The 7d is tempermental, but also forgiving
 

Attached is 7d frame grab. I've shot this same location several times with a Sony EX3 and JVC HD100, and the 7d handles it better in terms of exposing the subject and not washing out. For me, this is pretty good considering it's just existing light. I've used 5200k softboxes and an EX3 and still had trouble at this location.
Tamron 2.8, 28-75mm.

You need to examine your EX3 setup. That thing has at least 2 more stops of dynamic range than the 7D.

I realize the EX is highly regarded with an estimated 11 stop range, but I've seen it struggle in these harsh conditions, as would any vid cam. Exposing properly with the 7d is still guess work for me, unlike the EX which has that great double zebra metering, and yet the 7d just seems to cut you more slack, even without any viable built in method of judging exposure.

Interesting thing about the 7d is it has caused some reconsideration of value of all the quantitative metrics traditionally used to assess cameras. People point it at charts and say it sucks, people point it at a pretty girl and fall in love. Likewise, in a difficult location with ambient light blowing in from the windows, the 7d shrugs it off, but how could it? It's only 7 stops of range? Skeptics say that people are in denial and the figures don't lie, sure is a lot of love out there though.

People who shoot the charts, and the people who believe the charts aren't skeptics. They are people interested in what the optics are doing. I saw the charts on the 7D, weighed that against the images, and against what *I* need it for, and have decided to buy one. And yes, I've seen it alias on real images, and moire, and half-expose frames like any rolling shutter camera.

The problem is that those who point it at the pretty girl and fall in love come back on forums like this one when they have to shoot urban scenes with a moving camera and wonder why the camera is giving them fits. Or they show up to shoot an interview and the guest is wearing a silk shirt or tie, or they have pinstripes on the blouse and the camera image looks like it's swimming...

The charts merely tell you where things are going to fall down. And believing that doesn't make you a skeptic, it makes you informed. All these cameras have limitations. We utterly hammered the codec in the 5D on our shoot in October. It broke apart like an oil painting. But when we shot in a more "filmic" way, we were rewarded with some gorgeous imagery.

A man's got to know his limitations... ;)

Heheh... the 7d takes down another engineer. A man may need to know his limitations, but love conquers all... :)

Ha! That's the truth.

While they all seem a bit redundant now, I did read all the chart blogs just to fully understand what not to do or how not to shoot. Every camera has its flaws, but working creatively with your limitations always produces a better product. When the client loves the image, that's the proof of a job well done with whatever tool you use.

I'm going to start shooting commercials on my iPhone! ;-)

All this debate about which camera has the most latitude or the most resolution etc., is getting old.

I own both and here's the comparison I like the best: EX3 approximately $8K, 7D approximately $2K.

If the 7D doesn't have quite the resolution or latitude as the EX, but for most people the difference is imperceptible, that's OK because it doesn't have quite the price tag either.

Come on David, the Blackberry is much better than the iPhone...

A statement of fact is not an argument. And you're right, to most it's imperceptible. But you know what, I have a multi-part documentary going to PBS in 2010. And they care.

I'm really sorry these little squabbles trouble so many people as I know the majority of folks are mastering for YouTube or Vimeo and that's as far as it goes. A few master to DVD and fewer still BluRay for wedding or event clients.

But those folks shooting the charts aren't doing it for their health. They are doing it because it SUCKS to spend 4-8 months shooting a documentary only to have it turned down for broadcast because it doesn't meet spec. It matters to us and the people who pay us.

I would think Skywalker would have a little higher "specs" than your average PBS or most people for that matter.

The tale of Lucasfilm, Skywalker Ranch, Red Tails, Star Wars and Canon DSLRs on a 40 foot screen! | Philip Bloom

The great thing is, it will get better and making films will get less expensive and be in more reach for more people.

After 20 years shooting as a portrait studio owner the industry has been hit hard by "Moms with a Camera". I imagine the elite film makers are about to experience a similar breaking of barriers.

I am very familiar with the "specs" of film. The great thing about film is you get to select wardrobe, set, lighting, and every other aspect of the frame. It is COMPLETELY under the control of the filmmaker in narrative film.

