Sony Cyber-shot RX1R II added to studio test scene

Which is why we test most/all our DSLRs at standard focal lengths, with standard lenses.

Let me be even more clear: our studio scene is not a replacement for a proper lens test. We're looking at re-implementing a much larger target for proper lens tests, that will be placed at much further distances, to simulate near infinity results.

Our studio scene is more for assessing noise (ISO performance), color, noise reduction/detail retention, etc.

Again fair enough, but to play devil's advocate, if that is the goal then it seems rather arbitrary to take the photo at a distance that gets the edges of this scene in the frame

Heck, I'd seriously doubt any novice layperson would consider and RX1 Rii. And even if they happened to, had the means to, and their inability to understand this site led them not to buy one, i'd say that's a good outcome. Thanks Rishi for sharing the details. Let's see the full review soon!

It is intentional. Fixed prime lens cameras are designed similarly so that lens size, weight, and costs are minimized as much as possible. To get the quality you expect, the lens will be similar to a Zeiss Otus or a medium format lens.

Rishi,
When I say that I compared photos, I meant that I compared high ISO (3200, 6400, 12800, 25,600) RAW shots in the low light scene. It is here that blacks get affected by all kinds of weirdness: Chroma noise, Luma noise & brightness shift. The RX1r mk2 seems to hold blacks very well.

I don't believe you have a faulty unit with regards to lens softness, as I saw the same thing with the previous RX1R.

As someone who only shoots RAW and spends an awful lot of post-time in ACR and PS, I can tell you that holding the blacks is very important, and so does color-separation, for that matter. Sure, you can move a slider or stretch a curve, but the starting point on this camera should be great to begin with.

zodiacfml,
For the price of this camera, I expect Otus-like performance (or at least Batis). It's fixed and cannot be replaced, so if your're going to be "stuck" with it, it better be top-notch. For now, an A7r mk2 coupled with a Batis 25mm or the FE 55mm 1.8 would be a better combo.

You are seeing fuzzy image from imaging resource ?
Me, no.

0lf,
Yes. For example, open the Comparometer and put the still life shot @ 6400 ISO next to the A7R mk2 with the same still life shot@6400 ISO.
The shot from the RX1R mk2 has fuzzy/blurry edges.

I guess I am "simply massively confused" then Rishi.

I'd go further to say that if you claim anything about the corner sharpness of the RX1R2 based on Steve Huff's blog post you're plain stupid.

Why? He hasn't provided 100% versions of his images, or 100% crops of the corners.

I guess I am "plain stupid" Sidath.

Wow... And this is suppose to be an educational site...

I've owned the RX1 for nearly three years and consider it to be one of my favorite cameras. That said, the edge blurriness shown on the studio shots is an issue that should not be discounted. The 56" distance at which the studio shots were taken is not at all what I consider "close up," at least not in my use of the RX1. While one could resolve this by placing the focus point on the subject when the subject is not centered (and many of my subjects are not centered), moving the AF point around within the frame of the RX1 is an awkward and slow process. The only reasonable option is focus/recompose, which has its own issues, although it's worked pretty well for me with the RX1. (Maybe the focus shift created by focus/recompose somehow counteracts the focus shift caused by field curvature.) As for the Q "forcing" corrections on us, how could a raw file be corrected to resolve field curvature? I've never heard of a baked-in raw correction that changes the plane of focus.

Really? If you were doing a tight portrait, head and shoulders, you would stand two meters from the subject and use a 35mm lens? But then you wouldn't need edge sharpness, so I am trying to think of any reason to be that close to any subject for a landscape image when edge-edge was critical.

Besides, it's not a camera issue that's unique to Sony. This will be apparent shooting close with any WA lens, especially a fixed lens version. Unless you "fix" if similar to Leica.

Since I rarely carry more than one camera with me at a time, and that camera is usually the RX1, yes: I would take a portrait using a 35mm lens within two meters. And if the subject is not centered, shouldn't sharpness away from the center be a consideration, even at that distance? I want to emphasize, though, that this is not an issue I've particularly noticed in my use of the RX1. For someone choosing between the Q and the RX1, though, it clearly may matter.

