or Connect
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › motion flow test scenes?
New Posts  All Forums:Forum Nav:

motion flow test scenes? - Page 2

post #31 of 63
Quote:
Originally Posted by mark haflich View Post

Of course its still there. Its a one chip machine with a color wheel. It will always be there. The only issue is whether you can see it. But its still there. It will always be there. But shake your head violently from side to side, be careful not to hurt hurt yourself and I disclaim any responsibility if you do hurt hurt yourself, you have been warned, and you will probably see RBE. Better yet there are certain test signals which would make RBE visable to even a blind person. Sony used them at shows to show advertise against reap projection single DLP machines.

So what does this mean. Once you see RBE and convince yourself that its there even though you can't normally see it, return your machine and switch to a different technology machine. You may then see other beastly errors such as miscinvergence but WTF, you won't have invisable to your eyes RBE. smile.gif Message, unless you see RBE when watching content, don't worry about it. Enjoy.

The LED DLP that I have doesn't have a color wheel.
I was simply stating that may be a single chip thing.
I thought doing goofy things to find flaws was the mark of a true enthusiast?
post #32 of 63
There is no color wheel in a LED based DLP, but the color creation mechanism is still sequential.
Some people may still see the RBE.
It's just like a very very fast color wheel.
post #33 of 63
Quote:
Originally Posted by agrsiv95 View Post

I also have the HP 2.8. I will have to try it and see. One review said shaking your head can get them to show still so I guess their there.

Why would you go looking for RBE? I don't understand this, it's something that ruins projectors for most (many/some?) people who see it. So if you don't see it why would you want to go trying to see something that could ruin your projector for you?

It's one thing to measure/verify the calibration of your machine, something that you can fix, but it seems silly to go looking for things that you don't see (now) that could make you unhappy with your machine.
post #34 of 63
Quote:
Originally Posted by stanger89 View Post

Why would you go looking for RBE? I don't understand this, it's something that ruins projectors for most (many/some?) people who see it. So if you don't see it why would you want to go trying to see something that could ruin your projector for you?

It's one thing to measure/verify the calibration of your machine, something that you can fix, but it seems silly to go looking for things that you don't see (now) that could make you unhappy with your machine.

I'm not unhappy at all, I think it throws the best picture I've seen to date. I was just trying to see how bad, if any was noticeable and offer a first hand experience as to this projector.

The're weren't any that I noticed in Underworld Evolution but I'm sure I'm not as sensitive to as some.
Edited by agrsiv95 - 1/17/13 at 6:37am
post #35 of 63
Quote:
Originally Posted by opv View Post

There is no color wheel in a LED based DLP, but the color creation mechanism is still sequential.
Some people may still see the RBE.
It's just like a very very fast color wheel.

There will always be rbe unless the three primary colors are constantly being beamed, rbe is inplicit when the three colors can not be present at the same time and are flashed independently.
post #36 of 63
Quote:
Originally Posted by mark haflich View Post

There will always be rbe unless the three primary colors are constantly being beamed, rbe is inplicit when the three colors can not be present at the same time and are flashed independently.
Isn't that exactly what I said?
post #37 of 63

Let me throw my 2 cents in, I had a RS20 and after some time I became distracted more and more by the horizontal pan blurring on both hdtv and BR at 24fps.  I loved the black levels on the rs and was hesitant about the sony vw95 hearing so many saying it's black level was with an auto iris and still didn't have the black level of the newer JVC x55(?),  but I also read that sony's lcos panels are different than jvc and always handled motion much better.  At this point, I just didn't want to risk upgrading to a new jvc and still see the blur so I got the vw95.  It is way better in that area than the older jvc.  And black levels, after it was calibrated, I am totally delighted with its black level and shadow detail and I watch mostly scifi/horror/action with a lot of dark scenes.  It performs in all areas better than my rs20.

If you read the owners of the new jvc's obviously motion blur has improved but I've not read anyone saying it does better than the sonys. And black levels in very low light scenes with a very small amount of light the jvc 'outshines' the sony, but I've never been disappointed by the deep blacks of the sony in those few scenes some movies have.

post #38 of 63
Thanks for your input. I don't think anyone will argue that it will outperform the newer Sony models. I guess what I was really asking was if anyone thinks that the issue is to the point where it doesn't add noticeable artifacts to the image in regards to how motion looks. It seems that the consensus of owners of the most current models (this and last generation models (2012-2013)) is that motion blur is no longer an issue. I don't doubt that it's better. I will remain a skeptic of the claim that is it 100% a non-issue until I see for myself and if that's true the people of this fine forum will finally be happy that I've shut up about it. biggrin.gif

P.S. Zombie, check your Inbox.
post #39 of 63
To be very clear, the blur issue we are talking about here is the blur between image shots in the source material. We are not talking about blur caused by slow lcd panels or blur in each individual frame caused by a shutter speed too slow tp freeze the shot. This blur is not the fault of the projector and requires additional frames to be constructed by a computer (built into the projector but possible to do outside the projector). The more frames between original frames constructed, generally the better. What we are talking about is thus frame interpolation and how well each of thev various projectors does it. It is fair to say the Sony does better than the JVCs at this.

