it misinfomration/missleading? yes i like plasma but i still try to stick with facts.... lcd in general are not ideal to watch tv expereince comapred to plasma (flaws/issues wise)


Wow! You have a real fundamental misunderstanding of many aspects of display technology.

LCD is failed technolgy? For better or worse, it's dominating the market share and shows no signs of slowing.

Segmented LED backlights will not 'burn-in' like any phosphor based display (CRT, Plasma, OLED) can. (LEDs do age and fail, but not like phosphors.)

LCDs can be instant-on; ever heard of ferro-electric liquid crystal modes? Probably not... These are bi-stable, like a DLP. There is a huge ammount of R&D going into LC materials research; they will get better.

Plasma and LCDs both have their place. You're obviously a plasma fan-boy, but please stop posting mis-information you don't understand.

even if possible with the hz the speed is the issue with lcd, fix the speed of the ms and theirs no need for more hz ...

70% of motion blur in LCDs come from the SAH effect, only 30% comes from the lack of response time. Extremely high hz lowers the hold time reducing blur.

how is it misinfomration/missleading?

You said:

"they need instant on / off like plasma and its not possible because of the way the technology is."

Ferro-Electric Liquid Crystals have essentially "instant" response times (although they're bi-stable and must be temporally dithered) and Nematic mode LC response time has been improving greatly. Therefore, your statement is just plain worng.

"lcd in general is a failed technology for TV viewing"

As I said LCD has the lead in market share, so this is just plain wrong since most TV viewers choose LCD. Is it the best? no.. But a failure? hardly.

"also Led is another issue, it will cause Uneven wear like plasma"

Again, you are wrong. LEDs do not age like phosphors.

yes i like plasma but i still try to stick with facts...

lcd in general are not ideal to watch tv expereince comapred to plasma

Your opinion, not fact.

Just to set the record straight, I'm not an LCD fan boy, the only panel I have is on my lap-top. I have a direct-view CRT on my work computer and an SXRD projector in my Home theater. At work, we've designed display systems which use CRT, DLP, AM-LCD, Ferro-Electric LCD and LED and OLED display technologies (generally micro-displays, so obviously no plasma). All current display technologies have their strengths and weaknesses and I'm glad I have to ability to choose between them as the requirements of the application demand.

If you want all those things, then you are an OLED man, dude!

No blur and no fur
on this mighty beast!


OLED TVs: Technology Advancements Thread (http://www.avsforum.com/avs-vb/showthread.php?p=7714953#post7714953)


Exaggerating a bit are we ?

70% of motion blur in LCDs come from the SAH effect, only 30% comes from the lack of response time.

---------------------------------------------------------------

I don't believe 240hz is that much of an improvement can they make over 2,000 Hz lcd( yes two thousand Hz) or higher, whatever it takes for perfect motion?


Hertz by themselves don't reduce blur. They will increase blur in fact because after a point, the eye percieves high refresh intervals as sample-and-hold again (myself and others have tested this and it follows theory). Hz are only a meduim for the true effector, higher temporal frame rates. Frame rates are really what people are asking for, not higher hz. Knowing this can completely change the way you go about problem solving.
For instance all you really need to ask is what's the optimum frame rate - 75fps avoids flicker (though Oled/LCD will need BFI/scanning) so let's call that the goal . The best way to achieve it? Would it be so hard for broadcasters/movie studios to simply raise their standard by 15-20fps - apparently so.

I myself find it very interesting that people sometimes buy products in general with very little knowledge as to which is best. Lcd and plasma tech are ok but I see a trend lately which is one of the reasons i still watch a crt tube tv today and that problem is that every year vid manufactures try to fix the root causes by fixing the displays when actually the displays are much better than the signal fed to them. I am disappointed with hdtv standard when they developed it, they should have made it manditory for at least have original signal source recorded at 60 fps minimum at 1080p instead of haveing many standards for ATSC. Our source material is the main true culprit for our display problems not the displays themselves,manufactures just put band-aids on it!!!! WE here should lobby for higher quailty source materials and i bet most if not all major problems would be solved!!! thx

steve i'm still waiting for your response:) dont have any?

Without getting into a argument, give me a good reason(s) why lcd is better tv expereince for watching movies and games and regular programing then a plasma would be :)

Also led based can get uneven wear

Since this is the digital age we should have all film and broadcasters switch to an all and fully encompassing digital system simialar to the DCI system where source material is always recorded at or higher levels of std then what is available to consumers. I do not understand even today why develop HDTV (ATSC) std when film is still used today at a pathetic 24fps a 100 year old std in my opinion and I think the movie and broadcaster industry should do a complete radical update in using better quality tech which is available today cheers!!!

steve i'm still waiting for your response:) dont have any?

