The cruelly objective folks at DxOMark have tested the Sony RX1, a camera that has inspired Leica-like devotion (e.g., see the reader reviews at Amazon
So for the price of a Nikon D600 and a good 35mm lens you get the same image quality and seriously compromised picture-taking capabilities compared to a standard DSLR (see the conclusions in the dpreview.com review).
The Sony does seem like kind of a fun toy for people who love the 35mm lens perspective (I’m not one of them) but the test results don’t show it to be magic compared to a modern Nikon body and well-engineered 35mm prime lens (e.g., from Sigma!).
I would have thought the point of the camera is ‘smallest full frame’. If they made an interchangeable lens version (with light lenses) that might go a long way in the market… do you like carrying heavy camera equipment? I don’t and I moved to a Sony RX100.
Federico: I did love the old Canon Rebel G (film body) and the plastic 50/1.8 lens for the combination of superb image quality and light weight. That combination from the 1990s was 14 oz. for the camera and 4.6 oz for the lens, 18.6 oz. total for a full-frame SLR with a very high quality lens. The RX1 weighs 17 oz., which is admittedly much lighter than the D600 (30 oz. with battery) plus whatever lens, but it is not a substitute for the 8.5 oz. RX100.
Any thoughts on Sigma’s offerings along these lines? E.g., the DP2 Merrill?
James Duncan Davidson reviewed the RX1.
Please critique it.
Also he has retina enabled website so you might
want to visit using an ipad.
I think you have a strange way of reading the DO report , the conclusion is the sensor is better + lens is better than the canon and sigma and it was compared to a d3 as well as the d800 . The conclusion was “Its optical performance is outstanding, and particularly noteworthy for its consistent sharpness and homogenous imaging across the frame. With excellent image quality at maximum aperture becoming outstanding at f/2.8 and on, the Zeiss Sonnar T* 2/35 is likely to become a classic, against which all others are judged. Of course the lens can only be obtained with the purchase of the RX1, a camera that is not entirely without its own quirks and shortcomings. In spite of this, if you have the money to invest, then close to perfect imagery is assured.”. As you stated it doest have IS and its expensive. But who needs IS for a 35mm focal length other than marketing ; and expensive – compared to what? Its small and packed with a bunch of good stuff and IQ is excellent. As far as fixed focal length – i used one focal length from 1969 to 2010 bought the top canon stuff in 2010 and sold it this year. Too big! And found by studing my photos that i still used the same focal length even when i had a zoom. So I will pay more for less.
By the way i have this camera I use it every day – its now replaced my xpro1 (which i have already wore the finish off) and there are qualities about it that you just cant chart. My standard that i am looking for is a digital camera that is the same as the Contax G2 in function and build quality – it is my favorite camera. The RX-1 to me fits that requirement, the RX-1 is a digital blend of the Minolta cle and the Contax G2 film cameras and that is a good thing (my opinion). I think this is the beginning of some great things to come and the “dslr thing ” could become a dinosaur in time. They are so huge compare to the old slrs – they are computers with lenses. If there aren’t companies like sony and fuji the next feature canon and nikon will be marketing are wheels and motors to drive them around. So again i am biased IQ , convenience, size and weight matter. More than a whole list of marketing features and paper specs. And usually things that get boiled down to the essentials cost but in the long run you dont waste your money chasing rainbows.
another view: The Davidson review, http://duncandavidson.com/gear/sony/rx1/ , seems like a good one. It doesn’t contradict DxOMark or my posting. If you have enough money to buy a Nikon D600 and a good lens and then also buy a Sony RX1, the Sony is compact and delivers great image quality in those situations where its compromises (no optical viewfinder, mediocre autofocus system, no image stabilization) aren’t limiting. As Davidson notes “it doesn’t replace my Nikon D800”. Davidson does not suggest that there is some special magic to the 35/2 lens on the camera that has escape all the rest of the world’s designers of 35/2 lenses: “results arguably as good as any other camera fitted with a 35mm prime f/2.0 lens”.
Dailuo: I’m glad that you like the camera. Certainly we must be grateful to Sony for pushing the state of the art in digital sensors, most notably/practically in the Nikon DSLRs. The main point of my original posting was to show the disconnect between folks’ romantic notions about cameras like this and the Leicas versus the objective tests that show that the most commonly used (by advanced amateurs and professionals anyway) DSLRs and lenses perform equally well (and of course are greatly superior as picture-taking tools, assuming that you’re willing and able to carry the weight/bulk).
I agree with Dailuo, for some the DSLR may indeed be the superior picture taking tool, but, and I can’t attempt to quantify it, something is gained by carrying something smaller. To me this is significant and worth having both. If I had to get rid of one, it would be the bulky system.
Mike: Take that RX1 to the playground and see how its AF system deals with a gaggle of children running around! Then try the same experiment with a similarly priced DSLR and 35/2 lens combo.
So let’s wait for the RX2 with proper AF. Though we have been holding on to the Canon 5DMarkOne for years, despite its lousy AF performance, because it was a more compact FF SLR than the 1Ds series. Then the Nikon D700 began putting the glorious N brand back on track and spurred competition for being the #1 brand.
A compact FF camera has its market and Leica redefined the price bracket.
