Buying guides 2015

News Reviews Articles Buying Guide Sample Images Videos Cameras LensesPhonesPrinters Forums Galleries Challenges

Sony Cyber-shot RX10 III video overview & 1st look at Zoom Assist

Published Mar 30, 2016 | dpreview staff

Sony's new superzoom may look a lot like its predecessor, but an updated 24-600mm lens makes the Cyber-shot RX10 III a much more versatile camera - in theory. We've been able to spend a bit of time getting to know the camera's new features and putting its long zoom range to work. Take a look at what's new to the RX10 series in our video overview.

Tags: video, sony, sony-march-2016

Gear in this story

Sony Cyber-shot DSC-RX10 III

I own it

111

I want it

I had it

Discuss in the forums

Sony DSC-RX10 III Cyber-shot Digital Still Camera

Pre-order now

Not yet released

$1,499.99Shop now

Comments

Total comments: 92

Expect the Nikon DL 24-500 will become more attractive in the market because of its size, video, price and only 100 zoom short.

0 upvotes

I'm wondering about how the stabilization compares to the 5-axis? Also is the lens as weather sealed as the body?

Edited 1 minute after posting

0 upvotes

The image quality at 600mm equiv is unbelievably good considering that this is a superzoom camera with a 1" sensor.

3 upvotes

Rishi great review as usual, pointing things that others did not catch, like 1:2 Macro which is nice too.

4 upvotes

Silly me. I thought that as there was already an article about this as a stills camera, the video feature would feature the cameras video feature.

3 upvotes

Still don't understand the usefulness of focals above 200mm on a camera like this: outdoor sports? Wildlife?

Really?

4 upvotes

Yes, I would rather have had an f2.2 lens keeping the 24-200mm range. The 600mm might be useful in some instances but overall looks "showy" to me.

The 1" sensor would benefit from having more light and the option of shallower DOF.

3 upvotes

Just to confirm, is the manual focus behavior just as bad as the RX10 MKII? I find it unusable for video. I can't believe they are marketing video around this super duper lens and while keeping this approach for focus.

Also, why are they still playing the 30 minute recording time limit game? Yes, I know all the usual BS about taxes but let get real, I am not buying it because of the 30 minute limitation. How many sales could they add if an unlimited version was offered? I only need 1080p60 to be unlimited, not 4k for this to be a more useful tool for my business.

4 upvotes

What exactly about the MF behavior bugs you? I'll tell you what bugs me: having to assign a button to focus magnifier just to get focus magnification in video mode for MF. It's not engaged automatically when turning the focus wheel. And yet it is engaged automatically when doing so in stills shooting, which means that that focus magnifier button you've assigned for video becomes useless (redundant) in stills shooting... on a camera that has sparingly few customizable buttons to begin with! Well, at least the RX10 III has a couple more spare buttons than its predecessor.

0 upvotes

Still waiting to hear if the lens dislodges as the zoom goes past 136mm, and whether the zoom is too uneven to use during video. That's what I hear about the RX10 II.

0 upvotes

A petty criticism I know, but when describing the macro magnification, your reviewer/presenter described is as 0.49 (letter) X. It's not a letter X, it's a multiplication symbol, so the image on the sensor is 0.49 TIMES(about half) life-size.

2 upvotes

Turns out people actually do that on purpose, thinking it sounds better. (but in reality makes them sound stupid.)

Edited 2 minutes after posting

1 upvote

Did you ever pay attention in math class?

(the accusations of "stupid" in this forum reflect poorly on the accusers)

0 upvotes

Before anyone else derides use of the symbol "x" as somehow not meaning "times" take a minute to educate yourself:
https://en.wikipedia.org/wiki/List_of_mathematical_symbols

0 upvotes

For clarity, I wasn't deriding the use of the symbol 'x' for the multiplication sign, I was commenting that the presenter pronounced it as an 'x' (the letter) incorrectly; it would be like calling the plus sign "cross". If that was unclear in my original remark, I apologise.
And if we're going to Wikipedia for clarification, you might like to read, and pay attention to, the first line in this article:
https://en.wikipedia.org/wiki/Multiplication_sign

1 upvote

why didn't you criticize "point 49" as well? The "correct" enunciation would be "49 one hundredth"

http://www.sciencealert.com/people-who-constantly-pick-up-grammar-mistakes-are-kinda-jerks-scientists-find

0 upvotes

One of these and a RX1 Mk2 (X100t maybe) would be my travel kit of choice, I think it covers most things unless you are very specifically traveling to shoot landscape and wildlife, I think if the opportunistic sighting of wildlife and scenery urban and in the wilderness, for example if hiking, fishing or shopping, is your primary activity this would be a decent rig, although MFT aficionados may disagree.

wouldn't be able to afford to go far though, oh well Derbyshire gets some great light in the spring

Edited 11 seconds after posting

1 upvote

I shoot almost exclusively Sony at this point, but Canon does have to get the credit here for zoom assist. Sony has 100% copied that feature. Canon has had that feature for several years now on all their large zoom compact cameras including the G3 X, a direct competitor to this camera.

