24F and The Progressive Look at DVinfo.net

Barry Gribble · December 22nd, 2006, 11:14 AM

I've heard a lot thrown around about CCUs and encoding... but can anyone tell me if the 24f frames look 100% progressive? Can you visually tell the difference in a frame of a moving subject between a 24f frame and a "true" progressive frame?

Thanks in advance.

Raymond Toussaint · December 22nd, 2006, 11:36 AM

The motion blur you see in a (F) frame looks progressive, you don't see the 'jaggies' you normaly see on an interlaced field. It is shutter dependend.

Chris Hurd · December 22nd, 2006, 11:46 AM

Not only do the 24F frames look 100% progressive, they capture that way as well. Apple's Final Cut Pro, Adobe Premiere Pro, and Grass Valley / Canopus Edius all use 24P settings in order to capture 24F. They can't detect any difference between Frame mode and progressive scan, and neither can you. For all practical purposes, Frame mode *is* progressive scan.

Robert Sanders · December 22nd, 2006, 01:23 PM

Man, people need to get over this hang up with the "F".

The single biggest mistake Canon made was not using the "P" moniker like Sony did with their XDCAM products (which also use interlaced CCDs clocked at 48Hz and progressive scanned).

Bill Pryor · December 22nd, 2006, 01:50 PM

As I understand it, Sony's new V1 derives progressive frames from interlace chips much the same way Canon does. Not the same way, but they get P from I, which they call P, not F. It's the same damn thing. I'm shooting 24F(P), capturing that way in FCP, and I marked an in point, counted off 24 frames and set an out point and guess what: it's exactly 1 second. You shoot at 24F(P) and capture in HDV at 24P, you get 24 frames. It is progressive. It's not the pulldown thing with blurred frames between. It is progressive. It is 24 frames per second, it is not interlaced. Canon calls it F because they get progressive from interlace chips; Sony gets progressive from interlace chips and calls it P. Same thing.

Actually...one more thing. Here's a suggestion for somebody to make some money: Write a program that people can buy for 10 bucks. When they install it, every time there's an F after the number 24 on any web site, it automatically turns the F into a P. That should solve the problem once and for all.

Barry Green · December 22nd, 2006, 01:52 PM

V1 doesn't work that way; XDCAM-HD does, but the V1 is apparently progressive all the time.

But here's the bottom line -- I took the all-progressive all-the-time HVX, and the 24F-from-interlaced XHA1, put 'em side-by-side, split-screened the footage, single-frame-stepped through it, and -- 24F delivers 24P motion. It delivers the exact same motion cadence and the same motion signature.

Barry Gribble · December 22nd, 2006, 01:56 PM

Thanks Bill and Barry also... wow traffic is moving fast on this one.

I had seen a lot of threads go by on the codec and CCUs, none that were purely subjective on motion, though... and it's a tough thing to search for.

Bill Pryor · December 22nd, 2006, 02:07 PM

I knew the XDCAM HD did but I also thought the V1 used interlace chips. I stand corrected there. But XDCAM HD also says P, so the point is still valid. F is P and P is F and it's all the same.

Tom Roper · December 22nd, 2006, 03:12 PM

If the FX7 is using the same CMOS sensors as the V1U, is it to say that a 100% interlaced FX7 is using 100% progressive sensors? That would be odd.

If the CMOS sensors are bit addressable as stated, then you could scan any direction, even from the center to the corners. As a term, progressive scanned could be as outdated to describe the Sony CMOS sensors as 24F is to describe Canon's P mode.

The salient question is whether Sony's 24P manages to not lose 10.6% vertical resolution compared to 60i as happens with the Canon.

That question could be answered if someone would shoot an ISO 12233 test chart and email me (or post) the frame grabs, I will run it through the Imatest MTF50 and post the results.

Brent Graham · December 22nd, 2006, 03:13 PM

I like to kid that Canon calls it 'F' rather than 'P' because theres no 'rounding'...
get it...
an 'F' is a 'P' without the round part...
get it...?
ahahahahahah...
I am so funny it hurts.

