Less Rez in 16:9 in camera.
- Thread starter Elvis
- Start date
Z
Zoomforce
Guest
Re: Less Rez in 16:9 in camera.
it all depends if the camera crops it before or after it compresses it
it all depends if the camera crops it before or after it compresses it
Re: Less Rez in 16:9 in camera.
Actually there are a number of reason why 16:9 in-camera (squeeze mode) is superior to letterbox cropped in post. I've posted here about that a while ago, with some discussions concerning whether or not this could be true, but more or less it was left undecided by many (not me of course ;D). Anyway, answering a post over on 2-pop regarding this same issue I was asked by our beloved Barry Green to do a comparison shot. Not able to do exactly what he wanted because of the time of year (leaves on trees in Colorado ;D), I just did some frame grabs with photoshop using firewire.
Rather then go into the whole thing again here, might just be easier to go to 2-pop (sorry Jarrad) and take a look.
Here's the link: http://www.uemforums.com/2pop/ubbth...r=26304&page=0&view=collapsed&sb=5&o=&fpart=1
Hope this helps any who are wondering which to use for 16:9 shooting.
-Rodger
Actually there are a number of reason why 16:9 in-camera (squeeze mode) is superior to letterbox cropped in post. I've posted here about that a while ago, with some discussions concerning whether or not this could be true, but more or less it was left undecided by many (not me of course ;D). Anyway, answering a post over on 2-pop regarding this same issue I was asked by our beloved Barry Green to do a comparison shot. Not able to do exactly what he wanted because of the time of year (leaves on trees in Colorado ;D), I just did some frame grabs with photoshop using firewire.
Rather then go into the whole thing again here, might just be easier to go to 2-pop (sorry Jarrad) and take a look.
Here's the link: http://www.uemforums.com/2pop/ubbth...r=26304&page=0&view=collapsed&sb=5&o=&fpart=1
Hope this helps any who are wondering which to use for 16:9 shooting.
-Rodger
Z
Zoomforce
Guest
Re: Less Rez in 16:9 in camera.
Jan from Panasonic Confirmed with me this morning that the bars are added to the footage Before compression, so that is the way to go. The reason is, well I will repost my 2pop post here:
"Ok.. just to confuse people more...
DV and uses a lossy DCT codec. There are 8bit blocks with preset allocation to compress. This means that compression is confined per 8pixel blocks, depending on your codec either 8x4 or 8x8 blocks.
However, the new DV codec breakes the frame into 270 pieces, and divides those by 6(if I remember correctly), creating macroblocks. The algorithim can in fact look at those 6 areas of each video piece, and decide how much compression real-estate that piece needs to be artifact free. It then can take those savings and share them with the next video piece. This can be trouble if your codec is crap, as of the 2560 bits of info your frame has, If it decides all video pieces only need little compression, you can end up with only 1000bits of data after the algorithim,with the rest of the 2460bits of space empty which can be noticable, but nowadays is very rare.
Jarred Land
"
Jan from Panasonic Confirmed with me this morning that the bars are added to the footage Before compression, so that is the way to go. The reason is, well I will repost my 2pop post here:
"Ok.. just to confuse people more...
DV and uses a lossy DCT codec. There are 8bit blocks with preset allocation to compress. This means that compression is confined per 8pixel blocks, depending on your codec either 8x4 or 8x8 blocks.
However, the new DV codec breakes the frame into 270 pieces, and divides those by 6(if I remember correctly), creating macroblocks. The algorithim can in fact look at those 6 areas of each video piece, and decide how much compression real-estate that piece needs to be artifact free. It then can take those savings and share them with the next video piece. This can be trouble if your codec is crap, as of the 2560 bits of info your frame has, If it decides all video pieces only need little compression, you can end up with only 1000bits of data after the algorithim,with the rest of the 2460bits of space empty which can be noticable, but nowadays is very rare.
Jarred Land
"
Re: Less Rez in 16:9 in camera.
Jarred are you saying it will be a higher rez if I do it in camera then? Because the codec will look at the bars and not put a lot of compression on them? I have to say I'm not a tech head and this type of stuff gives me a headache. But thanks for the info!
