And I saw the "Skew" efffect during the quick pans.

Grazie

This looks more like interference from an electrical field of some sort, than a rolling shutter. The rolling shutter is something altogether different....

I've seen what looks like this, when dealing with power supplys & bad grounds, only the frequency was alot slower....

The rolling shutter effect can be discounted because the illumination is constant. It looks like a defect in readout, possibly confirmed if the reset fixed it. Electrical interference does cause such effects (radars particularly, and mobile transmitters) -- that can be checked by it being location dependant (unless it's your cell phone).

I've had this with the EX1, in daylight as well as under lights.

So far turning off Flicker reduction (Camera Menu) seems to have cured it, though I haven't used it much since reading about that fix.

It is a kind of "phasing" that I have seen when floros and cameras are mismatched. Likewise the Doppler effect (is it Doppler? maybe not . . ?) of helo blades. It is as if one set of scanning is only being "seen" by the viewer and then with the "reset" it is gone. Is it possible that the reset "removes" a charged item? I'm thinking of some kind of ground loop capacitance? The RADAR Magnetron used (use?) the looping effect of coil and capacitor to achieve the shorter wavelengths. Is it at all possible there is some non hard-wired transmission links WITHIN the device? And again, I am thinking just how DOES the CMOS rolling shutter work? This is way beyond me . . .

Interesting . .

Grazie

Hi,

These shots were taken outdoor and during perfectly steady lighting. So there is no risk for interference with pulsating light sources.

There is definitely something wrong with your cam. The rolling shutter cannot produce such artefacts during pure daylight lighting. Geometry dirstortion during panning is due to the inherent rolling shutter in most of the CMOS imagers, evident also in your shots. But that is something totally different.

The risk of interference from strong magnetic fields are close to nil if you stand outside far avay from any electronic equipment.

I would recommend that you send your camera for repair, together with a videoclip about the problem!!!

Christian

Before sending it for repair, as I said, try it with Flicker reduction turned off - it may be that you turned it off when re-setting to default. This has come up in several forums, and it seems that having Flicker reduction on when it is not needed can cause this.

So far turning it off has worked for me, but I haven't done enough shooting since then to be sure ...

[I]"And again, I am thinking just how DOES the CMOS rolling shutter work?"[/i]

Ever seen a really old stills camera with a roller blind shutter?
It was pretty much what the name implies, a roll of material with a slit in it. When you pressed the shutter release a spring caused the blind and slit to roll down exposing the film. Different spring tension changed the exposure time by making the blind move faster:

http://licm.org.uk/livingImage/Shutters-Blind.html

Now the problem was if an object moved past the camera as the film was being exposed the top of the object was exposed before the bottom so what would have been a vertical line could became a diagonal line.
CMOS imagers have the same kind of problem, the image is read one row at a time and it takes a largish period of time to read all the rows, 1/60th of a second in the case of the EX1. The bit that's hard to grasp is that the readout time is independant of the shutter speed, in fact a faster shutter speed will make the problem more obvious as a slower shutter speed will tend to mask the problem with motion blur.

The nastiest problem I've had is with off frequency fluro lights. You could probably imagine what happens. The light is on full brightness when the frame starts being scanned, the light is right off as the middle of the frame is scanned and back on at the bottom of the frame. Next frame the sequence repeats but starts with an offset. End result is waves of light that roll up/down through the frame.

Bob.

Thanks Bob. Yes I had been reading the DVInfo material for the past few months about this.

No, it isn't that I am considering that rolling shutter is the same or parrallel to this one, more that the actual process of having a looping function - rolling shutter - MIGHT have on this one. I repeat, that behind the CMOS, that there IS a function/program loop that is reacting WITH this. Again, not the actual "light" paths - but the actual "firmware" procedures being somehow phasically (IS that a word?) interrupted. And that the reset had somehow "shocked-it" into a better frame of mind!!!

Interesting.

Grazie

After notifying my "authorized Sony dealer" (Alpha Cine), I was told that they were sending this information the Sony engineers. Worse case scenario was they would replace the camera if necessary.

Since my use of the All Reset button, as I mentioned, I haven't had any problems. Nevertheless, that little doubt is still nagging me, it being so close to the recent experience.

OK,
now I understand what you're getting at and I've also managed to view the footage. You could be on the right path. The EX1 has many settings and a lot of feedback loops that come into play depending on what the camera has been told to do.

However those rolling bands are pretty fast, faster than what any of the normal camera feedback such as auto iris etc would be capable of. The Flicker filter could be in part the culprit, following a reset it is set to Auto, certainly worth getting it set to Off.

