Does the HDR-FX1 give 4:2:2 color?
 

Does the HDR-FX1 give 4:2:2 color? i heard it does but i can not find any info that supports this. please help!!

Close...
 

4:2:0 it's in the HDV spec I believe.

Troy

i looked
can you give me a link?

How about the actual specs? :)
 

Sony link.

Troy

thanks! missed that, even the online support guys couldn't tell me. oh well thanks again!

It is 4:2:0. The same a DVD and PAL.

HDV's 1080i 8 bits, 4:2:0, 25Mbps
HDCAM 1080i 8 bits, 3:1:1, 185Mbps
HDCAM SR 1080i 10-bit, 4:2:2 or 4:4:4, 600Mbps

I wonder what the color specs are of the signal measured from the component outputs when the camera is "live".

This is apparently BEFORE compression, which opens up a whole bunch of interesting questions... is this at a point in the signal path before color space reduction to 4:2:0?

Jim Arthurs

There has been some speculation as to the effective color sample in SD when HDV is ingested as 10-bit uncompressed. Some have maintained that in theory, you could end up with 4:2:2 with output to DVCPRO50 or Digibeta.

Barry Green explored this some time ago, perhaps he could elaborate now that he has the camera.

No answers. It seems fascinating, but I have no way in front of me to test it.

One thing I will say -- the FX1 appears to display the image on the LCD prior to compression, which may or may not be the way other cameras work, I don't know. But the FX1 definitely displays the LCD at a pre-compression stage, because when switching to 4:3 mode it adds "pillarbox" bars on the side, but they're transparent -- you can still see the image, meaning you can still see what's on the unused portion of the CCD. So you're seeing the effects of CineFrame, the effects of picture controls, etc... all before the signal is compressed. I don't know if that equates to the signal that's hitting the component output, or what color resolution or bit depth you're seeing on the LCD. Once HDV gets ahold of it it's decimated to 8-bit 4:2:0, but in DV mode we're clearly seeing the signal before compression and before subsampling the CCD to make the DV signal.

So, it warrants further investigation.