Messages

256

EOS Bodies / Re: Earthshatteringly Disappointed With 7D

« on: December 09, 2011, 01:12:22 PM »

However, I also agree with neuro - I thought the same thing when reading your initial post where you state you are mainly a landscape shooter. The 7D is a very good camera - no doubt. But landscape shooting was never going to be it's strongpoint. Full frame will usually handle landscapes more successfully.

I am mainly a landscape shooter, and I've purposely shot a 7D and 5D2 side by side to see if it was worth investing in the 5D2. At low to mid ISO I could not discern between 24" and 30" prints, and neither could anyone I showed them to.

The 7D is a fine landscape camera up to 30", which incidentally is the same maximum print size I would put on 5D2 files for critically reviewed landscape prints.

257

EOS Bodies / Re: Earthshatteringly Disappointed With 7D

« on: December 09, 2011, 01:09:33 PM »

My 7D will beat the crap out of any 5D you care to pick, at the image level...

The 5D will have lower ISO noise across the board...and that's pretty relevent to the issue the OP describes.

It will not have less noise in print, nor at ISO 3200. (I'm assuming you mean 5D and not 5D2.)

Still, armed with the info that the primary use was landscape photography, I would not have recommended the 7D.  Don't get me wrong - I think the 7D is a great camera.  But if I'm going out to shoot landscapes, I'll grab my 5DII every single time.

I've made comparison landscape prints at 24" and 30" from both bodies. At low ISO they are indistinguishable.

258

EOS Bodies / Re: Earthshatteringly Disappointed With 7D

« on: December 09, 2011, 01:00:52 PM »

Ok - here's one that has just received minor sharpening (what I would do for any shot coming in at ISO 200)

http://farm8.staticflickr.com/7011/6480510009_df4a26af47_o.jpg

And here's another that I've tried my hardest to "fixup" with sharpening and NR:

http://farm8.staticflickr.com/7006/6480518429_3cd975c2a1_o.jpg

Last night when I posted I had my MacBook Pro on my lap. I actually hate evaluating images on my MacBook screen because of the contrast and saturation, and the fact that the brightness is all over the place based on where I'm sitting.

This morning I took a 2nd look on my 23" HD monitor which is calibrated in all respects for photography and printing. Not just color, but brightness, contrast, and saturation are set to match printed output from my 3880 as closely as possible. Another important difference: I think last night I just clicked the image and used Preview. This morning, PS CS4. Screen view quality depends a lot on the scaling algorithm used by the viewing program, and a lot of programs produce scaled views which are grainier and softer than PS. Apple's Preview can be particularly bad with high resolution images scaled to certain sizes.

Looking at the first link...

* At 50% (22" print size on my screen) there is no noise visible, period.

* At 66% (36" print size) detail seems a bit rough, but there is no real noise to speak of.

* At 100% (54" print size) there is visible noise in some areas. But if you have any kind of film background, it's far less than the best 35mm films ever were, and even less than MF films printed this size. (Of course MF film would have much more detail at this size.)

Forget what I said last night that it seemed to have more noise then I would expect. Viewed in PS on a properly calibrated monitor it's fine. Exposing to the right would have helped with the noise at pixel peeping views, but made no real difference in print. The lens issues (CA and some softness on one side) stand out far more and at smaller sizes than any noise.

FYI, a standard Noise Ninja pass using their 7D ISO 200 profile eliminates pretty much all of what you see even at 100%. At that point you have to zoom to 200% (108" print) to see noise.

And this is all after you sharpened. If I use Noise Ninja, I use it before I apply local contrast enhancement and sharpening.

I trust PS scaling and my monitor 100%. What I see is what comes off my printer.

So how are you viewing and evaluating this?

259

EOS Bodies / Re: Earthshatteringly Disappointed With 7D

« on: December 09, 2011, 07:10:53 AM »

* Pixel peeping means greater magnification for the 7D image.

* Finer pixels also means that the 7D's noise is "sharper" than noise from lower resolution cameras even after you equalize the pixel dimensions and viewing conditions. It's not necessarily more noise and it's not necessarily apparent in even the largest prints, but it seems more obtrusive while pixel peeping. (Same thing with the D3x vs D3s.)

* I regularly make 24" and 30" landscape prints. I use sharpening, and sometimes also use NR. The prints are tack sharp, full of detail, and show no noise. In fact, I've made a few 20" portrait prints shot at ISO 800 and they are gorgeous.

* I've even got some 20" prints from cropped 7D images, surfing shots, at ISO 400. No noise on the print.

That said, there's a little noise even at low ISO while pixel peeping. But pixel peeping is like looking at a 60-70" print from a couple feet.

With all of that out of the way...your first sample does look like it has more noise then I would expect. Did you nail the exposure, or underexpose and adjust in post?

260

EOS Bodies / Re: How can Nikon remain in business with 12MP FF?