My response to what ? What is your question ?

Our source material is the main true culprit for our display problems not the displays themselves,manufactures just put band-aids on it!!!! WE here should lobby for higher quailty source materials and i bet most if not all major problems would be solved!!! thx

Yes!! the HD standard was crap as far as that was concerned. And band aid is a good word for what is happening now. Source frame rates are the real problem.
I'll be the first one in line :)

My response to what ? What is your question ?

Without getting into a argument, give me a good reason(s) why lcd is better tv expereince for watching movies and games and regular programing then a plasma would be :)

Without getting into a argument, give me a good reason(s) why lcd is better tv expereince for watching movies and games and regular programing then a plasma would be :)

I never made the claim that LCD was better for watching movies and games. I remarked because I couldn't believe that you were calling people names for owning a certain type of television. It's only a TV, relax.

I'm relaxed but what im trying to say is that lcd is in general worse then plasma for pretty much all tv expereince and i explianed along with the line of saying idiots and you responded so now im asking do you think lcd sucks in general? if so then you would have to agree to the points i mentioned why lcd is failed tech (of course keep the dead horse going and milking us for inferior technology since its not going anywhere)

Hertz by themselves don't reduce blur. They will increase blur in fact because after a point, the eye percieves high refresh intervals as sample-and-hold again (myself and others have tested this and it follows theory). Hz are only a meduim for the true effector, higher temporal frame rates. Frame rates are really what people are asking for, not higher hz. Knowing this can completely change the way you go about problem solving.
For instance all you really need to ask is what's the optimum frame rate - 75fps avoids flicker and blur (though Oled/LCD will need BFI/scanning) so let's call that the goal . The best way to achieve it? Would it be so hard for broadcasters/movie studios to simply raise their standard by 15-20fps - apparently so.

Thank you borf. You made my day. So if 75fps is the optimum frame rate, if I record at 1080p at 75fps with a video camera and then display it on a 120Hz lcd with BFI, but WITHOUT motion processing ( I think toshiba does straight 120hz, no 'motionflows') would you say I could then experience the best motion handling picture ever on an LCD. Or do the lcd tv's need to be further modified ( maybe 60 Hz + backlight scanning would be better, I remember a philips lcd with ClearLCD technology; it used scanning but I can't remember if at 60 or 120hz)?

And now a crazy idea! If I generated my own electricity with a generator and special converters to achieve 75 Hz electricity and have my lcd tv modified to work at 2* 75hz, that is 150 HZ !!! with BFI, then displaying 75fps recorded material, would I achieve LCD motion as smooth as cream? Does it work out like that???

Segmented LED backlights will not 'burn-in' like any phosphor based display (CRT, Plasma, OLED) can. (LEDs do age and fail, but not like phosphors.)


LED displays using local dim tech can be unevenly worn using this function. See my posts in the thread below, I explain how it can happen quite well.

Here is a short version of my posts.

They actually age faster than most plasma sets (half life of current LED displays is 50k hours (and use phosphor coating), back when plasma lasted that long, uneven wear became a problem in around 3 years of average TV watching. (black bar content) and will start to show uneven wear quite a bit sooner than your average plasma. (which are now up to around 100k hours to half life)

Now Sony is using a RGB LED display, if they aren't phosphor coated then you might see a reduction of uneven wear, but it would still be a factor as all light sources age and dim over time and LED local dim will cause the LED array to age at a different pace.

http://www.avsforum.com/avs-vb/showthread.php?p=14673484#post14673484

LED displays using local dim tech can be unevenly worn using this function. See my posts in the thread below, I explain how it can happen quite well.

http://www.avsforum.com/avs-vb/showthread.php?p=14673484#post14673484

I never said LEDs don't age; I've designed and worked with LED backlights (monochrome, RGB and false-white)since 2001 and have seen my fair share of failures; I just made the point that it's a different failure/degradation mechanisim than emmisive phoshor based displays and that gus's comments are out of line. Of course false white LED (which have a blue diode and yellow phosphor) will suffer luminance and chromaticity degradations, however since the backlight is not the image formation device, any burn-in will be more diffuse and less localized than the pixel-level burn-in that is exhibited on a direct phosphor based display (CRT, Plasma, OLED). How good are modern plasma? well if you take care of them probably fine (time will tell), but I still see quite a few with burned in images (probably driven in torch mode with static imagery, of course). I've also been been integrating OLED microdisplays for nearly ten years and burn-in is still an issue; OLEDs will find their place in the dispaly world, but they're not the flat panel holy grail some people think they are.

How good are modern plasma? well if you take care of them probably fine (time will tell), but I still see quite a few with burned in images (probably driven in torch mode with static imagery, of course).

