Sony A7r, gamechanger, to be announced Wed

Nikos · Expire: 2015-01-02

Please relax. This is not personal. Do not make it personal.

Advances in sensor technology can only be a good thing.

If you think you do not need a better sensor, this does not mean that sensor RnD should stop Wink

All of us who use 40 or 50-year-old lenses, we do it only for sharpness??
Of course not. It is the special rendering we love, the bokeh, color rendition, the vintage look blah blah.

Attila · Expire: 2025-11-18

Put on a lens what is perform crappy on m4/3 or APS-C and start to sign on Canon 5D2 ... I believe Nikon did ruin own reputation with this nee 16M full frame camera will rich low success just like Nikon V1, J1 Nikon didn't look at all what majority of customer want.. we will see how quickly will change to higher pixel or dissapear from market , crappy idea just like Pentax Q, or Pentax mirrorless etc

iangreenhalgh1 · Expire: 2014-01-07

Oh I'm relaxed, I'm just trying to understand the issue because I'm looking to buy a new camera in a few months and I don't know what sort of sensor would suit me best.

Realistically, I'm not going to be able to change my collection of lenses in order to suit the sensor, so it seems logical to me that I should chose a sensor that suits my lenses.

Let's say that my best lenses are resolving on their best day, 55-60lp/mm at most, that equates to a 4500x3000 sensor, which is just under 13mp.

So am I correct in thinking that, with the restriction imposed by what glass I own, any sensor that has at least 13mp is capable of extracting all of the resolution my glass is capable of?

Maybe my maths is wrong?

Nikos · Expire: 2015-01-02

Nordentro · Expire: 2015-01-29

I made this:

It is difficult to see the real difference in sensor area when both sensors are shown in landscape. It is easier to see the real difference if you tilt the APS-C sensor up in portrait mode over the FF sensor.

As you can see, the pixel size of the A7R is slightly bigger than a NEX 5n (compare fig. 1 to 2)
You can crop away all of the red area in figure 2 on a A7R and still have a NEX 5n picture left Wink

The widest part of a NEX 5n picture has the same resolution as the shortest part of a A7R picture.

iangreenhalgh1 · Expire: 2014-01-07

I don't see the point, just buy a cheaper 16-20mp camera. Larger pixels in theory, should offer advantages in noise, dynamic range etc.

I'd love to be able to try out 16, 24 and 36mp FF cameras and see which gives the best output with my favourite lenses, I have a feeling that the 16mp one would be everything I need or desire.

hoanpham · Expire: 2015-01-18

Cool visualisation.

I can confirm that D800 36Mpx is very useful for group shots.
Not many of my lenses can outresolve the d800's sensor.
Noise at 1:1 crop might not be as good as my pentax k5 nor 5dm2 nor nex5n, but

when resizing to 5dm2 size, the result is better regarding noise, and outperforms crop sensors.
Best group shots are from higher resolution, where i can still do some post processing.

If I can pick one of the nex FF, A7r is what I want, simply because I have more choices.

edit:
Most of my lenses outresolve 5d classic. Wide open is not counting.
Many lenses outresolve 5dm2.
D800 has a limited number of lenses to use, but many ais just blurred at 1:1 crop.

iangreenhalgh1 · Expire: 2014-01-07

So if I understand you right Hoan, most lenses can outresolve a 24mp FF sensor but resolve less than a 36mp FF sensor?

I've been trying to understand this issue, but it's fairly complex. This article has a lot of detailed info:

http://www.luminous-landscape.com/tutorials/resolution.shtml

Nikos · Expire: 2015-01-02

rbelyell · Expire: 2014-01-31

iangreenhalgh1 · Expire: 2014-01-07

hoanpham · Expire: 2015-01-18

Ian,

My last post is what I observed when using different lenses on different FF bodies, just daily use, nothing scientific.

When I am using the lenses, the f-stops are vary.
To see of a lens can outresolve the sensor, I need f5.6 or f8 or f11, hardly use f16, but sometimes f4 and f2.8.

Lenses that can be tested are all nikon mount, AF screw drive or MF, adapted to canon.

First time when I got the d800, noise from 36mpx at iso3200 is worse than 5dm2, nex5n and k5.
I shot raw, with all NR turned off, and do NR in post. That surprises me alot. But then I resized to 5dm2 resolution and other resolutions for crop sensor. NR applied, some what differently, to get the best result and compared with similar sizes, the 36mpx wins.

Same as 5dm2 when resized to 15-16Mpx.
Noise ratio at 1:1 crop might look the same, but NR can apply more on larger basis. Resizing will hide imperfection result after a NR.

iangreenhalgh1 · Expire: 2014-01-07

I see, so the downsides of the smaller pixels on a 36mp sensor can be overcome by downsizing and post-processing. That's interesting and something to bear in mind, cheers.

It will be interesting to see how results differ between the a7 and a7r once a few months have passed and a few people have been using them.

In the Nikon world, what has been the general opinion of the differences in output quality between the d600 and d800?

Obviously, the d800 is more demanding of the lens, but given the same lens, I wonder what would give the best quality, a straight 24mp file from the d600 or a 36mp file from the d800 downsized to 24mp after having noise reduction applied?

Apart from resolution and noise, there are also the factors of dynamic range and tonal quality, those, in theory, should be superior with the 24mp sensor due to it's larger pixels, I wonder if this is visible in real-world results?

hoanpham · Expire: 2015-01-18

The d800 is the only nikon i have.
For convenient, I use 5dm2 with nikon lenses as well.
With the same lens, it may not be comparable, 5dm2 is old, d600 is new.

5d/5dm2 has S-type screen, d800 has 1.7x eyepiece.
canon has smaller files. d800 has 45MBytes files.

For D-range, i use nex5n or d800. The d800 is Very good at d-range.

uhoh7 · Joined: 24 Nov 2010 Posts: 1300 Location: Idaho, USA

iangreenhalgh1 · Expire: 2014-01-07

Raquel Welch in 62.... Oh my, now that is something Wink

Nikos · Expire: 2015-01-02

ForenSeil · Joined: 15 Apr 2011 Posts: 2726 Location: Kiel, Germany.

Nordentro · Expire: 2015-01-29

iangreenhalgh1 · Expire: 2014-01-07

Nikos · Expire: 2015-01-02

iangreenhalgh1 · Expire: 2014-01-07

According to Dxomark results, 10ev is poor for an upto-date sensor.

I checked with Dxomark what the cameras I own have:

Sony NEX-3 12 EV
Canon EOS 450D 10.8 EV
Samsung NX100 10.7 EV
Nikon D50 10.8 EV

Some others:

Samsung NX200 12.6 EV
Canon EOS 6D 12.1 EV
Nikon D800E 14.3 EV

Dxomark gives 14.1 EV for the A7R and 14 EV for THE SLT-A99.

Therefore, I believe that the correct figure for the A7R is 14 EV, Dxomark doesn't have data for the A7.

fermy · Joined: 17 Feb 2012 Posts: 1974

10 EV sounds unbelievable. It's way too low for a modern sensor even on a crop, let alone the FF.

Nikos · Expire: 2015-01-02

It seems that the numbers from chip.de are not comparable to those of DxOMark.
Anyway, both rate a7r in the top 2-3 cameras.

iangreenhalgh1 · Expire: 2014-01-07

Yes, clearly the numbers are not the same. I think 14 EV sounds more like the accurate one.