HPX300 vx. EX3 Rolling shutter example

Based on this - the 300 looks much worse.
But honestly- who shoots like that? Whipping the lens back and forth?
 
Steve - how were these set up? How was the amount of movement determined?

The only way to do this, as far as I'm concerned, is to put the EX3 on top of the HPX300 (like Jim Arthurs did during his Red/EX1 shoot). Zoom in both of them so that they have identical horizontal framing. Then pan back and forth; with both cameras lined up on-center and framed up to the same width, you should be getting identical relative motion, and then your results should be indicitave of real-world performance. Split-screen 'em and let's see how it looks.

Also, run the test twice, once in 60i and once in 24p.
 
This test lacks lenses, framing and FR details and is there for inconclusive. Would really like to see it done with exact matching of all parameters between the 2 cameras (basically what Barry said...).

Not to mention the fact that the HPX300 is capable of AVCintra and DVCPRO HD at up to almost 3 times the bit depth of the EX3. With non long GOP recording and changeable lenses it stands up quite well needless to say in a grade.
All be it we are looking specifically at the skew a grade of the footage is needed to show what the footage will actually look like after being used and compressed. May even affect the appearance of the skew.
 
the test

the test

Here is how we did the test.
If you watch our video introducing the HPX300 you'll see when we go to the test portion of the video. I'm standing there with Ned from Panasonic and you see the two cameras mounted on tripods. I panned them both left and right, up and down at the same speed as best I could. I really didn't think that I was going to even use the footage. Someone asked for it so I posted it.

I think it's close enough to see that both of these cameras along with RED & DSLR cameras have this problem. If it's a deal breaker for you, then buy an HVX200 or an HPX500. This is why I keep telling everyone that every camera is not for every shooter or situation. Would this problem cause me issue in my videos, NO, would it if you where shooting a handheld fight scene, YES, so get the camera for that job that suits it best.

I think when buying a camera you need to realize that you are going to have to rent cameras from time to time to suit the job. Too many people are thinking that one camera is perfect for every kind of job and it's just not the case.

In the old days we used to rent 16mm/35mm film cameras, video cameras whatever was best for the job and budget. We understood that some cameras had limitations and if it couldn't do what we wanted we rented. We always owned a video camera and it was good for 75% of the jobs but the other 25% we rented. We understood if budget couldn't afford it then we made do with what we had.
Steve Weiss
 
Last edited:
Hi,

Frankly I didn't think the cameras moved siide to side at the same speed and I think the only way to do this is to have them on a singular tripod head in a side by side mount. And from the tests I have done they are virtually the same. The rolling shutter is the rolling shutter, the length of the lens and where you are in the zoom can certainly accentuate the issue. The swishing back and forth of course never happens in real life, a pan happens in real shooting. Knowing that the camera this ( and I mean either the EX1/3 or the HPX300) you shoot accordingly with care not to make it raise its head.

Anyhow the HPX300 has CMOS imagers, it has all the same characteristics of CMOS but none of the characteristics of long GOP.

Best,

Jan
 
I wouldn't be surprised if there was a difference. AFAIK Sony built in some compensation for the effect into the EX cameras, though I am not sure of how exactly it functions.

Regardless most of the time I have not come up against any problems with CMOS cameras rolling shutters. Most of the cause of issues such as the 'jello' effect are by camerawork that would result in a NG take even if the jello effect wasn't there.

For the few occasions where a problem is caused then we just have to accept that if we want perfection we have to pay for it. Much as a consumer camera doesn't have the dynamic range of 35mm film, we can't really complain about that.
 
Jan, not only is there a big difference between rolling shutter effects in the Zacuto demo, but there is also a large difference in rolling shutter effects between the EX cameras and the Canon cameras, and even, and I would like to put an emphasis on this, between other less expensive Sony camcorders.

I will try to find out more, but as I said AFAIK there is some form of compensation for the effect in the EX cameras. I will stand corrected if someone proves this incorrect, but the fact that there is a difference between the EX and many other cameras in terms of the 'jello' effect would bear this out.
 
I don't think Zacuto is biased, and I'm sure they're both fine cameras, but clearly no matter how you spin it the EX3 came out looking better in this test. Well, none looked 'good' under the circunstances, ... maybe "not as bad" would be a better way to put it, ha! Shirley it's not the end all of all comparisons but I doubt that all CMOS are the same and work the "exact same way". I may be ugly, but I'm stupid.
 
Last edited:
I'm going to bow to Barry on this one. I think he should do a scientific test between the two cameras. I'm removing my test because he thinks I should have been more scientific about it. I agree, I only had the camera for a few hours and really just panned and tilted the two cameras without really knowing how to setup the two cameras to make it apples to apples. I can see his point. So Barry, let's see the test. Steve
 
You must log in or register to reply here.
Back
Top