NEX-7 Rolling Review

[Wayne Fox]

I think the folks at Adobe (and folks named Schewe) might take issue with this POV.

I don't think Adobe nor Jeff have much to do with the camera engineers that developed the firmware to generate a SOOC file.  the only possible interpretation of this acronym is the jpeg file created in camera, because the raw file really is irrelevant to being "sooc"  ... it's just a bunch of data that needs rendered later in the workflow instead of at capture.

In fact, the default rendering of Lightroom from the Adobe engineers is far more useful than the default rendered by the cameras firmware.  And of course with Lightroom that's just the starting point, where as with the camera jpeg it's the end point.

[ErikKaffehr]

Hi,

My point is that I presume that most readers on this forum shoot raw, so we never see straight out of the camera images. SOOC is raw, and raw files need to be interpreted.

BR
Erik

I don't think Adobe nor Jeff have much to do with the camera engineers that developed the firmware to generate a SOOC file.  the only possible interpretation of this acronym is the jpeg file created in camera, because the raw file really is irrelevant to being "sooc"  ... it's just a bunch of data that needs rendered later in the workflow instead of at capture.

In fact, the default rendering of Lightroom from the Adobe engineers is far more useful than the default rendered by the cameras firmware.  And of course with Lightroom that's just the starting point, where as with the camera jpeg it's the end point.

[bobtowery]

Hi,

My point is that I presume that most readers on this forum shoot raw, so we never see straight out of the cmera images. SOOC is raw, and raw files need to be interpreted.

BR
Erik

Exactly.

[Wayne Fox]

Well, I respectfully disagree SOOC couldn't be raw, since it does need rendered and the acronym implies a useable file straight from the camera without further work , so  SOOC could only refer to a jpeg file ... which I completely agree has little interest to those here on LuLa.

I was confused how the Adobe and Jeff tie in was even made and why it was there, as I thought it was pretty clear the only ones influencing the "SOOC" jpeg would be the camera engineers, otherwise I wouldn't have made a follow up comment.

[douglasf13]

Agreed. SOOC has always meant Jpeg out of camera without further processing (but I'm a raw shooter.)

[Jonathan Cross]

The rolling review is great.  I hope that IQ will be covered.  I mention this in terms of the recent article on this site about whether sensors 'outresolve' lenses.  It would be nice to know how Sony lenses work in terms of IQ on the NEX-7 with its large number of pixels.  Does Sony glass do justice to the number of pixels on the sensor?

Jonathan Cross

[Peter McLennan]

Waiting with great interest for the video portion of the review. 
Hopefully, some sound recording aspects, too.

[Ray]

The performance of the NEX-7 seems extraordinary. Now that the DXO results are out, we can see this is a great example of increased pixel count not having the negative consequences of increased noise, as a result of improving technology, comparing equal size images of course.

Nevertheless, it should be mentioned that the D7000 still retains a 1/2 stop lead in DR at base ISO. On the other hand, the NEX-7 appears to have about 1/2 a stop better DR at ISO 12,800 through some electronic trickery. But I would consider the 1/2 stop advantage of the D7000 at ISO 100 to be more useful, although a 1/2 stop may be near the borderline of significance in any case.

We should also not forget the advantages of the larger sensor. The full-frame D3X of similar pixel count, but older technology, still retains an advantage regards SNR at 18% grey (about the level of skin tones), tonal range and color sensitivity, although this wouldn't be the case if we had a 54mp full-frame camera consisting of NEX-7 pixels. By the time we get such a camera, technology will have advanced again in other respects.

[BernardLanguillier]

I am glad to see that the DxOMark results are now widely accepted as representing the truth about camera sensors. Heck, even Phaseone is now using these results in their commercial mailings (like the one I received 2 days ago). 

http://www.dxomark.com/index.php/Cameras/Compare-Camera-Sensors/Compare-cameras-side-by-side/%28appareil1%29/736|0/%28brand%29/Sony/%28appareil2%29/485|0/%28brand2%29/Nikon/%28appareil3%29/371|0/%28brand3%29/Sony

Cheers,
Bernard

[Jeff Kott]

Nevertheless, it should be mentioned that the D7000 still retains a 1/2 stop lead in DR at base ISO. On the other hand, the NEX-7 appears to have about 1/2 a stop better DR at ISO 12,800 through some electronic trickery. But I would consider the 1/2 stop advantage of the D7000 at ISO 100 to be more useful, although a 1/2 stop may be near the borderline of significance in any case.

