Lee Jay: There are many root problems with this approach:
- File sizes and the resulting processing time- Video compression and likely no raw mode- Videos are often shot with slow shutter speeds to make for smooth video- Slow video shutter speed often results in motion blur as seen in the first image- Little or no ability to use fill-light- What little fill a video light can provide pales compared to what a flash can do- Can't use a flash to freeze motion- Even 30fps is often too slow to catch the moment. Anticipating a single shot can be more reliable.- Stuck using an EVF
I've taken stills from video before, but the situations in which it is a useful thing to do are extremely rare.
The article already makes this point.
4K Photo is about making it easier to use the 4K capability to get stills. It does little to mitigate the conflict between the needs of stills and video. Maybe I need to make this more explicit.
DigitalWalnut: I shoot with RED cameras regularly, which can do something similar (using the HDRx feature to record a second RAW track with shorter shutter speeds, from which you can extract stills), and I've found its usefulness to be fairly limited. Unless you're shooting a thoroughly choreographed scene with marks for your actors and an AC pulling focus, video/cinema work is usually done at slower apertures for the sake of having a greater depth of field (since you have to keep things in focus continually, not just for a frame or two). When your lighting or ND filters are attuned to the shutter speeds you'll be using for video, that puts even further constraints on the potential for good-looking stills from the same camera. I guess there might some situations when you absolutely have to have photos and stills from the same event and can't stop recording to whip out a DSLR/mirrorless/whatever, but I can't imagine it comes up too often.
I tried to stress in the article that there are tensions between trying to shoot video for its own sake and trying to shoot video with the intention of extracting stills.
This 4K Photo mode makes it easier to shoot video for stills extraction but it does little to mitigate the conflict between the two requirements.
Nikguy: Please stop calling the Rx10 an enthusiasts camera. Rent one go on a shoot like a grandsons baseball game or a wedding inside a church with no flash to be used. This camera fits the moment either video or stills. It' the best travel etc. camera I've ever had. And now Sony has made it better. I guess I'm enthusiastic about it!
@brownie314: we use enthusiast camera to mean a camera aimed at dedicated photographers who are likely to want to take control over their cameras. We use it to mean it's a *serious* camera.
electronbee: Only thing I don't like is the 1" sensor marketing ploy. The 1" sensor was originally devloped to replace a 1" imaging tube and the sensor is nowhere near an inch on any dimension. Ok, I'm off my pedastal.
electronbee: it is a terrible naming system but it should be stressed that it's almost universally used for sensors smaller than APS-C.
We're open to ideas about what else to call the different sensor sizes (not metric dimensions, though: we need a more sensible system of names for classes).
Petroglyph: I noticed the specs say Raw (ARW 2.3) any significance to the ver 2.3? Compression changes, for instance.
The original a7 is also listed as v2.3
jon404: Richard -- am I right in saying that, until you get into very high ISOs, there isn't much difference between large-or-small pixel counts, for equal size sensors -- BUT that the small-pixel variety, like the 36 MP Nikon, gives you more room to crop and also to deliver a good image through reducing it?
And so... that the small-pixel cameras, like my RX100 with its 20 MP on a 1" sensor ... might be the most useful, the most practical for professional photographers, mainly because of the extra cropping that their great MP count allows?
Broadly speaking, yes.
A high pixel count camera might drop a little behind at very high ISOs but will often match a lower pixel-count camera and, at low ISOs will offer more detail and opportunity to crop.
dosdan: The comparison of the noise of the D810 and the A7S in the 2nd page of the article was not a good choice, as the A7S is likely to be using Aptina's (now ON Semiconductors) DR-Pix technology.
There is a step in the read noise between ISO1600 and IS2000+, with the higher ISOs show less RN than expected. DR-Pix technology allows the changing of the gain in the sensor, so the sensor designer's trade-off between Full Well Capacity, Conversion Gain & Read Noise can be altered in-camera.
A better pixel-size comparison choice would have been between the Pentax K-5 (16MP) & K-3 (24MP), both APS-C cameras.
On the DR graph you can use the Screen tab to seen the pixel-level performance, while the Print tab shows the result using the same output size e.g. same print size.
We were aware something unusual was going on with the a7S but weren't sure it was that. But the fact that, even with a design that gives a non-linear performance, it's still not *dramatically* better than the D810, means it still gets the point across.
Having used the D810 in the first example, I wanted to continue the theme, rather than risk appearing to cherry-pick examples that fit with the theory. Furthermore, there's a general assumption that the a7S is the best high ISO camera imaginable, so it's still interesting to see how it compares.
zakoops: Will be interested to find out if BSI-stacked sensor of RX100M4 ($1000) has about twice the image quality of BSI Nikon J5 ($500), both being 1” and 20MB sensors.
Upgraded video is quite nice, but first what about image quality?
It depends (it's hard to quantify image quality). With the kit zoom on the J5, the Sony is likely to be significantly better, though.