When shooting documentary you don't necessarily get that luxury. So sometimes things get ugly. If you take a look at the films released in the past few years that have been shot digitally (LucasFilm is big on that), you'll note that they still go to real film for certain sequences that just don't work on video at this point.

Why is it that if someone points out that a camera isn't perfect, despite the fact that it makes lovely pictures, that somehow you are bashing it? Or saying that it's faulty, or not good enough? I mean really. Every camera that I've owned has had issues. The hate spew that went around on the EX1 was incredible. The DVX made LOVELY pictures, but when folks pointed out that it juddered (and it does) they were dismissed as haters. They were simply pointing out the truth. Didn't stop me from buying either of those cameras. My Panasonic S-VHS camera had flaws. My Sony Hi8 camera had flaws, my Nikon 4s had an autofocus that you could time with a calender. My EOS-10s has a body made of paper it seemed. My Canon T70 was a dog. SO WHAT?!

Yes, the 7D makes lovely pictures. It really does. For many applications it's good enough. But for some, it simply is NOT. I don't make those rules. Discovery-HD, the BBC, PBS, and others make those rules and if we want to submit our work to them, we have to play by their rules. I am not George Lucas, and I can't put my work out to the world on my own money. So I have to follow the same rules that everyone else does.

Oh, and attend a film screening some time. You'd be *amazed* at what "film specs" are. Lucas himself was run up a flagpole for his digitally shot Star Wars episode.

I'm done with this.

The difference is narrative vs. documentary. For narrative, you can control things and avoid rolling shutter and aliasing problems. And, if you screw up, you can do it over. With documentary, you have one chance to capture a moment in time - and you have little control over how it will play out.

So just out of curiosity how would the footage from "The Ranch" be classified?

Narrative or Documentary?

And I'm pretty sure that if that footage was of a random farm in the midwest and was included in a documentary for PBS it would have NO problem passing any QC.

I bet if you told them that it was shot of F900 they couldn't tell the difference.

Perrone, I know you're done with this, but I'd like to chime in and say great post! You're so right.

It's like getting mad because your neighbour says your hammer won't put in railway ties. Well, sometimes you need a sledge. It's just a fact, not a flaw. IF PBS wants you to use a different hammer, just do it if you want your stuff on their channel.

After all, I am confused.

According DxO Labs The Dinamic Range of Canon 7D is 11.7: EOS 7D.

According Barry Green (was found on another forum):
- "For comparison, the Red One was found to deliver 11.3 stops of total dynamic range from top to bottom; an EX1 does about 10.5 stops, an HPX300 does about 10 stops".
- "The Red One resolves about 11.3 stops, but most users seem to put the "usable" range at about 9. On the 7D, I'd say it resolves 9 stops, with a "usable" range of about 7.5, maybe 8".
- "But, anyway, back to the 7D... I configured the 7D as Stu Maschwitz recommends on his ProLost blog for flattening the 5D. Contrast -2, highlight tone priority on, and Neutral style. I put the camera on 1080/24p mode, used the stock lens, ISO 200".

So, the difference in the dinamic range of Canon 7D:
- by DxO Labs - 11.7,
- by Barry Green - 10 (total) or 8 (usable).

Are a methodics is different ?

Details details details! Congrats! Also, I'd like to know what kind of feedback PBS gave with regards to their standards vis a vis the EX and Canon. I understand that the 5d hit the wall with the BBC and that the EX had its ticket punched at Discovery and NGHD. First hand experiences would be interesting to hear.

Narrative. He controlled every shot and everything IN every shot. God it was beautiful.

They ask. And your choice is either to tell the truth and have them tell you no, or to lie, hope they don't actually look too closely and maybe get it on. Or have them find out, by looking at it on a decent monitor, and you not be able to submit again.

When I told the local PBS producer I had a show I was putting together for them, and asked if their specs had changed after the digital switchover, she said no. Just follow Red Book. Simple as that. My EX1 passes for SD delivery and that is exactly what I am going to hand them. Unless I can get a NanoFlash in time and get a passable HD signal.

Why? Because you saw a 4Mbs proxy on Vimeo and you thought that looked great? You think that was a critical look? You think that's the same as some engineer putting your footage up on a calibrated 30" monitor and scopes?

