|
|||||||||
|
Thread Tools | Search this Thread |
June 10th, 2011, 07:11 AM | #1 |
New Boot
Join Date: Apr 2011
Location: Johannesburg, South Africa
Posts: 14
|
Example 25i vs 50p
This short video shows the difference between interlaced and progressive video rather well.
|
June 10th, 2011, 08:22 AM | #2 |
Major Player
Join Date: Jan 2005
Location: Ottawa, ON
Posts: 471
|
Re: Example 25i vs 50p
Could you please elaborate. Though the two shots are distinctly different, how do you attribute 'interlace' to the distinction?
Cheers, GB |
June 10th, 2011, 01:24 PM | #3 |
New Boot
Join Date: Apr 2011
Location: Johannesburg, South Africa
Posts: 14
|
Re: Example 25i vs 50p
The two white helicopters in synchronised flight was shot in 25i, (interlaced) meaning 25 "half" frames per second, whereas the single red/white helicopter was shot in 50p (progressive) meaning 50 full frames per second
The props of the two white choppers appear to be standing still as opposed to the realistic spinning motion of that of the red/white chopper. This merely underscores that fast action needs to be shot in progressive mode. |
June 10th, 2011, 06:03 PM | #4 |
Major Player
Join Date: Jan 2005
Location: Ottawa, ON
Posts: 471
|
Re: Example 25i vs 50p
I did wonder if that was what you were referring to, but I can't see how the motion artifact shown can possibly be caused by 'interlace'.
The effect you see in the video where the prop appears 'still' is a common harmonic artifact, and is frequently found in cinema -- the wagon wheel in the western, the props spinning up in the WW2 film -- despite the fact that film is a 'progressive' format. It is not an 'interlace' effect -- it is an effect caused by a shutter speed, and maybe incidently a framerate, but only a framerate in so much as it affect the shutter speed (more correctly shutter angle) selection. In fact, the blurred motion you prefer in the second is the simple consequence of a slower shutter speed -- an effect more easily achieved in an interlaced format than a progressive one, in so much as slower shutter speeds are available. If you shoot -- all else being equal -- 50i and 50p, both will have identical motion behaviours. The interlaced format will have a vertical resolution that is half that of the progressive format, but otherwise nothing changes. A field from the interlace video would match a frame from the progressive one -- though it would be reduced resolution, it would be identical all else. In truth, interlace is often accused of effects it doesn't have. Most certainly 50i and 50p have identical motion handling abilities. Even the oft-cited combing artifacts that are attributable to interlace are really caused not by interlace, but by an improper attempt to force two interlaced fields to display as a single frame -- played back properly, sequentially, the two fields become two half resolution frames and all is well. The helicopter clip you link demonstrates an effect you can achieve with either interlace or progressive by playing with the shutter speed and finding one that either freezes the blades or more likely establish an accidental harmony that creates the illusion that the blades are frozen. Cheers, GB |
June 10th, 2011, 10:31 PM | #5 |
New Boot
Join Date: Apr 2011
Location: Johannesburg, South Africa
Posts: 14
|
Re: Example 25i vs 50p
Geoff, thanks for putting me straight. Your explanation has been highly informative.
The 50p part of the video was my first shoot with the Sony HXR-NX5E. I love the full manual functionality that it offers and I shall be experimenting more with the manual settings - frame rate and shutter speed. |
June 11th, 2011, 12:16 AM | #6 |
Major Player
Join Date: Mar 2011
Location: London, UK
Posts: 353
|
Re: Example 25i vs 50p
Just to be slightly pedantic, 25i is the correct term here for 25 (full) interlaced frames per second. 50i is a misnomer, I'm not aware of any regular camcorders that will shoot 50 frames per second, (i.e.100 fields per second). The same is true for 30i (or 29.97 or whatever).
Steve |
June 11th, 2011, 03:43 AM | #7 |
Major Player
Join Date: Jan 2005
Location: Ottawa, ON
Posts: 471
|
Re: Example 25i vs 50p
Steve, I think you will find you are barking into the wind on that one -- Sony lists the specs for the NX5 very clearly as '50i' or '60i', and that terminology is used across the board. This link to the HDTV primer site uses the terminology as I did:
What exactly is ATSC Cheers, GB |
June 11th, 2011, 04:15 AM | #8 |
Trustee
Join Date: Dec 2007
Location: Glasgow, Scotland
Posts: 1,546
|
Re: Example 25i vs 50p
Didn't the original poster mean to say 50i and 25p?
Can't say I'd ever heard of 25i before and it's not mentioned in the link quoted What exactly is ATSC Anyway Geoff's right about the interference pattern thingy. |
June 11th, 2011, 05:00 AM | #9 |
Major Player
Join Date: Feb 2010
Location: London
Posts: 302
|
Re: Example 25i vs 50p
Hi
The EBU (European Broadcasting Union) uses the frame rate and not the field rate when discussing formats and rates, so 1080/50i would be written as 1080i/25. It's in the wiki at 1080i - Wikipedia, the free encyclopedia I think 1080i/25 doesn't convey the fact you have more temporal resolution than 1080p/25, and to display 1080i/25 on an HD TV sees it being de-interlaced to 50 fps to keep the temporal detail, but that's Europe for you ;-) Regards Phil |
June 11th, 2011, 05:14 AM | #10 |
Major Player
Join Date: Jan 2005
Location: Ottawa, ON
Posts: 471
|
Re: Example 25i vs 50p
There are very few camcorders capable of 50 (or 60) p -- and the NX5 is not one of them. But there are some, and the Panasonic HDC-TM700 is one.
