[shoplink 26006 ebay]
[/shoplink]
Current A6000 with [shoplink 26006 ebay]Gariz case[/shoplink].
The last 48 hours I got two messages from a trusted and a new anonymous source:
- Trusted source: He confirmed the new camera will be named “A6100” and it will ship in March. He said latest he heard from his Sony source was that it uses a new 36 megapixel sensor. He said he will try to confirm the sensor resolution info soon.
- New source: Also that source said the new camera is named A6100 and it will feature a 36 megapixel BSI sensor.
To sum up:
SR5: Three trusted and one anonymous source now confirmed the A6xxx is shipping in March
SR3: Has new 36 megapixel sensor.
It’s unclear if the camera will be announced on Tuesday Jan 5 at the Sony press event (follow live her eon SAR at 5pm Pacific time). Keep in mind the Cp+ photo show event is taking place by end of February…
One more thing: A source told me the A6xxx replacement camera got delayed twice last year simply because there was no “serious” competition. Sony had it ready for launch but at the end decided to launch it in early 2016 only.
213 Responses
Sony introduces another class leading 36 Megapixel BSI sensor, the hits just keep on coming.
Non-overheating 4k please. I’ll gladly pay over $1000 for that.
A7RII doesn’t overheat anymore since firmware 3.0
Yeah, but the a6xxx would most likely come in a smaller body (though I wouldnt mind if it does turn out to be a7-sized), and would have a harder time dispersing heat.
Sony RX100 MK 4 is still limitied to 5 mins of 4k footage
Commlite is reporting some sort of problem with version 3.0 firmware…I do not yet know what the issue is or if it effects anything other than the Commlite adapter.
Admin is currently investigating the issue.
also need mic jack and rec level adjustment manualy 🙂
Yeah that too. Practically, I just want an APSC MILC version of the Rx10 ii.
I’ll be happy with phones out with the XLR adapter, but mic jack in would be practical for a lot of people
Is there a chance for high bitrate 4K video and 180 º tiltable screen?
High bitrate for 4k is quite relative when we’re talking about Sony cameras up to A7s/rII. Even their 100Mbps 4k is barely a minimum bitrate to work with 4k. Take an equivalently priced BlackMagic camera and the bitrate is about 50x bigger for 4k, which at this point is getting close to standard for true cinema. So there is still a long way before any Sony consumer camera can be called “high bitrate 4k camera”, even though it might still be sufficient for many usages.
Yeah, Sony video has a habit of breaking up in a way a proper intermediate codec at high bitrate doesn’t…
Plus users will have heating issues in a body so small. BM fix this with Peltier sensor coolers, but thats the difference between video as a primary function vs video as an add-on to stills; DSLRs simply limit the play time (or accept variable thermal noise).
Plus I’m pretty sure you still won’t be able to tell the difference between properly exposed and graded BM 1080p Prores/raw upscaled to 4k youtube vs same output from most DSLRs with native 4k in non intermediate. In-camera thermal noise and compression will kill it.
BM URSA Mini 4k is at $3000, Sony A7r/sII is at $3200, both offer 4k and are similarly priced so if you want to make a comparison, you should compare those two/three cameras. And the best 4k A7s/rII format (XAVC S 4K 100Mbps) is just slightly better than the least quality 4k format on the BM URSA Mini (which is used for proxy actually). So in pure bitrate comparison, the BM kills the best Sony ILCE. But then you are limite to ISO800 on the BM, no IBIS and other things that help catching-up on footage quality.
Yeah, I know; most DSLRs only output what a typical Resolve users would wince to use as stand-in footage 🙂
Im more looking at people who want 4k on an A6000, my point being that If they want decent quality video, well exposed 1080p prores HQ on even a Blackmagic pocket will beat the 4k XAVC-S a6100. With XAVC-S, half the time the extra pixels will be capturing nothing but macro blocking and compression artefacts!
unlikely. At least as far as the 180 degree tilting screen. Hard to do with an EVF with an eye cup. I doubt you’d get many votes to trade the EVF for a selfie setup. 😉
lol @SAR_Admin
It’s not a6000 replacement. The price will be significantly higher. Same situation as with a5000 and a5100.
