24p and Frame mode?
- Thread starter Caution
- Start date
DC
Cinematographiliac
Re: 24p and Frame mode?
"24P" is the popular shooting mode of the Panasonic DVX100/100A camera (within this forum) that really gives that film-look. "Frame Mode" is a shooting mode commonly associated with the Canon XL-1 camera. However, Canon is about to release the XL-2, which has a wealth of new features including 24P. Personally, I'm very happy with my tried and true DVX100A but it'll be interesting to see how the XL-2 really competes.
Here's some info (some a bit dated) to help explain so you can prepare your projects accordingly.
__________
The Frame Movie Mode Explained
an article by Michael Pappas
Michael Pappas is a Filmmaker at Arrival Entertainment Pictures. He is also involved in research on new image technologies that filmmakers can use to make films. He lives in Orange County, California.
FRAME MOVIE MODE on the Canon XL1
reprinted here without permission
In the past 11 years I have been doing research on getting video to have the same frame rate feel as 30 and 24fps film. In the end of October I heard about a feature that was on the XL1 that was going to allow the capture of 30 full frames a second. A bell went off in my head, if Canon was about to do what I was suspecting they were going to do this might be one hell of a feature. I have been able to film around 30 minutes of material in the frame movie mode indoors and out, in all kinds of lighting conditions. The feature worked great.
This feature is very important, not just to Canon but to the whole industry. Here is the side bar for all to read. This article is copy righted, if you want to repost it, e-mail me for permission. -- Michael Pappas.
Digital Cinematography has a new player... It's called the XL1.
The new Canon XL1 has a wealth of features that independent filmmakers to videographers will benefit from. One item that I've found that will be of great interest to independent filmmakers is a feature called Frame Movie mode. This allows you to adjust the way the CCD chip captures an image.
Normal video is made up of 60 half resolution frames a second or also referred to as fields. In the Frame Movie Mode, the XL1 captures 30 full frames instead of 60 half resolution frames a second. There are many beneficial results because of this special capture mode. First the captured frames vertical resolution is increased because the two interlaced fields of a single frame capture are not scanned at separate points of time like normal 60 field video captures. The odd and even fields are simultaneously captured all at once at 1/60 of a second in the Frame Movie Mode.
In normal video mode the CCD would capture the odd field at 1/60 of second, then capture the following even field at 1/60 of a second. If there is any movement in between the time when these frames were captured, there would be a discrepancy between the two separately captured fields resulting in what is called motion flicker. There is also aliasing in the video because the fields are containing information of different movements recorded at those different times of the captures. Since the XL1 can capture a full frame not half fields, aliasing is dramatically reduced.
Another by-product of the Frame Movie Mode is you get what has been referred to as the "film look". Video that is recorded in this mode would contain the same amount of different frames that a 30fps film (Like an Iwerks movie) transferred to tape would have. Another big advantage to this full frame capture is if the video was transferred to film, the video shot would contain a higher vertical resolution because you are transferring a full frame, not a field to film. One would only have to drop six full frames in total, or in other words, drop every 5th full frame and you would have 24 full DV frames transferred to film. If you took it to the next stage and have it line doubled and transferred to HD or another digital format before its transfer to film, your results will even be better.
Today's acclaimed Digital Betacam cameras can't even give you the option to record 30 full frames, they only do 60 field video. If Digital Betacams did offer this, it would indeed make the final product even look sharper and much easier to transfer to film. The Canon XL1's Frame Movie Mode feature is going to set a new standard in the way digital video images are captured. The Full frame capture feature could turn out to be a perfect tool for filming television productions, documentaries and feature films. -- Michael Pappas
_______________
"24P" is the popular shooting mode of the Panasonic DVX100/100A camera (within this forum) that really gives that film-look. "Frame Mode" is a shooting mode commonly associated with the Canon XL-1 camera. However, Canon is about to release the XL-2, which has a wealth of new features including 24P. Personally, I'm very happy with my tried and true DVX100A but it'll be interesting to see how the XL-2 really competes.
Here's some info (some a bit dated) to help explain so you can prepare your projects accordingly.
