That pretty much sums it up. For off sticks you need some form of prosthetic device.

Bob.

Adam Wilt's review mentions the hand holding issue and I've used that approach. A useful addition is to tilt the EVF up and press the back of the camera onto your chest; works best with the standard battery (BP-U30) because that fits flush with the back of the camera. Certainly impossible for me to hand hold supported only by the grip, but I've never used one hand support anyway on any camera and generally use sticks or shoulder/body mounts.

Thanks for the responses as always.

Regards,

John

well, this is a serious cam....enough variables to get even a serious hobbyist in trouble. The title of this thread says it all...lol..many buyers are gonna need a LOT of hand holding. Hope all you measurbators are up for it.

well, this is a serious cam....enough variables to get even a serious hobbyist in trouble. The title of this thread says it all...lol..many buyers are gonna need a LOT of hand holding. Hope all you measurbators are up for it.
=================
Whew! Thank God I'm not a serious hobbyist Bill.

John

You must have the shoulder brace counter balanced by the batteries. Didn't Bob post a picture of the set up? Also see Dough Jensen's vid for pictures of how to hand hold for ENG

with teh rolling shutter skewing issues of this camera, i wouldnt recomend going handheld without any support

>>>with the rolling shutter skewing issues of this camera, i wouldn't recommend going handheld without any support<<<<

Heavens-to-bettsy! Hand held doesn't mean hand waving. No one who generally waves cameras around on the end of their arm is going to buy or want to use an EX (or anything like it). No imperfections of hand holding will generate rolling shutter artifacts, and any such motion would render the footage unwatchable.
There are situations requiring hand held for cinematic reasons as well as practical situations where there is no other way. Once again, Adam Wilt discussed the magnitude of rolling shutter artifacts for the EX; let's keep it real

DJ: Have you studied just what it takes to capture the effect of the rolling shutter. You need a mega fast pan with a mega fast shutter. It's difficult to capture it when trying - you wont get it with a little hand-held shake.

"let's keep it real"

I am..
I wouldnt mention it if it wasnt an issue for some.

An associate of mine returned his 2 units. Admittadly, the shoots were using that deliberate Miami Vice handheld shooting technique, and it IS noticable considering the circular motion of these shots with this camera.

In addition, a car scene where we hung out of a sunroof of a Rav4 was virtually unusable. I believe this was a long gop issue with motion, but he believes it was the skewing... we're testing teh theory with the A1 this weekend if the weather holds out

As for me, for what i need, the EX would be perfect, but it really is too costly for my market

Go to the DVi EX1 forum. There is a thread on this with pictures and test charts etc and they shows exactly when and by how much the shutter skewing is. The problem is there as it is with all CMOS sensors and I would not use this camera for fast sporting events, but for doc and film making I don't think it would be an issue ever.

That's hardly a handholding issue. If the transmission time of the rolling shutter is a problem for the particular shot, it will be the same on sticks. Shooting with the camera vibrating (as in rough car dolly) probably is a problem.

sorry for being so discriminatory...there's some "pro's" that need hand holding, as well

Sadly this camera, despite its grunt, is nto suitable for certain purposes.
Long gop and slower shutters in progressive scan are not the best mix IMO
(1/25 and 1/50th) cause issues of their own with the codec

Im yet to see skewing in handheld shots with this camera, but i do see the notable motion artefacts akin to the long gop nuances we see in hdv

in any case, id live an ex myself, but for what i do, i just cant justify the cost

Padre, fair statement. Film is still superior!

Bill, also fair. Kodak used to do all the sensor matrix adjustments and it was just a matter of us testing and selecting the appropriate stock and using it appropriately. There are a lot of video DPs unsure about setting up cameras (as you see on CML) and the EX allows all the mal-adjustments to be made. Instructional DVDs will need to go way way beyond "this is the zoom lever".

