NEX-7 Rolling Review

[bobtowery]

Michael:

Thanks for the first installment. Good call to break up the review into numerous sections. FWIW my vote for the next one would be the "Nex-7, M9, Leica lenses."

Hope your trip to SMA was uneventful and that it is good to be back.

Question, which you probably covered at some point way back regarding your reviews... Are all of the images SOOC?

Thanks again, Bob Towery.

[michael]

Ain't no such thing as SOOC (which I presume means Straight out of Camera). Unless it's for a technical comparison every file I publish has been worked on in one way or another.

Michael

[barryfitzgerald]

Good luck with the bracketing request it's been a bugbear on many Sony models for years now even the A65 has a mere 0.7 +/-. Not sure what the problem is it's a dead easy thing to address.

[billh]

I’m with Bob - some of the upcoming subjects caught my eye, and the comparison between the M9 and NEX7 using Leica lenses is one of them. I am also curious to learn how the two 24mm lenses compare. I’ve been using a 5n exclusively with Leica lenses, and I love being able to use them on this tiny, capable camera - I sure wish the NEX7 was available.

The explanation and workaround the playback of either/or still or video files was appreciated. It has annoyed, and initially alarmed me when it looked like the stills had vanished. I did discover that combining peaking to outline an area where I want to prefocus with the high speed shutter (10 fps) gives me a reasonable chance of getting one in focus image of action like a dog jumping a log in the woods. I posted a few photos I took during the first few days I had the camera, as well as some showing the 5n with various M lenses to give some friends a sense of the size of this combination.

http://s1140.photobucket.com/albums/n561/billh96007/Sony%20NEX-5n%20with%20Leica%20M%20lenses/

I’m looking forward to hearing more adventures in Mexico and insights into the NEX7 from Michel!

(for people looking for information about the specifications and capability of the NEX7, a Sony rep gave a complete look into the features of it and the NEX5n, http://www.youtube.com/watch?v=co2Qa9y_1Ns&feature=player_detailpage )

[douglasf13]

Ain't no such thing as SOOC (which I presume means Straight out of Camera). Unless it's for a technical comparison every file I publish has been worked on in one way or another.

Michael

Hi, Michael.  I look forward to more of your rolling review.  I wanted to quickly mention that, assuming you're a raw only shooter (so jpeg settings don't matter,) try turning your picture style contrast all the way down to -3.  That will open up shadows quite a bit in the EVF.  My 5N has been set this way for weeks, and it works well.  A nice by product of this is that it'll give you a histogram reading that is a little closer to the actual raw file.

p.s. you may want to turn peaking and sharpness up a bit to compensate for focus peaking when doing the above.

[ErikKaffehr]

Hi,

SOOC is raw I pressume?

BR
Erik

Ain't no such thing as SOOC (which I presume means Straight out of Camera). Unless it's for a technical comparison every file I publish has been worked on in one way or another.

Michael

[bobtowery]

Hi,

SOOC is raw I pressume?

BR
Erik

Eriik, no, as Michael surmised, SOOC means "Straight Out Of the Camera." Although yes, in a sense RAW is more SOOC than a JPG! It just means you haven't tweaked your image in LR, Photoshop, etc.

[madmanchan]

Like Michael says, no such thing.     Even if you're shooting raw and haven't touched any sliders in the raw converter, there has been already some processing applied (just not by you!).  For example: default color mapping, default tone curve, default sharpening, default noise reduction, etc.

[michael]

More to the point, beyond what Eric has mentioned, who would want an SOOC image? If that's what one wants then shoot jpgs and set the camera's  "style" to whatever suites your fancy.

But if one shoots raw then sharpening is needed, as well as a wide range of other adjustments. Essentially, as Eric points out, there's no such thing as SOOC.

Michael

[ErikKaffehr]

Michael,

You got my point, exactly! ;-)

Best regards
Erik

More to the point, beyond what Eric has mentioned, who would want an SOOC image? If that's what one wants then shoot jpgs and set the camera's  "style" to whatever suites your fancy.

But if one shoots raw then sharpening is needed, as well as a wide range of other adjustments. Essentially, as Eric points out, there's no such thing as SOOC.

Michael

[Wayne Fox]

Eriik, no, as Michael surmised, SOOC means "Straight Out Of the Camera." Although yes, in a sense RAW is more SOOC than a JPG! It just means you haven't tweaked your image in LR, Photoshop, etc.

My interpretation of SOOC is some engineer(s) somewhere who probably aren't photographers (and under direction of some other person or guideline) have "tweaked" the image based on guessing what the subject matter is about , and trying to deliver an "acceptable" image on average as often as possible.  Perhaps OK for snapshots, not much use otherwise.

[bobtowery]

My interpretation of SOOC is some engineer(s) somewhere who probably aren't photographers (and under direction of some other person or guideline) have "tweaked" the image based on guessing what the subject matter is about , and trying to deliver an "acceptable" image on average as often as possible.  Perhaps OK for snapshots, not much use otherwise.

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

Back to the original post, I was really just asking the question, not intending to start a debate regarding the merits of SOOC vs processed. And I get that a digital image is really a file with a bunch of data, open to the interpretation of whichever program I choose to open that file with.  I also have rarely used an image that was not processed. Looking at a 640px image in a browser doesn't tell us that much anyway, I much more value Michael's written opinion.

So I don't think we have an arguable point here, or at least that was not my intention  .

[JimU]

Interesting installment regarding The User Interface today.