__________
The Frame Movie Mode Explained
an article by Michael Pappas
Michael Pappas is a Filmmaker at Arrival Entertainment Pictures. He is also involved in research on new image technologies that filmmakers can use to make films. He lives in Orange County, California.
FRAME MOVIE MODE on the Canon XL1
reprinted here without permission
In the past 11 years I have been doing research on getting video to have the same frame rate feel as 30 and 24fps film. In the end of October I heard about a feature that was on the XL1 that was going to allow the capture of 30 full frames a second. A bell went off in my head, if Canon was about to do what I was suspecting they were going to do this might be one hell of a feature. I have been able to film around 30 minutes of material in the frame movie mode indoors and out, in all kinds of lighting conditions. The feature worked great.
This feature is very important, not just to Canon but to the whole industry. Here is the side bar for all to read. This article is copy righted, if you want to repost it, e-mail me for permission. -- Michael Pappas.
Digital Cinematography has a new player... It's called the XL1.
The new Canon XL1 has a wealth of features that independent filmmakers to videographers will benefit from. One item that I've found that will be of great interest to independent filmmakers is a feature called Frame Movie mode. This allows you to adjust the way the CCD chip captures an image.
Normal video is made up of 60 half resolution frames a second or also referred to as fields. In the Frame Movie Mode, the XL1 captures 30 full frames instead of 60 half resolution frames a second. There are many beneficial results because of this special capture mode. First the captured frames vertical resolution is increased because the two interlaced fields of a single frame capture are not scanned at separate points of time like normal 60 field video captures. The odd and even fields are simultaneously captured all at once at 1/60 of a second in the Frame Movie Mode.
In normal video mode the CCD would capture the odd field at 1/60 of second, then capture the following even field at 1/60 of a second. If there is any movement in between the time when these frames were captured, there would be a discrepancy between the two separately captured fields resulting in what is called motion flicker. There is also aliasing in the video because the fields are containing information of different movements recorded at those different times of the captures. Since the XL1 can capture a full frame not half fields, aliasing is dramatically reduced.
Another by-product of the Frame Movie Mode is you get what has been referred to as the "film look". Video that is recorded in this mode would contain the same amount of different frames that a 30fps film (Like an Iwerks movie) transferred to tape would have. Another big advantage to this full frame capture is if the video was transferred to film, the video shot would contain a higher vertical resolution because you are transferring a full frame, not a field to film. One would only have to drop six full frames in total, or in other words, drop every 5th full frame and you would have 24 full DV frames transferred to film. If you took it to the next stage and have it line doubled and transferred to HD or another digital format before its transfer to film, your results will even be better.
Today's acclaimed Digital Betacam cameras can't even give you the option to record 30 full frames, they only do 60 field video. If Digital Betacams did offer this, it would indeed make the final product even look sharper and much easier to transfer to film. The Canon XL1's Frame Movie Mode feature is going to set a new standard in the way digital video images are captured. The Full frame capture feature could turn out to be a perfect tool for filming television productions, documentaries and feature films. -- Michael Pappas
_______________
DC
Cinematographiliac
Re: 24p and Frame mode?
Adam Wilt writes the following on the DVX100/100A's shooting modes (reprinted here without permission):
How is 24p recorded? What's the difference between 24p and 24p Advanced?
The camera can capture interlaced footage at 59.94 fields/second or 29.97 frames/second ("60i"), 29.97 progressive-scanned frames/second ("30p"), or 24 progressive-scanned frames/second ("24p").
The recorder (and indeed all the video I/O) runs at 29.97 frames/second, interlaced, all the time, regardless of what the camera is doing.
When the camera is running in 60i or 30p, the incoming video is placed into a frame buffer, compressed with the DV codec, and written to tape and/or sent out FireWire. That same frame buffer feeds the analog outputs, one field after the other. Whether the images come in interlaced or progressive, every 30th of a second – well, every 29.97th, but let's keep it simple for now – an image gets written to the frame buffer, and an image gets played out: two fields compressed as a frame for FireWire and recording, and two fields sent out one after the other for interlaced analog I/O.
