Videomaker – Learn video production and editing, camera reviews › Forums › Cameras and Camcorders › Professional Camcorders › 2x Z5U – Same Camera – Same Settings – Bad color matching
- December 1, 2010 at 9:28 PM #48877
I have had my 2 Z5s for a while, and while the picture between the 2 cameras always seemed a bit off even though every setting was completely identical, I just assumed this was caused by different angles in the cameras for my shoots. For the most part, I was able to fix this in post to at least be watchable.
After I did a test shoot for a new TV show, everything seemed to be much more out of whack. I had similar angles, yet the the color matching between the 2 cameras was pretty bad.The browns where especially bad, and I am having a heck of a time trying to balance everything in post.
Both cameras have all the exact same settings. I have tried resetting it to factory defaults, and then reloading a profile saved to a memory stick and transferred the exact same profile to the other camera.
This morning I did a controlled test to see exactly what was going on. I printed out a SMPTE color bar chart and put it on the wall. I then placed each camera one at a time and tried a few test shots.
First I started with -3db gain. Most the colors seemed to match pretty well, but if you look at the blacks, one of them has a bit of a red tinge and the other has a bit of a green tinge to it (compared to the other camera). As I added gain, ALL the colors seemed to becomre more and more different. I know I will never shoot anything with 21db gain if I can ever help it, but this shows what is going on with my camera.
I know there are cameras with Black Balance, but the Z5U does not have that option as far as I can tell, and I am not even sure if that is the issue.
It is quite irritating that I have 2 exact same pro cameras that I paid a pretty hefty chunk of $ for, with every single setting set identical, but yet I still wrestle with color matching. This makes it especially difficult to edit a multi camera shoot as I cannot just universally lay down some color correction and color curve changes to taste. I know there will always be some tweaking here and there between shots but having to go to each and every shot and re-adjust it is painfully time consuming. Vegas removes all previous effects added to a track when you combine for multi camera editing , and from experience if you re separate the tracks afterwards, the transitions don’t seem to work 100% correctly if there are some color correction or curve changes.
Here is the video I shot. I put a mask of the 2nd camera in the middle – you should be able to tell which is which!
NOTE: Even though I have the later shots labled as color corrected, it was VERY basic – and not fully tweaked. I just used the dropper tool from Vegas to pinpoint the white, the grey and the black.
When you have this problem, it’s best to get your cameras matched up before the shoot. If you have access to a waveform monitor then it’s just a matter of tweaking the settings until they match. If you don’t, using a monitor like we discussed in another post you’ll just have to tweak your second camera’s settings and ‘eyeball’ them as close as possible. An exact match would be great. Close ‘enough’ will work but it will mean some tweaking during color correction time in post.
Thanks again. Can SonyVegas do this live? (I don’t have my camera with me at the moment). I guess that means diving into the picture profile settings which scares me a bit. Probably just messing with the blacks somewhere?
I am still unsure how to use a waveform monitor to tweak colors. All I have used them for in the past is to adjust/monitor IRE values.
I don’t know if Vegas can do live video calibration. If it can, take a look at these pages to see where your signals should be when using a waveform monitor and a vectorscope.
I usually use either an old copy of DV Magic or it’s reincarnation On-Location to calibrate cameras in the field.
Did a quick overview and will read it thoroughly tomorrow. But from what I gather it mostly talks about the color bars from the equipment. If I take color bar samples from both my cameras, they are exactly the same (when produced from the camera) and fit the patterns I see on those pages. Something is happening with my recorded signal. The color bars I posted in the video were from a printed out chart on my wall with the camera aimed at it, and not produced from the camera itself.
A thought. It looks like these record to tape. If I am wrong just ignore this post.
Could this be an artifact of the tape recorders themselves? A test that might be worth doing is a repeat of your color bar test but capturing the video directly off camera to pc bypassing the tape. This might further prove if it is a sensor calabration issue or recorder issue.
The flipside would be to record the camera generated color bars to tape and see if the colors match.