October 20, 2014, 06:17:49 AM

Author Topic: DxOMark posted it's A77 results  (Read 3771 times)

JonJT

  • Guest
DxOMark posted it's A77 results
« on: September 24, 2011, 10:38:29 AM »
I know this article:
http://www.dxomark.com/index.php/News/DxOMark-news/Sony-A77-measurements-and-review-of-the-world-first-24-MP-APS-C-camera
Concerns a Sony camera but, I find the results to be encouraging for all of us.  Sony has demonstrated that, even with the high photosite densities of a 24.3 megapixel APS-C sensor, one can ascertain very respectable dynamic range, SNR and good high ISO performance, on paper anyway.  Perhaps this will assuage the fears of those with high megapixel fears.

And so, I'm excited to see what other manufacturers, particularly Canon, can do with their new sensors. What does canonrumors think?

canon rumors FORUM

DxOMark posted it's A77 results
« on: September 24, 2011, 10:38:29 AM »

Flake

  • Guest
Re: DxOMark posted it's A77 results
« Reply #1 on: September 24, 2011, 12:28:22 PM »
One of the big issue with high MP count sensors is diffraction, and with this camera diffraction begins at only f/5.6, many will say that it only begins there, but you really will notice the effect at f/11 and probably at f/8.  Cameras like the 5D MkII begin diffraction at f/13. 

The problem then is control of DoF, fine if you're using fast primes, but as a landscape camera?  Well maybe then it's not the best choice to make

Meh

  • 5D Mark III
  • ******
  • Posts: 700
    • View Profile
Re: DxOMark posted it's A77 results
« Reply #2 on: September 24, 2011, 12:42:23 PM »
I know this article:
http://www.dxomark.com/index.php/News/DxOMark-news/Sony-A77-measurements-and-review-of-the-world-first-24-MP-APS-C-camera
Concerns a Sony camera but, I find the results to be encouraging for all of us.  Sony has demonstrated that, even with the high photosite densities of a 24.3 megapixel APS-C sensor, one can ascertain very respectable dynamic range, SNR and good high ISO performance, on paper anyway.  Perhaps this will assuage the fears of those with high megapixel fears.

And so, I'm excited to see what other manufacturers, particularly Canon, can do with their new sensors. What does canonrumors think?

Very good results indeed.  Interesting.  At 24MP Sony has been able to keep the SNR about the same as what was achieved in the 16MP sensor used in the Nikon D7000 which has to mean they've either lowered the noise in each photosite or they've managed to maintain the size of the photosites despite a 50% increase in the number of photosites.  The latter would be possible if they've been able to reduce the amount/size of circuitry that sits beside each photosite and if that's the case hopefully Canon has made similar improvements and we'll see a jump up in MP and DR while maintaining or improving SNR.

Overall the performance curves are not better and sometimes worse than the D7000 and 7D but that's a good result given the increase to 24MP.   A77 is still worse than the D700 and 5D2 as expected but pretty close which is interesting given the FF equivalent resolution would be about 60MP.

JonJT

  • Guest
Re: DxOMark posted it's A77 results
« Reply #3 on: September 24, 2011, 01:04:03 PM »
One of the big issue with high MP count sensors is diffraction, and with this camera diffraction begins at only f/5.6, many will say that it only begins there, but you really will notice the effect at f/11 and probably at f/8.  Cameras like the 5D MkII begin diffraction at f/13. 

The problem then is control of DoF, fine if you're using fast primes, but as a landscape camera?  Well maybe then it's not the best choice to make
Indeed, a reason why I want to go FF some time in the future. 

In any case, the diffraction issue won't preclude the use of this camera for some people.  It's also encouraging that this level of performance could be obtained with such densities.  What if Canon, say, kept the 7D successor at 18 megapixels and focused on other areas of performance.  Might be better than this camera, even.

x-vision

  • 6D
  • *****
  • Posts: 478
    • View Profile
Re: DxOMark posted it's A77 results
« Reply #4 on: September 24, 2011, 01:16:37 PM »
Quote
I find the results to be encouraging for all of us. 

Interesting, I find the exact opposite.

I just compared the A77 to the D7000, which has the 16mp Sony sensor.
The D7000 came on top on all DxO measurements: SNR, DR, Tonal Range, and Color Sensitivity.
That's on the 'print' tab, which equalizes resolution; on the 'screen' tab, the D7000 is even better.

Sony pulled a Canon with the A77 - sacrificing image quality for the extra 'umph' of more megapixels.
Canon did it with the 50D and 7D, now Sony is doing with the A77.

Unlike Canon, however, Sony also updated their 16mp sensor (the NEX-5N) and here's the result:

http://www.eoshd.com/wp-content/uploads/2011/08/comparison.jpg

This is nicely corroborated by the DxO measurements, btw.