Jarred are you saying it will be a higher rez if I do it in camera then? Because the codec will look at the bars and not put a lot of compression on them? I have to say I'm not a tech head and this type of stuff gives me a headache. But thanks for the info!
Re: Less Rez in 16:9 in camera.
[quote author=Elvis link=board=Events;num=1080232816;start=0#6 date=03/25/04 at 10:40:01]Jarred are you saying it will be a higher rez if I do it in camera then? Because the codec will look at the bars and not put a lot of compression on them? I have to say I'm not a tech head and this type of stuff gives me a headache. But thanks for the info![/quote]
There was a good deal of debate concern just that when the DVX100 was just realease... Actually, I think it was even before it was released and just a bit after showing at WEVA in 2002. On 2-pop I can remember going into a long spiel why it should be better using in-camera letterbox .vs cropping 4:3 in post. Much for the same reasons that DCT made allowances for modified algorithms to make use of adjacent macroblocks for added compression information when data space was available.
No one knew the exact answer then, much as now. It was just surmised that because this *could* be done that maybe it *was* being done. Still a gray area, but the fact that a solid color will compress in less space, it only seems logical that at the very least the codac will have an easier (more) time to work on the active area and should give better results.
-Rodger
[quote author=Elvis link=board=Events;num=1080232816;start=0#6 date=03/25/04 at 10:40:01]Jarred are you saying it will be a higher rez if I do it in camera then? Because the codec will look at the bars and not put a lot of compression on them? I have to say I'm not a tech head and this type of stuff gives me a headache. But thanks for the info![/quote]
There was a good deal of debate concern just that when the DVX100 was just realease... Actually, I think it was even before it was released and just a bit after showing at WEVA in 2002. On 2-pop I can remember going into a long spiel why it should be better using in-camera letterbox .vs cropping 4:3 in post. Much for the same reasons that DCT made allowances for modified algorithms to make use of adjacent macroblocks for added compression information when data space was available.
No one knew the exact answer then, much as now. It was just surmised that because this *could* be done that maybe it *was* being done. Still a gray area, but the fact that a solid color will compress in less space, it only seems logical that at the very least the codac will have an easier (more) time to work on the active area and should give better results.
-Rodger
Re: Less Rez in 16:9 in camera.
Shouldn't shooting in Thin mode while using16x9 in camera be of the utmost in importance if concerned with a high res look? It appears to have made quite a difference in res in my "on the water footage versus my Thick mode setting footage footage!
Shouldn't shooting in Thin mode while using16x9 in camera be of the utmost in importance if concerned with a high res look? It appears to have made quite a difference in res in my "on the water footage versus my Thick mode setting footage footage!
Barry_Green
Moderator
Re: Less Rez in 16:9 in camera.
Using thin mode, in combination with in-camera 16:9 (squeeze) mode, would be of utmost importance, yes.
Using thin mode, in combination with in-camera 16:9 (squeeze) mode, would be of utmost importance, yes.
Re: Less Rez in 16:9 in camera.
Maybe this is too simple...
Shoot something in all modes and compare. What do your eyes tell you? What is the easiest way to work with the footage in post? Make decisions based off of those and make tradeoffs where necessary. I realize you are just trying to determine what's the best...But (IMHO) I wouldn't get hung up on it and just get out there and test/shoot.
For me, I use Squeeze mode, vertical detail to thin, 24PA and run to an external monitor that does 16:9 vertical squeeze so I can view it in the correct aspect ratio while shooting. I edit in FCP4 on a 23.98 anamorphic preset.
Andrew
Crystal Clear Media
Maybe this is too simple...
Shoot something in all modes and compare. What do your eyes tell you? What is the easiest way to work with the footage in post? Make decisions based off of those and make tradeoffs where necessary. I realize you are just trying to determine what's the best...But (IMHO) I wouldn't get hung up on it and just get out there and test/shoot.
For me, I use Squeeze mode, vertical detail to thin, 24PA and run to an external monitor that does 16:9 vertical squeeze so I can view it in the correct aspect ratio while shooting. I edit in FCP4 on a 23.98 anamorphic preset.
Andrew
Crystal Clear Media