What’s BSI?
BS Inside.
Back Side Illumination ….exmor r
More efective at gathering light ..higher iso…than normal sensors …exmor…especially for small pixels…as in smaller sensors and high resolution sensors.
The advantages for lrge sensors with bigger pixels are minimal, therefore older lower res sensors were not bsi, and a7sii is not bsi…but new higher res sensors will benefit greatly from bsi.
Its is not a new technology in that smaller sensors have used bsi for quite some time now.
Latest 100mp Sony Sensor
https://www.phaseone.com/en/Products/Camera-Systems/XF100MP.aspx?
full frame 645!!!!
YES, FINALLY!
Now I can wait about 20mp to be able to afford one of those.
Way too much resolution for an apsc-sensor 😀 In my opinion, the actuelly 24mp are good enough for these cameras. I would prefer better low-light-perfomance over a few pixels more in this case
See my response to duott.
They increase the resolution to keep the low light advantage and slight dynamic range advantage of FF sensors constant, I think. I’m not an expert, and I know that the popular view is that increase of resolution does not degrade overall image quality, but I still think if they made a 14/16mp APS-C sensor with current state of art technology, it would have better performance in all aspects (other than resolution) than the current 24mp sensor or the new 28/36/whatever mp sensor coming. I would like to see a 5200, too, with nex 5 shape, RX like RGBW tilting (downwards, too) capacitive touchscreen, flash shoe, top command dial and top fn button, ael bitton, electronic level and fast switch on time (below 1 sec), but I do not think they make this. The nex 5-5n-5r-5t-alpha 5100 “development” line represents what I hate in Sony’s product development and this is unlikely to change (first add a new sensor and an excellent capacitive touchscreen, then add wifi, top dial, fn button, and electronic level, but degrade the touchscreen and limit the functionality of the fn button, then add nfc only, then take away everything added from the time of the original nex 5, but the touchscreen and wifi, also take away the electronic flash shoe, but add new sensor and autofocus, so in 4 years we are back to square 1 albeit with better sensor and autofocus). Given that a sensible 5200 is unlikely I wait for the 6100. What I would like to see is addition of a sunlight legible touchscreen, electronic level and much faster operation (switch on time and wake up from sleep). A better sensor is always nice, but is not essential for me. IBIS would also be nice, but not at the expense of size/weight and only with minimum shutter speed settings and/or exposure calculation taking into account the stabilisation (as far as I remember neither of my nexes took into account the OSS when calculating exposure). Finally, for me weather sealing does not really count, but I am not against (I have used my nexes in various conditions, like desert, rain, snow and had no issues despite the lack of weather sealing, so I think even a plactics alpha would be capable enough for me). And yes, tri-navi, or put it simply, two unlabelled, programmable dials would be welcome instead of one dial and one mode dial (I guess many people use predominantly one mode, so the mode dial is rarely used), but that is again highly unlikely (I think they see it as a very high end feature, plus probably the mode dial is important marketing-wise).
I agree. I would rather see less noise.
Definitely agree.
Why. Just why.
They seriously think more Mp = better?
on sensor HDR, 18mp effective.
Lets see
3mp was enough..no. one needs more
6 mp was too much
8 mp was too much
10mp was too much
12mp was too much
24mp was too much
36mp is too much
42mp is too much
50 mp is too much
Backwards thinking like this is too much.
The new
Quanta sensor will have over a billion jots, read every photon, and readeat hundreds of thousands of times per second…mayby 10 years away.
For me that sounds just right !
And where are you going to store all that data? In buffer, card and hard drive?
