Official, Sony A7RIII announced, 400mm prime and 24-105 f4

Andy Trowbridge wrote:

TNeilyb wrote:What I like with Sony is that when the A7rII came out we asked for a touch screen and a joystick to help with AF selection. We asked for more battery life. We need two card slots... and these things happen. A7RIII is a very appealing camera for more than just landscapes.

Didn't get lossless compression of 14-bit RAW files, they obviously weren't listening to that. So it shoots 10fps but only in 12-bit.

That is a good point.

Bill Chambers wrote:E.J., where the figures in your comparison come from?  Are they Sony figures or an objective third party?

If you are referring to the table that I posted, it's straight from the a7R Mk III marketing materials.

For E.J. and other techno wizs:

OK, I confused, which, when it comes to tech stuff, is the norm.  I have TWO questions about the pixel shift technology.  According to the A7RIII write up on B&H it says:

Pixel Shift Multi Shooting

This unique compositing mode allows you to achieve even greater resolution than the 42.4MP sensor affords. Working in conjunction with the sensor-shift image stabilization, this mode shifts the sensor while making four consecutive exposures in order to acquire approximately 169.6MP of information for greater color accuracy and detail than possible with a single exposure. These files can then be merged together during post-production by using the Sony Imaging software suite.

If the only difference is just moving the sensor 1 pixel, how does that make it capture any more DR?  I mean, each successive image is recording basically the same data as far as exposure, correct?  Where, or why, does this give you additional DR?

Second question, it says the 4 images can be merged in Sony Imaging software suite.  Does this mean I would have to add yet another software processing program to my list of post processing software, e.g., merge the 4 images in Sony program, then process in Capture One, then finish of in CS6?  That's pretty cumbersome.

Thanks in advance for your thoughts and info.

Bill, check out this video with shows how the pixel shift works. It should answer most of your questions,
https://www.youtube.com/watch?time_cont ... -811lVi0Qc

signgrap wrote:Bill, check out this video with shows how the pixel shift works. It should answer most of your questions,
https://www.youtube.com/watch?time_continue=6&v=6-811lVi0Qc

Thank you!!

E.J. Peiker wrote:15 stops of DR is physically impossible with just 14 bit recording, it's marketing mumbo jumbo.

Only if the data is stored in a linear fashion.  For example, a 8 bit jpg can contain more than 8 bits of DR.

Question: My understanding is that Dynamic Range is greater when photographing at 14 bit versus 12 bit. If this correct, how would the Sony have about the same Dynamic Range as the 850, which is using 14 bit Raw?

Gene Gwin wrote:Question:  My understanding is that Dynamic Range is greater when photographing at 14 bit versus 12 bit.  If this correct, how would the Sony have about the same Dynamic Range as the 850, which is using 14 bit Raw?

Both cameras use 14 bit RAW.

On the subject of dynamic range, here are the pixel level Dmax values as measured for the a7r3 and the D850:
https://www.dpreview.com/news/430214940 ... nikon-d850

Look at the pixel level values, not the bogus 8 megapixel print values.

The dual gain past ISO640 (or there abouts) could be very useful in many situations. Actually being able to shoot at ISO 640 and just slide the exposure up and bring highlights down... Something impossible on a Canon as DR drops in the higher ISO realm. But will an e-viewfinder show the reality or a very underexposed shot? Cannot remember how my old Sony worked.

Neilyb wrote:The dual gain past ISO640 (or there abouts) could be very useful in many situations. Actually being able to shoot at ISO 640 and just slide the exposure up and bring highlights down... Something impossible on a Canon as DR drops in the higher ISO realm. But will an e-viewfinder show the reality or a very underexposed shot? Cannot remember how my old Sony worked.

Preliminary data doesn't show the A7RIII having any great DR advantage over the Canon 5D Mark IV from ISO640 and above.
http://www.photonstophotos.net/Charts/PDR.htm
 
Also DPREVIEW found both these cameras pretty much ISO invariant from ISO400 - https://www.dpreview.com/reviews/canon-eos-5d-mark-iv/11

Yeah but that's comparing a 30mp camera with much larger pixels to a 42 mp camera. That's 40% more pixels while still having the same or better DR depending on ISO.  Perhaps a fairer comparison is the 5DSR to the a7R3 which is only a 19% increase and the differences there, especially at lower ISO is stark.  Or compared to the D850, the 5DSR only has 11% more pixels yet has vastly inferior dynamic range.  Canon still lags in sensor technology compared to the entire rest of the photographic industry, regardless of whose sensor tech they use - still no BSI, still no stacked sensor, most Canon sensors still don't have on chip A/D - some of these items have been in competitors sensors for nearly a decade now.  Also if you look at the quality of the information coming off of the sensors in the deep shadows, most Canon sensors still show fairly strong pattern noise, something largely missing from other sensor manufacturers.  Canon sensors have suffered this since the inception of digital photography but others have largely eliminated it.

Andy Trowbridge wrote:

Neilyb wrote:The dual gain past ISO640 (or there abouts) could be very useful in many situations. Actually being able to shoot at ISO 640 and just slide the exposure up and bring highlights down... Something impossible on a Canon as DR drops in the higher ISO realm. But will an e-viewfinder show the reality or a very underexposed shot? Cannot remember how my old Sony worked.