I wish my Panasonic GH2's UI was as customizable.  I'd say that the GH2's UI is horendously horendous.

Supposedly the GH2's buttons are configurable, but I've yet to be able to find out how to set one of the buttons as an aperature button.

[John Camp]

On the NEX noise comparison:

I completely agree with the point about the emphasis on high ISOs. If you run into a guy who can't possibly use a camera because it doesn't have a clean ISO 6400, you've just met a guy who shouldn't be a photographer. Fifteen years ago, I would have killed for a good clean 800...in film. To me, a clean 3200, with adjustable white balance yet, is like a miracle. I've been shooting a Panny GH2 with the Nocton f0.95 and the lens is soft, but jeez, I don't care -- I can shoot in the dark.

At the margins, the 5 is cleaner than the 7, but I doubt most would see much of that in a print. But one thing that has bothered me about good, clean, say ISO 1600 photos with high-resolution cameras, is that you can see *everything.* Like dandruff. The 5's smaller size looks cleaner partly, I think, simply because you don't see so much crap sticking to the targets. It's really there, but who needs to see it?

Is it possible that the real sweet spot for all 35mm-category cameras (that is, everything from the Nikkon V1s to FF), might be about 18mp?

I have to say that the commentary I've seen lately suggesting that 16-18mp is better than 24mp because of space and processing requirements, is silly. Storage and processing speed and ram are cheap.

 

 

[ErikKaffehr]

Hi,

What Michel shows is that noise is very similar between the two models when the images are scaled to the same scale. Michael noted some advantage for the NEX 5N at highest ISO, just about an imperial quibble. This is pretty much what basic theory says, and also DxO-mark. It's nice to have it shown unequivocally by an authority like Michael.

The two enclosed screen dumps show "tonal range" as measured by DxO-mark in screen mode (actual pixels), and print mode (sized to same scale). The DxO data is essentially consistent with Michael's findings.

I have taken the Alpha 55 and the Alpha 77 for this comparison because I assume that the same sensors are used in the NEX 7 and NEX5N, the NEX 7 has not yet been tested.


Best regards
Erik

On the NEX noise comparison:

I completely agree with the point about the emphasis on high ISOs. If you run into a guy who can't possibly use a camera because it doesn't have a clean ISO 6400, you've just met a guy who shouldn't be a photographer. Fifteen years ago, I would have killed for a good clean 800...in film. To me, a clean 3200, with adjustable white balance yet, is like a miracle. I've been shooting a Panny GH2 with the Nocton f0.95 and the lens is soft, but jeez, I don't care -- I can shoot in the dark.

At the margins, the 5 is cleaner than the 7, but I doubt most would see much of that in a print. But one thing that has bothered me about good, clean, say ISO 1600 photos with high-resolution cameras, is that you can see *everything.* Like dandruff. The 5's smaller size looks cleaner partly, I think, simply because you don't see so much crap sticking to the targets. It's really there, but who needs to see it?

Is it possible that the real sweet spot for all 35mm-category cameras (that is, everything from the Nikkon V1s to FF), might be about 18mp?

I have to say that the commentary I've seen lately suggesting that 16-18mp is better than 24mp because of space and processing requirements, is silly. Storage and processing speed and ram are cheap.

 

 

[KirbyKrieger]

Hi,

What Michel shows is that noise is very similar between the two models when the images are scaled to the same scale.

That's how I interpreted the "data".  But I'm stuck on Michael's description of this.  He says, in conclusion:
"What this means is that in practical terms, the NEX-7 gives the photographer the choice of higher resolution and larger prints, or comparable resolution and comparable noise characteristics when compared to the NEX-5n, and both are presented at the same size."

I might be suffering the lingering effects of too much turkey (though, honestly, the elderflower martinis probably play a role), but I can't understand the first phrase without adding the word "noisier" as in " ... and larger, noisier, prints".  My take-away is that, normalized for noise, the cameras perform identically across the range of ISO's they have in common.  The NEX-7 records more information per unit area than does the NEX-5N, but at a cost of higher noise.

[michael]

Yes and no.

What it means is that when shooting with the NEX-7 you have a higher resolution (bigger print) image, but you do not give up anything in terms of noise performance compared to a NEX-5n if you scale the file down.

With the NEX-5n you do not have the inverse, the choice of "scaling up" without a quality loss.

Michael

[douglasf13]

Yes and no.

What it means is that when shooting with the NEX-7 you have a higher resolution (bigger print) image, but you do not give up anything in terms of noise performance compared to a NEX-5n if you scale the file down.

With the NEX-5n you do not have the inverse, the choice of "scaling up" without a quality loss.

Michael

Yep, that's why, assuming identical sensor technology, more megapixels is always a net advantage, outside of slower processing speed and taking up more storage space.  

Of course, preliminary tests of the NEX-7 are showing that it doesn't handle wide angle rangefinder lenses as well as the 5N, and that could be related to the smaller pixels.  I can't wait to see Michael's wide rangefinder tests on the NEX-7 vs. 5N.

[madmanchan]

The tricky bit is that it's hard to have identical sensor technology when scaling up the res (while keeping the overall sensor size the same, like we have here with the NEX 5N and NEX 7).  Scaling up the # of pixels means smaller pixels, which itself presents more challenges (besides noise and speed).

[Jeff Kott]

I can't wait to see Michael's wide rangefinder tests on the NEX-7 vs. 5N.

+1

Yep, that's the only question in my mind as to whether to get a NEX-7.