I imagine this camera you’re talking about… with 0,001 FPS and a SATA interface for interchangeable external hard drives, one per every shot.
All of that will be especially useful if you never print, mostly use high ISO and only need ~3Mp for Web.
He’s talking 10 years ahead, so if you follow his logic, you also need to think 10 years ahead for storage technologies and web requirements. From there you can easily expect storages to have grown exponentially higher than from today (where dozens of TB storages are made common things) and where web standards have grown from low res files to much high than “retina” and HiDPI files, maybe even to vectorized photography.
Okay, but they already make those rediculous 42Mp bodies today.
When they have those amazing 12Mp sensors… paired with CDAF *facepalm*
That’s another topic. As of today, 42mp sensors have pros and cons versus 12mp sensors, which gives both sensors a place to fit in the market. The means to produce them are the “same” so we should expect the overall pro/con ratio to be the same.
If you compare a 10 year-old 12mp sensor to a modern 12mp sensor however, there is a difference. So in the end it’s not a matter of pixel-count race but time and technology advancements.
Right… but Sony pairs 42Mp with their top PDAF and puts 12Mp in what is positioned as mostly a videocamera.
Where is the most intuitive pairing for reportage, the 12Mp + top PDAF? (rhetorical question… Sony hasn’t been making sense for a long time now)
Something tells me you haven’t used a Sony A7rII.
I’m glad you asked that.
We are talking in excess of a terabyte of information per second..current processors simply cannot process that.
Quantum processors are currently being developed…the most advanced one I know of is 1000 bits.
These are not binary like current processors, instead the also indicate quantum state.
With a single + or – quantum state..and normal. N…..we now have a base 4 bit instead of a base 2
Binary is 0 or 1
Quantum is 0N or 1N or 1+ or 1-..N is normal quantum state.
Color data could be determined
ON…no photon
1- red photo
1+ blue photo
1N green photon
A pixel would be generated by a 3d grid of Jots
For example
100 jots wide x 100 jots tall x 100 readouts would generate a single pixel
Note iso is irrelevent, exposure is irrelevent, and dynamic range is practcally infinite.
A few years ago a Sony marketing pyramid said
Ultimate speed
Ultimate resolution
Ultimate sensitivity
Clearly this is the technology they are working on…. Plus they are known to be involved in Quanta tech.
As for the efectiviness of quantom processing we are talking base 4 compared to base 2….so each byte doubles the capacity compared to binary..
2x 4x 8x 16x 32x 64x etc
Cleary a quantum processor is magnitudes faster than binary, and its development will probably be linked to Quanta sensor development.
But they are producing 42Mp monstrosities today, and 36Mp APS-C tomorrow, not in 10 years.
I see, and how many pixels is your brand new car capturing? ;~) Let’s see, I know of several autos with six or more 13mp sensors running simultaneously. Monstrosity!?!?!
Quanta sensors..in about 10 years…will have over a billion jots on a standard ff sensor …and you are calling 42mp a monstrocity.
Other people will look back at your comments and laugh at how ridiculous they were.
I won’t, I will start laughing at you now.
16Mp bsi with 500 pdaf af and fullreadout 4k@60p.
This is song of 2025+.
Where? You’re asking someone who hand soldered his first computer with 256 bytes of memory? Who had the very first 5MB hard drive ever made (prototype) in his office?
Two things you don’t currently bet against in technology: the persistent increase in bandwidth, and the persistent increase in storage capacities. You probably shouldn’t bet against transistor size yet, either, which means you can’t bet against complexity of size of a current level processor.
Too many people think that “oh that’s enough” and think that technology will just stop. Nope. It keeps going, and it enables new things. So sometimes things morph. My current smartphone is so vastly superior to the CDC6600 that I started computing on it’s not even funny. My current smartphone barely takes up my pocket. That CDC was bigger than my house, literally (which is one reason why I started building my own computers).