Yes, Nikon has always been very clever with their electronics and usually seems to be able to squeeze a little more out of the Sony sensors than others - although I have not seen Nikon be so clever at taking their electronics and squeezing them into a compact package, an area where it seems Sony has a one generation lead considering the size of the NEX cameras, quality of the sensor and the class leading EVF.

I never thought I would feel this way, but after using the Sony EVF and then using my D300 and Pentax K5, I think I actually prefer using the EVF, at least for all manual focus lenses. Being able to see a live histogram in the EVF,adjusting ISO and EC without taking the camera from my eye, focus magnification and focus peaking in combination are very compelling.

[douglasf13]

...Nevertheless, it should be mentioned that the D7000 still retains a 1/2 stop lead in DR at base ISO. On the other hand, the NEX-7 appears to have about 1/2 a stop better DR at ISO 12,800 through some electronic trickery. But I would consider the 1/2 stop advantage of the D7000 at ISO 100 to be more useful, although a 1/2 stop may be near the borderline of significance in any case.

There are two reasons that I know of that cause this discrepancy, outside of power supply conditioning, software, etc. 

First of all, Sony uses a more dense color filter than Pentax, Nikon or Canon.  This leads to better color and metamerism in general for Sony compared to the competition, but the disadvantage is that less light hits the sensor, which requires a little more gain from the camera.  So, it's a trade off.  If you compare cameras in DxO Mark, you can't see the "Color Response" tab, but you can see the tab if you look at the camera individually.  Sony cameras nearly always have the highest metamerism.

The second reason is the 14 bit process that Nikon uses, which is believed to be some kind of oversampling or multisampling with EXMOR based cameras. 

I'm not sure if I'd want to make the color trade off, but I do wish that Sony would use the 14 bit method, now that the advantage is starting to become tangible.

[dreed]

As a side bar to this rolling review, over at DxO they've tested both the A77 and NEX-7 (which we assume have the same sensor) and found that the NEX-7 narrowly bests the A77 in all of their tests. Presumably this gives us an idea of the actual impact that the translucent mirror really makes on photos - aside from light loss.

DxO Labs - NEX-7 vs A77

[bobtowery]

Re: The size comparison video...

You sure all that noise in the background isn't a bunch of photographers trying to storm the gate and their hands on all your toys? DROOL!!!!!!!!!!!!!!!!!!

[Ray]

First of all, Sony uses a more dense color filter than Pentax, Nikon or Canon.  This leads to better color and metamerism in general for Sony compared to the competition, but the disadvantage is that less light hits the sensor, which requires a little more gain from the camera.  So, it's a trade off.  If you compare cameras in DxO Mark, you can't see the "Color Response" tab, but you can see the tab if you look at the camera individually.  Sony cameras nearly always have the highest metamerism.

That's a good point, but what's not clear is the significance in practice of that higher SMI figure. I imagine it's something I might be concerned about if I were employed to produce the most accurate color possible when, for example, producing a catalogue for a paint company or fashion house.

As it is, when processing a RAW image I find that I'm always always changing the white balance or vibrancy or hue and saturation of colors to produce the most pleasing result to my eye. I don't really care whether or not in reality the sky was exactly the same shade and intensity of blue as it is in my rendition. It's what looks best that counts.

The easiest way to compare cameras in DXOMark is to use the option to compare up to 3 cameras side by side, and view 3 separate graphs on each page under the various measurement headings. However, such headings do not include Color Response where the SMI is found. I wonder why.

I did a search amongst the various explanations on the DXOMark site, and found the following comment which seems relevant.

