Red,
Here you've got 2 cameras set to auto. In other words, the camera is making all the decisions. That's as close to null as you're gonna get. The 700 doesn't have auto mode, so it has to be hand managed.
All three shots were within a couple inches at best of each other. Yes, they're hand held. However, the area in which I was standing didn't change all that much. What DID change, is the fish are flopping and in the last shot, it's deliberately underexposed a bit. People who know much about vid and shoot know why this was done.
What is clearly evident, is how the cameras reacted to the same footage, shot at the same time, compressed to the same format, output on the same timeline. If you can't see the value in comparing that, then well.....
It wasn't intended as a scientific test. I had 3 guys with 3 cams in the same place while I was on vacation. I wasn't really interested in taking time out to shoot charts or build a triple head tripod which still doesn't do much good due to curves in the glass, while I was on my vacation. Nor was I interested in blowing off the opportunity to capture a scene while I had all three cams present. The only real way to achieve this is to measure the distance of the apex in the glass curve and be sure all three cams are situated in that space at the same height and same angle. Putting "3 cameras as close together as possible and shooting in the same light at exactly the same subject" doesn't provide much more information than I've provided here. In addition, most folks have seen the HDV images of various compressions (or not) on the VASST site that I put up back in November, so it's not a question of what the format could do. This is a comparison of all 3 compressed to MPEG and then viewed in their rendered state to compare how they all hold up at the SD resolution. While it's not scientific, it's also very telling. If it doesn't make sense to you, then don't look.
BTW, rental with operator on the 700 is 2500.00 a day. I DID take advantage of what I had. I was damn lucky, being able to spend a day with Hui Kang, who is a very well known ASC. I shot this footage, but he was there with his cam. I'm grateful for the opportunity, and wanted to pass on what was seen.
Next time, I'll either ask the fish to hold still or better yet, not waste my time sharing what we discover during these little opportunities.

Spot, this is really appreciated!

I'm sure there will be many tests in all kinds of circumstances throughout the year, but we are blessed here in this forum to get the first comparisons that really show something.

I take it the 700 is the Sony HDW-F700?

Thanks again Spot, this is very cool!

Indeed, it's the HDW F 700. "Cheap" lens on it though, although you could buy 3 Z1's at retail to the cost of one Fuji HD18x 7 lens.

"Next time, I'll either ask the fish to hold still or better yet, not waste my time sharing what we discover during these little opportunities." are we getting a little stressed out, Spot. Since you are promoting a book on "HDV", "RED" comments are not out of line. I support your logic about opportunity, but you have been hammering "HDV" about facts or lack of them, so be respective to "red" as you demand of "HDV" towards Barry. Remember Spot, your one of the good guys who just happens to sell books and loves teaching and training people in Vegas and other things,( plus has a soul of a musicial artist).

OT: apit, check your email.

<< if someone could place the 3 cameras as close together as possible, in the same lighting and shooting the same subject material, and vary this setup under different subject motions and different lighting situations, then basically all variables would be equal in different circumstances, where the only different variables would be the camera itself, then this comparison could be a lot more reliable to draw conclusions. >>

That's exactly what I did. Three cameras, mounted to one tripod, shooting the same framing, same subject. Different places, different lighting conditions, shooting with/without 24p, 30p, 60i, cf24, cf30, shooting DV and HDV, downconverting HDV to SD in-camera and via firewire... static shots, motion shots, panning shots, resolution charts, color charts... 16:9 and 4:3... always the three cameras pointing at the same subject and as best as we could get, the same framing, and same exposure (but never using gain, and having all menu settings set at their middle/neutral values). Then in post I lined up the footage to where they were all frame-accurate to each other, split/screened it and made a DVD. That was the purpose of the whole comparison.

(edit: the charts were not shot side-by-side obviously, we set up each camera to frame the charts properly, which isn't entirely possible on the 3-camera-side-by-side rig, as parallax error enters into the equation).

Thanks SPOT,
I'd be interested to see them. The footage I've seen certainly isn't as bad as what I've seen other mpeg-2 encoders do in similar situations, there does seem to be some smarts coming into play to prevent thing sgetting really ugly. That's not to say it wasn't noticeable on a big screen though. Not that I think fast pans on a big screen a good idea anyway, both the problem shots I saw were static but with a lot of full frame motion. I'll admit, at the time I really didn't perticularly notice what was happening, it was only after someone mentioned those particular shots that I though ah, OK, something going on there.

One other problem we're just finding with the FX1 and audio is the on camera mic picks up everything, particularly the sound of you driving the focus-zoom rings. Nearly finished your HDV book BTW, good stuff in there, should have had some high quality color pages though for screen grabs, even the B&W images about interlacing, the printing quality makes it hard to see. I know better paper/printing means big price hike, sigh, and even adding a CD/DVD with stuff on it I'd imagine adds $5 to the cost .

Bob.

Thanks for the further explaination Spot. I probably read too much into it, from reading the other responses. This sounds like a a decent setup, where you definately tried to minimize outside variables. The only question I have is that you say all shots where taken at the same time, yet you handheld the camera. Just wondering how you could hand hold three cameras standing in the same place and shooting the same subject all at the same time? Or is it better to say you shot the same thing with each camera at "relatively" the same time?

My appologies, nice work.

Red

Well, if you'd read the bottom of the page, it says that the shots were taken approximately 5 mins apart, which to me would indicate at the same approximate time. In my lexicon, that's what it means. If I'd meant the 3 cameras were synced, I'd say that, too. Or say they were locked. They weren't. Further, I went on to say that I didn't feel like taking the time to build a triple head tripod. No worries though, Red, the images are down. If they're not of benefit, there's no point in wasting the bandwidth.

How come when I go to these pages, I get the HDV page with the Z1 pic on top and a little disclaimer and then two "page deleted" entries below that.
Am I somehow linking incorrectly?

"No worries though, Red, the images are down. If they're not of benefit, there's no point in wasting the bandwidth."

My My....touchy aren't we? I was just trying to understand the setup before I even looked at the images. Sorry, that's what one does before drawing any conclusions when doing A/B comparisons or in your study A/B/C comparisons. I didn't go to your page and read all the details....because you came to this forum and posted the message, thus I thought this would be the place to ask questions. I take it this was done in an open environment, thus is effected by the environment due to cloud movement and an ever changing light source. One should be aware of these circumstances when evaluating your study, thus why I was asking.

I was not criticizing the work you had done, just trying to understand the test setup, so that one camera was not biased one way or the other.
It's ok though, you take your ball, go home and pout, the rest of us will find a new game to play.

OK, I gotta ask, even at the risk of opening a can of worms.
How was the HDCAM footage from the 700 bought into Vegas?
I'm guessing 1394 from JH 3?
If that's the case that might have thrown a spanner in the works. We'd thought seriously about that deck with the 1394 option board however I was told that the downscaling is pretty poor so we scrapped that idea, even the HD outputs are nowhere near as good as a full on HDCAM deck. The last bit I find really hard to grasp, if it's SDTI it should be a straight data copy.
Now maybe I was fed bum info (wouldn't be the first time), if I wasn't though might mean the results are kinda skewed.

Bob.

My first guess would be a Convergent box with capture to disk using a suitable codec, then letting Vegas pick that up.

Nah, we used an Aja card on a G5, captured that bit as uncompressed QT since it was so short in length, and then downsampled it to DV in Vegas. I wanted all the media downsampled to DV so that the encoder was seeing the same information.
All things considered, a baseline had to be set somewhere, and the HDV/DV to MPEG discussions people keep having seemed to make that the best starting point.