Chris Hurd · December 22nd, 2006, 03:39 PM

I'm looking at Brent's reply just above this one, and I'm looking at Tom's reply just above his, and I'm thinking, you know, we have a pretty diverse group of people on this site. Which is always a good thing. Happy Holidays, everyone!

Barry Gribble · December 22nd, 2006, 03:41 PM

LOL, very un-grinchly of you Chris, but I do agree. And Brent's response was precisely the type of subjective analysis I was looking for :O.

Thanks as always for your hard work on the site.

Chris Hurd · December 22nd, 2006, 04:05 PM

Heh. Getting Cindy Lou a glass of water was your first clue that the Grinch was redeemable after all.

Soeren Mueller · December 24th, 2006, 04:24 PM

To be honest... Barry is the only one that I'd trust on this issue.... on the contrary any mod from the Camcorderinfo board would be the least person I'd trust on anything in regard to interlace... they still claim that you can deinterlace footage that was captured interlaced and retrieve perfectly the same progressive frames with the same resolution as when you'd have captured progressive frames in the first place.. bruawhahahaha :->

and lol @ diversity

December 22nd, 2006, 11:14 AM	#1
Barry Gribble Major Player Join Date: Aug 2003 Location: Los Angeles, CA Posts: 817	24F and The Progressive Look I've heard a lot thrown around about CCUs and encoding... but can anyone tell me if the 24f frames look 100% progressive? Can you visually tell the difference in a frame of a moving subject between a 24f frame and a "true" progressive frame? Thanks in advance. __________________ Barry Gribble Integral Arts, IMDB

December 22nd, 2006, 11:46 AM	#3
Chris Hurd Obstreperous Rex Join Date: Jan 2001 Location: San Marcos, TX Posts: 27,368 Images: 513	Not only do the 24F frames look 100% progressive, they capture that way as well. Apple's Final Cut Pro, Adobe Premiere Pro, and Grass Valley / Canopus Edius all use 24P settings in order to capture 24F. They can't detect any difference between Frame mode and progressive scan, and neither can you. For all practical purposes, Frame mode is progressive scan. __________________ CH Search DV Info Net \| 20 years of DVi \| ...Tuesday is Soylent Green Day!

December 22nd, 2006, 01:23 PM	#4
Robert Sanders Starway Pictures Join Date: Jul 2005 Location: Studio City Posts: 581	Man, people need to get over this hang up with the "F". The single biggest mistake Canon made was not using the "P" moniker like Sony did with their XDCAM products (which also use interlaced CCDs clocked at 48Hz and progressive scanned). __________________ http://www.starwaypictures.com http://starwaypictures.blogspot.com/

December 22nd, 2006, 01:56 PM	#7
Barry Gribble Major Player Join Date: Aug 2003 Location: Los Angeles, CA Posts: 817	Thanks Bill and Barry also... wow traffic is moving fast on this one. I had seen a lot of threads go by on the codec and CCUs, none that were purely subjective on motion, though... and it's a tough thing to search for. __________________ Barry Gribble Integral Arts, IMDB

December 22nd, 2006, 03:39 PM	#11
Chris Hurd Obstreperous Rex Join Date: Jan 2001 Location: San Marcos, TX Posts: 27,368 Images: 513	I'm looking at Brent's reply just above this one, and I'm looking at Tom's reply just above his, and I'm thinking, you know, we have a pretty diverse group of people on this site. Which is always a good thing. Happy Holidays, everyone! __________________ CH Search DV Info Net \| 20 years of DVi \| ...Tuesday is Soylent Green Day!

December 22nd, 2006, 11:36 AM	#2
Raymond Toussaint Regular Crew Join Date: Nov 2005 Posts: 173	The motion blur you see in a (F) frame looks progressive, you don't see the 'jaggies' you normaly see on an interlaced field. It is shutter dependend.

