D600 close to D4 at high ISO

Started Sep 28, 2012 | Discussions
John Motts Veteran Member • Posts: 5,438
Re: D600 close to D4 at high ISO
2

gl2k wrote:

John Motts wrote:

gl2k wrote:

Now we have it officially :

D600 is very close to D4. Much closer than D800 is to D4.

http://www.dpreview.com/reviews/nikon-d600/13

If you download the 3200 file and view them at the same size, the D800 actually shows slightly better noise than the D600.


D600 on left

Comparing images is not about resizing until you get what you want.

Nonsense. What on earth is the point of comparing cameras "at pixel level"? It achieves nothing and puts the higher pixel sensor at a disadvantage every time. You compare images at the same size if you want to assess how two cameras compare noise-wise.

If you want to compare the output using prints, you use prints of the same size. You don't use completely different sizes of prints, which is effectively what you are saying we should do.

If you wanted to compare the noise performance of a D800 with that of a D700, would you compare a 24x16 print from the D800 with a 14x9 print from a D700? Of course you wouldn't. To get the most meaningful comparison you'd compare 24x16 prints from both.

FYI the comparison that I posted was with the D600 image enlarged to the same size as the D800 image. I also did the opposite and reduced the D800 image, but the results were just the same.

If you wanted to compare the image quality of a 5x4 camera with that of a 35mm camera and produced a 6x4 print from the 35mm camera, a 20x16 print from the 5x4, and examined them equally closely, you would be able to say that there is no difference in quality between the output from the two cameras. Pretty ridiculous way of comparing them, and yet that is exactly what you are suggesting.

John Motts Veteran Member • Posts: 5,438
Re: D600 close to D4 at high ISO

gl2k wrote:

John Motts wrote:

gl2k wrote:

Now we have it officially :

D600 is very close to D4. Much closer than D800 is to D4.

http://www.dpreview.com/reviews/nikon-d600/13

If you download the 3200 file and view them at the same size, the D800 actually shows slightly better noise than the D600.


D600 on left

By further resizing the images even my old Panasonic TZ5 might give similar results.

Quite the opposite.

If you want to get your Panasonic to look as close as possible to a D800 you would compare them at pixel level. If you want to see real difference you compare them at the same size.

olyflyer
olyflyer Forum Pro • Posts: 24,195
Re: D600 close to D4 at high ISO

gl2k wrote:

olyflyer wrote:

gl2k wrote:

I'm blown. The D600 is pretty close to the D4. D800 looks weak compared to D600.

Scratching my head. Nikon puts a better sensor in a worse body compared to the D800. Kinda weird logic in some way.

Don't scratch you head any longer...





Now we have it officially :

D600 is very close to D4. Much closer than D800 is to D4.

http://www.dpreview.com/reviews/nikon-d600/13

Yes, but you can see that in my crops as well, though "much closer to D4" is not the same as better high ISO than the D800...

olyflyer
olyflyer Forum Pro • Posts: 24,195
Re: D600 close to D4 at high ISO

John Motts wrote:

gl2k wrote:

Now we have it officially :

D600 is very close to D4. Much closer than D800 is to D4.

http://www.dpreview.com/reviews/nikon-d600/13

If you download the 3200 file and view them at the same size, the D800 actually shows slightly better noise than the D600.

D600 on left

Exactly. That's what I did as well. Downloaded the raw files of the three cameras and developed in CNX2, resized to the same sizes. Any way I do the resizing the D800 ends up to be the best in terms of noise.

olyflyer
olyflyer Forum Pro • Posts: 24,195
Re: D600 close to D4 at high ISO

gl2k wrote:

The D600 delivers cleaner images at pixel level. Making it a great high ISO cam since you get the same image quality at 33% less pixels compared to a D800.

No, that's wrong. Didn't you read my post?

John Motts Veteran Member • Posts: 5,438
Re: D600 close to D4 at high ISO

coudet wrote:

gl2k wrote:

Sure it is. A downsampled image always look better than a comparable image viewed at 100%.

I compared it too (2 different raw convertors, didn't downsample).