Very true Serena but what we want is not more "instructional videos" but a manual from Sony

Sony has provided a manual but I guess you mean something more detailed. The knowledge needed for precisely setting up a "look" is a profession in itself.

Hand holding and rolling shutter: Nothing I'd shot to date suggested there is a problem with rolling shutter artifacts, so I've just been out to a nearby highway to explore this specifically. Hand held without attempt at steadiness, poles and high picket fences in the background, road markings, vehicles moving in both directions. I've tracked vehicles, I've wobbled and dutched the camera, I've panned in opposition to motion, I've done walking tracking, all at 1080P 25 fps 1/50th sec and 1/25 sec. Fast panning in opposition to motion produced distorted images that were blurred anyway, but nowhere did I get bent power poles, roadside signs, overhead wires, etc. The tall picket fences showed the characteristic of rolling shutter, but only in single frame examination. So no need to think that hand holding will generate problems and even bad hand holding generates blur that makes any oblique lines invisible to the viewer in real time.
Padre, what were you doing? A frame grab of the bad result would be useful.

"even bad hand holding generates blur that makes any oblique lines invisible to the viewer in real time."

Its not eeh rolling shutter or what have you, its the way in which the camera is used for the purposes i described.
The camera just cannot cope with it.

Its a culmination of long gop, rolling shutter and physical motion of the camera itself.
This motion on the outset pushes the codec.
Thats one thing.
The rolling shutter makes it worse because as the codec is pushed, motion information is lost. This is normal
BUT Add the distorted image caused by the rolling shutter
And inject THAT frame information INTO the lost information, and the problem is profound (FOR CERTAIN TYPES OF SHOTS)

Now not eveyone is going to do a shoot on a moving trailer into a car. Not everyone os going to be whip panning through bushes and trees and pylons and to be honest, the human eye probably wouldnt pik up on it anyway.

The rolling shutter issue is not too much of a a problem IMO.
The long gop codec however confounds the way in which that data is distributed across the GOP.
In turn, the rolling shutter LOOKS worse than it is becuase the codec makes it so.
Now if the EX1 was using an Intra codec as opposed to long gop, i bet your bottom dollar that the rolling shutter wouldnt even be noticed.

To be honest, i dont think its a problem even with long gop, but people are purists and stupid about it.

If the EX is outputting uncompressed through SDI, the rolling shutter wouldnt even be seen.
It wouldnt "last" for as long as an intra codec would show the one affected skewed frame, BUT thats ALL it would show as the cameras refresh would "catch up" by the time the next cycle comes to past, based on teh shutter speed

Now if that "skew" information (distroterd rolling shutter image) is within an I frame, then that entire gop sequence is affected with the same information..

And THIS is the problem.

See where im going with this?
Its not the camera CMOS per se.
One frame out isnt gona kill anyone.
BUT if that frame affects its siblings, then there WILL be a problem and i think THAT is the issue
Problem is, people havent even bothered to consider that this is probably the major influence of how the skew is "managed" so they automatically blame the CMOS

The only way this one frame affects its simbling is simply because of the codec in use (XDCam)

Couple long gop with progressive at 1/25 handheld on THIS heavy beasty, and there is potential for trouble. And thats what we saw. Mind you we were shooting at 1/125 and 1/50 when we saw it

Thats not to say its a bad camera, Its not.

But specifically, codec and frame management play hand in hand, and if one doesnt play nice, the other is affected.
I believe THIS is waht people fail to see or even consider with this problem.
If the skew is that bad for peoples shots, use another codec.
More than likely the "problem" wil lbe gone.

Thats my point.

Serena, please dont get so defensive as i noticed you jumped pretty high when i first bought this up..
I discuss all cameras and issues equally. I have no bias toward or against any camera.