Modern Plasma sets have pretty much licked standard burn in (takes real abuse these days) but uneven wear is still an issue, and this is what I am talking about, most plasma owners know that uneven wear is potentually a problem, so they take it into consideration, LCD owners think that this is imposible on this type of display not realizing that this local dim tech can unevenly age the backlight.

Say someone has a current LCD (from 05 for example) and 25% of their viewing is 4:3 40% is scope widescreen, and the rest is full screen 16:9 and they decide to upgrade to a LED display. Now odds are they won't change their viewing habbits now in the same time frame, you will start to see some uneven wear occur with these displays. And they will probably end up here wondering why their LCD has "burn-in"

Now only time will tell if this will be a real problem but everything I have seen and read so far is pointing to these early generation LED displays having real issues in a few years when some people will start to notice it on their 81 series LCD displays.

lcd in general is a failed technology for TV viewing and compared to plasma and the only reason it still exisit is because

of idiots who belive lcd is better


And the professional musicians who listen to music on cheap stereos (like my uncle) - are they idiots too? Most people are more interested in the content than their TV's phosphor trails or motion lag.

if I record at 1080p at 75fps with a video camera and then display it on a 120Hz lcd with BFI, but WITHOUT motion processing would you say I could then experience the best motion handling picture ever on an LCD?

I'll try to answer this just in case you plan on going out and building your own display :) Replace BFI with backlight scanning and you have a winner (but watch out for judder). BFI would not work here unless the screen were inactive during most of the 120hz duty cycle - that may contradict the very definition of "duty cycle" i don't know. Maybe someone else can answer that?

As mentioned, in the above scenario you're going to get 3:2 type judder here because of the hz / fps differential. Its optimal to sync these two together (both @ 75hz) to avoid this.

If I generated my own electricity with a generator and special converters to achieve 75 Hz electricity and have my lcd tv modified to work at 2* 75hz, that is 150 HZ !!! with BFI, then displaying 75fps recorded material, would I achieve LCD motion as smooth as cream? Does it work out like that???

I'd give same answer here.

Imo I think the goal should be a temporal frame rate somewhere around 75fps, displayed at at the same rate, with an appropriately long "down time" (similar to CRT) and you have perfect motion (close enough anyway as i can vouch for my Crt).

How good are modern plasma? well if you take care of them probably fine (time will tell), but I still see quite a few with burned in images (probably driven in torch mode with static imagery, of course). Image sticking/retention is commonly mistaken for burn-in on plasma screens. People just can't get seem to grasp the concept of plasma image retention and how it has absolutely nothing to do with phosphor aging.

Imo I think the goal should be a temporal frame rate somewhere around 75fps, displayed at at the same rate, with an appropriately long "down time" (similar to CRT) and you have perfect motion (close enough anyway as i can vouch for my Crt).

72 seems like the logical choice; at least it is a multiple of 24. I'm not exactly sure though, I wonder if the "optimal" frame rate is going to differ from person to person, e.g. what one person perceives as SAH another will perceive as flickering. Wikipedia lists the actual flicker fusion threshold as being 16hz in humans, meaning that 50% of people in trials indicated they saw no flicker at that rate. Obviously that is not practical for TV applications, I'm just citing it to illustrate that there could be a wide range of "optimal" frequencies. I don't really know if it would be worth it to pick a frequency out and say to hell with all the old standards.

I've been giving a lot of thought to duty cycle. I often see it measured in percentage, which, to me, doesn't really seem universal enough as it relates to blur. What is 10% at 30fps(~0.0033 seconds) would be 20% at 60fps. The percentages are different, but the time in seconds is the same. What I'm driving at is that it seems to make more sense to measure it in terms of time or maybe that what really matters to SAH blur is hold time(controlled by the duty cycle in some cases and fps in others) as it is the same in each of the 2 examples above. I am working under the assumption that blur would be equal in those examples, which I'm pretty sure is true, but you could throw everything I just said out the window if that is not the case. To sum it all up, a hold time of 0.0033 seconds would result in the same amount of blur across all display types, regardless of fps or duty cycle percentage, so wouldn't it be more useful to use that as the measuring stick for duty cycle? I suppose that maybe it is, and I just haven't noticed it used in that way, but I would like to see it used more frequently. Cycle seems to indicate a percentage though so what are you gonna do? Can we say duty time?

In any case fall times become increasingly important as we attempt to deal with blur.

I'll try to answer this just in case you plan on going out and building your own display :) Replace BFI with backlight scanning and you have a winner (but watch out for judder). BFI would not work here unless the screen were inactive during most of the 120hz duty cycle - that may contradict the very definition of "duty cycle" i don't know. Maybe someone else can answer that?

