Messages

226

EOS Bodies - For Stills / Re: Do FF cameras perform better in low light conditions?

« on: February 05, 2012, 02:19:30 PM »

Do FF cameras perform better in low light conditions such as in a high school gym or at an outdoor stadium at night? Would a 5D Mark II be my best choice for upgrading from a 40D? I was considering a 7D, however, I have read a couple of entries on this forum where it is mentioned that the 7D is not as clear as the 40D.

The 7D out performs the (still quite good) 40D in every way. But people have a tendency to compare images at 100% in PS. They fail to understand that higher resolution at 100% = greater magnification of the APS-C or FF image. At a target print or screen size the higher resolution image will be better even though it may look "worse" at 100% where any softness, blur, or noise is magnified more.

The 5D2 has about a 1 to 1.5 stop noise advantage at high ISO (i.e. 3200, 6400). I don't find this to be noticeable in small and medium prints. ISO 1600 and 3200 are perfectly usable on the 7D. I've printed plenty of ISO 3200 8x10's from HS gym basketball games and they are clean. The few 16x20's I've done do show some noise, but they're still very good prints.

The 5D2 would show an IQ advantage for a 16x20 under gym conditions, but I'm not confident the AF would hold up. Nor do I think I would trade 8 fps for 1 stop lower noise.

FYI, in a gym I generally shoot my 85 f/1.8 at f/2 or f/2.8 with ISO ranging from 1600-3200.

227

Third Party Manufacturers / Re: D800 WILL BE ANNOUNCED IN THE FIRST WEEK OF FEB.

« on: February 02, 2012, 11:15:01 AM »

Why would you say that based on that picture?  Unless you are into really large print, you will not see much different between 22MP and 36MP.  You will see a big difference however in the better ISO performance from a 5DmkIII for the every day use (hopefully anyway).

This is a misconception. Sensor size and technology level are the primary drivers of high ISO performance, not pixel size. I wouldn't bet on a 5D3 having better high ISO just based on pixel size.

I do agree that for normal print sizes you would not notice a difference between 22 and 36 MP. However, for those of us who do like to make large prints, the difference will be there and might be enough to force a switch if Canon does not answer Nikon quickly.

Even if the 5D3 has superior high ISO, how many people will see it "every day"? Most shots are made in the ISO 100-800 range, not 6400 and 12,800. I don't want to sacrifice total resolution for a little less noise at an ISO I'll never use.

228

EOS Bodies - For Stills / Re: 5D Mark II or 1D Mark III

« on: January 22, 2012, 11:17:58 AM »

If you are intent on parting with some $$$, then I'm not sure the 10-22 or Tokina will get you the image quality you may want, if you're used to L lenses.

The 10-22 is at least as good as the 17-40L. The Tokina 11-16 is better, the crop equivalent of the 16-35L II in terms of IQ.

The 5D Mark ii would be an excellent choice, and the Full Frame with your existing 17-40 would be as wide as you'd ever need.

The 5D2 is a great camera. However, a 60D or 7D + Tokina 11-16 will out perform the 5D2 + 17-40L at low to mid ISO. I have both those lenses and have done this comparison. On FF the 17-40L doesn't have the edge to edge sharpness / fine detail of the Tokina on crop.

That said, this is almost splitting hairs. You can make good large prints with either combo. But the 60D + Tokina is cheaper than the 5D2 by itself, yet will actually edge it out in landscapes given the 17-40L on FF.

229

EOS Bodies / Re: 5d or 1ds

« on: January 04, 2012, 07:36:06 PM »

Hmmm, interesting. Pity I don't like the photos out of the 7d, otherwise it would be a great choice. To me though, having FF or APS-H at the very least is important for low light use and lens choice. I actually DON'T like the extra reach you get with crop bodies. I suppose it's what I deserve for shooting film

Thanks for your suggestion,
Thomas.

+1 ... and I thought I was alone with this opinion

+2 ... you're not even close to the only one.

For those who haven't used these cameras, go check out the DP Review and Imaging Resource studio scenes and samples. 7D shots look as good as 5D shots out of camera across the board, except of course for having more fine detail.

If sports is a requirement it's just foolish to overlook the 7D. And if sports is not a requirement, it's foolish to overlook the 60D which will produce larger landscape prints with more fine detail than a 5D, and do so with cheaper glass on the UWA end. Both just edge out the 5D at high ISO (less color blotching). The 5D is close, but those who worship 36x24mm have the wrong sensor as the winner.

BTW, the 1Ds isn't even an option if low light is a requirement. It is worse at ISO 1250, its max ISO, then the 60D / 7D or 5D at ISO 3200! The technology is just too old there for it to be competitive.