I could understand the need to do that if we were getting, like… 3D photography, or some kind of Litro thing with changing focus. But regular photography, 42Mp sensors?! 36Mp APS-C? It’s just like, WHY. Even if operation speed and storage weren’t an issue.
I realize there are some people who need high resolution, like this 100Mp monster in the next rumor, but A6100 is supposed to be general purpose camera!
You could not understand that or anything else…you are too locked into the fact that you dont want better quality, and cannot understand that other people do.
Those who can settle for lesser quality are deserting the camera market in droves and using ever improving cellphones. Those of us that are left demand more from thier cameras…lots more.
Higher resolution does NOT equal better quality
Yes it does…!!!!
And like any quality improvemen, to fully benefit from it improvrmrnts across the board need to accompany it.
Any chain is only as strong as its weakest link.
I am on the same page here. There are more ways how to improve IQ: better lenses (today), different sensor tech, Foveon alike (future?). There are many people who are “crippling” their 16/20/24MPx photos to 2MPx or 8MPx to be in optimal resolution on their FullHD TV, UHDTD respectively. I am satisfied with current “sensible” native resolution (16MPx). There always should be more product linies like Sony did with (F)E-mount series. One for pro’s (they needed) and measurebators (they just want it) and one for “ordinary” users.
So you’re okay with 16 bit PCM truncated at 22Khz for your music and 1080P video, too?
With Apple now building 5K displays, what do you think your 12mp image is going to look like filling the display?
Even the “good enough” standards keep moving up over time.
I’ve always thought the opposite. 3mp sampling is not enough. 6mp sampling is not enough. Thus 50mp sampling isn’t enough, either. Fossum’s work on the Jot is intriguing and highly welcome and likely to be once again seminal. In theory, you record every photon.
But…now the sampling becomes frequency and angle based. I’ll want a tighter and more precise clock to know when the photon arrived, and I’ll want a tighter and more precise angle calculation to know where it arrived from. In other words, we’ll just move the sampling problem from one thing to another ;~).
Well, in terms of computing power, there’s no limit to how fast I want things to be. But in terms of resolution, there is a limit where my eyes and brain aren’t actually able to sense more information.
Just as with sound, where you don’t really need more than 16bit dynamic range and 48khz sampling frequency if your sound is properly bandlimited. Even a 16mp photo can fill most of my field of view without a noticeable lack of detail.
To me, low light sensitivity and dynamic range are the important frontiers of image quality today, not the number of pixels.
It is? Except for video if!!! there is no full sensor readout. The A7rII would best the A7sII if it would be able to have a full sensor readout… The A7sII can do it in FF, the A7rII only in APS-C mode… In FF mode, the A7rII skips pixel (hence no full readout) which explains the worse ISO. Apart from that, there is no reason to miss on more pixel. Although I think we are coming in a physical limit soon with APS-C. 36 MP on a APS-C sensor are that dense, you need incredible glass like the Zeiss Otus to have a big benefit. But then, why would I use an Otus on APS-C?
>It is? Except for video if!!!
Maybe if you’re only doing unhurried photography at ISO100 for large prints or trying to crop a lot… That’s a lot of ‘ifs’…
How about sensor-shift IS?
24mp does seem excessive for APSC. I’d much rather have a faster camera than those extra pixels at that size.
I was somewhat concerned about older rumors of 28mp which is not much of a step up from 24 mp.
If 36mp bsi is true I am a definite buyer.
Also hoping for burst rate of 12 fps and much better autofocus with adapted lenses and long lenses.
22% resolution increase. Just above the typically-cited threshold-of-visibility level. But I’d also be concerned about diffraction impacts. Depending upon how you calculate CoC, we’re diffraction limited at about f/4 on a modest sized print. My bigger problem is that all the E-mount lenses I own except two already show clear issues at 24mp. Those issues are just going to get more clearly resolved at 36mp.