24p is a bit more complex. [Of course, it's really 23.976p, the same way 30p is really 29.97p and 60i is really 59.94i, and even those numbers are approximations. When the National Television Systems Committee added color to the existing monochrome standard in 1953, they had to fudge either the picture frequency or the audio subcarrier frequency by one part in a thousand, to keep the color subcarrier from interfering with the sound subcarrier. Sound won; NTSC pictures now run at 30 * 1000/1001 fps (and that number is exact), and ensuing generations have had to struggle with the peculiarities of drop-frame time code and resetting their master clocks every day to account for drift. But I digress...] 24 does not fit so cleanly into 30; for every four frames of 24p video that go by, five frames of 30p (or 60i) video go by. Another way to look at it is that 24 frames of 24p video have to be squeezed, stretched, and mashed into 60 fields of NTSC-compatible video; that's four frames of 24p per ten fields of 60i.
There are several ways to do this; the AG-DVX100P and AJ-SDX900 offer two: 24p, which I call 24p Standard, and 24p Advanced.
24p Standard utilizes the same 2:3 or 3:2 pulldown cadence long used to transfer 24fps film to NTSC video. The first 24p frame is written to two fields of 60i video, the next is written to three, the next to two, and the next to three again, as shown in the graphic.
This cadence offers the smoothest possible raw conversion between the frame rates, and gives us the familiar judder of film transferred to tape. There are a couple of things to notice about it:
1) The "A frame" is defined as the only frame of 24p that occupies exactly one whole frame of 60i video with no overlap into adjacent frames.
2) The A, B, and D frames can be recovered by using two fields from the same 60i frame. The C frame cannot be; it is split across field 2 of the third (green) 60i frame and field 1 of the fourth (magenta) 60i frame. In an intraframe-compressed format like DV, that's an important distinction: the two fields of each 60i frame are compressed together, as a frame. If the two fields are very similar, as happens in the red, yellow, and blue frames, the compression uses a comparatively efficient "8x8 DCT" mode. If the fields are very different, as happens in the green and magenta frames when there's a lot of motion between the B, C, and D frames, a less efficient 2x4x8 DCT is used, possibly leading to more image degradation in those frames compared to their red, yellow, and blue companions.
Furthermore, the original A, B, and D frames can be copied in their compressed form from the 60i video data into a new 24p data file, but recovering the C frame requires decompressing the green and magenta frames and recompressing them into a new, 24p DV frame. That puts the C frame a generation down compared to A, B, and D.
If the 24p extraction tool uses white instead of superwhite codec ranges, as Cinema Tools does, it will clip whites in doing so – you really have to preprocess all your footage prior to reverse telecine to pull superwhites and saturated colors into range. Note that this is not necessarily an indictment of such tools: codecs that run in a white-only range instead of a superwhite range can eke a little bit more quality out of the in-range image, at the expense of the out-of-range data. However, if you're using the extended dynamic range that superwhites give you in acquisition, it means you have to add that extra processing step to pull your superwhites back into range before converting to 24p, lest the conversion hard-clip your C frames and remove that creative control from your domain.
24p Advanced uses a syncopated 2:3:3:2 pulldown cadence to stuff 24 frames into 60 fields. It's ever so slightly different in its playback; the standard cadence of 2:3:2:3:2:3:2:3... evenly intersperses the "short" and "long" frames, while 2:3:3:2:2:3:3:2... lumps two "short" frames together followed by two "long" frames. The difference is subtle, but can be seen on smooth pans or on regular in-frame motion, like the passage of a train. In my experience so far, about half the people looking at a 24p Advanced clip can see that the motion is a bit different, and half cannot.
But 24p Advanced isn't intended for making the 60i video look like film; it's designed to allow the best possible recovery of the original 24 frames. You'll note that all four original frames can be recovered from self-contained 60i frames; the green frame in 60i now contains the "extra" B and C fields and can be discarded, since all the information for B is contained in the yellow frame, and all the information for C is in the magenta frame.