As long as a FF has sufficient resolution to crop to the FOV you desire in the event you run out of focal length, there is precisely ZERO benefit to a crop body.

But 12 MP FF doesn't have sufficient resolution unless you're only making small prints. Right out of the gate the 5D is down to 4.8 MP at APS-C crop. There's no contest in the detail or sharpness between a 5D 4.8 MP crop and a 7D 18 MP file. One will just stretch to 11x14, the other will make a fine 20x30. If you have to crop any more you're done with the 5D. And that's assuming the 5D held focus.

The "extra reach" is an illusion.  It is a net detriment, not a benefit.  There is no sane argument to the contrary.

I've got 9 MP crop, tack sharp, detailed and noise free 20" surfing prints off the 7D that would be 4x6 prints at best had I been using a 5D and cropping to match magnification. Evening the odds would only require a $10,000 500L lens.

Even at the wide end there can be benefits to crop. When I tested them I couldn't see the difference in print between a 5D2 + 16-35L II and a 7D + Tokina 11-16 at 24" print size, but the Tokina lens was $900 cheaper. And against the price competitive 17-40L (which I own) it's no contest: the crop combo yields higher IQ.

All things being equal a FF sensor has the performance advantage in non focal length limited scenarios. But all things are rarely equal.

230

Lenses / Re: 85mm f/1.8 or 300mm f/4 for baby photos

« on: December 20, 2011, 03:42:45 PM »

I own both of these lenses. Let me add yet another vote for the 85 f/1.8. The 300 f/4, while a great lens, is not suitable for photographing babies, especially indoors.

When I photograph relatives during holidays and other family gatherings, I'm often using my Sigma 50 f/1.4 on crop, which is about the same FoV as the 85mm on FF. I also use the 85 f/1.8 on crop, but I tend to reach for the 50mm. It's about perfect for indoor portraits. You'll love the 85mm on FF.

Also: use your flash if needed, just bounce it off the ceiling.

231

EOS Bodies / Re: 5d mkII vs 50D in terms of AF

« on: December 20, 2011, 09:56:32 AM »

I'm mainly looking to get a FF for less diffraction with smaller apertures. The Zeiss 18mm seems like a great lens, but I'm also open to other suggestions for ultra wide lenses that will accept filters. And eventually the Canon TS-E 24mm f/3.5.

Diffraction doesn't impact any format more than any other for the same FoV and DoF. This is true from 4/3rds through large format.

That said, there are some lens options which call for FF. If you buy a 17mm T/S or a 24 f/1.4L, you probably want the stated focal length, not the cropped focal length. And T/S lenses don't work well with the flash housing on the crop bodies. (FF bodies don't have a built in flash.)

There are of course excellent UWA zoom options available now for crop if the goal is simply to go wide. Tokina 11-16 f/2.8, Canon 10-22, Sigma 8-16, etc.

232

EOS Bodies / Re: Earthshatteringly Disappointed With 7D

« on: December 19, 2011, 04:51:05 PM »

Thanks... i was hoping with the much lower pixel density it would be lower like F16, but F10 i suppose is a stop and a third more roughly... saving up for your 1dx? 

Note that diffraction does not impact any format more than any other for the same FoV and DoF. Also, as neuroanatomist pointed out, it's a subtle effect at first and becomes progressively stronger as you continue to stop down.

On crop at the 7D's resolution I don't really notice it at all until f/11, where sharpening offsets it nicely. f/16 is still usable for most intents (subject matter / print size combos), but I will not stop down to f/22.

233

EOS Bodies / Re: Earthshatteringly Disappointed With 7D

« on: December 19, 2011, 04:45:06 PM »

Yeah, I was curoius about that too.  I don't see how that can be the case.  Certainly for the same framing and aperture, the 7D will have less OOF blur - but bokeh is the quality of that blur, and I had thought that was a property of the lens for the most part.  The one area in which sensor size might play a role in bokeh is the 'cat's-eye' shapes of OOF highlights, which result from optical vignetting - but in that case, the crop sensor would be better than FF (assuming we're comparing the same lens, but that's the only possible relevant comparison for bokeh), since the smaller sensor reduces vignetting.

Yes, with one caveat: if the lenses being used on FF and crop have the same diameter of aperture, and the background is far enough in the distance, then background blur will be identical. Blur near the plane of focus is controlled by DoF. Blur far outside the plane of focus is controlled by physical aperture size. (Not the f-stop, but the actual size of the opening for the lens at the given f-stop.)

This doesn't often happen indoors because the background is generally too close. For outdoor shots though you can end up with the same background blur on FF and crop. Subject DoF will still be more shallow on FF, though that may or may not be desirable.