2 upvotes

I'm not sure what was the first camera with zoom assist feature, but the first time I saw it was in 2012 with Nokia's Pureview 808.

Now the next commenter can tell us where Nokia copied it from :)

0 upvotes

Eh, i don't think the Pureview 808 actually has any zoom, so that would be crop assist, and that indeed isn't anything new or unusual.

But i agree that it doesn't matter who was 'first'.

Edited 2 times; latest 2 minutes since posting

0 upvotes

You're right, it was more like "digital zoom assist" or crop assist as you said, but nevertheless what you saw on screen was just like how zoom assist works on the cameras mentioned by EricoftheNorth.

0 upvotes

I wonder how this one compares to A6300. I just sold the A6000 after 6 months because it drain battery too fast, shutter click to too loud, EVF lags and the 16-50mm kit lens starts to make funny noise at power on. Looking for replacement and considering A6300, this one and Nikon DL 24-500.

0 upvotes

a6300 EVF lag is very, very short, almost imperceptible at this point. Make sure you set the EVF refresh to 120Hz, and Display Quality to 'Standard' (not 'High').

Just FYI, the RX10 III doesn't really have much of a shutter sound, because of the leaf shutter.

4 upvotes

The shot of Alcatraz is way better than what I get with GX7 and PZ 45-175. Of course my setup only cost 1/3 but still, RX10 have amazing lens and sensor.

6 upvotes

But how does Sony implement the Zoom Assist feature? The EVF, LCD are all fed off the lens' view, unless you zoom out or have a separate lens giving the wider view, how can the view change??
Anyone?

0 upvotes

He says the lens zooms out momentarily. It then probably uses a digital crop to simulate zooming in.

Edited 2 times; latest 2 minutes since posting

1 upvote

@fatdeeman That'd make sense. What I wasn't sure about is when Rishi says "zoom in" literally means the lens zoom out or just the screen view "zoom in". The reason is that it's supposed to take "4 secs" to go from 24-600mm according to the DPR handson. That's quite slow to zoom, but in the video, it seems instantaneous.

0 upvotes

It zooms the lens back out, and then there's a little frame that gets smaller and as you turn the zoom ring that tells you what the FOV will be if you zoom into that focal length.

In other words, while in this mode, as you 'zoom in' by turning the zoom dial, the lens doesn't actually zoom in (it's zoomed out to 24mm, remember); instead, it just shows a frame of what the FOV will be. Then when you let go of the Zoom Assist button, the camera zooms into the preselected focal length, which you've determined by turning the ring until seeing the framing you want.

Should be pretty self-explanatory in the video where we show Carey using the feature.

6 upvotes

@Rishi, Thx for the detailed explanation. After retracting and selecting the view from the wider position of the lens, does it cause much delay to zoom back to the 600mm position? The optical quality of the lens at 600mm seems really good, gives Olympus 300mm F4 a run for money. The only weakness is the lack of PDAF or the Sony 4D tracking. So it can't be effective for moving objects in spite of the Zoom Assist

Edited 1 minute after posting

0 upvotes

It seems fairly fast looking back in our video at 1:27 and 1:34. Not sure we filmed how long it took to use go from zoomed out to 600mm in the Zoom Assist mode - sorry. Will try and go back through the footage to see if we have that.

Yeah, it does seem optically impressive. Would love to see some quantitative data on this lens for comparison (DXO?) - it'll still take us some time to get our lens tests back up and running (I'm working on it though).

Some sort of dual pixel architecture is really necessary on these things for phase detection, particularly at long focal lengths. One thing I wonder is how on-sensor phase-detect systems will deal with the large phase differences of longer focal length lenses. And how they'll be able to make measurements from extreme defocus.

One thing's sure: almost anything is possible, and I believe engineers will find some way to crack this problem. :)

3 upvotes

The 4secs num is from here: http://www.dpreview.com/news/9082856164/hands-on-with-sony-cyber-shot-rx10-iii?slide=4
If you work on the test score on long lenses, I strongly suggest including a test of sharpness at various distances, say at intervals from 10-50yards, e.g. nail an eye chart on an side wall in a parking lot. I can't believe not a single site does something like that. Different lenses' resolving power at different distances can vary dramatically, we need some quantitative measurements. If you do it, that alone will make DPR future test report worthy of reading.