Yeah – Sony has the Sensors (APS-C) and Fuji the lenses. A combination of both would be great (but I guess Fuji is at least going to 24MP this year)
Thom,
would this resolution not give similar results to mFT lenses ? As it would have
a similar pixel density looking at the various resolution tests on lenstip ,
diffraction { which obviously affects all apertures } really does not become an
issue till F/11. The weakness of several native e-mount lenses is a different matter lol
Yes, it would if the lenses were good enough.
Thing is, the smaller the capture area, the better you need to design lenses. Olympus–and to some degree Panasonic–is designing well above what they currently need, and it shows. Sony in APS is designing below what they currently need, and that, too, shows.
so there aren’t any APS-C lenses around, that can match the µFTs? 😀
A couple of the primes come close. But in my experience, the Sony E-type lenses are designed for consumer use, not highest possible optical capability.
yes, that’s my experience too.
But I wasn’t thinking about E-Mount exclusively, there are APS-C lenses from Canon, Nikon, Fuji, Samsung …
is there a general understanding, that µFT lenses are superior to those?
All of this ‘resolution’ is only availible at basic ISOs. It’s mostly marketing, but does actually make RAW files larger, slowing down the workflow and taking up more space.
The files are bigger…so what…deal with it…or get an older obsolete camera while the rest of the world moves on.
How much of the “rest of the world” actually needs higher resolution? At the very least they must also produce new bodies with the feature of low resolution and all the benefits that come with it.
But then again, Sony hasn’t been making any sense for a long time now… concentrating on niche products like A7x. So I wouldn’t be surprised by a 36Mp A6100.
This is bsi so iso and dynamic range will be improved…as is usuall the case when newer higher resolution sensors are introduced.
Compare the A7ii to A7Rii for example.
But how much better would those cameras be if they combined all those new enchancements AND the feature of low resolution? I hope they make at least one body like that, although Sony seems to be head over heels in the marketing Mp race
Low res is not a feature but an unfurtanate limitation that stands in the way of and handicaps those who want biger, better, sharper, crisper, more usable , functional pictures.
So I understand why you like low res.
How is a 3Mp image downsized from a 42Mp image better, sharper etc. than a 3Mp image captured on a 3Mp sensor of the same generation and physical size?
If you want 3mp use your celphone you don’t need a camera
Right… because ISO, AF, DR, lenses, flashes and everything on a smartphone is the same. Just the resolution is different.
You have the state of the Sony a7sii that should be everything you want in a camera.
So why do you need to complain that Sony is also making cameras for people who need more than you.
Because they coupled their best sensor with CDAF in what is essentially a video camera, and put their top AF in a 42Mp monster. Maybe there’s also an A9 coming… but nobody knows for sure. So far the only current Sony camera designed for reportage is A77II, from the discontinued A-Mount line.
Stop whining, Duott.
there is image blur even when you shoot 3 mp display with 3 mp sensor, blur amounts to half pixel size. so you get sharper edges
also with more pixels you get rid of demosaic artifacts.
see also my comment about pairwise comparison above
http://www.dxomark.com/Cameras/Compare/Side-by-side/Sony-A7R-II-versus-Nikon-D7200-versus-Nikon-D800___1035_1020_792
one of them is BSI – which has the most dynamic range? 😉
I think for larger sensors that BS1 is more to do with enhancing speed of sensor than image quality the differnces are at best marginal. On the upside from your link Sony now own Toshiba the makers of the D7200 sensor
Pushing the resolution 1.5 times is gonna take a toll at image quality at higher iso. A7r jumped from 36 to 42MP and still brought in iso improvements, but I have doubts about achieving same results with this resolution bump up. 28 MP would have been better with higher dynamic range and iso performance, or even better would have been to leave it at 24MP and bring even better performance with larger photosites
Some users need all thee resolution and image quality possible…and will buy the latest greatest equipment to achieve that.
