[Guest]

[quote name='Brightcolours' timestamp='1291987515' post='4924']

First of all, it is NOT about affecting so much, as it is about cheating.



Cheating? Yes, Cheating.



Just read the reviews, rating cameras that filter noise better, while you can achieve better results without the NR (doing NR yourself as needed).



1. How does long exposure NR affect shooting? By... for instance... getting rid of details? By... for instance... messing up photos of the night skies (stars and such)?



2. Of course NR is visible on prints. Take the desaturating NR from the Nikon D300 for instance. Or simply the removal of details, they do not appear magically back on prints.



3. From what camera?

If you look at the D7000 and K5 results on dpreview next to the 60D results, yes you clearly see an adverse effect from the NR.



It would help to just use one's eyes..

[/quote]



Well, I do understand the bit about cheating - I'm sure all Canon buffs feel annoyed when the main competitor's camera got better reviews by slightly cheating :-)



However I was more referring to "I know how to work the camera to get the results" bit. For some reason I was under impression that Nikon's "stealth" long exposure NR was detected by data analysis software and otherwise is not so obvious in the actual pictures?



1. Apart from the astronomers who examine the pictures on a pixel level because they *need* that data as accurate as possible, do you really say you would notice the difference in photo of the night sky?



2. Just to reiterate - I am talking about the "stealth" NR, not the standard NR feature.



3. Ok, I've checked "Compared to (RAW)" part and selected ISO 3200 for D7000 and 60D. Maybe it's indeed something wrong with my eyes, but I can't see big difference between two of them. Could you be more specific what I suppose to be seeing there?

[Guest]

[quote name='Lomskij' timestamp='1291988858' post='4925']



3. Ok, I've checked "Compared to (RAW)" part and selected ISO 3200 for D7000 and 60D. Maybe it's indeed something wrong with my eyes, but I can't see big difference between two of them. Could you be more specific what I suppose to be seeing there?

[/quote]

The only difference I can see is in the black box at the bottom of the image that is filled with rolls of colourful thread. You can see in the Nikon image that the colours are a little more saturated while they appear less saturated with the canon. In every other part of the images the two cameras are practically indistinguihable at iso 3200. Note that the colour desaturation in the deep shadows may be caused by the raw converter. I dont know.

Anyhow you are certianly right in saying that the two cameras deliver almost equal IQ at high iso in general. The only practical relevant difference is pointed out at page 17 of the dpreview.

Having worked with the raw files I can confirm that the Nikon really has an advantage if you brighten the shodows of low-iso images in post or in camera by using the A-DL function.

[genotypewriter]

[quote name='Lomskij' timestamp='1291982182' post='4922']

Ok, just out of curiosity:



1. Some NR at long exposure shots would affect your shooting - how?



2. Would it be visible on prints?



3. Did you actually see the NR effect on the images or you're just referring to the FT analysis of the noise distribution?

[/quote]

First of all, I got a DSLR because I wanted to be in control of things. Second of all, I have lenses that resolve lots of detail that get lost when NR is applied.



So:



1. Yes, I wouldn't want NR on any of my RAWs. Manufacturers may have a NoobRAW option to force NR but leave RAW raw.



2 & 3. Yes, NR is visible when the lenses resolve the detail. Doesn't matter whether it's done in camera or in post. How NR works is mathematical and understood... not a mathemagical.



GTW

[genotypewriter]

[quote name='Lomskij' timestamp='1291988858' post='4925']

However I was more referring to "I know how to work the camera to get the results" bit. For some reason I was under impression that Nikon's "stealth" long exposure NR was detected by data analysis software and otherwise is not so obvious in the actual pictures?

[/quote]

Let's put it this way... can you show images from a D3x or an A900 that are sharper than these 5D2 images:



1. http://www.flickr.com/photos/genotypewriter/5080867269

2. http://www.flickr.com/photos/genotypewriter/3903290645

3. http://www.flickr.com/photos/genotypewriter/3904062188

4. http://www.flickr.com/photos/genotypewriter/3395784426

5. http://www.flickr.com/photos/genotypewriter/3678794978



<img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/wink.gif' class='bbc_emoticon' alt='' />



Note: Except for 1 and 2, others don't even have adequate sharpening applied.