You have portrait opportunities with flat field subjects where you need edge sharpness? Because that is ALL this test will show.

Really, I avoid shooting ANY human subject THAT close with any lens. Unless it's a spur of the moment family type shot, a WA lens is a terrible choice for a close portrait. Any such images would have a near-100% chance of being throw away.

re: Choice between Leica and RX - Again, this doesn't prove anything, unless you have a need to shoot fine art with a FL that WILL distort the image.

Look at the Image Resource samples for a realistic idea of real world photography with this camera.

Rishi, this may be a preposterous notion, but I'm really interested in the possibility that focus and recompose with the RX1 at these closer distances may somehow compensate for field curvature. I use focus/recompose almost exclusively with the RX1, mainly because I find flexible spot to be cumbersome and less accurate than center point AF, and yet I haven't noticed the off-center lack of sharpness shown in your studio tests. It would be an interesting exercise to test this by using center point to focus on the edge of your studio scene, then to recompose and check what the edge sharpness looks like then. Just a suggestion, though I'm sure you have plenty of other tasks to occupy your time.

@Chris Crevasse - Rishi didn't suggest focusing and recomposing: he suggested composing and then selecting and off-centre focus point. This would force the camera to find optimal focus for the off-centre position.

And yes, manually selecting off-centre points can be a a bit annoying with the RX1. We'll be looking at whether then RX1R II's lock-on focus and AF-C will allow you to specify the focus subject, recompose and have the camera re-focus on a now off-centre point.

Ultimately, you're only likely to see this curvature of field if you shoot flat subjects at relatively close distances with high-contrast detail in the corners. Most real-world situations are less demanding and the level of detail being argued over is irrelevant.

Thanks, Richard. I understand what Rishi is saying. I'm just curious about this idea: if field curvature shifts the plane of focus, and focus/recompose shifts the plane of focus, is it possible that focus/recompose will compensate for field curvature at these near distances? Ideally, the answer is "yes," so you don't have to bother with placing the focus point on an off-center subject.

You are trying to say that a test scene shoud be adjusted to camera?!
Even if they do adjust a test scene,could we do the same thing in real world use like nature and streets?
Utter bs.

A camera shoud be able to take a picture of anything you piont her to,like all other cameras here. End of story.

Are you on right forum? Or web site,better to say!?

Look Zoro, I'm not sure what alternate realty your floating around in but you are exactly correct when you say;" A camera should be able to take a picture of anything you point her to, like all other cameras here. End of story." Which is exactly the point of my post; they pointed the camera at the scene snapped a picture, and its mush. What parts of that scenario are having the most difficulty with?

Read the comments by the DPreview team below. They aren't disappointing results. They are to be expected for a wide angle lens that has to be right up close to the test scene for it to fill the frame. The edges are much further from the lens than the centre

Revisit the Sigma DP2Quattro test scene then come back with the conclusion.

The DP2 has APSc sensor, so 1+ stop more DOF, longer 24x36 equivqlent focal length, and the picture show also more sharpening. Or may be you are telling us DPR staf don't know what their are talking about ?

I just believe what I see with my own eyes. DP1 Quattro with the lens of 28mm in 35 mm terms is closer (and looks better than the RX1R2).
If that is not enough, DP0 Quattro shows clearly what a great wide angle lens can do. Of course Sony is more sophisticated camera then the Quattro line but if we are talking IQ (not to mention respective prices), new child of Sony has nothing to write home about.

"Shutter Shock" is very much shutter speed dependent.

True, but DPR use LR / ACR for all tests (unless testing something specific with a different converter).

While I'm no optics expert, I don't understand how a correction baked into a raw file could correct for field curvature. The edges of the frame are not in the same plane of focus, and extra sharpening of mush is just going to create oversharpened mush. Rather than the Q auto-correcting the raw file, my guess is that the lens on the Q is corrected for field curvature. Assuming that is true, that would be a big advantage of the Q over the RX1, especially since adjusting the focus point on the RX1 is a cumbersome PITA.