Phototaker41: I like this camera but I don't want to upgrade from my old Sony RX100 because Image Quality is not improved much. So far I am satisfied with my RX100.and waiting for a new RX 100 with longer zoom (ex 24-200mm) and better image qaulity
An RX100 with 24-200mm lens? Isn't that an RX10?
falconeyes: Thanks for the review.
A few minor points of criticism though ...
- Another category "Entry Level Large Sensor Compact Camera"? Really?IMHO, all Large Sensor Compacts are enthusiast by definition. Who else pays 1000$ in this category?
- The Multi USB port physically should support mic input (I believe an older version of the port actually reserved pins for mic input). No ext. mic therefore should trigger a more prominent CONS comment.
- The RX100m3/4 lost a hot shoe but don't feature a sync port. And there is no excuse why the internal camera flash can't be switched to manual (w/o TTL pre flash) such that AT LEAST a remote flash can be triggerd optically. It cannot by simple oversight of Sony's part. No manual flash therefore should trigger a more prominent CONS comment.
Re: the scoring category, the 'Entry level...' category was created when it seemed possible that somebody might create such a camera. As you say, at nearly $1000, the RX100 IV isn't it. I've re-assigned it to the correct category.
nicoboston: I think DPReview should give 3 independent scores:- 1 score for still image quality- 1 score for video quality- 1 score for everything else
I have no doubt that this camera is "the most responsive and enjoyable". It is clearly a very interesting ans innovative imaging instrument.IMO, there is a disconnect between the final score "85% Gold Reward" and IQ. This unique score is misleading because many readers think that best score = best IQ.
Again, I truly believe this is a great and innovative tool; but I also believe that there are less fancy compact cameras that generate better still images. Your scoring method is too simplistic for an instrument with so many features.
Just my 2 cents ;-)
I'm struggling to think of another compact that comes close in terms of image quality. In daylight there are cameras that will make more attractively coloured JPEGs but the large sensor and fast lens mean that the low light capability and dynamic range of the RX100 IV are really hard to beat.
martinj68: looking forwards to the RX10M2 review, any ideas when?
It'll be a while yet: I think reviewable RX10 II's are only now starting to become available.
sportyaccordy: All the studio pics look a little out of focus.......
We shot more than one sample on more that one occasion. We believe any softness (and slight differences in where the sharpest points are), is within the realms of sample variation.
wandiba56: This is the case of money buys praise...what is extraordinary about these shots? Those with lifted shadows look very unnatural!!
bartolyni - We've provided all the testing to let you see why we've drawn the conclusions we've drawn (and explained the aspects of the camera we didn't like).
You're welcome to disagree with our conclusions but it's untrue/dishonest to suggest that this means we were paid-off. We did not receive any money for this or any other review.
berndimax: I cannot download the dpreview video samples.When I try, I get an error message:<code>Access denied </code> ....Anyone else got this problem?
Please try again: it should be fixed now.
Cheng Bao: many links are 404 error, I tried links under What's New -> Auto Focus, none of them are working right now
That should now be fixed, sorry about that.
utomo99: sony need to give the software bundle to maximize it
Capture One Express is a good start.
Allen Yang: I hope Panasonic will upgrade the FZ1000 with a touchscreen and 600mm focal length.
You wouldn't rather have a lens that's a whole stop faster?
Because the FZ1000's lens isn't *that much* shorter, but it is [significantly faster](http://www.dpreview.com/articles/4788790204/shooting-with-the-canon-powershot-g3-x?slide=3).
meanwhile: "but it also shows that 600mm equiv isn't quite as big a leap up from the FZ1000's 400mm equiv as the numbers imply"
Err, the graph shows that because the scale is wrong.
J A C S - It's Log2(x), so a halving of the field of view is represented by (roughly) the same distance on the scale. It seems like a fairly obvious choice.
J A C S: "In video capture, stabilization becomes active in 5 axes, gaining vertical and lateral translation, plus rotation around the focal axis. "
Is that still in-lens 5 axes stabilization? Or there is IBIS as well? It seems unlikely to do 5 axss in the lens.
J A C S: it's two axes from the lens (pitch and yaw), then three additional axes (horizontal and vertical translation, and rotation around the optical axis) from digital correction.
Another anlge: Another angle at the aperture. Non-professional. (Part 2 of 2)
What aperture shows, actually, is how much of incoming light (illuminance, light “quantity” per unit of area) is passed to the sensor. Same aperture in different systems (with similar properties of lenses – refraction, transparency etc.) will provide roughly the same illuminance. And now is the sensor, its physical properties, which is important.
I disagree. I'd suggest that illuminance seems important because the system used to describe exposure is based around illuminance. It has the advantage that you don't have to re-learn anything if you change between formats.