GTW

[Guest]

[quote name='genotypewriter' timestamp='1292200551' post='4957']

Let's put it this way... can you show images from a D3x or an A900 that are sharper than these 5D2 images:



1. http://www.flickr.com/photos/genotypewriter/5080867269

2. http://www.flickr.com/photos/genotypewriter/3903290645

3. http://www.flickr.com/photos/genotypewriter/3904062188

4. http://www.flickr.com/photos/genotypewriter/3395784426

5. http://www.flickr.com/photos/genotypewriter/3678794978



<img src='http://forum.photozone.de/public/style_emoticons/<#EMO_DIR#>/wink.gif' class='bbc_emoticon' alt='' />



Note: Except for 1 and 2, others don't even have adequate sharpening applied.



GTW

[/quote]



No problem whatsoever:



http://www.opticallimits.com/active/magic/get.jsp?id=840612746_oorib&format=v

http://www.opticallimits.com/active/magic/get.jsp?id=819402060_Dvine&format=v

[genotypewriter]

[quote name='Lomskij' timestamp='1292233107' post='4961']

No problem whatsoever:



http://www.opticallimits.com/active/magic/get.jsp?id=840612746_oorib&format=v

http://www.opticallimits.com/active/magic/get.jsp?id=819402060_Dvine&format=v

[/quote]

I beg to differ...



The first pic of the bird gives a false sense of sharpness because we as humans think if we can see the feathers (or similar things that we find difficult to see with the naked eye), it must be sharp. But if you really look, the bird is filling the entire frame and there's hardly any 1 pixel wide detail.



The second pic has the same issues... black fur is not really a good thing for measuring detail.



GTW

[Guest]

[quote name='genotypewriter' timestamp='1292238250' post='4963']

I beg to differ...



The first pic of the bird gives a false sense of sharpness because we as humans think if we can see the feathers (or similar things that we find difficult to see with the naked eye), it must be sharp. But if you really look, the bird is filling the entire frame and there's hardly any 1 pixel wide detail.



The second pic has the same issues... black fur is not really a good thing for measuring detail.



GTW

[/quote]



Ah, ok. Because I was comparing them with your samples and these two look sharper than yours to me, sorry :-)



For instance, 3903290645 - I don't really understand why you call the picture sharp? Or do you refer to the chroma noise where skin texture supposed to be as details?



Also, if fur is not good for measuring detail, then what is? As I understand, you don't agree with the LPH results measured by review sites?



P.s. Regarding that bird / feathers picture, what's wrong with this?

[genotypewriter]

[quote name='Lomskij' timestamp='1292242160' post='4964']

For instance, 3903290645 - I don't really understand why you call the picture sharp? Or do you refer to the chroma noise where skin texture supposed to be as details?[/quote]

Looking at it another way, it's got more detail than a "less noisy" D3s shot and just as much detail if not more than a D3x shot (where the latter has a slight MP advantage) at the same ISOs.



The D3x shots you showed were shot at ISO100, f/4 and f/3.2, 1/1250 and 1/1000 respectively. The ones I showed were shot wide open at f/2 and except for the first one, the others were shot at higher ISOs and at slower shutter speeds than 1/1000 too. The smaller apertures/greater DOFs in the D3x shots you showed also give a sense of more sharpness naturally because there are more things in focus.



Also there's some excessive noise on the face in that particular shot is because it was shot at around ISO800 (no flash) and vignetting was corrected which pushes the corners at least a stop further in post. So you're essentially comparing ISO100 center crops of a D3x (isn't the base ISO200?) against ISO800-ISO1600 border crops here. Also none of these shots were done on a tripod or with flash... so you have to take those factors in to consideration when talking about obtainable detail too.





[quote name='Lomskij' timestamp='1292242160' post='4964']

Also, if fur is not good for measuring detail, then what is? As I understand, you don't agree with the LPH results measured by review sites?



P.s. Regarding that bird / feathers picture, what's wrong with this?





[/quote]

The problem is not exactly with fur/feathers itself, it's the magnification. Even when we see fur at a large magnification, the first thing that pops in to the minds of many is "sharp" because of the conditioning we've received from photography sites/magazines. Try this for now... resize one of the images so their sizes match, take equal-sized crops from the bird shot and my fabric shot (pay special attention to DOF) and compare them side by side.



I do agree with LPH as a proper measurement but I don't agree that it's fully indicative (the correctness of how lens review sites do their testing aside).



GTW

[Guest]

Can sombody explain to me what is going on here? What are you guys trying to prove here? That all cameras can produce sharp images? That some sharp images have croma noise while others dont? Is this discussion at all relvant for anybody planing to use a camera for shooting pictures? On these forum I sometimes dont know whether I should laugh or cry.

[mst]

[quote name='genotypewriter' timestamp='1292279567' post='4972']

D3x (isn't the base ISO200?)

[/quote]



No, base ISO of the D3x is 100.



-- Markus