0 upvotes

The 4s was a very rough estimate; just a rough guess on the spot, not measured. Also, that was when using the rocker. It may be quite different with Zoom Assist. We'll remember to measure it once we get one in our offices.

As for various subject distances - I've wondered this as well. However, I have yet to see conclusive quantitative evidence that the results change drastically with different distances beyond a certain threshold distance, say, 30m or so. Between 40x focal length and infinity, I have a feeling that results don't drastically change - do you have any evidence to the contrary?

Results can certainly change at closer focus distances.

The thing is, it would be very difficult to test multiple focus distances, & I am yet to be convinced there are huge differences - but that's a chicken & egg problem. Until you have the automation and resources to test multiple lenses at multiple distances, you just won't know how much of an issue this is.

1 upvote

More important to me, actually, is showing results for both focus at the center (flat-field results), vs. focusing at each measurement point in the frame. Else, for lenses that have a lot of field curvature, you risk underestimating the lens performance for portraiture - where you'd actually focus an off-center subject using an off-center point.

Even if we did do one subject distance, though, I feel like there's so much value add we could have. First of all, testing representative copies, showing both visual and MTF results, presenting them in an easily comparable manner (a la our old lens widget), for perhaps a couple different bodies, etc., would still provide something unfortunately I can't find elsewhere.

If you feel thorough comparative results exist elsewhere, though, please do let me know - I'd like to know if I've overlooked something or some site. :)

1 upvote

@Rishi: I have some anecdotal evidences re. resolving power at different distances. For images taken close by, these two images have similar details, all taken at 300mm equivalent:
http://4.bp.blogspot.com/-7ICtHbUxmm4/Usl5BjXg8wI/AAAAAAAAcq0/XSYII8vvNNc/s1600/P1010317.jpg

http://1.bp.blogspot.com/-Fsmyu1PZSLs/VqOE9a-LTzI/AAAAAAAAs0M/6DqygaMi6FA/s1600/P1230747a.jpg
by two lenses that can't be more different (btw not my pics). And I'm certain if the bird is placed 30-50 yards out, the results will be drastically different even though I don't have the images to prove. And you'll agree to that too once you find out what the lenses are (I'll give answer tmrw :). The surprise is how similar the two images are.

0 upvotes

Hi BostonC - not sure what to conclude from those. Mind PMing me any details about what I should extrapolate? I'm working on designing lens tests at the moment. Thanks!

0 upvotes

Obviously, when looking at picture #74, this camera does not feature Eye-AF or the algorithm is not designed to recognize peacock head...

1 upvote

What??? No bird face recognition??? Useless!!!

9 upvotes

This is why pros use Canon.

1 upvote

hold the phone, Canon got bird face recognition? i am in love. <3

2 upvotes

And rhino face

2 upvotes

Dang. I'm selling the A7Rii and getting this beast. Seriously, I am impressed with the image quality all the way to 600mm.

0 upvotes

i am looking for a7r ii, how much u want for it? what condition is it in? thx

0 upvotes

Wait for FZ1000 II to top that, along w DFD focusing, touch screen. Hopefully it comes w better lens spec, and AF assist (kudos to Sony for that). And $500 less.

7 upvotes

Meant Zoom Assist. Still can't figure out how it's implemented.

0 upvotes

And weather sealing.

1 upvote

Why shouldn't I wait for the RX10IV with 1000mm and APC sized sensor by then?

2 upvotes

Because a 1000mm f/4 on an APS-C would be the size of a Canon 500mm f/4 with an EOS-M on the end.

0 upvotes

I'm just pointing out the fallacy of waiting on purchasing a real camera by talking about imaginary specs of an imaginary camera. 😜

1 upvote

Sony is marketing this as having "pro-grade" video capabilities, but the usability regarding focus is anything but.

A fly-by-wire manual focusing ring, CDAF, and no touch screen make it difficult to do ANY follow focus/ focus pulling effects (manual or auto).

3 upvotes

Pro video people don't use AF.

4 upvotes

Heard about WiFi and remote control through smartphone/tablet Apps? No chance of inducing wobble in the process (at least not in the camera ;-)

0 upvotes

"Pro video people" can't use fly-by-wire for manual focus either.

2 upvotes

Yeah they can. It's much much easier to adjust to millisecond delays than it is to rely on a computer to guess where you want the focus to be or hope that our fat human fingers can pick out what small fine detail we want to focus on from a tiny screen. Unless you're constantly going to extreme opposing planes every half second, wire focusing shouldn't be a problem.

