Jogger: Just wondering which compact camera DPR would recommend in place of the RX100/2 .. serious question. DPR reviews are increasingly irrelevant.
You don't seem to understand what I just said. Only if you zoom to 100%, you can bypass scaling. Showing the full (scene of the) picture on a monitor, involves automatic scaling. So in my first mention of scaling, of course I didn't refer to zooming, because who in their right mind takes pictures to only look at 100% crops rather than the full scene. You don't either, so my comment about scaling for monitor purposes still applies.
M DeNero: Another niche model from Sony that is too expensive to generate strong sales and will contribute to Sony's financial problems. They have too many niche products.
Yes, because one camera per mount really builds a system that people would faithfully "invest" in. ;-)
What do you think happens when you print at a given (standard) size with 300 DPI? Resampling/remapping done by the print software on your computer or the printer itself. The chances of hitting exactly 300 DPI or say 150 DPI at the print size chosen with the resolution used are usually close to zero.
Same with displaying on your monitor, because I have yet to see a consumer use a monitor exactly matching the resolution of his/her modern camera. No zooming required. In fact, only if you always display your pictures by zooming to 100%, you'd bypass resampling/scaling, but in that case, you'd always end up with a tiny portion of the original scene being displayed. Not plausible.
Except that it seems to sell well, which isn't too surprising after the success of the predecessor.
It's compact cameras like these that can really set themselves apart from cellphone cameras and therefore will be able to dodge the fall of compact camera sales in general, much longer too.
gavp: It does indeed look very capable in low light. Probably far less so in bright light - a base ISO of 160 and only 1/2000th on offer will potentially be quite limiting. Something like an LX7, with ISO80, 1/4000th and built in ND filter certainly trumps it easily there.
First, there were several (non OEM) solutions for the old model that did not involve glue and yes, Sony offers its own solution for the II now (again without the need for glue).
It's hard to take people seriously who call anything that involves scaling "tricks".
Because practically every picture you have ever taken, was subject to scaling/resampling at one point before ending up as a physical print or a picture displayed on your monitor. Even without your consent.
If you don't use an external ND or even polarizer filter that is...
First, there is no RAW ISO 6400 for the X20 in the database, so you're comparing ISO 6400 vs ISO 3200.
Second, scaled to the same output size, the RX100 II is both quite a bit cleaner and has a lot more resolution left, despite the stop difference. As expected of course.
Greynerd: A valiant attempt by Sony to match the G1X but spoilt as usual in true Sony compact fashion by too many pixels when the light dims. The razor sharp image of the Queen's head if you select the G1X on the noise page, as the Sony descends into mush when you raise the ISO is pretty compelling. Especially as the ISO performance of the enormous sensor in the RX100 is supposed to offset the slow fully zoomed lens.
It's one of the most efficient sensors out there, regardless of the amount of pixels.
The G1X sensor is well over 2 times larger still, so you'd expect it to be over a stop cleaner... if the sensor were to be as efficient, which it is not. Add the faster lens of the RX100 II and you have a tool that should at least be similarly capable in low light within its range of FL's.
Marvol: From the introduction "The RX100 II has a list price of $750 - $100 more than that of the original RX100"
From the review "Sony promised better low light performance, and it has indeed been delivered. Is it $150 better?"
That, DPR, is a cheap shot and well below the belt. You of all people should know not to compare introduction prices with current market prices. You can make any camera look good or bad by comparing it to cherry-picked current prices of any other model.
That is even before considering the fact that the "$150" does not only go towards better low light quality, which the reviewer also conveniently ignores there.
This sounds like the reviewer was looking for something to justify the pre-concluded Silver Award. Very unprofessional.
@ tokugawa :But that's a $100 difference, not $150....
Niala2: Fujifilm X-M1 seems sooo much better in all comparaison shots with the RX100 II (RAW, JPEG, Low-light, Day-light, all ASA settings)...
So either I made a mistake, or it is truely so and not "tolerable" that this is not pointed out allready directly in the review.
Because I beleve the Fujifilm X series are not even full frame, and have (excellent) interchangable lenses...Where am I wrong ?
The X-M1 has a sensor that is over 3 times larger and even without a lens the Fuji is larger and heavier and by no means a pants pocket sized camera.
In other words, you're comparing completely different cameras. If you were to equip the Fuji with a lens covering a similar range, you'd also be shooting up to 2 stops higher ISO's in low light due to the slower lens at the wider and medium FL's.
rhlpetrus: The 5s is not well aligned, check corners. For my tastes the 5s is best in terms of natural reproduction, very good for a smartphone.
Or more likely the lens elements aren't well aligned, which is more of a QC problem at manufacturer level.