However I side with Dailuo in interpreting the info from DxOMark about the lens and the chip. Sorry to slightly disagree, but this is your second article about a camera, that is now available. I will pass buying for now, hoping that SONY has the technology to speed up the AF of their fixed-lens camera, just as Fuji recently did with their cropped-format one.
I think you need to reread the dxo lens tests.
First they are updated for the 5d mark III.
Second the scores dont even come close, no matter which canon 35/2 you use.
http://www.dxomark.com/index.php/Lenses/Compare-Camera-Lenses/Compare-lenses/(lens1)/1019/(brand)/Sony/(camera1)/833/(lens2)/474/(brand2)/Canon/(camera2)/436/(lens3)/1086/(brand3)/Canon/(camera3)/483
In fact the canon 35/2s are among the most terrible lenses in the database. There are high end Nikon and Canon l zooms that outperform it at 35. Both the canon 35L and the Nikon 35 1.4 outperform the canon when stopped down to 2. Miserable lenses.
Third, go to dpreview which has the exact same dxo test but without the dxomark obscure numbers bs — actual breakdowns by distance and a way better comparator. With it you can see that high end 35s are comparable to the Zeiss objectively, though they give and take (better than the zeiss at the distortion and holding back vignetting, worse at the acuity and abberation control). Makes sense: the former can be compensated for in a raw processor (easy since the lens and body are mated) the latter not so much and usually you get in lenses made for longer register distances (slr).
There are many reasons not to plunk down 30 large on a rx1 plus accessories (cost, no interchangeability, short battery life, contrast af without phase detect pixels in sensor, outrageous price for essential accessories like lens hood battery charger), no need to start grasping at areas where the rx1 is clearly superior to a full fram slr.
Terry: I followed the link that you sent. The newer image-stabilized Canon 35/2 (“IS”) clocks in at 17 perceptual megapixels, the RX1’s 35/2 at 18 MPix. Both are a lot better than a wide-range consumer price zoom (down around 6-9 Mpix) but I don’t think 17 versus 18 is significant. The Canon lens has less distortion (0.4% versus 0.7%). It has about the same vignetting (both are down nearly 2 f-stops in the corners!) and less chromatic aberration. Unless there is some important characteristic that DxO is not measuring, I would rather have the slight reduction in resolution and gain image stabilization (to reduce motion blur on handheld images) by using the Canon rather than the Sony/Zeiss lens.
Philip,
That’s the problem with the dxo summary numbers. They are amalgam of numbers averaged together to provide a comparative estimate. For instance for zooms, that is across the range instead of at 35mm. In the case of the Sony, the total numbers are distorted by the fact that the lens design favors sharpness including corner contrast and flatness of field (the latter is not measured by dxo) over vignetting and distortion. This is because the lesser register distance means there is no retrofocal. The design which make light hit evenly on the film plane.
Note that the latter two are easier to adjust in post (for instance who cares if there is a 2.5 stop loss at the edges if the contrast is enough for le sfix to recover it.)
Here is the exact same data broken down without the summary numbers
http://www.dpreview.com/reviews/lens-widget-fullscreen?compare=true&lensId=sony_35_2_rx1&cameraId=sony_dscrx1&fl=35&av=2&view=mtf-ca&lensId2=canon_35_2_is&cameraId2=canon_eos5dmkii&fl2=35&av2=2
Note that in this, you can pull up say the canon 24-70L zoom at 35/2 and see its performance without aggregate numbers.
It’s fun to troll the database. Try to find a lens that at 35mm and f2 that compares to the rx1 in sharpness. You can’t. But you will see almost all the lenses crush the rx1 in holding back distortions.
This is a trait of sonnar designs which are unique to rangefinder class cameras: http://en.wikipedia.org/wiki/Zeiss_Sonnar
On an slr, a 35mm lens is wide angle enough that all lens designs covering this focal length will need some retrofocal element inside it to either traverse the mirror box or to gather angles before correcting for distortion. These are basically doublegauss lenses (Zeiss calls that planar) followed by some retrofocus group. If Zeiss were to make the they’d be distagons. (Note: almost all fast normal lenses are juat double gauss lenses. this explains why they are cheaper than 35mms)
http://en.wikipedia.org/wiki/Angénieux_retrofocus
these extra lens elements impose a cost on the overall sharpness as it introduces extra glass air interfaces and necessitates more elements or fancier glass materials to correct the chromatic abberations. Aspherical could make the design more compact, but they don’t improve sharpness. (The rx1 has three. Clearly because it needs to be more compact).
Btw there are added restrictions in lens design to support IS and autofocus because here needs to be groups that are small and adjustable. That’s why I think 35mm IS is a bad idea. It limits the design in ways that compromise sharpness and increase cost. Why use a fixed if the sharpness plus cost penalty means your better off with a zoom?
Thanks, Terry, for the dpreview link. It seems that the Sony/Zeiss lens has better corner sharpness at f/2 and f/2.8. I would probably be using the lens primarily at f/4, though, (especially since the lens has IS) and the sharpness curves are almost identical there. Both lenses are substantially better in corner sharpness at f/4 than the Canon 24-70/2.8 L II zoom at 35/4. So there is an advantage to using a prime, but it is only at f/2.8 and f/2 where the differences between these particular primes are apparent in the numbers.