« on: December 07, 2011, 05:20:09 AM »

In the example below (100% crops from the 5DII, processed with DPP using no NR or other adjustments), the light source illumination was the same.  For the shot on the right, ND filters were used reduce illumination by a fixed amount to match exposure, giving (IMO) a more common real-world situation where you might need to use ISO 6400.

Imaging Resource tests don't show as dramatic a difference as shutter speed drops. Are you sure you got the exposure right?

261

EOS Bodies / Re: How can Nikon remain in business with 12MP FF?

« on: December 07, 2011, 04:47:25 AM »

12 mp is plenty for almost any usage. In fact for most pro photographers more than 12 mp serves literally no purpose whatsoever. To this majority therefore Nikon provides cameras that have much higher ISO performance, much better A/F and much better ergonomics than Canon, Nikon devastates Canon in performance on it's bodies.

Hyperbole on every point. The D3s has better high ISO (not much better) because it's a newer sensor design than the 5D2. The D700 does not have better high ISO despite being 12 MP.

AF is better on the D700. With Canon you can get a low cost 21 MP FF body, with Nikon you can get low cost pro AF in a FF body. Only their sales departments know which differentiation is better. (That said, I don't understand it when competitors fail to take open shots. Canon should have put pro AF in the 5D2.)

Ergonomics is very much a matter of personal preference. There are things I like and dislike about both.

That said, 12 MP is enough for many uses, so Nikon is not at a huge disadvantage. But there are some popular uses which require more MP. I would rather shoot landscapes with my 7D than with my friend's D700.

Canon's dynamic range and USO are laughable in comparison.

A roughly 1 stop difference on either is not laughable, especially when Nikon has younger sensor designs. Let's see what the differences are with Canon's newest FF sensor.

262

United States / Re: How Many MP's On A FF Equals a 35mm Canon AE1 Film Camera?

« on: December 06, 2011, 05:47:54 AM »

Digital is certainly still behind film in available dynamic range.

That depends entirely on the film. Velvia? DSLRs have a much wider DR. Portra? They're not quite there yet. The others fall somewhere in between. With B&W it depends on the emulsion and processing. Some film and developer combos can yield incredible DR.

One thing you'll find is that there is usually a trade off with film, or at least with color film, between resolution and DR. Velvia can hold its own on a good scanner in terms of resolution, but has much less DR. Portra can exceed a DSLR's DR, but can't touch it in terms of resolution. The combination of resolution, dynamic range, and high ISO capability in today's DSLRs is simply unmatched in 35mm film.

263

EOS Bodies - For Stills / Re: Dynamic Range, Grad ND Filters and Canon vs. Nikon

« on: December 05, 2011, 11:31:16 PM »

This got me thinking, so I looked at DxOMark to compare the 60D (which I use) against the Nikon D7000.  The D7000 is rated at 2 stops more DR.

DxOMark is horribly off in every DR test I've ever taken the time to double check.

This is not an attempt to stir the cauldron of anti-DxO commenters;

Do you want accurate information, or fantasy?

The 60D offers about 10 stops when processed for maximum DR in RAW. The D7000 offers about 11.

Note that DR measurements vary based on what you consider to be an acceptable noise floor. If you are willing to accept more noise, you can recover more shadow detail. This is part of the reason why double processing a RAW file and blending them yields greater DR then a single pass at the file. But the difference between the two bodies is still going to be small for the same technique (JPEG; RAW; RAW x2 pass).

For those of you serious landscape photographers, what is your "gold standard" for DR?  What scene do you need to capture, without a GND filter, to declare that your camera has "enough" DR?  (and if you tell me you want to point directly at the sun and still see detail in a gopher hole I won't take you seriously. )

An 18 stop sensor would eliminate most of the exposure blending I currently do when shooting sunset landscapes. I don't think the D7000 wouldn't eliminate any.

264

Lenses / Re: Canon 85mm 1.2LII vs Sigma 85mm 1.4

« on: December 05, 2011, 06:55:27 PM »

I have found my "hit rate" with f/1.4- f/2 lenses (50 f/1.4 and 135 f/2) improves radically if I shoot using AI focus and shot at 8 fps on the MkIV, particularly when shooting people handheld. Just that tiny rocking on your feet is enough to shift focus from the eyes to somewhere useless.

I think it's more than just front/back movement, though that's an important point. I get the feeling that one shot AF quits a split second too soon with fast primes, or at least with Sigma fast primes. AI Servo keeps trying even if you and the subject are stationary. I get a higher hit rate if I use AI Servo and wait to fire until the AF has settled.

265

United States / Re: How Many MP's On A FF Equals a 35mm Canon AE1 Film Camera?

« on: December 05, 2011, 05:16:24 PM »

I would like to know how many MP's does a FF sensor need to be equal to a 35mm Film Camera?

I did some pretty extensive comparisons when I picked up my 7D. The 7D matches Imacon scanned, 35mm Velvia 50 on high contrast detail and beats it on low contrast and color detail. (Film's ability to resolve detail varies based on detail contrast to a greater degree than digital.) I would put 35mm Velvia 50 at 12-18 MP depending on target contrast. For most photographic scenes I would put it at around 12 MP.