Extracting a true 24p clip from a 60i recording simply requires copying the raw data for the red, yellow, magenta, and blue frames into a new 24p file, skipping the green frame altogether. No decompression or recompression is required, and all recovered 24p frames retain first-generation quality. No clipping or other loss is incurred; you are still working with all your frames in their first-generation glory in 24p.
Indeed, Cinema Tools goes one step farther: it gives you the choice of either creating a new 24p file, or simply rewriting the Quicktime frame pointers in the original file to skip over the green frames! Rewriting the header is much faster (not that Cinema Tools is slow in writing a new file), but your "new" 24p file is about 25% larger than it needs to be, since it still contains the discarded frames, and you can no longer go back to that file and play it as a 60i file: Cinema Tools has no "undo" for rewriting the frame pointers.
When to shoot 24p? 24p Advanced? 30p? 60i?
The general rule is to shoot 24p Advanced if you want to extract the original 24 frames/second for a 24fps edit or film-out. Shoot 24p Standard if you are going to stay on video and edit at 30 frames/seconds (60 fields/second, i.e., plain ol' video at NTSC frame rates), without extracting the original 24 frames into a 24fps timeline. In more detail:
Shoot 24p Advanced for:
Post-production using tools that understand Advanced pulldown. 24p Advanced footage can be turned into pure 24p footage more cleanly than 24p standard footage, because every frame in the pure 24p timeline is pulled from a whole frame in the 24p Advanced footage, whereas the C frame in 24p Standard footage is split across two different source frames as discussed above.
Shoot 24p Standard for:
Getting the “film look” on video when you're staying on video and editing at 29.97.
Intercutting with film transfers also using 3:2 pulldown, and staying on video at 29.97.
Working with traditional film-on-tape tools that understand 3:2 pulldown, but not 2:3:3:2 advanced pulldown, when you need to extract the 24p footage for true 24p processing.
Shoot 30p for:
Getting true progressive pictures with a 30 fps frame rate, as when pulling stills for motion analysis.
Working alongside Canons and older Panasonics in Frame Movie Mode, when you want to match their motion rendering.
You want the slightly "filmic" motion of 30p, but don't want to go to 24p, and you aren't concerned about ever going to film or converting to PAL.
Shoot 60i for:
Anything that you want to use as plain ol' video at NTSC frame rates: in 60i, this camera makes pix that look (from a motion-rendering standpoint) just like the pix from any other video camera. In other words, use 60i for everything that isn't supposed to “look like film” and doesn't need progressive scan!
Again, these are general guidelines, not firm rules. However, I will state one fairly firm rule: never use 30p if you think you might go out to film or convert to PAL. In these cases stay with 60i or use one of the 24p modes. DVFilm in Austin says they'll give you good looking 24fps film outs from 30p footage, but it's much harder to do so and get clean motion, and even DVFilm recommends against it. I don't know any other film-out place that will even try.
_______________
Adam Wilt writes the following on the DVX100/100A's shooting modes (reprinted here without permission):
How is 24p recorded? What's the difference between 24p and 24p Advanced?
The camera can capture interlaced footage at 59.94 fields/second or 29.97 frames/second ("60i"), 29.97 progressive-scanned frames/second ("30p"), or 24 progressive-scanned frames/second ("24p").
The recorder (and indeed all the video I/O) runs at 29.97 frames/second, interlaced, all the time, regardless of what the camera is doing.
When the camera is running in 60i or 30p, the incoming video is placed into a frame buffer, compressed with the DV codec, and written to tape and/or sent out FireWire. That same frame buffer feeds the analog outputs, one field after the other. Whether the images come in interlaced or progressive, every 30th of a second – well, every 29.97th, but let's keep it simple for now – an image gets written to the frame buffer, and an image gets played out: two fields compressed as a frame for FireWire and recording, and two fields sent out one after the other for interlaced analog I/O.
