Canon 7D, meet the Billups VFX and ToNi charts (dynamic range)

[philiplipetz]

All roughly same price point, but different compromises. None ideal. So it really is a test of what is most important, IQ, usability, overheating, sound, low light, form factor, interchangeable lenses, resolving power or recording format. At this price point you must give up something and the cost/benefit ratios are not well known.

Great to hear that you had no problem with interviewees when using the 7d. That was a big worry.

Good to know that there are some having to make the choice between those cameras. The HMC-40, as great a little camera as it is, just seems an odd duck compared to the 7D and GH1.

And as for your question…

I recently shot an interview with two 7Ds. We did get a bit of, "wow, you're shooting with a stills camera?" But once I turned the camera around and played back the clip, the response was just, "wow."

And on the same shoot, we walk in the door to scout the location, and the first thing we hear (from the manager of the interviewee) is, "cool, you guys are shooting with the 7D? I've been thinking of getting one."

 

[Terry_Lasater]

Thanks, fellas!

An image sensor is designed for a certain level of sensitivity. This is the "native ISO." To get all other ISO values, including higher AND lower ones, gains and offsets are used, which add noise. Certain values require more complicated adjustments and thus result in more noise.

On the 5D and 7D, the "native ISO" is speculated to be 320, based on noise tests. 160, 640, 1250, and 2500 offer the least noise, compared to 100, 200, 400, 800, 1600, and 3200.

http://www.dvxuser.com/V6/showthread.php?t=185471 -- This thread honestly deserves a sticky as it needs to be covered and understood by everyone.

Not only that, my running theory also considers that higher and cleaner ISO's retain more highlight detail than lower ISOs. However, I don't have the technical know how to pull it off.

It'll be interesting to see what Barry has discovered about these three things in conjunction (native ISO, HTP, Light Optimizer) in conjunction and contrast/comparison to the GH-1 and HMC-40.

My initial guess-- since Barry won't tell us anything, that's all we can do for now-- is that the 7D @ ISO 320 with the other options disabled permits the 7D to capture an atonishing view of another dimension located within the crevices of cardboard boxes found behind Wal-Mart Superstore dumpsters.

 

[Stephen Mick]

(I'm not trying to sidetrack this thread, trust me. But…)

On the same recent shoot, we had the 7D mounted on a Porta-Jib Traveler set on a Sachtler 20 tripod. It looked ridiculous. Until we looked at the footage.

Hopefully Barry will have some sun soon to re-run his tests. I know many are on the edge of their seats.

 

[ydgmdlu]

I bet it's the GH-1. Honestly, if it were not for the codec and mudd issues I think it could be capable of more (in terms of its image). 1080p looks pretty sharp.

I'm not sure that Barry has his GH1 back in his hands yet.

 

[Barry_Green]

but are there really a large group of people trying to decide between the 7D (or GH1) and the HMC-40?

Not sure if there are any, but the reason I drag the HMC40 into this comparison is because it's been the only native-res camera I have had on hand.

Secondly, I'm very interested in seeing how the DSLR compares to a conventional video camera; has it "arrived"? Is it truly time to consider retiring conventional video in favor of this new paradigm? If the DSLR does everything the conventional video camera does, as well or better, then yeah, open the floodgates. But if it doesn't, if it is inferior in some ways, then we now have a new data point to consider and something to mark progress by.

Third, as you said, I think it's a little fun to look at both these hybrid products and ask -- how does a still camera that shoots video, compare to a video camera that shoots stills? If someone had $2,000 and they needed both, which way is a better way to spend their dough? But, heck, I may be the only one interested in that; I find the whole "convergence" thing interesting; will it really be that long before our video-enabled DSLR also makes phone calls and surfs the web? And if so, what are we giving up along the way?

