Celsus: Hey DPR, if this is news, then why don't you have an article raving about the amazing 85 score for the DxO One? It is much bigger news that a 1" sensor can match a 50MP FF sensor! Or don't you believe that either? Then you should do a fact check job on it, right? Right?
That's the multi exposure score, which of course will score higher than single exposures for getting more signal. The single exposure score is comparable to the Sony and Canon cameras with the 1 inch BSI sensor.
Androole: I'm interested to see the 120fps footage, since it'll be much better quality.
The 960 fps is super impressive, but the quality is not really usable for larger than a phone screen (or Youtube videos, I guess). Maybe for scientific applications on a low budget, where the aesthetic quality isn't as important. Or heavily filtered, perhaps?
Is that the same 5% group? No.The point is, the tiny percentage that has a phone capable of 240fps, will hardly if even have any impact on those looking to buy a compact with this feature, since that group is part of the *whole* market.
So really, what you mentioned lacks a point.
<5% has a phone that supports 240 fps.
munro harrap: And as aforementioned in MANY complaints, SHUTTER LAG figures, accurate ones please.
Since all the pinpoint eye-focus (if not wearing sunglasses, spectacles or patterned clothing etc), accuracy in the whole world, even in a perfect system is of NO use outside of studio and your test conditions, if, when you then press the shutter, your £2,600 Sony body then takes a sixth of a second before it releases the shutter after you press it, as did the A7R.
You can already predict the outcome somewhat. The A7R couldn't manage an electronic front curtain yet, which explains the relatively long shutterlag (prefocused too) of about 0.16 seconds. Look at the A7, A7S or A7II (electronic first curtain) to get an idea when prefocused: between 0.02 and 0.05 seconds.
Utterlyotter: Magnificent camera, but not for me for the forseeable future even though it does give me a touch of GAS.I really really hope this sensortech will be implemented in the upcoming A6100 or A7000. (End of the year or 1st quarter of next is my guess)Teething issues - wich most new tech has - should hopefully be ironed out by then.
+ Quiet shutter, built in ND capability and Ibis would be nice, as long as it doesn´t make it significantly larger..
The "significantly worse" is based on a test of a single copy. DXOmark tests (they test multiple copies) disagree.
But I guess there's a big difference between >2x the price and "just">1.86 times the price....
And what's the point? Ever thought about size and weight? The Sony F4 zoom is significantly smaller and about half the weight of the Canon F2.8 lens. It's even smaller in volume and much lighter (the Canon F4 weighs 40% more) and optically no worse than the Canon F4.Not to mention the extra functionality with native lenses, mentioned in the article above.
The weakest link usually isn't the lens, it's the person trying to operate the camera. Unless you're planning to shoot at larger apertures, there's likely little difference in the output, ceteris paribus. Horses for courses.
Last night it was listed at $892 on Amazon, today for $909, not far off your ~$800. And it's similar in performance to the Canon F/4 which is still listed at $1k.In other words, overpriced is pretty relative here....
"Use Canon FF zooms in the meantime. They (most importantly, the Canon EF 24 - 70mm / 2,8L USM II) are _far_ better on the A7 series than any Sony normal zooms, incl. the expensive Zeiss 24-70mm/4."
Here he goes again, mentions a lens costing more than twice as much as the Sony 24-70, but calls the latter "expensive".
PerL: So would this system enable Sony cameras to shot a low light indoor sports game with a 200 2.0 lens wide open at 10 fps with a higher number of keepers than a Canon 1Dx or a Nikon D4s, or is it just nerd stuff?
"for those really needing AF-performance"
A rather vague and subjective term. I see AF performance in that specific scenario (medium to good light, not sports but still moving human subjects and camera, large apertures) that many other FF cameras struggle with (see 5DS and D810), right there. Improving keepers rate in such scenarios doesn't sound like "nerd stuff" to me at all.
Sounds like a regular use case for most ILC users indeed.
And then the alarm clock wakes you up to reality.
The reasoning that if it doesn't satisfy a niche within a niche of niche users, it must therefore me nerd stuff, escapes me.
Class A: Could you not have chosen an AF point manually in your first test? The vast amount of shots were unusable with the nose being in focus but not the eyes.
It would be extremely disappointing if this were everything the camera could achieve with third-party glass.
BTW, if a lens requires AF microadjustments with a regular PDAF system then it will need the same adjustment with a sensor-based PDAF system. The only problem the latter can avoid is a constant offset which a regular PDAF system can compensate with one global adjustment as well.
If the A7R II is immune against FF/BF issues then because it employs CDAF in the final stages of focus acquisition.
Thermal changes and a constantly moving mirror can make minute difference in distance between the optical path to the imaging sensor versus the path to the AF sensor. And those small differences in distance, can make quite a difference for the accuracy of AF measurement in practice.
naththo: Fairly soft image in RAW Im afraid. 1" sensor and over 20mp is not ideal. Noise is pretty good but I think for detail and 1" sensor keep it down to at least 10mp or so.
No, you still haven't offered a shred of support for:"So, the point is valid that it has too much MP than what is meaningful or useful."
It still adds detail and it has never hindered its noise performance (quantum efficiency has been classleading, read noise exceptionally low too). So there is no downside.
"This is also the reason why Nikon started the 1 series with a moderate MP count."
The fallacy in the first part of your post was that you change several factors and then attempt to draw a conclusion.