The RX100 IV has a 24-70mm equiv F1.8-2.8 zoom, rather than the ~27-81mm equiv F3.5-5.6 lens on the Nikon.
Even if they used the same sensor, that would give the Sony around a 2EV (4x) advantage across its range.
random78: Could the preview folks please elaborate a bit more on the statement "Focus appeared to be on par with a native Canon body". Does it mean on-par with native Canon bodies in live view, or on par with native canon bodies in non-liveview PDAF mode.
I'm afraid we don't have the camera here (Seattle) and the guys in New York with the camera don't have a Tamron lens. We'll make a note to check, when we receive the camera, though.
We're being told it appears comparable with OVF, secondary sensor PDAF.
Martian Keyboard: I have to re-quote the link that The-Bunker posted below because it seems so worthy to know about Stacked Sensors.
It could be the next generation of sensor technology, - I'm not sure.
If it is what I seem to understand, it will even make phone cameras significantly improved.
In fact you'd expect it to have more of an effect in phone sensors, since the circuitry is a larger proportion of each pixel.
However, in this instance, Sony seems to be making creative use of the extra space this frees-up for circuitry, as well as maximising the size of the photosensitive area, to give a much faster chip, capable of really interesting things.
KonstantinosK: 42.4MP BSI CMOS! GULP! Sony has them licked... Any hint of 14bit RAW?
Sony classifies its existing models as 14-bit as well.
Everything we've heard so far has suggested it is still compressing the data within its Raw files but we're seeking clarification.
rarawin: Does the electronic silent shutter mean that the camera is capable of faster flash sync speed than 1/250s?
I'm afraid not: it's the physical shutter defines the flash sync speed. It's 1/250th of a second in this case.
slightlystippled: This is not the first camera to have XAVC-S, the RX10 had it long ago.
Not true, I'm afraid.
The RX10 was launched before the a7S, but didn't offer XAVC-S when it first hit the market.
XAVC-S capability was added in a firmware update in [August 2014](http://www.dpreview.com/articles/9371229209), some 10 months after launch.
bobn2: You're still saying this is a Sony sensor. I'm pretty sure it's a Toshiba like the D7100.
BarnET: the Toshiba sensor in the D7100 did, but it could be it's been fixed between that generation and this.
onesoul: Interesting that you cannot do a direct camera vs camera comparison between the D7200 and the SAMSUNG NX1. Why can't I compare these two competitors head to head? Is this a political thing?
No, it's a category thing. The NX1 is in the semi-pro category, whereas the D7200 is at the top of the mid-level category.
It's not perfect: you have to draw lines somewhere and wherever that ends up being it becomes awkward around the boundaries.
(The D7200 is mid-level because the D7000 was a down-specced camera from Nikon's semi-pro D300S, before you ask).
If you click the 'Include all categories' box next to the scoring comparison, you'll be able to see them side-by-side. The semi-pro class isn't held to a *much* higher bar, so that comparison should still show the relative strengths and weaknesses.
Lin Evans: I'm having difficulty recconsiling this statement: " Despite the relatively low resolution of this sensor in the D7200, it will still stick to your subject in a wide variety of scenarios better than any competitor's APS-C DSLR, and the dedicated phase-detect module means the camera is also good at actually focusing on that tracked subject, even in continuous shooting. The improved low-light performance (down to -3EV) means you can continue to focus, and track, subjects in very dim light." and this statement: "If you're specifically after a DSLR and you mainly want to shoot stills, then the D7200 is an obvious choice:" Why would a camera which has the best AF tracking available in an APS C dSLR be considered one which is an obvious choice if " you mainly want to shoot stills???" Perhaps I"m misunderstanding what you mean by "stills." To me stills are non action - I assume you mean this to be non-video?
Still*s* images, as distinct from video frames.
Not images of still subjects. Sorry for any confusion.
Randy Veeman: I see different numbers here. Can someone confirm what is correct? In the article it says Sony mirrorless revenue is up over the past year and DSLR revenue is down. That does not mean unit sales and could be misleading.Someone said for 2015 (6 months not a full year), DSLR unit sales are growing (or not falling) faster than mirrorless. Maybe that is units shipped though?
So far the most telling number if correct is Sony having an 11% market share. In 2006 Sony predicted a 20% market share by 2011 and Wikipedia says they had a 13% market share in 2008.
You're looking at the past four months, I've used data for the previous 12 months.
The article is based on the *value* figures from Sony. These aren't misleading, but they do tell an incomplete story.
For this reason I've included CIPA shipment numbers for both value and volumes.
theprehistorian: Looks nice. Now how about some decent DX primes?
@Long Tom: it is.
Camera offering shallow depth-of-field can be stopped down to offer deeper depth-of-field (until the lens can't be stopped-down any further). The opposite may not be true.
The same depth-of-field will be achieved when using [equivalent apertures](http://bit.ly/equivap). And if you use the same shutter speeds then you're likely to get very similar image quality, regardless of which ISO setting you have to use, since the noise will primarily be determined by [shot noise](http://bit.ly/shotnoise).