Maybe you're right. Maybe the engineers at PBS and other broadcast stations don't know their business and can't tell. Maybe that $200k of gear they have in the room is just for looks. Hell, what do I know. Maybe I could just slip it by them. They probably won't notice.

Unless I can get the NanoFlash, it's SD delivery off the EX1. I didn't mention the Canon because I won't have it in time to even bother. My local PBS is still SD broadcast, so although I will create an HD master, I'll be making an SD master for them.

I'll have to get clarification on whether the 5D/7D will pass muster for HD broadcast. But since the published Red Book hasn't changed since 2007, I am going to bet no.

Must be; I don't know how they conducted their test but -- 2.5 stops different is ... unheard of. My test was conducted in two ways: I lit up a scene with way too much dynamic range (I think it was 14 stops of total dynamic range, if I remember properly) and shot it on the 7D and then checked what stayed in range and what strayed out of range, and came up with about 9 stops. But that was less scientific than I'd like, so I next got a Stouffer 41-step calibrated grayscale chart, which shows a series of translucent "steps"; each step is equivalent to 1/3 of a stop. You look to see how much you can discern, and when it can't discern any more, you stop counting. On the 7D, that was about 8.3 steps.

And my results correspond pretty much identically with those at dpreview.

So I don't know what DxO is doing differently. They got 10 stops of dynamic range out of a GH1, whereas I got 8.3. I don't know what their methodology is or how they're arriving at the different results.

Thanks Barry!

According dpReview all DSLRs are quite close in Dinamic Range at ISO 200:
- Nikon D300s usable range - 8.4,
- Canon 5D Mark II - 8.4,
- Pentax K-7 - 8.4,
- Canon 7D - 8.3,
- Lumix GH1 - 7.7.

Of course I wonder what the dynamic range would show EX1 on their methodology?
Probably no one will answer this question.

Really? I know Mr Bloom is a talented DP with obviously a strong reputation with the likes of George Lucus, but unless he's got a direct line to God all of the footage I saw was natural light and shot very much the same way that I shoot documentaries.

No. Because I have met Rick McCullum working on a project and he is not known as someone who is that easily impressed. Philip mentioned how impressed he and the others were who viewed this on a 40 FOOT SCREEN. Do you really think that ILM would even consider meeting with Philip if they didn't think mixing the cockpit footage from the 7D/5dMkII with the footage from the F35 wasn't going to work. Maybe you should just pass along the contact info of your PBS producer and Lucas can call him directly and save everyone a lot of time and money.

Also, the bureaucrats at PBS are not the only people who know how to use a waveform and vector scope, in fact I have a couple of those and a 50" calibrated monitor and I know how to use them.

And yes there are plenty of PBS stations that have $200K of worthless SD equipment. KQED wanted a BetaSP for the distribution of a project I produced for the Pacific Mountain Network. I had to go to the local broadcast museum to rent one.

I would call it narrative too in the sense that Mr. Bloom had 100% discretion over which shot to use -- and presumably Perrone's point is that in a documentary you're often beholden to a particular "Live" event as it unfolds. If moire or some other unwanted artifact pops up you're stuck with it. Bloom can choose the perfect shot, but the guy filming a SEAL raid in Somalia is "In the moment" of a dynamic situation and has zero control with just moments to capture the story.

There are all kinds of documentaries and different ways to shoot them. Sure there are times when you have little or no immediate control over a situation, but rarely, if ever, have I not had the opportunity to go to get B-roll footage were I did have "enough" discretion over the shots I thought I needed.

On many of the shots at the Ranch, Philip got to choose the lens and he had a tripod, it didn't appear that he had much more than that. He also had to meet, and work with several people to demonstrate the capabilities of these camera's, its not like they hired him to simply photograph the ranch. I'm sure there was plenty of hospitality but I'm guessing he was pretty busy while he was there and might not of had as much "discretion" as you might think.

I'm not sure its mutually exclusive but I'd characterize that shoot more like a documentary than a narrative.