234

EOS Bodies / Re: Earthshatteringly Disappointed With 7D

« on: December 19, 2011, 04:35:18 PM »

I regret buying my 7Ds as, yes I got more in focus than the 5DII but each 5DII image that is in focus is far better than those from the 7D - especially the bokeh from the 7D which is horrible in comparison.

LOL! Bokeh is entirely a quality of the LENS.

If you don't know this then don't bother to offer an opinion on how the bodies compare.

The one beauty about the 5DII is that you can use the centre point for focus then crop to get the AF point off centre and still have a better IQ than any 7D will give.

A demonstrably false statement since the two are very evenly matched at low to mid ISO when the 5D2 has all of its pixels, something that can be confirmed by comparing test images and measurements at sites like DP Review and Imaging Resource (to eliminate any chance of operator error).

235

EOS Bodies / Re: Earthshatteringly Disappointed With 7D

« on: December 18, 2011, 03:21:35 AM »

1. I can see the noise from an ISO 100 shot on my 7D when looking at the entire photo on my 27" screen.  That is... NOT a 100% crop.  Quit saying that noise is only an issue at 100%.

I cannot see this on screen or in print, and my printer can most certainly out resolve my screen. If you are seeing noise at 27" equivalent print size, then your exposure was off, or some action in post brought it out, or the scaling algorithm is messing up.

I've got ISO 800 shots which look great on my monitor and print great at 24". Granted they've had a bit of Noise Ninja filtering, but still.

2.  Next you will say "calibrate your monitor!".  That has nothing to do with it... as my XSi shots look fine when viewing the whole image on the screen.

Though the XSi does have measurably less noise at ISO 100, it's not significant and should not be human observable even when pixel peeping. The Imaging Resource comparometer bears this out. You have to shoot under identical conditions, and when shot under identical conditions these cameras do not differ in noise at ISO 100.

2.5 (Ninja Edit) Extreme resolution monitors are on the horizon (most likely this year). There are already talks of a 2880x1800 Macbook Pro... and an iMac could come out with 5120x2800!  At those resolutions npise is going to start to be easily visible in a lot of photos!

There's no noise to speak of from either camera at ISO 100, even when pixel peeping, given a proper exposure.

3.  Cropping:  one of the main reasons to move up to a higher MP is so you can crop a bit closer.  If your image has a ton of noise in it that negates a lot of your ability to crop.

I've made 16x20" surfing prints from 8-10 MP crops of 7D files, with ISOs up to 800. Noise ranges from non-obtrusive to none at all. Again this assumes proper exposure and some post work, but it shows what can be done.

4.  Different people have different purposes for their photography.  If you are doing portraits, you might leave the final picture a bit soft so as to downplay facial blemishes.  However, if you are shooting landscapes you want to bring out every detail and that means "sharpening".  If there is a ton of noise hanging around it might not matter to the person doing portraits, however if you are trying to sharpen tiny details in your photo (that you bought an 18MP camera to capture) noise makes life extremely difficult.

You yourself produced a nice, sharp, detailed landscape with the 7D and it had no real noise to speak of.

I don't know what you're looking at that you think the 7D has any real noise at lower ISOs. I can only imagine that exposures are off, or you're mistaking surface texture for detail, or the scaling algorithm is messing up. (Apple's preview app can really screw up the scaling of anything, even a 5D2 image. Not sure why. It seems to be random, but it happens.)

236

EOS Bodies / Re: Earthshatteringly Disappointed With 7D

« on: December 15, 2011, 05:19:24 PM »

Huh? I shoot everything with my 7D and never do any noise reduction in post process, I mainly shoot raw and have never had any problems with noise on ISOs up to 1600... I've printed my pictures on sizes up to 40''x56''...here's a gallery of some of my pics (landscapes, portraits, action, macro, you name it...)

http://www.500px.com/ruddyflorentino

Nice work!

(sarcasm on) But you better zoom in to 300% because I think there's noise and you need a D3s  (sarcasm off)

237

EOS Bodies / Re: Earthshatteringly Disappointed With 7D

« on: December 15, 2011, 05:17:05 PM »

There are WAY too many 7D users complaining about the noise.

There are also way too many people who spend all day studying mundane snapshots at 100% so they can fight about equipment in online forums. Personally I'm far more interested in the opinion of photographers making prints and online galleries, i.e. producing and viewing photographs at normal sizes for the purpose of enjoying photography.

How do you sharpen the subject without sharpening the noise too?

One technique I occasionally use is to apply sharpening from within Noise Ninja, apart from any separate NR step. I turn down the noise sliders so that they just balance out the sharpening. As long as I balance the two correctly it ends up working pretty well and sharpens detail without enhancing grain or excessively smoothing anything out.

Bottom line, viewing the 7D at 100% is SAND PAPER.

http://web.canon.jp/imaging/eosd/samples/eos7d/downloads/001.jpg

Bottom line is you don't know what you're talking about.