Another major issue is ones susceptability to seeing motion blur. If you don't see it, there is no reason to switch on motion curing algs because these generate artifacts themselves, not being real frames, which may be more objectionable than the blur itself. Take the Sony's for example. You can shut the motion blur fixers off, or set them to say low or high. Maybe low, medium, and high. why are there choices beyond off or on?. Its the artifacts and the dgree one sees those artifacts.

Could the algs be better and would they be better with more computational power assuming you deal with the processing lag. the answer is clearly yes.

The real fix lies in improving the source, more frames.

There is a temptation to switch on motion handeling in the Sonys.

I watch a lot of sports. Action where I would think I should see blur. Its probably my eyes. I just don't see any blur. But i still get tempted to turn it to low. Its there to use and i know i should be seeing bvlur (but i don't0, so since I should then I should switch it one. So sometimes I switch it on. but I see or rather I don't see any differences.
post #40 of 63
Wait, I'm confused. Who is talking about Creative Frame Interpolation. I've been talking about native motion handling (without turning on any creative frame interpolation) the entire time. I thought the rest of us were too?
post #41 of 63
Maybe you are. But exactly what is that. The projector can not make a blur in a captred frame unblurred. A projector could because of the particular implementation of technology (namely lcd0 cause an unblurred frame to be displayed with a blur or smear.


But I think what you guys are talking about is that moving object appears blurred when in real life your eyes would see it without blur. If the source frame is captured without blur, the projector with the right stuff could basically replicated whatr your eye would see in real liofe but this replication could never be native.

So set me straight. What kind of blurs are we talking about?
post #42 of 63
A dumb question about Sony's 240 hz panels vs JVC's 120 Hz panels....when playing back a 24p source are the Sony panels driven with 10 consecutively identical frames or is there some interpolation taking place between the 24Hz frames? If there is no interpolation (ie, one can totally switch it off) why should this be "better" than a JVC showing 24p at 120 Hz rate (ie, 5 consecutively identical frames)? It seems plausible that the JVC panels could still be "settling" when a new 24Hz frame occurs such that there is a difference between the 1st and 2nd frame of the supposedly 5 "consecutively identical frames". Equally plausible is the Sony settling time causes a difference between the 1st and 2nd frame of the supposedly 10 "consecutively identical frames". In essence the error due to this settling time in the Sony would be 1/2 of that of the JVC. I take this is what the discussion centers around...no?
post #43 of 63
Quote:
Originally Posted by mark haflich View Post

Maybe you are. But exactly what is that. The projector can not make a blur in a captred frame unblurred. A projector could because of the particular implementation of technology (namely lcd0 cause an unblurred frame to be displayed with a blur or smear.


But I think what you guys are talking about is that moving object appears blurred when in real life your eyes would see it without blur. If the source frame is captured without blur, the projector with the right stuff could basically replicated whatr your eye would see in real liofe but this replication could never be native.

So set me straight. What kind of blurs are we talking about?

Not all LCDs are the same. This is also true about LCOS panels. Response times vary between different iterations of LCD and LCOS technology. Typically the faster the response time of the panel the better they resolve motion and motion resolution. One of the reasons JVC's LCOS panels have such a hard time dealing with motion natively is that their panels have a slow native response time. Without any sort of FI to add extra frames in, these panels add extra motion blur to the image. They add blur that wasn't there to begin with. This is what this thread is all about and what they OP is asking. Although some are reporting this isn't too much of an issue anymore with this years current models. I want to see it for myself. I'm a born skeptic and I always like to see things for myself. smile.gif


The reason DLP is better with motion is because the panels native response time is more than 1000 times faster than current LCD and LCOS panels. As long as there are no issues with video processing, motion on a DLP system will look how it was supposed to. If there is any kind of blur captured it will still show up on a DLP projector, but they do not add any extra.
post #44 of 63
OK. We are on the same pageBut I don't think it is motion blur. If the frame had no blur, it was shot at a shutter speed that stop any action in the frame (but it is likely that because of the wider lens opening required to get enough light for the exposure at the higher shutter speeds, the back ground and fore ground will be out of focus and look blurry), as long as the panel rrespose time was faster than the time the frame was on the screen, there should be no motion blur. Now certain LCD panels are just not sharp at the trailing or leading edges, I forget which. If one examines the chips using a camera aimed through the projector lens at the panels, the edge sot of looks amorphous. And not all three panels are driven the same. With a DLP (one chip or three), the edges are of course sharp.