4 upvotes

An f2.8 lens made a lot of sense in front of the 1" sensor. You got effective light gathering and DoF control that was on-par with an APSC DSLR with superzoom - With overall IQ being probably a bit better, good build quality and a more convenient package.

With this lens, the camera is back in traditional bridge-camera territory. A good bridge camera, but no longer a DSLR alternative.

2 upvotes

Ok, I take it back.

220mm (actual) at f4 is equivalent to an APSC camera with a 371mm f7.2 lens.

So, take a Nikon DSLR, a D7200 say, and add a Nikon 18-300mm f3.5-6.3G ED VR Lens.

That's a very close match, with the Nikon being just 1/3rd stop faster.

Price for the Nikon set-up? $1793.

0 upvotes

This is not a replacement for RX II...You can still buy 2.8 but of course not 600mm.

2 upvotes

RX10 III vs. your example:

24-600mm vs. 27-450mm
1051g (2.32lb) vs. 1225g (2.70lb)
5" vs. ~6.9" depth

The Nikon depth is just adding the two depths together ignoring the overlap that goes into the body.

3 upvotes

Compare it to a 4K DSLR and 24-600mm eqiv zoom lenses. The RX10 III #1 feature is 25x zoom for 4K UHD video. If you just want to shoot stills get the A7 or a DSLR.

3 upvotes

Totally agree.

0 upvotes

It's about F11 full-frame equivalent on the tele end.

2 upvotes

A better comparison would be a GH4 with 12-35mm f2.8, some mid tele like 35-100mm plus 100-300mm.

0 upvotes

It would be interesting to see a FF 600 f4 lens shot at f11 to f16 to compare with the RX10 III wide open.
Perhaps include something like that in the full DP test ?

0 upvotes

"It's about F11 full-frame equivalent on the tele end."

What means "equivalent" here?
The crop factor of the 1'' sensor is 2.7.
It's the DOF (depth of field) which is equivalent to about 600mm full frame at F10.8 - with the CoC (Circle of Confusion) also adjusted according to the larger sensor, usually it is 1/1500 of the sensor diagonal.
But the 220mm 4.0 of the RX10 III is still F4.0, not F11.

DOF-Example:
220mm Tele at F4.0 in 10 meters focus distance in front of a 1 inch sensor leads to a calculated DOF of about 172 mm.

The full frame DOF equivalent is about 600mm Tele at F10.8 in 10 meters distance: Calculated DOF about 166 mm.
But the background (in the same distance behind the focus plane) is more blurred compared to the 220 mm lens.

And 600mm with a full frame cam in 10 meters focus distance at F4.0 results in 61 mm DOF. But of course, the lens diameter would be very large then (600mm / 4 = 150mm).

Edited 2 times; latest 8 minutes since posting

0 upvotes

Usually 'equivalent' when discussing aperture relates to amount of light let on to the sensor, which also approximates to shallowness of DoF.

I use this calculator to work out lens properties between sensor sizes: http://www.pointsinfocus.com/tools/depth-of-field-and-equivalent-lens-calculator

Edited 2 times; latest 1 minute since posting

0 upvotes

@The Squire:

Thanks for the link to pointsinfoucs.com's DOF calculator!

I checked it with the following data:
Full frame cam (Nikon FX, crop factor 1.0)
Circle of Confusion (CoC, sharpness criterion): 0.03 mm
Lens: 600 mm
f-number: 10.0
Object distance: 10 meters (10'000 mm)

Results for DOF:
pointinfocus online calculator: 178 mm
DOFMaster online calculator: 0.16 m (=160 mm)
Excel sheet with DOFMaster formulas: 157 mm

Why the deviation with the pointsinfocus-calculator?

I checked the formulas there and found an error in the formula for the near distance of DOF:

In the denominator (beneath the fraction line) it says:
H + s + 2f

It should be (compare DOFMaster formulas!):
H + s - 2f

Otherwise: Nice calculator!

Edited 2 times; latest 2 minutes since posting

0 upvotes

I assume that the 0.49x magnification is a 35mm film (FF) equivalent, because 0.49X magnification on a 1 inch sensor would be super close. Is it a 35mm film equivalent magnification?

0 upvotes

It is equivalent.

1 upvote

way too pricey for most of us folks. Perhaps if it had an ASPC sensor it would be worth it but with a 1" sensor no thanks.

Edited 46 seconds after posting

6 upvotes

Agree100% 1" sensor is total cr-p at that price your better with a 24meg sensor aps-c far far better IQ
Tom G

2 upvotes

And how much you would pay for the lenses to cover that FL range?