Fourth -- how good is good? We've all seen amazing footage from the 5D, 7D, and GH1, but -- is it really, truly, anything better than what we've had before? The BMX video -- could that have been done on an HMC150? I think probably so. So what's the real, tangible improvement here -- if any? Obviously the price tag, yes, and the convenience of not hauling around an adapter, heck yes. But while everyone's assuming we've taken giant steps forward, and the new DSLRs are... well, "of course they're sharper, of course they have more dynamic range, of course they're this and that" -- are they? Or have we taken a step backwards? I'd like to know. So when I find that the 7D has a stop less dynamic range and significantly lower resolution than the bottom-of-the-line pro video camera, I think that's newsworthy. Not to discount other aspects that are way in the SLR's favor -- interchangeable lenses and low-noise high-ISO and native shallow DOF being the three most obvious and well-known, but -- everyone talks about those, nobody talks about the others, they all just seem to assume. Assuming is not good, generally.

When Barry gets his GH1 back…well those are tests that are certainly going to be more consistent with my needs.

I'm certain many folks will be more interested in that particular comparison, for sure. And that's what I wish I could have been using as the comparison all this time; I really only pulled the HMC40 into the discussion because it was the only other 1920x1080 product I had around at the time.

Yes, it's raining and cloudy. Most disappointing. But, I may have found a way to at least verify whether my original test was conducted correctly or not, so I'm gonna fiddle around with that for a little bit and if I can determine that I definitely didn't make the mistake I'm afraid I might have made, then I can post the results later today.

 

[Stephen Mick]

I'm very interested in seeing how the DSLR compares to a conventional video camera; has it "arrived"? Is it truly time to consider retiring conventional video in favor of this new paradigm? If the DSLR does everything the conventional video camera does, as well or better, then yeah, open the floodgates. But if it doesn't, if it is inferior in some ways, then we now have a new data point to consider and something to mark progress by.

I'm with you on this, for sure. It's intriguing to see how any cameras stack up against each other, especially when they fit the same price point. And data is data, which is always good to have to judge where we are now against where we'll find ourselves in three months, six months or even a year.

But, as has been stated in other threads, as great and useful of tools as DSLRs are, their potential is not yet not "fully realized," in my mind. When they have proper audio handling, without a third-party software hack, no limits on recording times, and other features professional videographers and filmmakers come to depend on, then they will have "arrived," I believe.

I have used the GH1 and 7D on actual paying gigs, including one shooting with just the 7D and no camcorder "safety net." While it's certainly workable for specific jobs, I'm not retiring my EX1/NanoFlash just yet, that's for sure. And quite honestly, there are jobs where any DSLR as a primary acquisition source is just silly. (I've got 1.25TB of fly-fishing footage in DVCPRO-HD. Imagine how much storage I'd need for equivalent 7D files!)

Convergence is a beautiful thing. The ability to deliver the kind of images any of these cameras offer is truly amazing, more so when I think that only a few years ago I was shooting with a DVX. But there's still a long ways to go before the workflow and comfort level of using the 7D, the GH1 or any DSLR as an "A-Camera" catches up to the image quality. (Or at least that's my take.)

The paradigm may be shifting, but it seems to be moving very slowly.

 

[Barry_Green]

Okay, I have verified that I didn't blow it. So -- posts will be coming later today with footage; no need to reshoot.

 

[Stephen Mick]

I've got the popcorn in the microwave, waiting patiently.

 

[johnhafner]

Auto lighting ruiner +1 That was my thread on cinema 5d haha.

 

[johnhafner]

Barry what lens did you use on this test for the 7d?

I can tell the difference in sharpness between lenses even in Video mode so it makes a difference.

Also, did you ever re-run this test with Auto Lighting Optimizer off? ALO adds noise and overall bad image IMHO.

Finally, the default contrast settings CRUSH the blacks on the 7d with most lenses. I usually run the contrast 2-4 steps below the midpoint to bring back shadow detail.

I'd also interested to see the same test ran with a nice prime lens like the Pentax Takumar 50mm SMC 1.4 or Canon 1.8. Poor lenses DO reduce contrast the sensors sees and thus overall dynamic range can be limited by your lens choice. ( Most cheaper Zooms for the 7d have poor contrast and saturation IMHO)

 

[Barry_Green]

See the Stouffer thread where I re-ran the test using the "flatten your 7D" settings, contrast at minimum, and the Zeiss ZF 50mm/1.4 lens at f/5.6. Auto Lighting Optimizer is always always off.