In any event, the most troubling aspect regarding motion for most would be the blur in the source because of limited frames and how well FI in ach projector can correct or rather reduce the perceptable blur.
post #45 of 63
Quote:
Originally Posted by mark haflich View Post

OK. We are on the same pageBut I don't think it is motion blur. If the frame had no blur, it was shot at a shutter speed that stop any action in the frame (but it is likely that because of the wider lens opening required to get enough light for the exposure at the higher shutter speeds, the back ground and fore ground will be out of focus and look blurry), as long as the panel rrespose time was faster than the time the frame was on the screen, there should be no motion blur. Now certain LCD panels are just not sharp at the trailing or leading edges, I forget which. If one examines the chips using a camera aimed through the projector lens at the panels, the edge sot of looks amorphous. And not all three panels are driven the same. With a DLP (one chip or three), the edges are of course sharp.

In any event, the most troubling aspect regarding motion for most would be the blur in the source because of limited frames and how well FI in ach projector can correct or rather reduce the perceptable blur.

Yes, trailing edges that look like the image is smeared when certain types of motion occurs. It looks very similar to "blur". But in the end you are correct, motion blur is something else. This is image smearing hahah tongue.gif
post #46 of 63
Quote:
Originally Posted by Geof View Post

A dumb question about Sony's 240 hz panels vs JVC's 120 Hz panels....when playing back a 24p source are the Sony panels driven with 10 consecutively identical frames or is there some interpolation taking place between the 24Hz frames? If there is no interpolation (ie, one can totally switch it off) why should this be "better" than a JVC showing 24p at 120 Hz rate (ie, 5 consecutively identical frames)? It seems plausible that the JVC panels could still be "settling" when a new 24Hz frame occurs such that there is a difference between the 1st and 2nd frame of the supposedly 5 "consecutively identical frames". Equally plausible is the Sony settling time causes a difference between the 1st and 2nd frame of the supposedly 10 "consecutively identical frames". In essence the error due to this settling time in the Sony would be 1/2 of that of the JVC. I take this is what the discussion centers around...no?
I believe that Sony shows the same image 10 times if all the image "optimization" turned off. The biggest difference is that Sony uses analog/multilevel signal do drive the panels and JVC uses PWM/digital.
post #47 of 63
Its a problem with how the crystals are driven. It will reduce the total system MTF to some extent but it is a relatively minor fault in the great scheme of things. Its got to do with how each pixel crstal is driven. The sony panels are driven differently than the JVC. It may have something to do with anaslogue drivers instead of digital and several years ago there was quite a bit of discussion about this. But in reality, at the level we are at today I would think the image degradation caused by this is very minor.
Edited by mark haflich - 1/18/13 at 9:50am
post #48 of 63
Quote:
Originally Posted by Victor View Post

I believe that Sony shows the same image 10 times if all the image "optimization" turned off. The biggest difference is that Sony uses analog/multilevel signal do drive the panels and JVC uses PWM/digital.
Thanks!
Yes, I understand the difference between the Sony and JVC drive. If panel response were fast enough all 10 images (Sony) would be identical and likewise all 5 images would look identical in the JVC (regardless of how they're driven). But if the panel response is not fast enough it's possible that 9 images (Sony) or 4 images (JVC) differ from the first image. That would cause the JVC to look more smeared than the Sony because of the panel refresh rate. I can understand how some folks would not be able to distinguish this difference while others can.
post #49 of 63
I think it is not just simple classic "panel response time" - it is that and algorithm they use for PWM that causes choppy motion. I suspect that it works not just within one frame, but uses longer "integration time" otherwise it would work as good as Sony analog drive.
post #50 of 63
So it looks like we have another instance of the supriority of analogue over digital. I wonder if you take the right sea sickness pills, you can make the choppiness of digital drive planes bother you?

OK. i am going to watch Lawrence of Arabia from the new blue rasy upscaled to 3840 x 1920 in glorious 8 bit Rec. 709.

