Activity › Forums › Sony Cameras › 1080p or 720p video format for a PAL DVD project?
-
1080p or 720p video format for a PAL DVD project?
Posted by John Winski on February 1, 2010 at 2:17 pmHi, I am working on a project for a charity, which is helping build schools and helping to train teachers in Cambodia. I am shooting footage on an EX1R. The footage will be used for DVD training videos to be used in Cambodia (PAL, 230v 50hz). I’d like the option also to use some of the footage for public service announcements on TV in Singapore or in movie theaters (ads prior to beginning of movie). I am editing in FCP6.
Am I best off shooting in 1080p25, or should I shoot in 720p or some other format? Any other important settings to consider?
Thanks in advance for your help!
John Winski replied 16 years, 3 months ago 5 Members · 12 Replies -
12 Replies
-
Alister Chapman
February 1, 2010 at 6:20 pmIt depends on where your priorities are. If the primary use is SD DVD then I would shoot 720P as this tends to down convert with less aliasing. If you want the smooth motion look of 50i video then shoot 720P50. If you want a filmic look then 720P25. However if your primary concern is the quality on a big screen in HD then shoot 1080P25.
It also helps when down converting to soften the image just a tiny bit. Turning down the detail level in the picture profiles to -10 greatly reduces aliasing in both HD and SD with only a minimal effect on the HD images apparent sharpness.
Alister Chapman
http://www.xdcam-user.com -
Michael Slowe
February 1, 2010 at 6:27 pmAlister, you don’t think that shooting 1080i is a good idea at all?
Michael Slowe
-
Alister Chapman
February 1, 2010 at 6:40 pmNot for general software down conversion to SD. As the math doesn’t work well with odd numbers of lines in each field being discarded it can lead to nasty jaggies. Some software converters will make a total hash of down converting fields. I find that 720P50 is the most reliable way to get good 50i SD.
When down converting to SD having too much detail can be your enemy. The more detail you start off with the bigger the problem becomes. In HD if you have a gentle transition from light to dark over 4 pixels when you down convert that to SD, if your not careful you end up with an instant transition from light to dark over just 2 pixels. If there is any movement this then becomes pixels flashing on and off, it doesn’t look good and leads to jaggies and aliasing etc. A lot depends on the way the downconversion is done and whether it incorporates anti-aliasing or predictive/adaptive scaling.
Alister Chapman
http://www.xdcam-user.com -
Rafael Amador
February 2, 2010 at 5:13 amI agree with Alister.
Normally i end up in DVD and i shoot 720p25.
In theory is the flavor that allow more quality (data per pixel).
Soot interlaced only if so have been required.
rafael -
John Winski
February 2, 2010 at 6:02 amThank you both very much for the advice! I’ll go with 720p25!
-
Michael Slowe
February 2, 2010 at 10:38 amSo, Alister and Rafael, if one is intending to end up with SD DVD’s is there any point in shooting HD at all? I’ve been finding though that since I started shooting HD my down converted SD pictures are better. I either convert within Media 100 or, more often now, with the greatly improved BitVice. Although my material is interlaced the BitVice down convert must have the anti aliasing that Alister refers to because I now have very little trouble, used to though with previous down converts.
Michael Slowe
-
James Hilton
February 2, 2010 at 12:53 pmI have done it both ways, and unless you want some extra work, go with 720p50.
If you use 1080i25 you are best off using motion estimation to recreate 1080p50 then downcovert that to SD, it works well.
If you have 720p50 just convert to PAL SD 25i.
What you use to resize makes a big difference, I have actually found an AviSynth script with lanczos resize working on an uncompressed avi can often result in less aliasing than getting FCP or Premiere to do the same resizing for you…..it is a lengthy pc only option if you are having trouble with aliasing in your video though.
In my experience results usually look better if you resize the output from a good HD camera & codec than shooting with a good SD camera form the start, technology has moved on and downsized picture quality from your EX1R will potentially beet the majority of SD cameras out there.
-
Rafael Amador
February 2, 2010 at 4:36 pmHi Michael,
We start always with the 1920 x 1080 pixels of the CMOS, and whatever the workflow to SD, you have to downscale somewhere. and you know downscale is a critic operation.
Nowhere else (software, hardware solution, Teranex..) you will get the downsizing that you get in camera: at 12 or 14b and from a pristine-untouched signal right from the CMOS.
About quality, the best option would be shooting directly in SD, of course you would lose the possibilities of playing with a bigger picture.
720p is a middle point. You let the half of the downscale, but keep some room to re-frame, pan, zoom, etc. And as I said, 720p25, is the setting that offers more info/pixel.
720p50 have no much meaning for me: You end up trowing away half of the picture. Only if you want to make an slow motion without losing the audio.
rafael -
Alister Chapman
February 2, 2010 at 7:41 pmMy experience is that a high end SD camera such as a Digibeta etc can produce an image equal to a very well downconverted HD camera. The SD camera is optimised for SD, it has the appropriate optical low pass filtering for SD and the noise can be lower due to the larger pixels. However well done HD downconverted to SD should look very good.
The advantage to shooting HD is that you have so much more real estate to work with. This gives you the ability to crop in to the image or re-frame, it also helps future proof your rushes. It’s particularly useful for chroma key work where any small keying errors are less noticeable as they tend to be smaller. I shoot pretty much everything HD these days, downconverting when I need SD.
Alister Chapman
http://www.xdcam-user.com -
John Winski
February 2, 2010 at 11:04 pmHi Alister, I sent a note to Sony to ask them about the formats and to better understand the new format naming convention used in the ex1r. Here is their response: Unlike HDV and similar formats, the EX1R is actually a progressive imager and recording, you will not have any more aliasing in 1080 than you would have in 720. Shooting in 1080 will give you more temporal resolution and depth of modulation, which will actually yield a better down conversion.
“Unlike HDV and similar formats, the EX1R is actually a progressive imager and recording, you will not have any more aliasing in 1080 than you would have in 720. Shooting in 1080 will give you more temporal resolution and depth of modulation, which will actually yield a better down conversion. New to the EX1R are formats that were not previously available on the other EX cameras. This required a change in the naming of the formats. For example, in HQ mode, you can shoot in 1920 x 1080 or in 1440 x 1080 resolutions in 60i, 30P, and 24P. To differentiate these modes, they decided to use the horizontal resolution for naming instead of the vertical resolution as other cameras do. So now you see HQ 1920/60i and HQ 1440/60i. When you see a 1280/50i, that indicates you are recording at 1280 x 720 @ 50i.”
Would you agree on the aliasing issue?
Reply to this Discussion! Login or Sign Up