But I suspect you're right -- the original format comparison is probably between 25p and 50i, which likely hints at why the two shots look different: the 25p recording maybe defaulted to a shutter speed of 1/25 and generated the comfortable motion blur, the 50i shot was at 1/50 and 'froze' the blades. Cheers, GB |
June 12th, 2011, 01:05 AM | #11 |
New Boot
Join Date: Apr 2011
Location: Johannesburg, South Africa
Posts: 14
|
Re: Example 25i vs 50p
To clarify:
The first part of the video was shot with a Sony HDR-XR500E at its highest quality setting being 1920x1080 50i (not 25i as stated, I apologise for the error). The second part was shot with the Sony HXR-NX5E set at 1280x720 50p. |
June 12th, 2011, 04:05 PM | #12 |
Trustee
Join Date: Dec 2007
Location: Glasgow, Scotland
Posts: 1,546
|
Re: Example 25i vs 50p
|
June 15th, 2011, 02:16 AM | #13 |
Major Player
Join Date: Mar 2011
Location: London, UK
Posts: 353
|
Re: Example 25i vs 50p
In film and video, we normally talk of frames per second (fps), i.e. 18fps, 24fps for film, 24 (or 23.976) fps, 25 fps, 30 (or 29.97), 50 fps and 60 (59.94) fps for video. Fields are a video-only concept for interlacing. They don't add any visual information to the stream, they just package different parts of it at separate points in time, but there is no more real information there than the frame rate offers. this was first established in 1936 when what became known as System A television was launched in the UK as 405i/25. The EBU notation is much more logical.
The title of this thread is 'Example 25i vs 50p'. The HDTV primer reference posted doesn't even mention 25/50Hz field rates that are being discussed here,- not unusual for US documentation to ignore non-USA based technology. The ATSC table for '50Hz based video titled 'ATSC Standard A/63:1997 (Standard for Coding 25/50 Hz Video)' has a single column for frame rate (in Hz) with field rate added as supplementary information in parentheses. Also, the term '50i' does not appear anywhere in the table. An additional issue is Progressive Segmented Frame video (psf). Here, a true 25 (or 30/29.97) frames per second video stream is split into a 50 field per second interlaced stream for recording or transmission where 25 or 30/29.97 progressive capability is not available. The image temporal sequencing is transparent to the carrier, but a correctly set up display will reconstruct a true progressive video close to the original. This mechanism is used frequently in Europe, (and I assume in the '60Hz' world also). It would be totally inappropriate to call a 25psf stream 50i as that is just a carrier issue and transparent to the viewer. The current popular use of 50i, (meaning 25 interlace frames per second) is probably encouraged by camera manufacturers and vendors as it gives their product a higher number in advertising for those that do not appreciate the mixing of terms and are attracted by big numbers. Steve |
June 15th, 2011, 05:40 AM | #14 |
Major Player
Join Date: Jan 2005
Location: Ottawa, ON
Posts: 471
|
Re: Example 25i vs 50p
Steve, you are entirely correct & I thank you for the detail. My quip was intended only to reference the futility of flying in the face of popular convention -- being right is important, as is being understood. The manufacturers/marketing departments have chosen their own phraseology, and it isn't 'wrong' -- just not right!
Cheers, GB |
June 15th, 2011, 06:40 AM | #15 |
Inner Circle
Join Date: Mar 2003
Location: Ottawa, Ontario, Canada
Posts: 4,220
|
Re: Example 25i vs 50p
Steve I do not think you are correct. Interlace does in fact capture fields not split a frame into two fields. That is why progressive displays have a problem deinterlacing because the fields are exposures taken apart not at the same time. So just adding the fields creates the interlace artifacts everyone sees on a progressive display with simple deinterlacing. The jagged edges are motion differences between the fields. If the fields came from the same progressive frame image there would be no problems in adding them together. This is not the case and results in the difference between simple and expensive deinterlacing displays. The early electo/mechanical television systems were the way you describe though. The fact that the fields are taken at different time intervals means there is more temporal information. That wasn't the case in 1936. The motion characteristics of interlace are similar to motion characteristics of a progressive frame rate at the interlace rate. In my case that is 60i or 60P. I can demonstrate this with my own camera a Sony CX700 which will shoot 60i and 60p. The 60P image has twice the detail but the temporal motion is exactly the same. Displayed on my 240Hz SonyLCD there is little difference between 60i and 60P since the deinterlacing algorithms on the display make both similar. The extra detail can be seen more easily on my Panasonic plasma since it doesn't upscale the interlace image as much as the Sony but the motion is still the same.
Ron Evans |
| ||||||
|
|