Guys. Go goggle Wiki rec. 2020 for the new 4HD standard. Grab hold of the new color space, double the area of rec 709. Man getting enough light out with that space is going to be a real challenge. Goodby to present day projector plastforms. Going to have to be lasers.
post #51 of 63
Not to belabor this but I agree it is response time. That said if the same identical image is being "flashed" 5 times in a row the response time for the last 4 flashes should be close to nil because there is no change between these flashes -- there is nothing to respond to....The majority of the response time is incurred during the first of the 5 because that's where the change occurs. Likewise, with Sony the response time is greatest for the first flash and should be close to nil for the 9 remaining refreshes since nothing is changing. I believe the larger native contrast ratio is why JVC panels are limited to 120 Hz.
Edited by Geof - 1/18/13 at 7:31pm
post #52 of 63
Thread Starter 
Now my eyes are getting blurry reading all of this. I'll let y'all know what happens when i get my 50 next week.
post #53 of 63
I am not sure about what I just read. The JVC has a lower ref black level because there is less scatter caused becauses the crystals owhatever on one layer of the panel are better aligned in a DILA panel. Its not like the DILA has more steps of contrast. That to my non stdied analysis seems specious.. The same number or steps are involved regardless of the panel construction. Or do I have it wrong.
post #54 of 63
I think he's referring to grey to grey and black to white timing of the LCOS panels.
post #55 of 63
geof says the sony has less contrast range to respond to. That is wrong i think. the same levels are sent to each by a binary code. the code for the first step of gray above black will be the same. the measured value might be higher on the Sony due to high scatter from the layer on the chip substarte and there for the dynamic or static contrast range might measure less but the machine has the same amount of work to do, tatcheting between contrast steps. I do agree that if the fraame is repeated more in one unit the response time for the extra flashes will be less but I really thing this would have little to do with smear. The frame has to change on both, one just one flash later. would that affect smear? I dunno.
post #56 of 63
The binary code does not establish the dynamic range (contrast ratio) of the panel. Yes each panel operates from 16-235 but what differs is the amount of change between 16 & 17 - the Sony's don't go as black there. That said, I've modified my post to remove the offending sentences because I was getting astray from the point I was trying to make (and it was poorly worded as well).
Edited by Geof - 1/18/13 at 7:36pm
post #57 of 63
Thanks for the fix. Of course the dynamic range is established by the measured value at 100 IRE divided by the measured value at 0. with a greater dynamic rage, the delta value of each step will be higher but thast will have absolutely nothing to do with the spead it takes for each step. that is a characteristic of the particular design of the panel including how it is different. This is a healthy discussion and I mean no disrespective of my friend Geof.
post #58 of 63
Given identical lumen output at 100% the delta step value (for steps between 17 and 235) is the same for projectors even with with vastly different on/off ratios. The difference in On/Off is solely between steps 16 and 17.

Here's why: The Y value at any particular level (90%, 80%, etc) is calculated from the 100% value so these projectors should be outputting the same light level at every stimulus level if they have identical gamma's. However, a higher CR display will output less light at 0% than a lower CR display so the luminance step between 16 & 17 will vary depending on the CR of the display.
post #59 of 63
Are you saying that given your set of set up and projector assumptions, that the two projectors would measure the same between 17 and 18? If the near blacks measured the same, the onlysuperiority of the blacks of the JVC would be essentially at 0 or 16 and i don't think that is the case, or course i could be wrong. but the superiority of the blacks is evident in the near black as well. Remember on off CR is just a number. its not a magic something but it does reflect performance in the near blacks. I it is essentially irrelevant at high steps and is completely irrelevant if the high end lumens out are the same.
Edited by mark haflich - 1/19/13 at 6:46am
post #60 of 63
Just stating that the luminance steps between levels 17-18 or 18-19...or 234-235 would be equal IF two projectors were setup to output the same lumen output at 100% stimulus and IF they have identical gammas. The Contrast Ratio of the projector is irrelevant in this case. The contrast Ratio only changes the step size between level 16 and level 17: Higher CR projectors have a larger step between level 16 and 17 than lower CR projectors (IOW, level 17 is easier to see with higher CR projectors).

Sounds odd, I know, but it is because the Y value at any particular level (90%, 80%, etc) is calculated from the 100% value so if the 100% values match then the luminance step size is identical regardless of the projectors on/off ratio. Still find it hard to believe me? Read this post by Greg Rogers.

I didn't meant to take this thread off topic. But to get back on topic I think one main cause of motion smear in previous JVC was well document by Mark Petersen in his MTF study. It appears that JVC has made some improvement in that area since Mark authored that paper and that could explain why many think JVC's motion handling has improved since the RS25/35.
New Posts  All Forums:Forum Nav:
  Return Home
AVS › AVS Forum › Display Devices › Digital Hi-End Projectors - $3,000+ USD MSRP › motion flow test scenes?