As mentioned, in the above scenario you're going to get 3:2 type judder here because of the hz / fps differential. Its optimal to sync these two together (both @ 75hz) to avoid this.


Thank you borf. Yes, I want to have my own lcd tv built for me, I just need to know what specification will bring me the desired results, that is no motion blur , no sample and hold and little or no judder and as close to a crt as possible.

So far, let's say what blazerqb11 proposes, that 72fps is the optimum frame rate (it's good for Blu-ray movies at 24p to get rid of judder as 3x24p=72hz). Then I go for backlight scanning (let's forget about 120hz and BFI).
Then I generate my own electricity @72Hz and my LCD TV is modified to be a 72Hz lcd tv with backlight scanning. If I make a home movie recorded at 72fps in high definition and then play it on my 72hz tv I won't get any judder because ther will be NO hz/fps differential, yeah baby! And Blu-ray moveis at 24p will have no judder, either! Can you see any flaws in this plan? Thank you in advance

72 seems like the logical choice; at least it is a multiple of 24. I'm not exactly sure though, I wonder if the "optimal" frame rate is going to differ from person to person


I agree, 72hz is the most practical way to go if not ideal. I just mention ~ 75 as a minimum to shoot for because it works for me and always read its the best compromise considering various sensitivities and conditions (brightness, peripheral vision ect). Its far from a "magic" number. Plus broadcasters/film makers want to go with the least expensive / complex. Though i've heard some say they could see flicker well above 85hz, you gotta shoot for the majority.

I want to have my own lcd tv built for me, I just need to know what specification will bring me the desired results, that is no motion blur , no sample and hold and little or no judder and as close to a crt as possible.

So far, let's say what blazerqb11 proposes, that 72fps is the optimum frame rate (it's good for Blu-ray movies at 24p to get rid of judder as 3x24p=72hz). Then I go for backlight scanning (let's forget about 120hz and BFI).
Then I generate my own electricity @72Hz and my LCD TV is modified to be a 72Hz lcd tv with backlight scanning. If I make a home movie recorded at 72fps in high definition and then play it on my 72hz tv I won't get any judder because ther will be NO hz/fps differential, yeah baby! And Blu-ray moveis at 24p will have no judder, either! Can you see any flaws in this plan? Thank you in advance

Well that blows me away. First off, I don't see any flaws in that plan. I can say from practical experience it doesn't get any smoother / cleaner than 75fps @75hz (or somewhere around there). As far as the blu-ray at 72hz there will be a little judder since you're repeating frames (repeating frames is what causes judder), but this faithfully reproduces the cinema experience (24fps @ 48h-72hz) which many prefer over the "soap" look. Without the hz / fps differential you'd get the soap look in movies.
Secondly, there are more technically minded people on these forums - I can only give first hand experience. But speaking of 72hz there are some plasma and small LCD monitors (and i do believe now LCD TV) that operate at that frequency. They don't accept 72fps input - that might be a place to start i dunno!
I don't know your desired applications but if it doesn't work out, there's always been an open slot in the market for a gaming display over 42" without blur/lag.

Just fyi, the biggest hurdle might be overcoming the brightness reduction that happens when the duty cycle is reduced (since the backlight will be off most of the time the picture will be dimmer).

This refresh rate stuff, reminds me of when I was at a electronics store and to my eyes, the plasmas was flickering.

Its like all the time spent staring at lcds at work/home that now plasmas look like an crt running at 72hz instead of the 85hz's solid image.

As far as the blu-ray at 72hz there will be a little judder since you're repeating frames (repeating frames is what causes judder

If repeating frames is what causes judder wouldn't it be more optimal to control duty cycle instead of repeating frames. It seems to me that duty cycle and flicker would be inversely proportionate(more duty cycle=less flicker e.g. 100% duty cycle=0 flicker) meaning that if you can control the duty cycle why would you want to increase hz? If you can, just display 24fps for the same amount of time it would be displayed at 72hz, in one continuous duration instead of 3 separate ones, and you have equal motion handling and flicker with no judder.

(Phew, wow I think that makes sense now...)

If repeating frames is what causes judder wouldn't it be more optimal to control duty cycle instead of repeating frames. It seems to me that duty cycle and flicker would be inversely proportionate(more duty cycle=less flicker e.g. 100% duty cycle=0 flicker) meaning that if you can control the duty cycle why would you want to increase hz? If you can, just display 24fps for the same amount of time it would be displayed at 72hz, in one continuous duration instead of 3 separate ones, and you have equal motion handling and flicker with no judder.

(Phew, wow I think that makes sense now...)

