60i vs 30p vs 24p

[Ken Ross]

Originally Posted by Dave Rosky

Yes, this is why I haven't bought a TV yet. Obviously I'd like to get one that does a good job of deinterlacing both 24P 2:3 and 60i (perhaps even generating a good quality 60P stream from the 60i), but even though I'm a technical person and have some idea of what is technically possible, it's so hard to actually find out what products actually do what, and I don't have 8 hrs. a day do devote to research.

Actually picking the best TV is one of the easier jobs today. Every review has called the latest generation of Pioneer Kuro displays the 'best ever'. Seldom have I ever seen such universal praise for any AV product.

Owning two, I can attest to their unbelievable picture quality. And yes, they perfectly deinterlace, but you do get some motion jitter with 30p in a 60i stream. I don't believe any display will show buttery smooth motion from this 'wrapper'. However, you do get buttery smooth motion from every other source I've ever fed it. The Pioneers have something called "smooth" processing which even helps smooth out motion judder inherent in some 24p BR discs. It's almost weird to see such smooth playback of scenes you know contains this judder. That processing can be engaged or disengaged, your choice.

Highly recommended!

[Dave Rosky]

Originally Posted by Ken Ross

Actually picking the best TV is one of the easier jobs today. Every review has called the latest generation of Pioneer Kuro displays the 'best ever'. Seldom have I ever seen such universal praise for any AV product.

Owning two, I can attest to their unbelievable picture quality. And yes, they perfectly deinterlace, but you do get some motion jitter with 30p in a 60i stream. I don't believe any display will show buttery smooth motion from this 'wrapper'. However, you do get buttery smooth motion from every other source I've ever fed it. The Pioneers have something called "smooth" processing which even helps smooth out motion judder inherent in some 24p BR discs. It's almost weird to see such smooth playback of scenes you know contains this judder. That processing can be engaged or disengaged, your choice.

Highly recommended!

Thanks, Ken. Yes, the Kuros come highly recommended from a number of sources, so in that sense I guess it's an easy choice. But they are also near the top of the heap price-wise. I know you get what you pay for, but I was hoping to find something more mid-priced that still had decent signal processing performance. The new crop of Panasonics due out soon are rumored to have improved signal processing and might be more mid-priced, so I may wait a bit and check them out.

I might end up wanting a Kuro, but I will have to build up some courage before trying to get a $5,000 TV past the finance committee ;-)

[Ken Ross]

Yeah Dave, generally that 'finance committee' demands something in return. That can be a very expensive proposition! ;)

[Bob Kittleson]

Originally Posted by Aaron Courtney

60i is only yielding 30 real fps, as captured by the camcorder, any way you cut it on a progressive display.

Aaron, why do you state that as if it were an established fact? It hasn't been proven by the discussion and links in this thread. In fact, at least 2 of the linked articles have mentioned deinterlacing methods for 1080i that yield 60fps output.

Consider, for example, the digitalcontentproducer article. The descriptions given for the "Bob" and "2D FIR" methods specifically say that frames are output each field time, i.e. 60fps. Those methods by themselves don't seem very desirable, but how about this one:

"A Vector Adaptive interpolator uses memory to hold four fields. Logic measures motion between fields. For static video, weave is employed. For dynamic video, samples come from the current plus a previous and/or a future field."

The article unfortunately doesn't specifically say if this method would output frames at 30 or 60fps, but it 60fps seems feasible.

In any case this is all theoretical, because different displays and video processors certainly use different methods. I don't think anyone can make a blanket statement that all fixed pixel displays render 1080i as 30fps or 60fps. It seems that we're left to test our own particular equipment and judge the results for ourselves. It's unfortunate that this is not more standardized.

[Aaron Courtney]

Traditionally, de-interlacing for playback on a progressive display has always meant recombining the two fields that contain the original frame. The fact that several video processing manufacturers have moved well beyond that idea with proprietary algorithms designed to create 60 frames per second from 60 fields per second is really immaterial at least IMO. They are trying to create something that did not exist as these images were laid to tape, flash, HDD, or whatever in cam. I'm not saying it doesn't work, just that it is not the same thing as capturing 60 full rez frames per second.

[Ron Evans]

I think it is more appropriate to call interlace cameras 60 fields per second cameras because they never capture frames at all, or record frames to tape or whatever, its always fields. They take a field every 1/60 of a second. The field just happens to be half the vertical resolution in discrete horizontal strips( one odd the next field even). That is very different from a picture with complete cover at half resolution. Thus the scene capture rate is 60 fps not 30 fps its just that only half the horizontal strips are being recorded at a time. This of course is very different from a camera capturing 30 frames per second and then processing field information to record. With a CRT phosphor decay our eyes still see the last field when the next is displayed and that is why we perceive a full image at what appears to be 60 fps, not 30 fps. The screen refresh rate is 60hz and the image actually changes every 1/60 sec( just not all of it,but that is true of most images anyway and why it works so well) Our eyes/brain interpolate the missing/decaying horizontal lines and that is what I would hope modern electronics should do on a progressive display. It would be nice to know what manufacturers are creating 60fps though.

Ron Evans

[Ken Ross]

Ron, good explanation. People do get confused with frames and fields and that appears to be why many feel that half the resolution is lost when its not. I do believe that Pioneer at least is creating 60fps in their propietary 'smooth' mode.