Preliminary data doesn't show the A7RIII having any great DR advantage over the Canon 5D Mark IV from ISO640 and above.
http://www.photonstophotos.net/Charts/PDR.htm
 
Also DPREVIEW found both these cameras pretty much ISO invariant from ISO400 - https://www.dpreview.com/reviews/canon-eos-5d-mark-iv/11

Whilst I am happy to push a little with my 5D4 I do not find the ISO400 pushed 4-stops works quite as well as DPR will have us believe. I find there does tend to be a stronger colour cast than shooting the right ISO for a level exposure. I am forced to do so at times because of highlights but I do find 2-stops is about my limit and at ISO3200 upwards maybe 1-stop. I am however no scientist

Neil, if you look at the curve, ISO 400 pushed 4 stops will be much worse than ISO 640 pushed 3.33 stops due to Sony's dual gain sensor design.  The secret to ISO invariance on these cameras is to keep the push within the same sensor gain algorithm.  Sony uses one from ISO 100 to 500 and another from ISO 640 and up.  If you push across the two, you will have problems like you describe.  Redo your test with an ISO 640 shot and you will see that there is very little difference if you push that or just take the shot at a much higher ISO.

E.J. Peiker wrote:Neil, if you look at the curve, ISO 400 pushed 4 stops will be much worse than ISO 640 pushed 3.33 stops due to Sony's dual gain sensor design.  The secret to ISO invariance on these cameras is to keep the push within the same sensor gain algorithm.  Sony uses one from ISO 100 to 500 and another from ISO 640 and up.  If you push across the two, you will have problems like you describe.  Redo your test with an ISO 640 shot and you will see that there is very little difference if you push that or just take the shot at a much higher ISO.

Sorry for not being clear E.J. I was referring to the 5Dmk4 as in Andy's post. I will edit to be clearer

Back on to the the Sony, I do wonder now how many photographers will buy the cheaper, still 10fps, higher mega pixel camera with pretty much the same AF capability, big battery instead of the A9??

Neilyb wrote:Back on to the the Sony, I do wonder now how many photographers will buy the cheaper, still 10fps, higher mega pixel camera with pretty much the same AF capability, big battery instead of the A9??

I'd be looking very seriously at that A7RIII, if it had lossless RAW compression and shot at 10fps in 14-bit. 

Andy Trowbridge wrote:

Neilyb wrote:Back on to the the Sony, I do wonder now how many photographers will buy the cheaper, still 10fps, higher mega pixel camera with pretty much the same AF capability, big battery instead of the A9??

I'd be looking very seriously at that A7RIII, if it had lossless RAW compression and shot at 10fps in 14-bit. 

Do we know at what fps it loses the 14bit files? If I did buy one it would not be to shoot 10fps as I normally would max out at 6-8fps. I think my 5d4 is always in stealth continuous anyway. 

Andy Trowbridge wrote:

Neilyb wrote:Back on to the the Sony, I do wonder now how many photographers will buy the cheaper, still 10fps, higher mega pixel camera with pretty much the same AF capability, big battery instead of the A9??

I'd be looking very seriously at that A7RIII, if it had lossless RAW compression and shot at 10fps in 14-bit. 

According to several sites, it shoots 14 bit 10FPS with a buffer of 28 shots so 2.8 seconds.

It does not drop out of 14 bit mode in continuous shooting or with automatic bracketing.  The a7r2 did do that.

While I too wish it had lossless compression 14 bit file, it does have uncompressed 14 bit files like it's predecessor after an early FW update, why is not having lossless compressed files a showstopper for purchase?

E.J. Peiker wrote:

mortsgah wrote:I'm hoping the AF improvements will improve the "AF hunting" and slower AF than my Canon. I was recently in Costa Rica and tried to use my a7rII hand held with the Sony 90mm macro lens and off camera flash for frogs that were slowly moving up plants and sticks and would miss about 25% of the shots due to the AF not locking in. Tried a variety of the various AF settings but still found it tricky. Granted my 1Dx is designed for faster AF action, but I sure would like to see the new a7 be an improvement in that area. So much easier to hand hold for that kind of work than the big ole 1Dx.

You have a combination of a lens not designed for tracking with a camera not designed for tracking there.  The a9 would do much better, probably similar to your Canon but still, macro lenses are simply not designed to do that.   

I have moved completely from Canon to Sony. Earlier this year I sold my 1DX2 and some of my lenses.

On my trip to Namibia in June, I took my A9 and the 70-200GM. I also took the Canon 100-400L MkII with the Metabones adapter as the Sony version of this lens had not yet begun shipping.

The Sony combo (A9 plus 70-200GM) was absolutely amazing for action shots of birds diving for fish into the Walvis bay waters. We were on this small boat which was moving and jumping up and down. Tough with all of us cramped in there, but I managed to get really great shots which I am not sure would have been possible with the Canon system. I used only the medium speed since at high the camera really chews up the card. Yet, there were some wonderful captures.

I tried the Canon 100-400 , sadly it was unable to perform adequately, as expected. However, it did very well with slow moving or stationary subjects.

I have now ordered the A7Riii as an upgrade to my A7Rii since it checks all the boxes for me. Yes, would have liked it to be simply an A9 clone with more pixels to crop into, but that will happen with time. For now I am simply very happy that the fps has doubled, the AF is better, the battery is larger, the buffer is larger, continuous shooting is possible in silent mode and that the camera writes to the card so much faster. Everything else is bonus for me.

The 400 2.8 prime is not for me, too heavy and limited value. Would love to try Sony's 200-500 f4 if they ever come out with one.

Pradeep

Pradeep, there is a Sony FE 100-400 that is very very good!