Hi,

It sure does a lot of sense to me. Let's hope it's doable, as I'm only beginning to research electronics I can't be the judge of that. Hey guys, borf and blazer thanks for writing.

With all that in mind it also seems that any mention of hz as it relates to the flicker fusion threshold is virtually meaningless with out a duty cycle, or duty time figure. Nothing will flicker no matter how low the hz is if it has 100% duty cycle, while it seems that anything would flicker, regardless of how high the hz is, if the duty cycle is low enough. That could also be why the cited flicker fusion threshold on wikipedia is 16hz. It seems incredible low, but with a duty cycle of say 75%, it probably wouldn't flicker.

All this seems to be pointing me in the direction of one conclusion and that is that the only time that hz determines flicker is when duty cycle is determined by the physical properties of the display(as is the case with all current technologies I know of). If duty cycle is actively regulated than it is the only determining factor of flicker.

If you can, just display 24fps for the same amount of time it would be displayed at 72hz, in one continuous duration instead of 3 separate ones, and you have equal motion handling and flicker with no judder.



I read that a few times...it would eliminate flicker but not blur because you're talking about holding frames again which will get you sample and hold blur (If you going to sample and hold frames, they need to be displayed at a rate well over 75hz - to at least 240 i hear and they need to be unique frames BTW! ). Its much easier to bypass SAH all together with BFI or scanning.
I think judder might be gone for the most part in that scenario.

I agree about the duty cycle thing. The page below clarified all this stuff for me. Its a good read.

link (BTW!)

I read that a few times...it would eliminate flicker but not blur because you're talking about holding frames again which will get you sample and hold blur (If you going to sample and hold frames, they need to be displayed at a rate well over 75hz - to at least 240 i hear and they need to be unique frames BTW! ). Its much easier to bypass SAH all together with BFI or scanning.
I think judder might be gone for the most part in that scenario.

The page below clarified all this stuff for me. Its a good read.

link (BTW!)

But if the hold time is the same, regardless of whether it is split into 3 separate durations, motion handling is equal, right?

Let me throw out a theory that is developing slowly in my head. If you had very high fps coupled with a set duty cycle percentage, say 15%, couldn't you just about eliminate blur and flicker?

But if the hold time is the same, regardless of whether it is split into 3 separate durations, motion handling is equal, right?

I think you're saying judder should be gone. If so you have fallen victim to the rantings of frothy mouthed would-be cinema gurus proliferating the web with their pie in the sky "magic numbers" - sweet 5:5 @ 120hz and its wet-dream cousin, 3:3 @ 72hz. Funny eh?
Though these symmetrical cadences duplicate the cinema experience and are smoother than 3:2 judder more experienced will tell you they don't eliminate judder - not by a long shot because you still have repeat frames. Movie camera blur can cover some of it up but if the shutter is fast and the source is pristine (which HD camcorders now are capable of) it will stand out like a sore thumb.

Let me throw out a theory that is developing slowly in my head. If you had very high fps coupled with a set duty cycle percentage, say 15%, couldn't you just about eliminate blur and flicker?

Need more info but however you do it,
1 keep it above 72 hz to avoid flicker
2 syncronize fps with hz to avoid judder
3 display each frame (each one unique) less than 1/240s to avoid blur.

72 seems like the logical choice; at least it is a multiple of 24. I'm not exactly sure though, I wonder if the "optimal" frame rate is going to differ from person to person, e.g. what one person perceives as SAH another will perceive as flickering. Wikipedia lists the actual flicker fusion threshold as being 16hz in humans, meaning that 50% of people in trials indicated they saw no flicker at that rate. Obviously that is not practical for TV applications, I'm just citing it to illustrate that there could be a wide range of "optimal" frequencies. I don't really know if it would be worth it to pick a frequency out and say to hell with all the old standards.

I've been giving a lot of thought to duty cycle. I often see it measured in percentage, which, to me, doesn't really seem universal enough as it relates to blur. What is 10% at 30fps(~0.0033 seconds) would be 20% at 60fps. The percentages are different, but the time in seconds is the same. What I'm driving at is that it seems to make more sense to measure it in terms of time or maybe that what really matters to SAH blur is hold time(controlled by the duty cycle in some cases and fps in others) as it is the same in each of the 2 examples above. I am working under the assumption that blur would be equal in those examples, which I'm pretty sure is true, but you could throw everything I just said out the window if that is not the case. To sum it all up, a hold time of 0.0033 seconds would result in the same amount of blur across all display types, regardless of fps or duty cycle percentage, so wouldn't it be more useful to use that as the measuring stick for duty cycle? I suppose that maybe it is, and I just haven't noticed it used in that way, but I would like to see it used more frequently. Cycle seems to indicate a percentage though so what are you gonna do? Can we say duty time?