Velvia 50 is the highest resolution color film in production, and it pretty much matches the highest resolution B&W films available for general photography. (Special purpose B&W films can go higher, but they make lousy picture films.) So any other film, or any lesser scanner, and the number for film goes down considerably. Most films don't break 10 MP. Most scanners are even worse regardless of their reported dpi.

I don't believe there's any detail that can be recovered by projection that is not recovered by an Imacon.

266

EOS Bodies / Re: A tongue-in-cheek missive to those who want a high MP 5D3 :)

« on: December 05, 2011, 07:27:44 AM »

But I completely and totally reject the argument that its possible to ever have a sensor with too many megapixels. Yes, extreme MP counts create problems for the scientists and designers to overcome and I'm confident that they will overcome them.

When those 'problems' involve basic principles of physics, like diffraction, overcoming them may prove difficult...

Yes, diffraction imposes an upper limit on pixel density. No, we are not there yet, and a 35-45 MP FF sensor would yield benefits in fine detail and maximum print size.

267

Lenses / Re: Battle of the 50mm's - 1.8, 1.4 and 1.2L

« on: December 05, 2011, 07:24:36 AM »

How would you properly test out a 'fast 50mm' in a store in say less than 5mins with bad/insufficient lighting? Are there some of of 'tests' that you could do, e.g. focussing (both AF/M, there's going to be insufficient light meaning the lens will have trouble trying to focus.

I actually bought my Sigma used and spent less then 5 minutes testing it. I simply focused on various objects, in good light, shot, and checked the results. The results were reasonable for f/1.4. Most were in focus and those which were not I could get in focus with a slight variation on AF mode or technique. I own other fast lenses so I knew what to expect here.

If you lack confidence in your ability to judge on the fly like this, take a focus adjustment chart, set it up, and shoot it, then judge the results at maximum magnification on the LCD. Manually defocus, AF, and shoot, then repeat several times, and do so at a couple different distances. I wouldn't expect perfection. Any fast lens I've ever tried this with will be a bit off 1 or 2 times out of 10. But if it's good for most shots, or is good after a quick MFA, then you're good. If it's all over the place, then check another copy. And, again, I would actually try AI Servo and shooting after the AF has "settled". I actually like this technique for any fast prime, though it seems to make a larger difference with the Sigma (and maybe the other 50's?).

This can be done in 5 minutes.

268

EOS Bodies / Re: A tongue-in-cheek missive to those who want a high MP 5D3 :)

« on: December 05, 2011, 02:23:35 AM »

We are at the threshold if not well beyond lens resolution with some of the current sensors and lens combinations.

Nope. There are plenty of lenses which can resolve more then the 7D's sensor can deliver. At FF size the 7D's pixel pitch would result in 45 MP images. And I would gladly pay for a FF body with a 45 MP sensor that had the noise characteristics of the 7D. IQ at low ISO would be comparable to the 645D, and it would still be usable at ISO 3200 and 6400.

You're the guy bragging about the V8 in his car, talking trash about anyone with a V6 and claiming that they just can't squeeze the same power out of a V6, about to get his rear handed to him by a V6 with twin turbos.

269

Lenses / Re: Battle of the 50mm's - 1.8, 1.4 and 1.2L

« on: December 04, 2011, 10:09:20 PM »

Here's a solid vote for the Sigma 50 f/1.4. It's optically better at wide apertures than even the f/1.2L (check the TDP ISO crops). It is built very well with a true ring sonic motor. (Though it's not weather sealed. If you need that the L is the only option.) And it has absolutely amazing bokeh.

The Canon f/1.8 and f/1.4 50's are only usable wide open in an emergency. The Sigma is very good and usable wide open, excellent at f/2, and stunning at f/2.8. 20" and 30" portrait prints from a 7D and the Sigma are no problem.

The focus issues are exaggerated IMHO. Every fast 50 has "focus issues" because nailing focus at f/2 and wider is difficult. FYI I found the 50 f/1.8 to be far less precise and reliable in terms of focus than the Sigma 50 f/1.4.

If you're really worried, go to a good local shop and test a few samples before buying.

Fast lenses shot wide open challenge an AF system. When I'm working really close, I use spot AF. At other distances I use single point AF. I also use AI Servo with this lens even when shooting still subjects. I find that holding AF and waiting for the lens to "settle" gives me a higher keeper rate. It doesn't take very long, but it seems like one shot AF decides on focus just a fraction of a second too soon with this lens.

270

Lenses / Re: Sigma 50mm f1.4 EX DG HSM....are they any good?

« on: December 01, 2011, 07:21:19 PM »

But again, this guy is going shopping, not ordering online. He can cherry pick to get one with decent AF characteristics.

For himself, sure.  But he also has to deal with the bitching of people who are unhappy with the rental, which is why he warns potential renters up front.

Sorry...I ment OP's friend