December 22nd, 2006, 01:50 PM	#5
Bill Pryor Contributor Join Date: Jan 2003 Location: Kansas City, MO Posts: 4,449	As I understand it, Sony's new V1 derives progressive frames from interlace chips much the same way Canon does. Not the same way, but they get P from I, which they call P, not F. It's the same damn thing. I'm shooting 24F(P), capturing that way in FCP, and I marked an in point, counted off 24 frames and set an out point and guess what: it's exactly 1 second. You shoot at 24F(P) and capture in HDV at 24P, you get 24 frames. It is progressive. It's not the pulldown thing with blurred frames between. It is progressive. It is 24 frames per second, it is not interlaced. Canon calls it F because they get progressive from interlace chips; Sony gets progressive from interlace chips and calls it P. Same thing. Actually...one more thing. Here's a suggestion for somebody to make some money: Write a program that people can buy for 10 bucks. When they install it, every time there's an F after the number 24 on any web site, it automatically turns the F into a P. That should solve the problem once and for all.

December 22nd, 2006, 01:52 PM	#6
Barry Green Barry Wan Kenobi Join Date: Jul 2003 Location: North Carolina Posts: 3,863	V1 doesn't work that way; XDCAM-HD does, but the V1 is apparently progressive all the time. But here's the bottom line -- I took the all-progressive all-the-time HVX, and the 24F-from-interlaced XHA1, put 'em side-by-side, split-screened the footage, single-frame-stepped through it, and -- 24F delivers 24P motion. It delivers the exact same motion cadence and the same motion signature.

December 22nd, 2006, 02:07 PM	#8
Bill Pryor Contributor Join Date: Jan 2003 Location: Kansas City, MO Posts: 4,449	I knew the XDCAM HD did but I also thought the V1 used interlace chips. I stand corrected there. But XDCAM HD also says P, so the point is still valid. F is P and P is F and it's all the same.

December 22nd, 2006, 03:12 PM	#9
Tom Roper Trustee Join Date: Jun 2004 Location: Denver, Colorado Posts: 1,891	If the FX7 is using the same CMOS sensors as the V1U, is it to say that a 100% interlaced FX7 is using 100% progressive sensors? That would be odd. If the CMOS sensors are bit addressable as stated, then you could scan any direction, even from the center to the corners. As a term, progressive scanned could be as outdated to describe the Sony CMOS sensors as 24F is to describe Canon's P mode. The salient question is whether Sony's 24P manages to not lose 10.6% vertical resolution compared to 60i as happens with the Canon. That question could be answered if someone would shoot an ISO 12233 test chart and email me (or post) the frame grabs, I will run it through the Imatest MTF50 and post the results.

December 22nd, 2006, 03:13 PM	#10
Brent Graham Major Player Join Date: Oct 2006 Location: San Antonio, Texas Posts: 222	I like to kid that Canon calls it 'F' rather than 'P' because theres no 'rounding'... get it... an 'F' is a 'P' without the round part... get it...? ahahahahahah... I am so funny it hurts.

December 22nd, 2006, 03:41 PM	#12
Barry Gribble Major Player Join Date: Aug 2003 Location: Los Angeles, CA Posts: 817	LOL, very un-grinchly of you Chris, but I do agree. And Brent's response was precisely the type of subjective analysis I was looking for :O. Thanks as always for your hard work on the site. __________________ Barry Gribble Integral Arts, IMDB

December 22nd, 2006, 04:05 PM	#13
Chris Hurd Obstreperous Rex Join Date: Jan 2001 Location: San Marcos, TX Posts: 27,368 Images: 513	Heh. Getting Cindy Lou a glass of water was your first clue that the Grinch was redeemable after all. __________________ CH Search DV Info Net \| 20 years of DVi \| ...Tuesday is Soylent Green Day!

December 24th, 2006, 04:24 PM	#14
Soeren Mueller Regular Crew Join Date: Mar 2004 Location: Düsseldorf, Germany / Denver, CO Posts: 137	To be honest... Barry is the only one that I'd trust on this issue.... on the contrary any mod from the Camcorderinfo board would be the least person I'd trust on anything in regard to interlace... they still claim that you can deinterlace footage that was captured interlaced and retrieve perfectly the same progressive frames with the same resolution as when you'd have captured progressive frames in the first place.. bruawhahahaha :-> and lol @ diversity