However, this has the effect of hiding the effect of sensor size.
You essentially have two choices:
Think in terms of illuminance and have a separate acceptance that larger sensors give better image quality.
Or consider total light and realise that it tells you *why* larger sensors give better image quality (and the extent of that difference, such that you can establish whether one camera or sensor is over- or under-performing those expectations).
And yes, medium format and large format do have potential to be better than full frame. The 35mm format is only used as a reference point for historical reasons, not because it's *best* or inherently ideal.
What *f-number* related to is illuminance. Without more explanation, the word 'aperture' is ambiguous.
Another anlge: Another angle at the focal length. Non-professional.
Couple of words about focal length in mm of lenses. Honestly, who understands them? Only the ones like you. But for casual user it means nothing. You have to know the system (sensor size), although in fact it’s enough to know how much mm in that system is giving you magnification 1 – meaning, that picture taken will have field of view as seen by human eye. Wait a minute!.. Why don’t just producers mark the magnification of lenses relative to what human eye sees? So just divide the mm of lenses by what mm would give the magnification 1! Let’s test: what the lenses 0,5-2 is? Easy and understandable. Now, c’mon DPReview, you know you can do it – make the world better and brighter :).
If you can find a definitive figure for the angle of view offered by the human eye, then great.
My understanding is that it's not that simple: you have a region where you can see with some precision and a wider region (extrapolated and half-remembered from when you last looked there) out to a much wider angle of peripheral vision. Which figure would you use?
ThePhilips: @DPR, the GX8 in your camera DB is incorrectly listed as having a tilt display, not the "fully articulated" one. And the the "Environmentally sealed" should be "Yes" too.
That should now be fixed.
goshigoo: It is bigger and heavier than Sony A7 ?!
I always think m43 is about portability.........
At this price, wouldn't it better to consider A7 / A7II, which are selling at ~950 USD and 1400 USD only in Hong Kong...
@Martin Hind - you'd need to be more specific about what you mean by 'lets in *as much light* as a FF 2.8 lens'.
It lets in the same light-per-unit-area as *any* other F2.8 lens. But, if you consider how much light it lets in, in total, then you'd find that it illuminates the Four Thirds sensor with the same amount of light that would be seen by available to a full frame sensor at F5.6. This, in turn, affects [shot noise](http://bit.ly/shotnoise).
F2.8=F2.8 but ISO 100≠ISO 100, so the different effect of F2.8 on different systems is slightly obscured.
Or, looked at another way: its aperture (35.7mm diameter) placed in the 200mm lens to which you're saying it's equivalent, would only be 200/5.6.
yahoo2u: The EM-5 can still be had for $500. A lot better deal.
I don't think it's any surprise that a three-and-a-half-year old, probably out-of-production body can be picked-up more cheaply than a newly-announced one.
Buy, even just based on specifications, the GX8 would clearly be a better choice if you've any interest in video.
lacikuss: I can't find the "eye detection" feature in the camera, only face detection. What I'm missing?
There isn't an eye detection feature, *per se*, but the Eye AF feature is something that can be assigned to a Function button.
We'll discuss this more, including how it interacts with your other focus mode choices, in the full review. (Coming soon).
Timmbits: I am really not saying this to criticise, but I wouldn't mind having something like this at a lower resolution (like 10-12MP) so it has even larger photosites (with all the benefits that accompany that - less noise, better sensitivity... smaller file sizes).
The performance of this isn't _that_ impressive, when you consider that the Nikon1 was offering frame rates like this for years now (probably also using a Sony sensor). I suspect when Sony-semi signed Nikon there were certain performance non-compete clauses in effect for X-years.
Only 70mm lens equivalent (remember, that is like a 45mmAPSC). OK for portrait, but not for wildlife or sports (games, etc) and the other stuff. That is understandable, because if you want more, Sony wants to sell you the RX10, and if this did that, there may be no RX10.
Price will come down over time. Maybe I can afford one in a couple years. ;)
@Timmbits a lot of what's been said about the megapixel race (including on here) is wrong. There are various factors that have contributed to this but, although more, smaller pixels tend to look worse when viewed at 100%, in many cases they actually contribute less noise to the final image.
[Read this for a more in-depth explanation](http://www.dpreview.com/articles/5365920428/).
As I said in my last comment: If a 20MP sensor gives you 20MP resolution in good light and *the same or similar* noise when scaled down to 12MP (and they generally do), then why not move to higher pixel counts?
Increasing the pixel size on small sensors won't make up for the difference in sensor size. It would let you get the same image quality as a crop of the larger sensor, but if you view them at the same size, the larger sensor image will be less noisy and more detailed.