24p is a bit more complex. [Of course, it's really 23.976p, the same way 30p is really 29.97p and 60i is really 59.94i, and even those numbers are approximations. When the National Television Systems Committee added color to the existing monochrome standard in 1953, they had to fudge either the picture frequency or the audio subcarrier frequency by one part in a thousand, to keep the color subcarrier from interfering with the sound subcarrier. Sound won; NTSC pictures now run at 30 * 1000/1001 fps (and that number is exact), and ensuing generations have had to struggle with the peculiarities of drop-frame time code and resetting their master clocks every day to account for drift. But I digress...] 24 does not fit so cleanly into 30; for every four frames of 24p video that go by, five frames of 30p (or 60i) video go by. Another way to look at it is that 24 frames of 24p video have to be squeezed, stretched, and mashed into 60 fields of NTSC-compatible video; that's four frames of 24p per ten fields of 60i.
There are several ways to do this; the AG-DVX100P and AJ-SDX900 offer two: 24p, which I call 24p Standard, and 24p Advanced.
24p Standard utilizes the same 2:3 or 3:2 pulldown cadence long used to transfer 24fps film to NTSC video. The first 24p frame is written to two fields of 60i video, the next is written to three, the next to two, and the next to three again, as shown in the graphic.
This cadence offers the smoothest possible raw conversion between the frame rates, and gives us the familiar judder of film transferred to tape. There are a couple of things to notice about it:
1) The "A frame" is defined as the only frame of 24p that occupies exactly one whole frame of 60i video with no overlap into adjacent frames.
2) The A, B, and D frames can be recovered by using two fields from the same 60i frame. The C frame cannot be; it is split across field 2 of the third (green) 60i frame and field 1 of the fourth (magenta) 60i frame. In an intraframe-compressed format like DV, that's an important distinction: the two fields of each 60i frame are compressed together, as a frame. If the two fields are very similar, as happens in the red, yellow, and blue frames, the compression uses a comparatively efficient "8x8 DCT" mode. If the fields are very different, as happens in the green and magenta frames when there's a lot of motion between the B, C, and D frames, a less efficient 2x4x8 DCT is used, possibly leading to more image degradation in those frames compared to their red, yellow, and blue companions.
Furthermore, the original A, B, and D frames can be copied in their compressed form from the 60i video data into a new 24p data file, but recovering the C frame requires decompressing the green and magenta frames and recompressing them into a new, 24p DV frame. That puts the C frame a generation down compared to A, B, and D.
If the 24p extraction tool uses white instead of superwhite codec ranges, as Cinema Tools does, it will clip whites in doing so – you really have to preprocess all your footage prior to reverse telecine to pull superwhites and saturated colors into range. Note that this is not necessarily an indictment of such tools: codecs that run in a white-only range instead of a superwhite range can eke a little bit more quality out of the in-range image, at the expense of the out-of-range data. However, if you're using the extended dynamic range that superwhites give you in acquisition, it means you have to add that extra processing step to pull your superwhites back into range before converting to 24p, lest the conversion hard-clip your C frames and remove that creative control from your domain.
24p Advanced uses a syncopated 2:3:3:2 pulldown cadence to stuff 24 frames into 60 fields. It's ever so slightly different in its playback; the standard cadence of 2:3:2:3:2:3:2:3... evenly intersperses the "short" and "long" frames, while 2:3:3:2:2:3:3:2... lumps two "short" frames together followed by two "long" frames. The difference is subtle, but can be seen on smooth pans or on regular in-frame motion, like the passage of a train. In my experience so far, about half the people looking at a 24p Advanced clip can see that the motion is a bit different, and half cannot.
But 24p Advanced isn't intended for making the 60i video look like film; it's designed to allow the best possible recovery of the original 24 frames. You'll note that all four original frames can be recovered from self-contained 60i frames; the green frame in 60i now contains the "extra" B and C fields and can be discarded, since all the information for B is contained in the yellow frame, and all the information for C is in the magenta frame.
Extracting a true 24p clip from a 60i recording simply requires copying the raw data for the red, yellow, magenta, and blue frames into a new 24p file, skipping the green frame altogether. No decompression or recompression is required, and all recovered 24p frames retain first-generation quality. No clipping or other loss is incurred; you are still working with all your frames in their first-generation glory in 24p.