No I didn't downsample either. Where did I say that I did? I reproduced the two images at the same size, which is how any intelligent comparison is made.

RodluvanII
RodluvanII Senior Member • Posts: 2,094
Re: Vargen till Rödluvan...

olyflyer wrote:

RodluvanII wrote:

That said, I believe 36Mpx is a laugh and serves very little, if any, purpose for the wast majority of people.

Helt fel slutsats. (Wrong conclusion.)

You got me, I'm convinced now.

 RodluvanII's gear list:RodluvanII's gear list
Nikon D3X Nikon D810 Nikon AF-S Nikkor 24-70mm f/2.8G ED Nikon AF Nikkor 180mm f/2.8D ED-IF Leica Summarit-M 50mm f/2.5 +11 more
bobn2
bobn2 Forum Pro • Posts: 53,598
Re: D600 close to D4 at high ISO
2

btobey wrote:

Yes I agree. Perhaps I should perform an analysis to define the relationship between SNR and resizing. But I agree with you that it is common to resize images. However one needs to ask themselves why do you need 36MP then. I mean why put up with the inconvenience and inefficiency in workflow if all you intend to do is resize.

The question is, how many cameras do you want. I have a D800 a 5D. If I want an A2 image that is 36MP, the D800 does it without resampling. The 5D (or any other FF camera) needs upsampling (the D800 is really upsampled since each Bayer sub-array needs interpolation to yield a 36MP colour image).  An A3 image is 18MP. Things like a 2nd or later generation Canon, the 24MP Nikons and Sony's need downsampling, a D700 or my 5D need upsampling. At that size image the D800 produces a visibly superior image to any of them. An A4 image is 9MP. The D800 produces that very nicely with a direct half resolution merge of the Bayer quad, very fast indeed to process, and produces results better that the others which need interpolation and downsampling.

So everyone resizes always. If you don't want an array of cameras, one for each picture taking situation, the D800 covers it all, except for the highest frame rates.

Also, quite often I find myself revisiting and reworking stored files. Then quite often I find myself wishing I had captured all of the information that the D800 will capture.

Further, cropping is also a common technique. When you consider cropping, the advantages of resizing diminish.

Hardly, the advantages of having plenty of pixels to start with are most evident.

-- hide signature --

Bob

olyflyer
olyflyer Forum Pro • Posts: 24,195
Re: Vargen till Rödluvan...

RodluvanII wrote:

olyflyer wrote:

RodluvanII wrote:

That said, I believe 36Mpx is a laugh and serves very little, if any, purpose for the wast majority of people.

Helt fel slutsats. (Wrong conclusion.)

You got me, I'm convinced now.

I am not trying to convince you at all.

chlamchowder Senior Member • Posts: 2,083
Nikon did a good job

Good job Sony! D600, A99 and RX1 - all using the same sensor so one of them should win 2013 Camera of the Year...

Good job to Nikon, too, for extracting better image quality from the same sensor while offering it at a lower price point...

 chlamchowder's gear list:chlamchowder's gear list
Sony Alpha DSLR-A580 Nikon D600 Sony DT 18-55mm F3.5-5.6 SAM +8 more
RodluvanII
RodluvanII Senior Member • Posts: 2,094
Re: Vargen till Rödluvan...

olyflyer wrote:

I am not trying to convince you at all.

Mission accomplished.

 RodluvanII's gear list:RodluvanII's gear list
Nikon D3X Nikon D810 Nikon AF-S Nikkor 24-70mm f/2.8G ED Nikon AF Nikkor 180mm f/2.8D ED-IF Leica Summarit-M 50mm f/2.5 +11 more
Princess Leia Veteran Member • Posts: 6,956
Re: Nikon did a good job

chlamchowder wrote:

Good job Sony! D600, A99 and RX1 - all using the same sensor so one of them should win 2013 Camera of the Year...

Good job to Nikon, too, for extracting better image quality from the same sensor while offering it at a lower price point...

lower price because it is a dumb down camera...

Keyboard shortcuts:
FForum MMy threads