None of this matters, Perrone's point was that the 7D doesn't "measure up," it doesn't meet PBS's standards. Well I guess that depends on who's doing the shooting and who's doing the measuring. I get the impression from Philip's blog, and what I can see for myself, that he and others don't share that opinion.

There are also other people who are extremely unimpressed with DSLR projected big screen images. Although this could depend on where you sit in the theatre, things tend to look better the further back you sit.

Most cockpit material shouldn't expose too many of the DSLR weaknesses. In the past, 16mm cameras have been used to shoot aerial or with the old gun cameras on specialised mounts to be cut with 35mm footage.

Did not mean to upset your apple cart or drive you out of the conversation.

I hate it when people are such fanboys of Nikon over Canon or PC over Apple, it is stupid.

I see your point and have limited knowledge on this subject. I was simply engaging it to continue my learning.

The PBS Red Book is freely available for anyone to download here:

Red Book . Key Deliverables . PBS Online Web Site | PBS

The Section that pertains to Video quality is quoted below for convenience of the readers. Note that I will have to try to get an exception for the EX1 also (added emphasis is mine):

Scratch the Red One from PBS as it's not 3 chip and not CCD. Who writes this stuff!

Great! That is funny.

The info PDF I found on the site is dated 2007, which is pre RED One.

The BBC specs sheets do get up dated, but I suspect the current DSLRs would be counted as SD same as HDV and Super 16.

All the broadcasters in the UK have similar specifications when they are commissioning.

Our local PBS station (KOPB) gives so many bits to the alternate SD channels that "HD" is a joke. Every scene change looks like a mosaic wipe.

I guess Frontline never read the Red Book... High Definition War Footage Taken with Canon 5D Mark II - Obama's War - Gizmodo

Still, while I really like the 5D2, no way is it the best cam for every situation. It rocks for narrative on a budget though.

When I asked the local producer if there was an update, she said no, and that is the one she refers to for productions. So there you go.

It's pre RED One, it's Pre XDCamEX also which is why I can't hand off my EX1 stuff as HD. It passes muster for SD.

I know people who are shooting HD using the EX series fitted with a Nanoflash.

Right,

According to Red Book I still don't pass because I am shooting CMOS, but I think I can get them to look past that. But I can't hand them XDCamEX footage. HDCam would be ok, so I am trying to get a NanoFlash next fiscal so I can actually hand off "broadcast grade" footage.

The problem, is I have to choose between buying the Nano, or buying Avid... I can't get both... UGH!

Amazing that they count super 16 as SD when its resolution exceeds 1080p, not to mention the 14 stops of latitude.

It's the grain that is seemingly the problem for the engineers, it's too demanding on the transmission codec. 2 Perf 35mm is limited to 250 ASA stock. Although, I suppose by that reckoning 100 ASA or less Super 16 should be fine and of course you could de-grain.

There is a bit of engineering politics going on as well I suspect.

Just rent a Nano for an hour, shoot a little footage that you'll include in your doc, then when you submit say "Yeah, it's XDcam EX, but I used the Nanoflash and ummm, I think you'll be impressed with the look I was able to achieve and anyway who are you takin', Longhorns or the Tide in the BCS?". You wouldn't be lying, you did in fact shoot with the Nano, that 99% was NOT Nano footage is a mere caveat that you needn't elaborate upon.

Nano and the native 35mbs is pretty close from what I've seen.

The closest place to rent a Nano is about 6 states away. That's a no go. And I work with these people every month. Do you really think I want to misrepresent myself?

And you can't tell the difference between 280Mbs I-Frame and XDCamEX? Really? Did you see the Footage test I put on Vimeo when I got the EX1? Dude.. it wasn't pretty.

The BBC specs mean zilch, nada, nothing. I've worked on too many BBC shows that weren't even close to what's in their 'white paper'. If what you shoot is interesting in any way, it might as well be shot with an iPhone. Content beats tech specs 11 times out of 10.

That depends if they are HD shows or SD shows. Even then for HD they allow a percentage of SD content, however, there will always be exceptions if there is no other way of obtaining material.

I have no idea what ESPN HD's specs are but I do know they air some horrible looking SD mixed in the with there HD content and I also know a guy whose had stuff broadcast with ESPNHD that he shot on a crummy HDV camera.