In any case fall times become increasingly important as we attempt to deal with blur.duty cycle is just one metric that is used. Hold time is an even better metric but not many get the concept. To be even more accurate:

Hold-time
Retinal velocity
Distance from screen
Retinal persistence
Perception

.....Are what determines degree of motion blur from SAH.

Duty Cycle is just a factor that helps determine hold time. This is why interpolated frames with 100% duty cycle still works at reducing blur.

I think you're saying judder should be gone. If so you have fallen victim to the rantings of frothy mouthed would-be cinema gurus proliferating the web with their pie in the sky "magic numbers" - sweet 5:5 @ 120hz and its wet-dream cousin, 3:3 @ 72hz.

I am not suggesting 3:3 or 5:5, I am suggesting 1:1 with an adjustment in duty cycle to compensate for flicker. That way there is no frame repetition and I think that the blur should be the same as long as the hold time is.

duty cycle is just one metric that is used. Hold time is an even better metric but not many get the concept. To be even more accurate:

Hold-time
Retinal velocity
Distance from screen
Retinal persistence
Perception

.....Are what determines degree of motion blur from SAH.

Duty Cycle is just a factor that helps determine hold time. This is why interpolated frames with 100% duty cycle still works at reducing blur.

Thanks! That is what I was driving at, hold time, its just that my thoughts were a little jumbled.

Need more info but however you do it,
1 keep it above 72 hz to avoid flicker
2 syncronize fps with hz to avoid judder
3 display each frame (each one unique) less than 1/240s to avoid blur.

That is kind of where I disagree. It doesn't have to be above 72 hz to avoid flicker, the duty cycle just has to be high enough to avoid flicker. The higher fps is IMHO what would lower hold time and thus reduce blur. Am I missing something?

It doesn't have to be above 72 hz to avoid flicker, the duty cycle just has to be high enough to avoid flicker.


That's perfectly valid if taken by itself. Like you said, hz don't matter if the duty cyle is long.
The only reason you need 72hz is because the duty cycle must be short to avoid blur. So within the context of this theoretically motion-perfect LCD i'v been talking about, you need 72hz (minimum) to avoid flicker.

That's perfectly valid if taken by itself. Like you said, hz don't matter if the duty cyle is long.
The only reason you need 72hz is because the duty cycle must be short to avoid blur. So within the context of this theoretically motion-perfect LCD i'v been talking about, you need 72hz (minimum) to avoid flicker.
The following are my theories:

Firstly, what I was trying to establish with my last post is that the best way to display any given source with a set frame rate, sans interpolation of frames or another method to increase the number of unique frames, is to display each frame for the time necessary to have a high enough duty cycle to avoid flicker in one single duration, if this is possible. This is in opposition to what is the mainstream practice which involves repetition of frames, because current tech is incapable of actively altering duty cycle; it is set in stone, determined only by hz, because of the physical properties of the tech, namely fall time in a CRT. Here is a practical example: I'll continue to use 24fps. In order to display 24fps/hz with a 10% duty cycle you must display each individual frame/hz 0.00416(repeating) seconds. In this case that time is also equal to the hold time. Conversely with 24fps displayed at 72hz each hz must be displayed for 0.00138(repeating) seconds in order to accomplish a 10% duty cycle. This figure, however, does not represent the hold time as each frame is displayed 3 times meaning the hold time is equal to 0.00138+0.00138+0.00138 which is 0.00416(reapeating). Because hold time is equal to the 24fps example motion handling should be the same. Duty cycle is also equal, meaning the screen is lit for the same percentage of time and flicker is also equal.

The reason I went into such meticulous detail is because this also relates to my second point, that being that if duty cycle is actively managed hz has no bearing on flicker and that duty cycle does not have to be "long", it just has to be optimal. Duty cycle, as it relates to flicker, will also be optimal regardless of hz. In this scenario the number of unique frames will determine hold time and thus be the sole factor in determining blur. In LCD this could be accomplished with backlight scanning provided you have a fast enough rise and fall time. Another example: for this scenario let us say that 10% is optimal duty cycle for flicker simply because it will be easy to work with mathematically. As stated previously with 10% duty cycle 24hz's hold time will be 0.00416(repeating). Duty cycle could be either lowered or raised according to preference; lower duty cycle and hold time is lowered meaning motion handling is greater but more flicker will be evident. The most optimal scenario is one that involves optimal duty cycle for flicker with enough unique frames to lower hold time to the desired motion handling level. If the desired hold time is 1/240th of a second, and 10% is optimal duty cycle the number of fps needed is equal to 24(1/24=0.0416*0.1 is equal to 1/240. If flicker is evident at 10% duty cycle more unique frames would be needed, e.g. if 50% duty cycle is required the number of unique frames required to meet the quoted sufficiently low hold time is equal to 120(1/120=0.0083*0.5 = 1/240)

The most optimal scenario is one that involves optimal duty cycle for flicker with enough unique frames to lower hold time to the desired motion handling level.

