In case you skip the article, the time lapse scenes were shot with the 7D.

Excellent find!!

The camera is set up nicely and I LOVE that rolling rack for side to side "dolly" shots.
Gotta get it... yeh..(pant, pant) gotta get it all.

v

We bought the prototype sliders that Losmandy never put into production. Not too hard to roll your own. If that's too daunting there's at least one supplier of prebuilt kit at a reasonable price.
I really like sliders, quick and easy to setup unlike a full on track dolly.
Sliders would work really well with a 5D/7D as the basic camera is lightweight.

Bob.

Don't get your hopes up thinking that next Star Wars will be shot with 5D or 7D as the compression artifact would stand out like sore thumbs on the big screen. Those are amazing cameras that represent a huge leap through that will spearhead the revolution but it's still not there.

well said patryk,

i'm getting a bit tired of all this hyperbole and gushing praise of what is basically a 'prototype' full frame video camera.

there's no denying what's been done with it is pretty bloody impressive, but it isn't going to be making features, nor serious tv any time soon.

leslie

Yeah, some shots were quite impressive, but lots of them couldn't even get the whole frame in focus ;)))

Here's the next installment of a great viral video ad campaign:

http://gizmodo.com/5385446/canon-1d-mark-iv-the-5000-new-king-of-cameras

v

Some of the problems you have noticed may get back to this test:

http://thebrownings.name/WHP034/pdf/WHP034-ADD39_Canon_5D_DSLR.pdf

Many lowly video cameras perform better in these tests as the camera was designed from the ground up to shoot video. It's very easy to get misled by how images look out of the camera on a monitor. Transission chains and scalers can make a mess of images with technical problems that are hard for the eye to see.

I'm pretty certain some of the issues behind the never ending complaints about poor SD images out of Vegas are related to this problem as well.

Bob.

as the compression artifact would stand out like sore thumbs on the big screen.

The kids who lined up to see Paranormal Activity are not aware of this problem. To be sure George Lucas is aware of it -- but it's likely that if George were in his youth he'd be up late writing a script that could be shot on a used FX1 or a 5D and a $200 TrackGlide. Considering products like the 5D and Vegas too, it's not hype to say that there is more opportunity than EVER to make features on a micro budget. For those with little money but lots of energy, ideas, talent and something to say -- it's a Golden Age.

Vegas can do features

xberk, DVX100 came out what 6, 7 years ago. It WAS a revolution in filmmaking and yet how many films did you see shot with that revolutionary camera? And even more so how many of those were successful?

I assume Paranormal Activity was shot with some kind of HDV flavor as such its image did not exhibit the same compression artifacts (and beyond HEAVY anti-aliasing that Bob's article points out), thus the film's image albeit compressed didn't make some of the shots unwatchable.

Hi Patryk -

Hope this finds you in good spirits for the holidays.

To the public at large, let me state unequivacably that I think Patryk is one of the most innovative and exciting DP's on this forum. Though we've never met, I LOVE his lighting style, and have even sent my son to appretice under him. I'm a real fan, etc.

That being said, I'd like to query Patryk (or anyone else) directly.
Would you be kind enough to review the footage that I shot the second day I owned the camera at: www.VictorMilt.com and point out any artifacting that you see?

I just am blown away by this camera, so I'd like to see what I'm missing, as I'm planning to shoot a job with it right after the holidays. Forewarned is forearmed.

best to all,
Victor Milt

It's in my opinion impossible to tell image quality looking at footage that's been recompressed by YouTube. I wasted a bit of time trying to find the cause of some nasty artifacts in some footage I uploaded recently and as far as I can see it looks pristine before YouTube got to it. The worrying thing is it looks like interlace artifacts and the footage was interlaced so maybe, just maybe there's something left from the de-interlacing that I cannot see that's triggered a problem in the YouTube encoders.

The BBC tested the 5D and found it unsuitable for BBC funded projects. I posted a link to the test results above I think. The reason they freak out over seemingly invisible problems is because of what can happen when the footage goes through their transmission chain.

Bob.

Did anyone else read the Lucas article? I'm too lazy to re-read it but I got the impression they intended to use the 5D for the cockpit scenes of the Tuskagee Airmen. If the 5D were that bad, wouldn't they fabricate a prop cockpit that could accommodate a full-sized camera? I don't know. I'm but a student here.

Now here is a show-stopper that I would have been proud to participate in:
http://www.hurlbutvisuals.com/

In fairness, and I quote:
In “Navy Swimmer” Shane points out that these are the cameras used: “the slow-mo shot running towards the Helicopter was 35mm, aerial photography was Sony F23 with a gyro stabilized head, all the interviews were the Sony F 900, and the action shots were 5D.”

But whether you are using the 5D2 or you just love to be amazed, be sure to watch this spot.

v

I am utterly dumbfounded by all this.

Im going home to crush my EX-1 with a hammer immediately.

If I can't get images with no movement in them crystal clear and with less than 10% of the frame in focus there is now way I can produce anything meaningful.

I cried when that single leaf or whatever it was accidentially came into focus. Simply heart-rending.

Since I have been following Canon's line up for some time and very curious of the what the future holds I wanted to chime in. Yes, the new DSLR that record video are impressive and yes they have limitations. They rightfully should.

I have no proof, but my gut tells me that a replacement to the cameras such as the XHA1 will be a similar body, with a CMOS sensor that will allow SLR lenses to be mounted and have all the benefits with XLR audio etc. This may be wishful thinking, but it makes sense.

Chad

Vic, you video shows the camera in its best light and at its strongest. Fast moving shots, shallow DOF, when shots are static they are purposefully overexposed. As such it look like it totally fits your shooting style and thus my opinion doesn't really matter.

In all fairness despite all its limitations i was going to buy 7D as it fit a specific shooting style but last minute as the project felt through i decided not to, as i found that the camera is not versatile enough yet to pluck down the money as it will not get as much usage as my 35mm adapter rig does.

Now consider those 2 stills:
http://patrykrebisz.com/canon_test/tea_red.jpg
http://patrykrebisz.com/canon_test/livingroom_red.jpg

where i marked the areas where the compresion of the camera kicks in and the dancing around abstarct pattern of noise and compression prevents me from telling the story i want to tell -- a calm static, not-much-happening scene of a person drinking tea or just looking. Now if the scene had high motion, or the scene depended on overexposed whites then i probbaly wouldn't care and the camera would be fine, but in this case it failed to deliver. In other words the camera is fantastic in some areas but as i mentioned above is not versatile enough just yet for me to justify the $.

(thanks for kind compliments!)

Hi Patryk -

I'm not sure I fully understand the issue, but I really want to.
The stills you show are gorgeous.
So I'm assuming that you're getting some sort of "dancing" artifact noise in the areas of high contrast?
If that is so, would you be kind enough to post the moving shots of the scenes that you did the still grabs from, please?

I am most interested in your analysis, as I'm about to embark on a rather expensive shoot and want to know what to look out for.

In the past there have often been caveats and situations to watch out for. You would simply avoid them. For instance, in the early days of DV you'd avoid diagonal straight lines or suffer "stepping". In one inch video you didn't dare to have a houndstooth jacket (moire) or a red shirt (blooming). Even in the world of film, you had to watch the famous "spokes turning backwards" shots, and you knew that you had to pan slowly or not a all, due to strobing issues. And you NEVER did overcranked footage with HMI lights.

I'm not denying there are problems w 5D2 - I'm just taking advantage of this forum to predict and avoid similar situations.
So what specifically are you not happy with? I just don't see the problem in the still shots.
v