Richard, this is not quite their first large sensor model, in more than one was the GXR with A12 28mm and GXr A12 50mm were large sensor fixed, prime-lens cameras quite a while ago.
Most people just focused in presenting the GXR as a system camera where it was more a collection of fixed lens cameras.

You are right in that Ricoh had the GRD and film GR1 cameras way before Sigma and this is just an evolution of these cameras.

Pcmag have a review that includes impressive imatest results here: http://mobile.pcmag.com/?ref=310197&url=http%3A%2F%2Fwww.pcmag.com%2Farticle2%2F0%2C2817%2C2417599%2C00.asp

At f2.8 centre is 2105 lpp, corners 1927 lpp - pretty sharp! Corners become 2120 lpp by f4.

Also, using the 21mm adaptor gives pretty good results considering - 1906 lpp in the corners at f4.

Sounds like a great lightweight landscape tool.

You bring up a very good point. One to which it is very difficult to find answers. I suspect that when using smaller apertures, in this lens, and the pancake lenses, they retain only a "crop", or the center section, of what would otherwise have been a much larger lens. This makes for less light going in, as the outer parts of the lens don't need to be focused through that same size hole: they're not there. It makes it easier to design a sharp lens, and lowers production costs.
I'd very much like to see lux numbers, of a pancake lens compared to a conventional one, with the same field of view. (angle)

26.25mm

The converter is 0.75x so it'll be 0.75x35

There's a sample image in the gallery in the 35mm Crop mode. It's mentioned on page 1 of the preview and is listed as one of the options that can be assigned to one of the Fn buttons.

Try google instead of being confrontational. Sony quotes the WSJ in their marketing.

Marcus, I'm not sure which picture you have looked at, but those chechered corners, at f2.8, Nikon looks depressingly awful. I'm not sure what it means to you, but in the real world it means GR has a better lens (if nothing else). In other specs, both cameras are almost like twin brothers.

@Zvonimir Tosic

You are looking at a different page. The above poster is talking about page 8 (Studio Comparison - RAW) not page 9 (A vs GRD comparison) where for some strange reason, DPR only showed f2.8. (They should add f8 to the crops to the comparison on page 9).

The Studio Test Scene was shot a f8, and here the Coolpix A shows better corners than the X100s, RX1 and GRD (Notice the checkerboard and words Kodak Grey Scale on top border. see link below).

http://www.dpreview.com/previews/ricoh-gr/8

DPR mentions that by 1 or 2 EV from wide open, images are sharp across the frame on both cameras, but the Studio Test scene shows the extent to which the Nikon lens is more even (center, borders, and corners). But all anybody here wants to talk about is f2.8.

f4-f8, the sweet spot, is where you want corners and edges of the frame to be close to the center sharpness. DOF is so shallow at f2.8 corners are just not a priority.

In camera vs desktop RAW conversion makes a huge difference in some cases and if the only value in the RAW samples was to compare identical conversion between two cameras, then it should be limited to DNG files only since all other formats will have different code paths, whether that is inside an Adobe product or across two applications.

Photo sites should ban the word yawn. It seems to be the most frequently expressed sentiment, and, ironically, one of the most exhausting.

@Revenant: That may be how it usually works, but when amazon acquired dpr good old Phil said that there will be more reiews thanks to more staff and stuff. I.e. that they have easier access to test cameras. Doesn't look that way.

Not true at all that it makes "virtually no" difference.

The Pentax K-30 being only 12-bit RAW cost the K-30 at least 1 EV of Dynamic Range vs the K-5 when DxOMark tested them.

From DxOMark:

K5 14.1 EV DR

K-30 13 EV DR

K-01 12.9 EV DR

All three cameras use the same Sony 16 mp Exmor sensor, but the K30 / K01 aren't able to fully take advantage of the huge DR performance that sensor provides.

If you shoot and edit RAW, or photograph scenes with wide DR, 14-bit lossless RAWs are preferable for maximum latitude.

Note: the D7000 and Coolpix A, two 14-bit RAW capable cameras, scored 13.9 EV and 13.8 EV for DR (dynamic range) respectively and they use the same Sony Exmor sensor found in the 3 Pentax cameras.

Date can be found here:

http://www.dxomark.com/index.php/Cameras/Camera-Sensor-Ratings

@marike6

Jaques very clearly asked about "real world" performance, and you quote DxO mark...

Mark, while you are right that a 14 Bit DAC can make a small difference at low ISO DR, you are wrong when asserting that the sensors are equal except for the DAC.

As the 12/14 Bit DACs are actually embedded on-chip with the Sony sensors, they are clearly different beasts.

The difference a 14Bit DAC makes is smaller than 1EV in DR and depends on the sensor resolution and Full Well Capacity.

There is a lot of information in these previews. You just have to read it and think for yourself. You sound like most average people, who need to be spoon fed information, and need their conclusions drawn for them.

@R Butler: what would be the downside in publishing a schedule of coming reviews? It is not like you would be disclosing strategic insight to your "competitors". Such a schedule would help potential buyers better time their own purchases. Those of them who trust your methodology and share your views anyway. The indirect advantage of publishing that schedule would be to force you guys to feel the healthy heat of a public deadline... ;-)

Richard, since you're using the word "honest", you shouldn't really call the Sigma bit a preview but a "first look" or "beefed-up press release with some pics in it".

Bamboojled, this isn't a technical manual. It's a review. It's perfectly fine for the reviewer to show enthusiasm and voice his subjective opinions. It'll make for a very boring reading if the reviewer tries to write in a cold uninvolved language.

Studio shots taken at closed apertures. No wonder, there's practically no difference (other than one caused by different focus points — dpreview isn't careful enough about them).
In my opinion advantages of GR are:
1) Better body with better UI and better control
2) Much better lens, which kills Nikon.
3) Price
The only advantage of A is 14bit RAW.