3 upvotes

I'll match the video OOC against most APSC DSLR's, except the NX1 any day, and you would never tell which one was the 1" sensor! It's that good! Read the reviews at B&H from the pros!

0 upvotes

See thread above: A less capable Nikon APSC DSLR with a shorter superzoom would cost $1700+

0 upvotes

I know it's just because it's a 600mm lens, but that's quite a cool demonstration how the houses from Belvedere were compressed in above Alcatraz, and the contrast was decent considering the atmospheric effects that are common over the Bay.

0 upvotes

Thats a huge camera, is bigger than an A6300 with 18-200 attached

0 upvotes

Too bad the E mount doesn't have a 600mm equiv. lens.

3 upvotes

Not the same.

3 upvotes

GM 70-200 + 2x tele on the A6300?

2 upvotes

It's large, but the only reason for Sony to make this RX10III 24-600 f2.4-4.0, is that most of their competitors already have a such "compact" cameras

Nikon DL 24-500 2.8-5.6
Pana FZ1000 25-400 2.8-4.0
Canon G3X 24-600 2.8-5.6

It's actually only Canon that managed to make theirs small, but most likely by compromising IQ.

http://camerasize.com/compact/#665,672,556,626,ha,t

1 upvote

I can't think of any circumstance where 1 inch camera / lens size would make a difference though. All great cameras. Think about what choices we had 10 years ago.

Edited 30 seconds after posting

0 upvotes

Nice for the money!

4 upvotes

I like it!

4 upvotes

Thanks for the quick preview of the zoom assist feature! I was having a little difficulty visualizing how it worked.

4 upvotes

What camera did you use to record Rishi? It looks... bouncy. Otherwise, thank you, and I appreciate you answering a question I hadn't thought to ask - the non-bouncy flash is disappointing. I use that all the time on my a6000!

0 upvotes

Bounciness courtesy of a heavy lens on an a7R II, and a little too much coffee for the cameraman :)

7 upvotes

Nice snappy fast-paced video - the way it oughta be done!

4 upvotes

Zoom assist. Now that's a good idea.

6 upvotes

I agree. And people who complain because it doesn't have a mechanically coupled zoom obviously don't see the possibilities. (I also like being able to frame when using remote control apps.)

0 upvotes

Zoom assist is a very cool feature. Surprised it took this long for it to appear - is this the first example? However, where this is most needed, i.e. fast paced photography with a lot of scene changes, it probably cannot really be used.

0 upvotes

No one seems having asked the question: how is it implemented w/o a separate wider lens or zooming out?

0 upvotes

BostonC - it does actually zoom the lens back out when you press the Zoom Assist button.

0 upvotes

Looks interesting this new camera.

Given your comments on the focusing system, I guess it would be fair to assume that birds in flight style photography might be asking a little much from it?

1 upvote

You'll certainly get better results for BIF from a dedicated DSLR, yes, absolutely.

1 upvote

Total comments: 92

www.dpreview.com

connect.dpreview.com

Editorial content

Product database

Community

About us Work for us Advertise with us FAQ Feedback Interest-Based Ads Privacy Legal

All content, design, and layout are Copyright © 1998 - 2016 Digital Photography Review All Rights Reserved. Reproduction in whole or part in any form or medium without specific written permission is prohibited.

Mobile site


Lights... Camera... Stinging nettles? Our Sony a6300 video shooting experience Mar 31, 2016	First sample images from 20MP Sony Cyber-shot RX10 III Mar 30, 2016	Hands-on with the Sony RX10 III Mar 29, 2016	Sony Cyber-shot DSC-RX10 III puts emphasis on lens reach and video capabilities Mar 29, 2016

98	Camera hardware interest vs. actual photos
69	Disappointed A6300 High ISO
63	What's keeping you with m43 now that Sony has released a 100MP medium format sensor?
59	D5 Read Noise Continued.
48	85mm GM on the way. Test vs. 85mm Canon L on the weekend :-)
48	The new Panasonic GX80 is coming
44	Do you run uncompressed RAW,
42	Why do people ask for C&C and then get angry?
37	Have you been disappointed by a camera upgrade?
36	Rumor: Smaller GX8 to be announced soon?
36	D5 "Auto AF Fine Tune Failed"
34	Sony RX10M3-Camera of Year 2016?
32	Going from iPhone 5 to Galaxy S7 Edge
31	Pany 100-400 or Sony RX10 III
31	EM-1 MK II
30	Panasonic vs Olympus AF
30	70-300 comparisons
30	A Quick Side by Side comparison of Sharpening tools
29	16-35 FE Zoom lens.....3rd lens lucky! Passed brick wall test.
29	Nikon D5: first commercial shoot & initial thoughts