The proof is in the pudding.

Agreed. Which is why nobody talks about noise while looking at my 17x22" portfolio. I make prints. I don't spend hours obsessing over 100% views. You would be shocked at how much noise is literally invisible even at larger print sizes. As my post processing has improved I have reworked some images to further reduce noise, only to see no difference in a 24" print.

Obsessing over 100% views is the height of ignorance. A 100% view means nothing outside of the context of the total resolution and maximum possible print size. It's like looking at 35mm and 4x5 film under a microscope and saying 4x5 has just as much grain and softer details. Then spending all day arguing in forums that 4x5 is no good and people should just use 35mm. A person who did this would be laughed out of the forum for failing to realize the context of the microscopic view, and the fact that 4x5 can make much larger prints which will be sharper and cleaner than any 35mm print.

Yeah, these arguments about noise at 100% look that stupid.

238

EOS Bodies / Re: Can a 7D do Landscapes?

« on: December 12, 2011, 02:07:07 AM »

I tried tons of aperutres and f/8 gave the best results (I gave one comparison in the OP).  In the case of Shot1 I simply just focused too closely to be able to get everything "in focus" without losing a ton of detail overrall.  With my old camera it was more forgiving in that regard... because of that I grew quite sloppy in my hyperfocusing technique... guess I'll have to be more careful in the future!

I'm surprised you even needed f/8. How far from the barn were you?

As for the tetons at sunset / sunrise: they are spectacular!

Nice shots. I especially like the first one.

239

EOS Bodies / Re: Can a 7D do Landscapes?

« on: December 12, 2011, 02:02:33 AM »

Short Answer: Yes!

Congratulations on giving the 7D an honest run and discovering what it can do! Good shots, good comparisons, and good analysis. Your detailed post can serve as a guideline to future 7D/60D/T2i owners.

A couple comments:

* I often use Local Contrast Enhancement on my photos. http://www.luminous-landscape.com/tutorials/contrast-enhancement.shtml. I bring this up because this can also greatly increase the perception of sharpness and fine detail in some shots.

* Playing with an online DoF calculator is a great way to get a feel for what apertures are actually needed in different situations: http://www.dofmaster.com/dofjs.html.

* +1 on the LiveView focusing suggestion. I do it all the time. This also helps to achieve the optimum combination of focus and aperture as you can scroll around the scene and use DoF preview. Doing this rather than blindly stopping down can create a greater difference in fine detail then any difference between Canon's current APS-C and FF sensors.

I'm glad you took the effort to get the best out of the 7D. Enjoy it. Yes, eventually I think a >30 MP 5D3 will come out. But that could be a ways off and the 7D is a brilliant 24" - 30" print landscape camera in the mean time. You really could not get much more out of the 5D2 sensor for this purpose.

240

EOS Bodies / Re: 7D + 10-22mm or 5D III + 16-35mm L II?

« on: December 11, 2011, 06:43:05 PM »

But the OP was talking about waiting for the 5d3, and got a 7d whilst waiting when he could have gotten the 5d2 was my point.

jrista said his choice was based on wildlife shooting and on cost.

To say the 5d2 and 7d is comparable at all is just wrong, plain and simple.

Looks like I skewered a sacred cow 

I use both FF and 1,3 crop and that beautiful smooth bokeh and transitions of the FF is a big difference to the 1,3.

A fast prime on an APS-C camera can easily diffuse a background. For portraits the difference between something like an 85mm prime on crop and on FF is the difference between one eye in focus and one eyelash. I don't typically want one eyelash in focus, and usually stop down to f/2 or f/2.8 even with a 50mm on crop to make sure I get both eyes and most of the face in focus.

There are situations where the more shallow DoF of FF can be useful, but the difference is not nearly as dramatic as it is made out to be. At the end of the day we obsess too much about small differences. Shuffle up some prints shot with a wide open prime on both. See if anyone notices a difference and glows about one or the other without you even telling them the prints were made with two different cameras. They certainly would see a difference between those prints and prints made from, say, a P&S with the background in sharp focus. But probably not between the DSLR prints.

I would get FF just to have a 17-equiv TS lens.

I believe I said, before your post, that those with T/S lenses are best served by FF bodies.

No need to compare them,

I compared them because of the comments to jrista that he should sell the 7D and buy a 5D2+16-35 now because "a 7D isn't for landscapes." Contrary to the meme, landscape photography is an area where there is little to no difference between the sensors. If you've got a 7D, save the money and wait for a 5D3.

Now if jrista had posted that his 85 f/1.2L just wasn't giving him shallow enough DoF on his 7D, I would have told him to get a 5D2. I don't think I've ever heard someone make that complaint however.