No, not defensive at all. You did state that even hand-holding will be a problem with this camera, a quite significant claim. I'm always interested in people's experience and if there are problems with a bit of gear I want to know about it. That's why I went out to the highway, to test what you claimed and to check whether I'm going to strike problems in my yachting work. All of the issues (long GOP etc) are well understood as potential sources of problems and your exposition offers nothing new or specific to this camera (or any other). I reckon my quick highway test was pretty rugged, the background being cluttered and the motion of camera and vehicles being quick. I was expecting, from your experience, to see some worrying problems with distortion and image macro-blocking. Those tests showed nothing unexpected. In fact much better than I would have anticipated. So obviously there is a great difference in the way you were using the camera and what I was doing, so what is the significant difference?
I'm asking for specifics, not theoretical background. And an example of the unsatisfactory imaging.

I believe we were moving about 60kms an hour probably faster, shooting from the drivers side INTO the car to get a profile shot with a moving background thrugh the pasanger side window

" your exposition offers nothing new or specific to this camera (or any other). "
well it IS specific to THIS or any CMOS camera using long gop. Its not isolated, but it exists.

I dont doubt your tests were thorough, and i dont doubt it exceeded your expecttions. The camera exceeded my own and im a tuff nut to crack...

There's no connection between a rolling shutter and how the mpeg-2 codec is going to compress the image. With shutter off at 1/25 it's even easier on the codec.
How does camera weight come into play?
If anything a heavier camera means less vibration, that makes the job easier for the encoder and less jelo.
To get one frame affected by skew and not another in a GOP would seem to require a high enough G force to cause the camera let alone the operator to splatter.. I guess potentially it could be done in a G suit at the long end of the lens but you'd be getting a useless image anyway. I've shot with a Z1 from a pretty bouncy copter with the lens hood in the rotor wash and at the long end of the lens. I still didn't notice any macroblocking. The shot was useless but that's a given, trying to shoot closeups at 600' from a 2 blade copter without a gryro mount is a waste but I wanted to try.
There is an image sequence around that shows a pretty decent 'jelo' effect, induced by kicking the tripod. Thing is though apart from the almost artistic jelo affect the image holds up perfectly.

There are some interesting artifacts from CMOS imagers but have nothing to do with the encoder at all. Same affect has been seen in the Redone and ultimately worked out to be due to the lighting. Discharge lamps stobe enough to cause multiple exposures of the imager and as it's read from top to bottom with a fast moving subject you end of with multiple skewed images. Probably without a rolling shutter you wouldn't see this and you do have to look at it frame by frame to notice.

The one area that the long GOP could bring you unstuck is timelapse. Timelapse of heavy cloud cover could give you a frame sequence with almost no similarities between frames and really stress the encoder.

There's a lot of history over at Reduser regarding CMOS imagers and the rolling shutter issue. When the images from the prototypes were first posted and dissected the VFX guys were up in arms saying the camera would be useless to them and I could understand the problem. The skew cannot be correct in post. And yet just recently someone was asking how come the Redone is mostly being used for VFX shots ? :)
Now I'm not saying you've not seen something funky going on, there's many ways to make the EX1 do some ugly things but so far none of the uglies I've seen were related to the rolling shutter or the long GOP encoding. Perhaps if we could see what you're talking about we could avoid taking stabs in the dark as to what might be going on.

One thing I did just think of that can cause you grief with a lot of cameras is the OIS, it sounds contradictory but once you get over a certain threshold you do better with it off than on in fast movement / vibration.

Bob.

Ill see if i can get a clip but i really doubt it.

In regard to the artefacts. Its THOSE artfacts which are bought a lil more into the light when coupled with long gop. Thats all Im saying.
THAT was my point.
I didnt mean to start an entire debate about the EX

Like every camera there will be issues.
For the kind of stuff i do, the Ex is more than perfect, but my budget just wont justify it.. not yet...
For others, DVCProhHD is their avenue and all the power to them.
Each to their own.
I still feel that the EX lens, build, functions and moreso the CMOS itself far outweighs any slight negativiite elements which may affect one shot once in a blue moon.