Some do not currently need the highest quality, and if they make newer greater cameras for me there will be lesser obsolete cameras available at lower prices for you that will be perfectly adequate for your needs.
So you also benifit from the constant striving for better.
Based on these rumors this camera will be a welcome addition to my system.
Back in 2003 I wrote that 24mp APS/DX was were I calculated the gains of additional pixels to begin producing highly declining increases in ability. Up to that point we’d all see pretty clear benefits to increasing pixel count.
I’d love to be proven wrong, but diffraction, among other things starts to take a real toll above 24mp APS/DX. I suppose we could say that “flexible cropping” will be one of the reasons cited as a benefit for increasing the pixel count. But if you do a 1.5x crop on APS/DX you’re basically at m4/3. I’d rather have the m4/3 cameras with 20mp sensors and the very high caliber lenses we find there than the Sony at 36mp and the mostly poor lenses we’ve gotten for E so far.
Don’t get me wrong, I still will take any pixel count increase given to me. More sampling is better than less, all else equal. But I think we’re hitting the same problem the computer manufacturers did with CPU clock speed. At some point, the perceived benefit difference by users just isn’t high enough to justify upgrading. My suspicion is that other things about this sensor will be more welcome than the pixel count increase (e.g. 4K video).
Diffraction is more of an issue at small appertures .
At f 6.3 there should not be noticeable diffraction
At f8 some but minimal
At f 11 noticeable but tollerable
Smaller appertures will be more problematic.
Higher resolution will always give a better image than lower res, even when the optics are not as good as the sensor, and even with diffraction.
Sony also has firmware that reduces the effects of diffraction.
Now that higher resolution sensors are becoming a reality I would hope some smart engineer can come up with a way to reduce or eliiminate diffraction…I would imagine that by allowing normal photons in and filtering out diffracted photons diffraction could be eliminated…if someone can figure out how to do that.
Possibly by some sort of special modication to the edges of the diaphragm to transmit complete photons but block partial photons….the treatment would extend one wavelength from the edges of the diaphragm blades.
Yes, diffraction impacts increase with smaller and smaller apertures. But as anyone who’s ever done real testing of lenses with MTF charts knows, once diffraction limit is hit, you’ll have a clear chip off of contrast numbers produced by the lens. In a time when people are praising lenses like the Otus, diffraction can just take whatever that benefit they saw right back out if you’re not careful.
“Reduces effects of diffraction.” Well, if you mean that they oversharpen edges, sure. To my knowledge, no current camera has a deconvolution engine in it that can calculate and remove diffraction impacts. It takes too much computer to do that.
Photons follow quantum rules. A photon is both a particle and and wave. I suppose that someday we’ll have physics that better understands quantum impacts and can deal with side issues like diffraction, but I suspect that we’re going to go a different way before that, where we simply record where a photon came from and when. Work on “lens-less” systems seems more promising at the moment than anything else.
Are you guys still talking about cameras 🙂
Yes. Cameras record data. We’re talking about getting optimal data. Just as Ansel Adams did when he talked about chemical formulas and temperatures.
Is that fun because Ansel Adams talked about it? I prefer Ernst Haas to Ansel Adams.
I sm not talking about a firmware fix like sharpening or what Sony now does
I am talking about actually eliminating the problem of diffraction at the source, in tje lens at the diaphragm by blocking or not blocking photons completly.
The diffraction is not caused by a photon that is not blocked, nor is it caused by a photon tha is blocked.
It is caused by a photon that is partly blocked by sricking the edge of the diaphragm, therefore distorting its path.
If those partly blocked photns can be fully blocke or not blockrd at all there would be no diffraction.
I do not know how this can be done, but I am absolutely certain it can be done if someone figures out how.
Then our optics will be able to fully utilize the advances commiing in sensor technology..such as quanta.
I would suggest that this be the highest priority for research in new lens desighn.