JonJT

  • Guest
Re: DxOMark posted it's A77 results
« Reply #5 on: September 24, 2011, 01:25:55 PM »
Quote
I find the results to be encouraging for all of us. 

Interesting, I find the exact opposite.

I just compared the A77 to the D7000, which has the 16mp Sony sensor.
The D7000 came on top on all DxO measurements: SNR, DR, Tonal Range, and Color Sensitivity.
That's on the 'print' tab, which equalizes resolution; on the 'screen' tab, the D7000 is even better.

Sony pulled a Canon with the A77 - sacrificing image quality for the extra 'umph' of more megapixels.
Canon did it with the 50D and 7D, now Sony is doing with the A77.

Unlike Canon, however, Sony also updated their 16mp sensor (the NEX-5N) and here's the result:

http://www.eoshd.com/wp-content/uploads/2011/08/comparison.jpg

This is nicely corroborated by the DxO measurements, btw.

Hmm, the A77 pictures are quite poor.  But, that is really the reason why I said "on paper".  It was good to see that an increase in megapixels do not necessarily mean patently horrible performance in lab tests, in comparison to other, similarly sized sensors.  But, of course, that doesn't doesn't mean the quality for the final images is going to be good, as you showed.  Perhaps, Sony got it wrong here but someone else might. 

I just want to add that the NEX 5N picture is significantly cleaner than the 5D II picture but, the 5D really has more fine detail. 
« Last Edit: September 24, 2011, 01:31:38 PM by JonJT »

Meh

  • 5D Mark III
  • ******
  • Posts: 700
    • View Profile
Re: DxOMark posted it's A77 results
« Reply #6 on: September 24, 2011, 01:41:41 PM »
Interesting, I find the exact opposite.

I just compared the A77 to the D7000, which has the 16mp Sony sensor.
The D7000 came on top on all DxO measurements: SNR, DR, Tonal Range, and Color Sensitivity.

Yes, but only slightly worse.  It's remarkable that they maintained similar performance measurements given the 50% increase in resolution.  But good point that they did make the choice to increase resolution while maintaining SNR, DR, etc. as opposed to keeping resolution the same while increasing SNR, DR. etc.

I think the reason for that choice is the market position of the camera.  Sony A77 is clearly not a Pro camera and for average consumers Megapixels still rule the day.

canon rumors FORUM

Re: DxOMark posted it's A77 results
« Reply #6 on: September 24, 2011, 01:41:41 PM »

Meh

  • 5D Mark III
  • ******
  • Posts: 700
    • View Profile
Re: DxOMark posted it's A77 results
« Reply #7 on: September 24, 2011, 01:49:50 PM »
One of the big issue with high MP count sensors is diffraction, and with this camera diffraction begins at only f/5.6, many will say that it only begins there, but you really will notice the effect at f/11 and probably at f/8.  Cameras like the 5D MkII begin diffraction at f/13. 

The problem then is control of DoF, fine if you're using fast primes, but as a landscape camera?  Well maybe then it's not the best choice to make

Yes but just to clarify for people who aren't familiar with the physics.  Diffraction effects are caused by the lens, specifically the aperture (light bends when it passes an edge, creates bands through a slit, and rings through a small circular opening).  The smaller the aperture the greater the effects of diffraction.  The sensor itself does not cause any diffraction but as the resolution goes up for a given sensor size (smaller pixel pitch) the sensor will begin to resolve the diffraction effects and cause a softening of the image that offsets the increase in sharpness (and depth of field) that would otherwise be achieved by using the smaller aperture.  Therefore, for a given sensor it can be stated at what aperture this effect begins to be noticeable.

K-amps

  • 1D X
  • *******
  • Posts: 1519
  • Whatever looks great !
    • View Profile
Re: DxOMark posted it's A77 results
« Reply #8 on: September 29, 2011, 04:54:04 PM »

I just want to add that the NEX 5N picture is significantly cleaner than the 5D II picture but, the 5D really has more fine detail.

Exactly what I thought. NEX5n looks cleaner, less noise. But 5d2 is cleaner, on the bottle the "..EWERY" letters appear as that on the 5d3 but as "FWEBY" on the nEX5n.

Not sure if it's the IQ of the sensor or merely a more aggressive NR algorithm, this would explain both the cleaner but less clear rendition of the NEX5n
EOS-5D Mk.iii 
Sigma 24-105mm F4 ART; EF 70-200 F/2.8L Mk.II; EF 85mm L F/1.2 Mk. II; EF 100mm L F/2.8 IS Macro, 50mm F/1.8ii;  TC's 2x Mk.iii; 1.4x Mk.iii

canon rumors FORUM

Re: DxOMark posted it's A77 results
« Reply #8 on: September 29, 2011, 04:54:04 PM »