Photato: Sensors are just dumb areas where the photodiodes sit. So it has never been about sensor size but photodiode size.DSLRs capture better images not because the sensor is large, but because the photodiodes are large and contain many of them.I'd say from 3 to 6MP is a good number for small sensors considering in the poor conditions must shots are taken; low light, camera shake, bad focus, people moving, etcThe question could be rephrased as, how large the photodiodes should be?
@ AndreaV: First, I was obviously addressing the post I directly replied to from Photato,
Second, practise disagrees with all your theories. Simply because sensors with smaller pixels use technologies in practise (microlenses, blacklit technologies, stacking, etc.) to make up for efficiency losses due to wiring etc. Practise tells us that sensors using 1 micron pixels have a similar quantum effiency compared to sensors using 7 micron pixels, huge difference.
Third, the D7000 architecture at the pixel level is the same as the D800, irrespective of year of release. Supported by the fact that per given area, every measurement per given area gives equal results too.
Fourth, yes, processing can make a larger difference. But fact is that you don't have any practical examples to support your point, at the contrary.
So despite all your talk about studies, practise simply proves you wrong. And if you want to talk about studies, talk to mr. Fossum and see if he agrees. I know the answer.
gerard boulanger: So... considering the existing devices, going further than 3 M pixels makes no sense.I strongly believe manufacturers should stop the pixels race (we can't even display/share) and work on the IQ, especially in low light conditionsA slightly larger sensor and better optics on a 5-6 Mpixels could provide much better IQ, but... most "smart phone" consumers are still in the pixels race unfortunately.
HTC did that, it is comparable in low light to cameras with 3 times more pixels on the same surface. In good light, it lags behind.
By your logic, the D800 produces no better images than the D7000 in terms of noise etc, because what matters, the pixels, are of the same size and it's not about sensor size.
Disregarding the fact that despite that they are based on the same technology, the D800 is well over a stop cleaner at equal output size,....because the sensor is larger.
By your logic, the HTC just about hit the sweet spot in terms of pixel size for a small sensor, disregarding that blind tests done with hundreds of people prove it's no better in low light than say an S4 with much smaller pixels and much worse in medium to good light.
In practise there is no correlation worth mentioning between pixel size and noise for a given sensor area or output size.
Falko: The A77 autofocus and tracking are much faster than canon 70D. The better color and exposure accuracy of the Sony SLT-A77 is obvious too. At least for me, the Sony SLT model is a clear winner and more attractive proposition for video users.
Based on the videos posted, the A77 focuses faster when using video mode in low light too, it's the tracking in low light where it loses points. And I haven't seen any low light comparisons for focus speeds in still mode either. I do know that the A77 was slightly faster than the 7D in this regard.
The argument made by DXO that only part of the light hits the dedicated PDAF sensor in SLT's, is one sided too, because the PDAF pixels on main sensors only receive a smaller portion of the light as well. And even when using DSLR's in their regular stills mode (with OVF), the PDAF sensor only receives the light that passes the semi silvered center section of the main mirror, passed on to a smaller secondary mirror behind it and bounced to the AF sensors. So even there, we're talking about a smaller portion of the incoming light.
yabokkie: for video, Canon intentionally slowed down the AF speed for smoother, more "natural" focus.
I agree that the focus (no pun intended) on speed in general is misleading. Irregular hunting and hesitation to start (re)focusing, that's more interesting (read: more objectional for video).
The lagtest for live view focusing was the more interesting bit imho.
ianimal: Ain't the Sony SLT auto-focus limited to only f/3.5 aperture? And the Canon can in theory use any aperture? I just asking, don't blame me :)
Limited to the largest aperture but no larger than f/3.5. So with a f/5.6 lens it'll be ...just that.
Eleson: I kind of expected more from the shutter lag test. Better time rez than 1/10th, a bit more info on the set up. Does Sony use Electronic First Curtain for instance? In general, the setups would have been great to see for repeatable tests. AF mode etc.Also, I would like to see the result of the roller shutter test...
"Does Sony use Electronic First Curtain for instance?"
zigi_S: To all you sony fanboys out there. SLT isn't the same tech as on sensor pdaf. SLT still has a mirror! And it's nothing new. The SLT concept is old and only a desperate manufacturer like SONY would try to market it like inovation. It will be funny when mew canon M models will AF much better than the NEX system. Where will be the new tech from sony then? Sony's on sensor PDAF is inferior to CANON's. It loses light!
According to Imaging Resource tests, the A99 isn't faster, but I have no doubt that the extra PDAF on sensor that provides a depth map, will help a lot with tracking and accuracy.