Anyone noticed a weird interframe flicker/strobe on 7D panning shots?

powervideo

powervideo

Well-known member
It's not progressive flicker, but I see a slight movement/flicker "within" the frames when panning, even panning very slowly. It's not a solid progressive frame strobe (if that's the right description) like I get with Varicam or an SDX900. It's something else, and quite distracting. I've been testing at 50th sec 1080/25 ISO100, picture setting standard. A friend with a 5D remarked on the same thing with his 5D.

Peter
 
Just racing off for a three day shoot, but I will test more and post footage if I can't get to the bottom of it.
 
There's an issue where the 5D drops frames on slow pans that's been discussed over cine5D. Mine does it and I've yet to find a way around it. A few people have had luck resetting the camera to default or re-initializing their CF cards but it hasn't worked for me. A friend just got a 7D so I'll run some tests and see if it does the same thing.
 
squig said:
There's an issue where the 5D drops frames on slow pans that's been discussed over cine5D. Mine does it and I've yet to find a way around it. A few people have had luck resetting the camera to default or re-initializing their CF cards but it hasn't worked for me. A friend just got a 7D so I'll run some tests and see if it does the same thing.
Click to expand...


It's not dropping frames that's for sure. I'm using Sandisk Extreme IV cards. It's like a field delay issue within the frame - a sort of tiny time delay like interlace has within a 50i frame rate. It definitely doesn't look like DVCPROHD progressive. I'm watching on Sony L5 HD CRT monitor via SDI so issues like this seem to show up more than flat-panel monitoring.

The pictures look great and I'm very happy with the colormetrics and sharpness, but I'm sure there is a pecularity in the camera's technology that gives this effect on camera motion. It could the CMOS or rolling shutter related. It's not noticable at all on static shots so life goes on. I think I've become a pixel-peeper.:Drogar-Shock(DBG):

Peter
 
Ive been struggling with this - it most definitaly does not look good at some pan speeds..

Are you used to shooting 25p ??

Theories possibly all wrong..

-------
The 25p theory..

50i smears the frame and pans look good

30p has a short enough duration for pans to look good

25p can cause strobing in pans due to frame rate

Read up on panning speed, keep slow and all will be fine

------

The tripod and handling theories

Lack of mass on my miller arrow 25 designed for 25lb cams does not have enough inertia to get the fluid head working

Adittionally pulling focus my brain cant do two things so I judder my pans while pulling focus

----------

Camera is flawed theory

It does something when you load it with DATA

-----------

Viewing theory

Maybe some phase problem between frame rate and LCD monitor refresh rate

-------------

Could be any or all of the above

--------------

ALso I tested the camera in my office when I bought it just panning across my files on the wall - looked terrible

Actually using the camera with a person in the frame looks better

Just my thoughts

Try some pans at 50p and see the difference

S
 
I think it's just rolling shutter Peter. Some people are very sensitive to it and see it even in very slow camera motion or even when all motion is in frame and the camera is not moving. There is no visible distortion, but the frame is not as solid as the output of a CCD camcorder. This is natural because the rolling shutter is always active. A fast pan can produce a visible 5 or 10 degree shear, but even a 0.5 degree shear can be perceived as an unstable frame.
 
morgan_moore said:
Ive been struggling with this - it most definitaly does not look good at some pan speeds..

Are you used to shooting 25p ??
S
Click to expand...


All I shoot is broadcast 25P. I'm definitely shooting 25P. There is something that doesn't look just right. I don't think it's a fault in the camera, I think it's something else. Probably rolling shutter as Otis says.
 
Otis Grapsas said:
There is no visible distortion, but the frame is not as solid as the output of a CCD camcorder.
Click to expand...

i think this is it Otis. That describes it perfectly - the frame is not as "solid" as a CCD-based camcorder. I will work around it but it is something to be very aware of. I was about to shoot a commercial on the 7D tomorrow, but we've bailed and will be shooting on the 2700 Varicam. I just need to do some more infield testing on my own in-house stuff before I risk it on a shoot with a celebrity I won't be able to reshoot with.
 
morgan_moore said:
what is your 'normal camera'

my suggestion is to build a rig so that the 7d sits on top and do some tests
S
Click to expand...

I own a SDX900 with HJ21 HD, Sachtler 18P and the whole enchilada, but shoot with rentals for HD. For the 7D I have Zacuto Marksman, Z-Finder, Zeiss primes, etc,.
 
Barry_Green said:
My guess is that you're seeing the effects of the aliasing, but we won't know until you post footage.
Click to expand...

It's not aliasing Barry. I'm definite on that. I'm sure it's a CMOS/rolling shutter related issue. I'm away on shoot but will post footage ASAP. I find it strange that besides Otis no-one else is seeing this. If most guys here are doing webvids, they may not be experiencing my observations. It's night and day to me. Maybe it's because I'm viewing on a broadcast monitor via HD-SDI, but it sure as hell doesn't look like 25P prgressive from Varicam or XDCAM HD.

But as I said earlier, I'm not dissing the camera. There's lots to love, but this is an issue that affects the way one shoots for broadcast - no question.

Peter
 
Could be the aliasing biting in and rounding off details/edges unevenly between frames in slow pans. The faster you pan the more motion blur will smear out the image thereby lessen the effect of aliasing.
 
No it's not smear or aliased stepping. It more like a mild field reversal (but it's not) where the playback flickers. As I said it's subtle but not as solid as a CCD capture. I'll experiment with some other frame rates and 720 to see if it's a 25P thing.
 
powervideo said:
I find it strange that besides Otis no-one else is seeing this.
Click to expand...

I would say 90% of the working TV professionals that have only used shoulder mount CCD cameras notice it. Only the description varies. Only a tiny percentage in this sector is CMOS but those who tried it say the image is uneasy, some say it's aliasing or use other (completely unrelated) descriptions. You need to be aware of it, adjust your technique and rethink the camera motion. You can get used to it but if you mount two cameras on the same plate, one CCD and CMOS and start shooting, it is quite obvious that the CCD produces a very different motion quality. IMHO that's the main problem with CMOS. You can easily avoid the nightmare scenarios that produce wobble and large distortions, but you cannot avoid the rolling shutter character. When the camera moves, CCD gives you reality in motion, film is a little synthetic which could be desired in some cases, but CMOS is visibly artificial. The public is used to it by now because of the CMOS mobile phones and consumer camcorders, so it will not affect the market.

Nothing new. When audio CD came out, the consumers were very happy with the fast deep bass and the frequency range, but audiophiles noticed that the mastering and overall sound quality was junk. DSP improved n the following years and now we have 16/44 CD remasters that compete well with the analog originals. Today people get used to compressed recordings and low audio quality, so companies can produce TV sets with walkie talkie audio quality in 5.1 surround and nobody complains as long as the TV is thin and matches the decoration. Not many people care about the "little" details. Panasonic still have CCD consumer cameras though. There is still hope:)
 
You must log in or register to reply here.
Back
Top