However I have a theory.
We're all assuming it's not caused by the lighting. I don't know about that at all. Is sunlight perfect, it does come from one massive hydrogen bomb, that creaks and groans, bubbles and boils. It could be that it isn't as constant a light source as we're assuming.


Another theory.
A lot of light is reflected, off large buildings for example. They move and vibrate and could modulate light intensity.

Bob.

The Flicker filter could be in part the culprit, following a reset it is set to Auto, certainly worth getting it set to Off.

Bob, you've confused me (and that's not necessarily you're fault). If the problem was evident BEFORE a rest, and executing a reset fixed the problem (setting the flicker filter to auto, then why would one want to set the flicker filter to off?

Wouldn't turning it off open the door to repeating the problem?

> "Wouldn't turning it off open the door to repeating the problem?"

Jay, to answer your own question try with it off, then with it on, and see if there's a difference.

If it's set to "auto" I'm not sure what triggers it between off and on, so that makes it hard to be scientific about the cause.

Peter, I understand what you're saying--I even agree with your assessment. However, I'm a bit too paranoid right now to try anything for fear of screwing something up, now that's it's working. My confidence level with this new camera is not where it needs to be.

I'll look to see what the setting is currently and let you know.

While it's true that the Sun is very active, the output is absolutely stable over the timescales relevant to this discussion.

Currently, the "FlickerReduce" shows:

Mode: Auto
Frequency: 50Hz

Here, in the US, if I'm not mistaken, electrical current operates at 60Hz per second. But the sample video Ola provided was shot outdoors!

Not knowing what causes the problem makes it difficult at best to avoid the problem, hence my paranoia.

Jay, I agree that the concept of "Flicker Reduction" CAUSING flicker, or as you've better described it, "rolling flutter" is hard to mentally grapple with, but the advice I've followed is to switch it off unless there actually IS flicker, such as with a CRT, and this has worked for me, so I'm sticking to this fttb.

As far as the 60 or 50 setting is concerned, I've read:

a) Set it at your local refresh rate, or whatever that Hz voltage thing is

and I've also read

b) Set it at the opposite to your local refresh rate ...

Well, I believe that this setting only comes into play when Flicker Reduction is turned ON, so until this happens, I'm keeping it off. If I do need to turn it on, I'll try both and use the best result ...

btw, I don't think it's paranoia, just healthy curiosity!

OK,
thinking about this some more and checking the manual and reading some comments on Ola's thread.
Things that could cause this.
1) The Flicker Reduction setting. In Auto it's looking for Flicker to kill, seems to me that having it Off couldn't create a problem and having it On could.
2) ECS i.e. Electronic Clear Scan. This causes the shutter to try to sync to a flickering light source such as a CRT. Again I can't see anything to be gained by having it On so I'd make certain it's Off.

What concerns me about Ola's sample footage is the problem seems to occur after a rapid camera movement and the camera doesn't stay static for very long so it's hard to really know what's going on. Some of the auto circuits in the EX1 have a very slow response time by default.

What I'd suggest you try to do is recreate the problem. Then see if any of the camera settings that I mentioned make any difference to it. That way you'll know they were the culprit and you don't have a faulty camera. Even if Sony give you a replacement camera you still don't know for certain if it was a fault unless they tell you they found a fault and that's why they've given you another one.

Bob.

Okay... I took a deep breath, several actually, and turn OFF the Flicker Reduction. Then, with the camera running, I moved through several gyrations around the house and could neither see nor record any rolling flutter.

As the old adage says, time will tell.

What concerns me about Ola's sample footage is the problem seems to occur after a rapid camera movement and the camera doesn't stay static for very long so it's hard to really know what's going on.

Agreed! I noticed that the flutter seemed to happen after movement, but mine were slow, gentle pans. However, during one test, it showed up as soon as I turned the camera on while static.

To add another variable, last week I saw a video in which the EX1 had been strapped into an acrobatic aircraft. At times the plane was pulling ten Gs with great and sudden movements, yet nary a flutter. Go figure!

You can watch the video here: XDCAM Does 10Gs. (Have your barf bag handy!)

UPDATE:

Here is the answer I received from Alpha Cine (where I bought the EX3) in regards to this rolling flutter issue:

"My service guru tells me this is a problem that gets fixed at Sony. You should bring your camera in for my service guys to take a look at it and they will either give you a new camera or send it off to sony for a fix. Supposedly it's a bug where the shutter or the fake electronic frame divider is off."

The only problem is they are in New York and I'm in Miami.

"this is a firmware issue that an update can easily fix, in shop"

I feel kinda vindicated!

Grazie