[Dave Rosky]

Originally Posted by Ron Evans

I think it is more appropriate to call interlace cameras 60 fields per second cameras because they never capture frames at all, or record frames to tape or whatever, its always fields. They take a field every 1/60 of a second. The field just happens to be half the vertical resolution in discrete horizontal strips( one odd the next field even). That is very different from a picture with complete cover at half resolution. Thus the scene capture rate is 60 fps not 30 fps its just that only half the horizontal strips are being recorded at a time. This of course is very different from a camera capturing 30 frames per second and then processing field information to record. With a CRT phosphor decay our eyes still see the last field when the next is displayed and that is why we perceive a full image at what appears to be 60 fps, not 30 fps. The screen refresh rate is 60hz and the image actually changes every 1/60 sec( just not all of it,but that is true of most images anyway and why it works so well) Our eyes/brain interpolate the missing/decaying horizontal lines and that is what I would hope modern electronics should do on a progressive display. It would be nice to know what manufacturers are creating 60fps though.

Ron Evans

Ron, yes, the traditional summary way of saying what you said is that interlaced video transmits static or slower motion in full resolution at the frame rate, and fast motion in half the vertical resolution at the field rate.

That's the compromise of interlace. You can't have fast motion at full resolution unless you go to full progressive 60 frames per second.

And, yes, it sure would be nice to have a table or list showing which deinterlace methods are used by the various TV's. Someone linked to a page earlier in this thread that lists TVs known or believed to display 24P properly (by shifting the refresh rate, and the Kuros are on that list). Perhaps TVs that have higher quality 24P display also have higher quality deinterlacing in general.

[Ron Evans]

Also of interest may the latest Amberella info.
http://www.ambarella.com/news/press_...r_01072008.htm
Would be interesting to know which cameras use this latest chip too.
Ron Evans

[Dave Rosky]

Originally Posted by Ron Evans

Also of interest may the latest Amberella info.
http://www.ambarella.com/news/press_...r_01072008.htm
Would be interesting to know which cameras use this latest chip too.
Ron Evans

This is definitely a move in the right direction. It may be a long time before broadcasting moves to 1080/60P because of bandwidth limitations, but there is no reason at all that camcorders can't move in that direction. I don't know if the BD spec supports 1080/60P (anyone know?), but if it doesn't, it should be added. Right now, the 60 fps progressive capability of most newer TVs is being partially wasted (unless you've got it hooked to your computer and are playing video games).

The BD disc capacity is so large that there's no reason that movies can't also finally move up to 60 fps, except that the film industry is so attached to the ancient 24 fps look. It sounds stupid, but I'm afraid our great grandkids will still be watching movies at 24 fps 100 years from now.

Knowing how long product design cycles tend to be, I'd be surprised to see this chip even in 2009 model camcorders, but maybe...

[Aaron Courtney]

Dave, pretty sure the answer to BD spec and 1080p60 is "no" as of today. I guess that means all of the players today will be obsolete as soon as that format is ratified by the BDA because I highly doubt a simple firmware update is going to work. Guess we shall see in the next "X" number of years (before they get their act together).

Ron, GREAT find! A lot of that press release sounds pretty familiar, hmmm? Comparing it to the information presented in this thread, someone might get the idea that I actually wrote it, LOL! It's kinda comical to read between the lines and catch their pitch for their product.

[Dave Rosky]

Originally Posted by Aaron Courtney

Dave, pretty sure the answer to BD spec and 1080p60 is "no" as of today. I guess that means all of the players today will be obsolete as soon as that format is ratified by the BDA because I highly doubt a simple firmware update is going to work.

Yes, I would think decoding 1080/60P might require additional processing power, so it probably won't be just a firmware update, unless some manufacturers are already anticipating it. On the good side, in a year or two, that second BD player with new capabilities will cost a *lot* less than your first one did ;-)

This concurrent shift to both hi-def and progressive displays is the biggest shift in consumer video technology since TV was introduced, so unfortunately it's taking a few years for things to settle out.

[Dave Rosky]

There's an interesting white paper posted at the Amberella site:

http://www.ambarella.com/docs/1080p60.pdf

This talks about using 1080/60P for broadcasting. Initially, one would think that this would take nearly twice the bandwidth of 1080/60i, but it turns out that due to the higher level of vertical correlation within a frame, that it actually takes only around 20% more bandwidth for the same quality.

[Aaron Courtney]

Another nice find, Dave. Even though there are some typo's in that paper, I think it nicely summarizes a lot of the content of this thread and corroborates what the MIT expert wrote and argued in front of the ATSC adoption panel.

[Dave Rosky]

Originally Posted by Aaron Courtney

Another nice find, Dave. Even though there are some typo's in that paper, I think it nicely summarizes a lot of the content of this thread and corroborates what the MIT expert wrote and argued in front of the ATSC adoption panel.

Yes, to me one of the more interesting points of the paper is how broadcast infrastructure is now lagging consumer electronics technology where it has always been ahead in the past. The same thing is soon going to happen with color space as in a year or so most all TV's sold will be xvYCC capable. It will be rather ironic in a year or two when a consumer can go from production to display in 1080/60P xvYCC before broadcasting has even adopted one of those new technologies.

Of course the problem is that the broadcasting infrastructure is large and expensive to change, so it will happen more slowly. Nonetheless it's amusing to see that in some respects the envelope is now being pushed by the consumer end.