I agree... need to read the rest more though.

The following are my theories:

Firstly, what I was trying to establish with my last post is that the best way to display any given source with a set frame rate, sans interpolation of frames or another method to increase the number of unique frames, is to display each frame for the time necessary to have a high enough duty cycle to avoid flicker in one single duration, if this is possible. This is in opposition to what is the mainstream practice which involves repetition of frames, because current tech is incapable of actively altering duty cycle; it is set in stone, determined only by hz, because of the physical properties of the tech, namely fall time in a CRT. Here is a practical example: I'll continue to use 24fps. In order to display 24fps/hz with a 10% duty cycle you must display each individual frame/hz 0.00416(repeating) seconds. In this case that time is also equal to the hold time. Conversely with 24fps displayed at 72hz each hz must be displayed for 0.00138(repeating) seconds in order to accomplish a 10% duty cycle. This figure, however, does not represent the hold time as each frame is displayed 3 times meaning the hold time is equal to 0.00138+0.00138+0.00138 which is 0.00416(reapeating). Because hold time is equal to the 24fps example motion handling should be the same. Duty cycle is also equal, meaning the screen is lit for the same percentage of time and flicker is also equal. Not true. The case of 3 repeating frames with a 10% duty cycle gives an "effective" duty cycle much longer than 10%. There is no way they are the same regarding motion handling. I've shown figures on this very topic.

Edit: to be more clear,

(24Hz) 41.6ms frame at 10% duty has a hold time = 4.16ms

(72Hz) 13.8ms frame at 10% duty has a hold time = 1.38ms

If the 72Hz example is repeated 3 times the actual hold time = 4.16ms BUT the "effective" hold time is much longer since you are essentially pulsing the frame 3 times spead out over the 41.6ms

Not true. The case of 3 repeating frames with a 10% duty cycle gives an "effective" duty cycle much longer than 10%. There is no way they are the same regarding motion handling. I've shown figures on this very topic.

Please do explain. If hold time is equal why is motion handling not? Why should it matter if hold time is split over three durations or just one. What do you mean by "effective" duty cycle? Duty cycle is the percentage of time to screen is lit, isn't it?

Edit: to be more clear,

(24Hz) 41.6ms frame at 10% duty has a hold time = 4.16ms

(72Hz) 13.8ms frame at 10% duty has a hold time = 1.38ms

If the 72Hz example is repeated 3 times the actual hold time = 4.16ms BUT the "effective" hold time is much longer since you are essentially pulsing the frame 3 times spead out over the 41.6ms

Ok then, why is effective hold time longer, and that would mean that a single duration would have better motion handling, right?

For your convenience here is what I was thinking:

Here is a practical example: I'll continue to use 24fps. In order to display 24fps/hz with a 10% duty cycle you must display each individual frame/hz 0.00416(repeating) seconds. In this case that time is also equal to the hold time. Conversely with 24fps displayed at 72hz each hz must be displayed for 0.00138(repeating) seconds in order to accomplish a 10% duty cycle. This figure, however, does not represent the hold time as each frame is displayed 3 times meaning the hold time is equal to 0.00138+0.00138+0.00138 which is 0.00416(reapeating). Because hold time is equal to the 24fps example motion handling should be the same. Duty cycle is also equal, meaning the screen is lit for the same percentage of time and flicker is also equal.

No SAH? 240Hz LCD's will show exactly the same SAH blur as 60Hz or 120Hz ones. It doesn't care for an LCD if you draw one frame of 1/60s or 4 frames of 1/240s, the result is exactly the same. Unless motion interpolation takes place, obviously :rolleyes:

SAH is inherent to all LCD's. It's just that 240Hz TV's will be able to do a motion interpolation way much more precise than 120Hz ones, and therefore the result is, may be, more natural looking, and pleasing.

In the same way 120Hz tv's only improve blurring when frame interpolation is active, 240Hz tvs, or even 2400Hz tvs will/would need motion interpolation algorithms to reduce SAH blur.

Please do explain. If hold time is equal why is motion handling not? Why should it matter if hold time is split over three durations or just one. What do you mean by "effective" duty cycle? Duty cycle is the percentage of time to screen is lit, isn't it? Because in the case of 3 seperate pulses you are spreading the information over a much longer period. The effective (percieved) hold time and duty cycle will now be much longer.

