|
|||||||||
|
Thread Tools | Search this Thread |
March 25th, 2009, 09:27 AM | #1 |
Major Player
Join Date: Jun 2003
Location: Orlando, FL
Posts: 494
|
Question about rolling shudder issue
I've been looking at the FX7 and the V1u as a new camera. I was pretty much sold on the FX7 till I found out that it doesn't have lanc control. I shoot a lot of dance performances and have a rear control for zooming in and out is crucial. Now I have a question about rolling shutter. One of my yearly performances uses a strobe light in one part. Has anybody shot a stage type event that had a strobe and what were the results? I've seen examples of flashes going off near CMOS cameras, and a strobe would be similar. But I'm also thinking that from distance and the intensity of the flashing might not produce the rolling shutter. Am I correct in my thought process?
As always, thanks in advance for any input. |
March 25th, 2009, 10:37 AM | #2 |
Trustee
Join Date: Oct 2006
Location: Antrim, Northern Ireland
Posts: 1,569
|
David,
I had to check this on sony.com - I didn't remember Lanc being absent on the FX7. The FX7 manual there (search the site for fx7u) shows a Lanc socket, see page 119 of the manual. So you can add a Bogen, Bebob or other Lanc control. Check out Spot's "V1 versus FX7" link at the top of this forum. Confirm this before you buy in case the Lanc has disappeared, but I would find that unlikely. Not sure about strobe lights though, I've never tried it.
__________________
Irish Railway Videos on Youtube |
March 25th, 2009, 10:45 AM | #3 |
Major Player
Join Date: Jun 2003
Location: Orlando, FL
Posts: 494
|
You may be right. I was looking at the comparison chart between the 2 cameras and saw that the FX7 didn't have the iLINK external rec, I think that's where I messed up. Thanks for clearing up that part of my question. Now I'm still wondering about the rolling shutter from a stage strobe light.
|
March 25th, 2009, 12:36 PM | #4 |
Inner Circle
Join Date: Feb 2007
Location: Apple Valley CA
Posts: 4,874
|
It SHOULD have a LANC... about 99% sure of that... std 2.5mm, towards the back IIRC.
As for the strobe, it will depend somewhat on the situation. Partial frame exposures are a part of life with a CMOS based camera. The best way to explain the potential problem is that if the strobe and the refresh on the CMOS aren't in perfect sync (virtually a lock on THAT!), you will probably see frames with part illuminated by the strobe, and part not, as the sensor refresh catches portions of the frame while the strobe is active, and part while it's not, if that makes sense. Adding to the challenge is the duration and cycling rate of the strobing. I shot some video with an HC7 (smaller CMOS, same generation as the FX7) with a DJ strobe going, very low light situation, so wasn't usable for that reason, but it seemed to be OK considering. I think a lot will depend on the overall lighting scheme and the strobe intensity/duration/cycle. Hope that makes sense and at least helps you contemplate the situation to see if it might work. |
March 25th, 2009, 12:51 PM | #5 |
Major Player
Join Date: Oct 2002
Location: Albuquerque, New Mexico USA
Posts: 333
|
First of all, I shoot a lot of fast action stuff with CMOS cameras and have never noticed the rolling shutter effect in the real world. Strobes are another matter altogether.
CCD cameras take a "snapshot" of the entire frame and then read it out, CMOS cameras read a frame sequentially, line by line, just like old tube cameras. Strobes have always been problematic for video regardless of the image sensor technology. On a CCD camera, strobes or camera flashes are generally so short that they occur on one field of a frame or occasionally don't show up at all if the CCD happens to be in between field snapshots. Essentially, if the CCD sensor catches it, it will usually illuminate the entire field, but it may miss some flashes which will cause a temporal stutter. Since CMOS is reading each line sequentially, depending on the duration of the flash, it may illuminate only some part of the frame depending on when the flash fired, how long it lasted and what part of the frame the CMOS sensor was scanning during the flash. It may be that the rate at which the strobe fires might illuminate the top third of the image and the next flash migh illuminate the middle third and the next, the bottom third. However, without having the strobe and the camera synchronized in some way, there is no way of knowing. (Tube cameras had a lot of persistence in their phosphors, so it wasn't as much of a problem.) Regardless of the image sensor technology, the rate at which the stobe fires and the frame rate of the camera will interact and cause some sort of unpredictable temporal artifacts. Another thing to consider which could be much more trouble than CCD vs. CMOS, is MPEG2 or AVCHD encoding. Both technologies gain most of their compression efficiency by using the available bitrate to encode only the differences between frames. If you are shooting a stobe, one frame will be radically different than the one that preceeded it and this will swamp the encoder and produce severe artifacts, usually 8x8 pixel sharply defined blocks. Shooting strobes is a problem, always has been. I've had better luck creating the effect in post where I can at least synchronize the flash rate with the frame rate, although it doesn't sound like you will have that option. MPEG2 or AVCHD encoding still isn't going to like it. |
March 25th, 2009, 01:15 PM | #6 |
Major Player
Join Date: Jun 2003
Location: Orlando, FL
Posts: 494
|
Thanks for everybody's input. That's what I was a little afraid of. I shoot a "Nutcracker Ballet" performance every year for a dance studio and they have a scene where the stage goes black, a single strobe comes on and the Nutcracker grows to full height. Currently I shoot with a Pany DVC-80, but it's time to start thinking of the replacements. Last year I was able to shoot a different performance with a Z1U and they had a strobe scene that didn't come out that bad. I've been debating between HDV or AVCHD and which cameras would work, plus of course price. Then of course it's also CCD vs. CMOS. I've not shot with a camera that has CMOS chips so that's why I'm trying to take in all the info I can. I've been shooting with a Z1U for a couple of years but mostly studio stuff where I can control the lighting. Thanks again.
|
March 25th, 2009, 07:18 PM | #7 |
Trustee
Join Date: May 2004
Location: Knoxville, Tennessee
Posts: 1,669
|
I have an FX7 - it has LANC. It also has peaking display, despite what Spot's chart says.
|
March 25th, 2009, 08:27 PM | #8 |
Inner Circle
Join Date: Feb 2007
Location: Apple Valley CA
Posts: 4,874
|
Z1U is of course CCD, not CMOS, why not pick up a second one of those? Or an FX1 if you want to save a few $...
Not that the FX7/V1U are bad, I like the smaller size myself, but if you know there's a "problem area", and already shoot the Z1, I would think it would be a good choice? |
March 26th, 2009, 10:25 AM | #9 | |
Major Player
Join Date: Jun 2003
Location: Orlando, FL
Posts: 494
|
Quote:
|
|
March 26th, 2009, 01:53 PM | #10 |
Inner Circle
Join Date: Feb 2007
Location: Apple Valley CA
Posts: 4,874
|
Hi David -
You might look for some posts by Mark Von Lanken - I couldn't find them offhand, but he's done some comparisons of the HMC150 and the Z5 and posted in another thread here on DVi - worth a look to help decide whether you can live with RS or not. |
| ||||||
|
|