In practice, the SMI for DSLRs ranges between 75 and 85, and is not very discriminating. It is different for low-end cameras (such as camera phones), which typically have a SMI of about 40. For this reason, we give this measurement as an indication but do not integrate it in DxO Mark.

[douglasf13]

That's a good point, but what's not clear is the significance in practice of that higher SMI figure. I imagine it's something I might be concerned about if I were employed to produce the most accurate color possible when, for example, producing a catalogue for a paint company or fashion house.

As it is, when processing a RAW image I find that I'm always always changing the white balance or vibrancy or hue and saturation of colors to produce the most pleasing result to my eye. I don't really care whether or not in reality the sky was exactly the same shade and intensity of blue as it is in my rendition. It's what looks best that counts.

The easiest way to compare cameras in DXOMark is to use the option to compare up to 3 cameras side by side, and view 3 separate graphs on each page under the various measurement headings. However, such headings do not include Color Response where the SMI is found. I wonder why.

I did a search amongst the various explanations on the DXOMark site, and found the following comment which seems relevant.

  I think, at this point, we're all starting to split hairs between DR numbers, SNR, color, etc.  FWIW, if you talk to Iliah Borg, he'll tell you that the color separation difference is most noticeable in greens when comparing Nikons and Sonys, and the Sonys show more green hue resolution.  He still generally prefers his A900s over his D3x for landscapes for this reason, despite the D3x having other advantages.

[John Camp]

For me, the size comparisons are absolutely critical. When I'm traveling by car, and only moving around on foot for short distances, I prefer a FF or APS-C DSLR (specifically, now, the Pentax K5) because I like the viewfinders better than EVFs. However, when I'm traveling by air, or walking longer distances, or doing certain kinds of street photography, I take the m4/3 system, the Panasonic. I wish the comparison had shown the GF1 rather than the GH2, because that's the more size-critical version of the Panasonic cameras. But, I understand that you can't have everything in practical reviews.

Some questions, if Michael is inclined to answer them:

-Is the "bag" of gear you carry, with native-mount lenses, much different in terms of weight and size between the Panasonic and the NEX cameras? Could you stow both bags in the overhead in regional jets?
-Is there a sharp difference between the Panasonics and the NEXs in print quality for, say, 19-inch prints?
-Leica lenses aside, how do you judge the lens quality of the two systems?
-Do the Leica lenses provide an increase in performance that is apparent in prints?

[Craig Arnold]

Thanks for the size comparison Michael !

As an X100 fan it's only fair to point out that you had the X100 hood attached the whole time, which makes it look a lot bigger.

But the NEX7 does look like a very nice little camera. Really cleverly done, lots of good, even brilliant technical and design touches all round. Probably the single most impressive piece of high-tech camera design I have ever seen.

If I didn't hate Sony Group so much* this camera would be very tempting indeed.



*Completely off-topic rant...
I swore after their rootkit fiasco with their music CDs in 2005 that I would never buy another Sony product unless there was no alternative. So far I haven't - and a lot of people I know in IT made similar pledges. It was in my view simply unforgivable. The biggest single act of cynical abuse of customers' online security ever committed by any corporation. Sony officials should have gone to prison. I do accept that Sony BMG have nothing to do with the camera and electronics division.

[erictoddjohnson]

Michael,

Thanks for the very thorough and informative rolling review of the NEX-7. In your size comparison video, you talked about the 18-200mm Sony lens as a video lens, and described its benefits for that mode.  I don't shoot much video, but am more interested in the NEX system as an alternative to a FF DSLR for times when size and weight are an issue, e.g., travel or hiking.  I already have an NEX-5 with the 18-55, and find it ideal for those situations.  Could you give us your impressions of the IQ and overall utility of the 18-200 for shooting stills in one of your future installments? 

Thanks!

Eric

[michael]

The 18-200mm is a great lens for stills as well. I should have made that clear in the video.

As long as you have decent light, it's the lens to have mounted on the camera for spontaneous shooting. The IS works really well.

Michael

[grzybu]

Michael,

Is Nex-7 viewfinder much better than one in GH2?
I know resolution is higher, but does it make really big difference?

Grzegorz