Indeed, Cinema Tools goes one step farther: it gives you the choice of either creating a new 24p file, or simply rewriting the Quicktime frame pointers in the original file to skip over the green frames! Rewriting the header is much faster (not that Cinema Tools is slow in writing a new file), but your "new" 24p file is about 25% larger than it needs to be, since it still contains the discarded frames, and you can no longer go back to that file and play it as a 60i file: Cinema Tools has no "undo" for rewriting the frame pointers.
When to shoot 24p? 24p Advanced? 30p? 60i?
The general rule is to shoot 24p Advanced if you want to extract the original 24 frames/second for a 24fps edit or film-out. Shoot 24p Standard if you are going to stay on video and edit at 30 frames/seconds (60 fields/second, i.e., plain ol' video at NTSC frame rates), without extracting the original 24 frames into a 24fps timeline. In more detail:
Shoot 24p Advanced for:
Post-production using tools that understand Advanced pulldown. 24p Advanced footage can be turned into pure 24p footage more cleanly than 24p standard footage, because every frame in the pure 24p timeline is pulled from a whole frame in the 24p Advanced footage, whereas the C frame in 24p Standard footage is split across two different source frames as discussed above.
Shoot 24p Standard for:
Getting the “film look” on video when you're staying on video and editing at 29.97.
Intercutting with film transfers also using 3:2 pulldown, and staying on video at 29.97.
Working with traditional film-on-tape tools that understand 3:2 pulldown, but not 2:3:3:2 advanced pulldown, when you need to extract the 24p footage for true 24p processing.
Shoot 30p for:
Getting true progressive pictures with a 30 fps frame rate, as when pulling stills for motion analysis.
Working alongside Canons and older Panasonics in Frame Movie Mode, when you want to match their motion rendering.
You want the slightly "filmic" motion of 30p, but don't want to go to 24p, and you aren't concerned about ever going to film or converting to PAL.
Shoot 60i for:
Anything that you want to use as plain ol' video at NTSC frame rates: in 60i, this camera makes pix that look (from a motion-rendering standpoint) just like the pix from any other video camera. In other words, use 60i for everything that isn't supposed to “look like film” and doesn't need progressive scan!
Again, these are general guidelines, not firm rules. However, I will state one fairly firm rule: never use 30p if you think you might go out to film or convert to PAL. In these cases stay with 60i or use one of the 24p modes. DVFilm in Austin says they'll give you good looking 24fps film outs from 30p footage, but it's much harder to do so and get clean motion, and even DVFilm recommends against it. I don't know any other film-out place that will even try.
_______________
Caution
Active member
Re: 24p and Frame mode?
Thanx DC that helped me a lot on understanding the difference between the two, but what should i do when bringing in my footage that is shot 30p on frame mode in to avid xpress pro? AVX pro only gives me options of 30i 24p and 25p ??? so wht do I do? (its progresive footage at 30fps shot on frame mode and i bring it in as 30i ? wouldnt this result in poor quality ?)
???
Thanx DC that helped me a lot on understanding the difference between the two, but what should i do when bringing in my footage that is shot 30p on frame mode in to avid xpress pro? AVX pro only gives me options of 30i 24p and 25p ??? so wht do I do? (its progresive footage at 30fps shot on frame mode and i bring it in as 30i ? wouldnt this result in poor quality ?)
???
Re: 24p and Frame mode?
30p will work just fine in a 30i project and it will not reduce the quality of the footage alredy shot. The only difference will be rendering effects that have and edge and is keframed to move - for example - a picture in picture that is moving across the screen will move in 1/60th increments rather than 1/30th resulting in a smoother move. All blended types of effects such as supers, dissolves, fades. keys will have no difference in quality because of the interlace project type.
Michael
30p will work just fine in a 30i project and it will not reduce the quality of the footage alredy shot. The only difference will be rendering effects that have and edge and is keframed to move - for example - a picture in picture that is moving across the screen will move in 1/60th increments rather than 1/30th resulting in a smoother move. All blended types of effects such as supers, dissolves, fades. keys will have no difference in quality because of the interlace project type.
Michael