Lets say you split it into 10,000 parts each with a 10% duty cycle. Your effective duty cycle and hold time will be ~99% (close to 41.8ms) with a true hold time still = 4.18ms

No SAH? 240Hz LCD's will show exactly the same SAH blur as 60Hz or 120Hz ones. It doesn't care for an LCD if you draw one frame of 1/60s or 4 frames of 1/240s, the result is exactly the same. Unless motion interpolation takes place, obviously :rolleyes:

SAH is inherent to all LCD's. It's just that 240Hz TV's will be able to do a motion interpolation way much more precise than 120Hz ones, and therefore the result is, may be, more natural looking, and pleasing.

In the same way 120Hz tv's only improve blurring when frame interpolation is active, 240Hz tvs, or even 2400Hz tvs will/would need motion interpolation algorithms to reduce SAH blur.I may be wrong but I don't think anyone was saying as such. I think everyone agrees on this point (At least I hope so :))

Because in the case of 3 seperate pulses you are spreading the information over a much longer period. The effective (percieved) hold time and duty cycle will now be much longer.

Lets say you split it into 10,000 parts each with a 10% duty cycle. Your effective duty cycle and hold time will be ~99% (close to 41.8ms) with a true hold time still = 4.18ms

If I understand this correctly(not sure I understand all the reason behind it 100%, but I think i get the gist of it) this will only apply to repetition of frames, and not to an increase of unique frames. Is that right?

I believe so. This subtle distinction between unique/repeated frames makes all the difference in the world.

It's amazing how complex this is.

Do either of you know what the average fall times for CRT and plasma are?

Hi,

I would like to know the fall time for CRT. CRT motion quality is the holy grail. Thanks guys:cool:

No SAH? 240Hz LCD's will show exactly the same SAH blur as 60Hz or 120Hz ones. It doesn't care for an LCD if you draw one frame of 1/60s or 4 frames of 1/240s, the result is exactly the same. Unless motion interpolation takes place, obviously :rolleyes:

SAH is inherent to all LCD's. It's just that 240Hz TV's will be able to do a motion interpolation way much more precise than 120Hz ones, and therefore the result is, may be, more natural looking, and pleasing.

In the same way 120Hz tv's only improve blurring when frame interpolation is active, 240Hz tvs, or even 2400Hz tvs will/would need motion interpolation algorithms to reduce SAH blur.

If you double the processing per second aren't you going to increase gaming lag?

Hey fellas i know i'm preaching to the choir, but i made a couple videos explaining exactly what we've been talking about. If anybody knows a good ftp site to upload them to you would find them interesting, because they surprised me a little. The vids are ~ 23mb (x2)

I've heard Crt rise + fall times ranging from microseconds to just under a millisecond. I think it depends on the particular grade of phosphors used.

Hey fellas i know i'm preaching to the choir, but i made a couple videos explaining exactly what we've been talking about. If anybody knows a good ftp site to upload them to you would find them interesting, because they surprised me a little. The vids are ~ 23mb (x2)


borf, please do upload. I'm very interested. But I don't know any ftp sites.:)

Hey fellas i know i'm preaching to the choir, but i made a couple videos explaining exactly what we've been talking about. If anybody knows a good ftp site to upload them to you would find them interesting, because they surprised me a little. The vids are ~ 23mb (x2)

I've heard Crt rise + fall times ranging from microseconds to just under a millisecond. I think it depends on the particular grade of phosphors used.

I don't know any ftp sites either, you could upload to something like megaupload.com and post the link though.

I went with Mediafire. Take a look and let me know how the downloading goes!

AVS link (http://www.avsforum.com/avs-vb/showthread.php?p=14727232#post14727232)

I went with Mediafire. Take a look and let me know how the downloading goes!

AVS link (http://www.avsforum.com/avs-vb/showthread.php?p=14727232#post14727232)

It works! Easy to download, easy to play. Thanks!

Thanks for your specs moreHD. A little ancient, but i didn't expect such problems. Hopefully more vids today (with lower res).

I think in general people here know that the displays are far more advanced and than source material that we feed even today BD or hdtv is only 8 bit while we have displays that are 10 bit native and higher with the ability to do higher framerates than 24p at 1080 res and we should all lobby for better source material that is native 1080p at 60fps or higher therefore all the band-aid methods utilized by display manufactures today wouldnt complicate and degrade image quility!!!! I bet most of the problems we see today in displays would disappear if we truly had good source material to begin with this is my opinion ( right now we have garbage entering our displays and display manufactures are trying to band-aid it and as I see it this is not going in the direction that I like!!!!) LOTS OF MONEY SPENT ON BAND-AID SOLUTIONS BY MANUFACTURES!!!!