But the quote above makes it rather ironic, because the Nikon ended up with worse low light results and less usable resolution. Using your logic, we should conclude: more resolution means better low light results too.
Rob: I expected a lot more from the BSI sensor. If it's not going to provide at least a whole stop of better high ISO performance, then it doesn't deserve the hype that Sony's marketing is giving it.
Sony isn't hyping it at all as a sensor that gives a great high ISO performance boost. They're hyping it for read out speed (to improve video quality, add faster video framerates and provide faster focusing, to name some of the key elements).
And who says the LX7 has a BSI sensor? Haven't seen proof for this.
falconeyes: I don't know if the m4 sensor is visibly noisier than the m3 sensor. I have trouble to see it and prefer to wait for a lab testing.
However, IIRC what Sony said in a recent interview, Sony moved the ADCs from the sensor chip off to the stacked logic layer, using copper vias. This way, Sony can incorporate more ADCs and increase readout speed by a significant factor. Which is used by a much faster electronic shutter and video.
However, it made me wonder immediately if readout noise wouldn't increase marginally. I am especially looking forward to DxO testing on DR at base ISO to have a closer look at that. The DRAM chip though may have nothing to do with this.
If Sony managed to incorporate the ADC array onto a separate logic layer for their larger chips w/o any negative impact on image quality then it will allow for a whole new generation of image sensors. E.g., they could implement arbitrarily low ISO and arbitrarily high dynamic range in the digital domain of the stacked sensor.
Depends where you look. In the green channel there appears to be less smoothing (as in more grainy noise and more detail):http://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=daylight&attr13_0=sony_dscrx100m4&attr13_1=sony_dscrx100m3&attr13_2=panasonic_dmclx100&attr13_3=canon_g7x&attr15_0=jpeg&attr15_1=jpeg&attr15_2=jpeg&attr15_3=jpeg&attr16_0=6400&attr16_1=6400&attr16_2=6400&attr16_3=6400&normalization=print&widget=1&x=0.8473594976452121&y=0.55320938054751
Seems like they dialed down edge sharpening at higher ISO's a bit too, which obviously doesn't help these comparisons when focusing on detail. But it does reduce sharpening halos in contrasty areas, see the link in my next post:
Paul Farace: Another reason my telescope is collecting dust in the closet. I remember looking at the moon on the summer nights in 1969 and thinking that for now, it's a virgin moon, and if, God willing, our Astronauts were successful in a few weeks, the moon would forever be changed. I wish NASA could bottle that excitement from that long ago summer, and use it today.
"I saw a documentary film that clearly shows that the entire American moon landing was a colossal hoax."
You must have missed the end titles, which gave away that the documentary was a hoax, to show how gullible the hoax thinkers are, rather than those listening to scientists who worked on the projects or those that actually landed on the moon.
Your "dead giveaways" regarding the flag and shadows have been debunked as theories about a thousand times now.
Low light scene shows that there's a decent boost in shadow noise performance at high ISO again (hints at less read noise, which would make sense, considering the new design). The rest seems about equal, well within what could be sample variation. Still a nice bonus and this is also the area where the Exmor R sensor showed the largest gains over the original frontside illuminated sensor in the RX100.
In terms of quantum effieciency (which mainly translates to noise performance in midtones and highlights), there was little to be gained anyway as it was already at the top in the industry.
If you select the low light scene (different light temperature with less light in the blue spectrum), you'll notice quite an improvement in read noise (shadow noise, see decrease in blue/purple noise) at high ISO with the new sensor. That's where the main benefit seems to be in terms of noise performance, but that's pretty much a bonus, rather than the main design goal (speed) of this new sensor.
It's was also the main boost in noise performance of the backside illuminated (Exmor R) over the first generation front side illuminated version (found in the RX100).
Shadow noise at high ISO is now at or past the level of the 20MP Sony APS-C sensor. Which is quite a feat, considering the fact that the latter is 3 times larger in size.
@EskeRahn: Sony said in an interview that they opted for slightly less (and a different algorithm too) NR at higher ISO's, to preserve a little more detail in the new Jpeg engine. Which explains why we can see a little more grain in jpegs in the RX100IV.
nicolaiecostel: The 7D mk2 is an action camera. With 10 fps and class leading auto focus, the camera is forgiven for low DR and not that good high ISO performance.
In contrast, a camera like the D800E or D810 is a slow, high megapixel camera with great image quality made especially for applications where a great image with lots of detail is required.
What canon has done is it has made a D810 style camera with the upscaled 7Dmk2 sensor.
And this just doesn't make any sense.
I see some used A7S bodies go for more than what some have paid for them new (during discount periods). Meaning, if the A7RII does offer a compelling combination of different strengths from several previous A7 versions (great video, great resolution, great DR) and add a few unique tricks (faster AF and PDAF with many types of lenses, better viewfinder, in body 4K, IBIS) there's a good chance it will hold value as well as the A7S.
oselimg: Once again the self righteous gear lovers are in the business wishing Canon vanished from the face of the earth. First of all camera buying people aren't aware that you exist. Secondly if canon or Nikon were go out of business you would pay double the amount of what you pay your favourite camera. Thirdly EF lenses would double in value so you lose again. Be careful what you wish for.
If by "until recently" you mean more than 5 years ago. Yes current usable DR approaches that of some APS-C cameras released over 5 years ago (NEX 5 etc). And that's ignoring noise patterns in low ISO shadows, which would still favor the Exmor too.