Well, if you’re a fan of Manhattan (tv show on the Manhattan Project), you probably have heard about shaped charges creating a lens like effect to cancel shockwaves. Shockwaves could be considered a bit like diffraction in one sense, especially when they begin interacting. If I weren’t so congested at the moment, maybe my brain could imagine a way to modify diffraction via some complex nano-level process. But at the moment, it can’t.
But the problem is a little more complex than you suggest: if you BLOCK a photon, you’re blocking some light from getting to the sensor and being recorded. So more diffraction produces less light. Less light produces a lower signal-to-noise ratio ;~). There’s no free lunch.
Further, there’s the issue that we’re talking about a quantum effect here. We still have Heisenberg’s principle to deal with, among others. I’ve recently been reading about quantum physics, and it’s fascinating, but the math and theories are starting to get out of my range to handle.
Personally, I still think that a different approach to capturing light is going to be the solution (e.g. lens-less systems, or at least lens-less in the way we think of lenses today).
2 kinds of photons pass through a diaphragm.
Good photons pass unobstructed in thier native form.
Bad photons pass too close to the diaphragm edges and are distorted…refracted.
Ther has to be some way to discriminate between them and eliminate the bad photons.
A lot of very smart folk have tackled that problem, folk that understand quantum effects far better than I do. It’s not nearly as simple a problem as it sounds.
I did not say it was simple.
I did say it can be done.
Sorry, but I think you should reread an optics or physics book, especially the Huygens principle applied to light. What do you mean by partly blocked photon?
I mean that the “bad” photon has passed so close to the edge of the diaphragm that its direction of travel has been altered.
These bad photons should somehow be differrent from the ” good ” photons that have not passed close enough to the edge to have thier direction altered.
By discriminating between these two types of photons it should be possible to block, eliminate , cancel or ignore the bad photons , thereby eliminating the effects of diffraction.
I have not changed any optical principle…the diffraction effect still exists and is happening, but I am talking about eliminating the effects by eliminating just the bad photons.
Again..I do not know how to do this but believe it can be done.
Someone just has to figure out how..and it would bring the quality of optics to a new level.
Lenses would no longer be diffraction limited.
You don’t get diffraction at f/4, it’s not a linear thing that starts to happen and appear at any f/number. Neither is is a ratio based on pixel count and f number, what most calculators do are wrong. Just to prove that, a 100mm lens at f/4 has an aperture of 2.5cm (an diameter at which light goes through without being spread), whereas a 10mm lens at f/4 gets a 2.5mm wide aperture (which starts to be a bit tighter for light to pass through flawlessly).
So getting this hypothetical 36mp aps-c sensor doesn’t mean you won’t be able to shoot above f/4 without getting diffraction. You can still go much higher than f4 according to your lens and get no diffraction on your sensor. Which is pretty logic if you just think about the fact that a lens that starts to diffract light will do it at a certain f-stop will do it whatever is the sensor you put behind.
But the point you’re making right is that the higher the pixel count, the sooner the diffraction will become an issue, ***when it starts to occur*** however.
Actually, you get diffraction at all apertures. A common term that evolved was “diffraction limited,” which has a number of different definitions. The definition I tend to use is that, in a Bayer sensor, the Airy disc is fully recorded in the data in a way that it will clearly be demosaiced to visibility, and thus the resulting impage with reduce recorded contrast. Back in the early days of DSLRs, diffraction impacts weren’t well recorded: the photosite size was far larger than the Airy disc. The demosaic tended to hide any edge impacts of diffraction. When I made my calculations in 2003, diffraction impact was one of the variables, but so were a number of others. As I note in my OP, I’ll take more sampling, I just don’t expect as much out of it as I did back when we going 3mp->6mp->12mp, for example.
There’s nothing particularly wrong with shooting at a diffraction limited aperture if you need the DOF. But edge acuity and overall contrast will not be as high in a diffraction limited scene than it is with a non-diffraction limited aperture with the same lens/sensor combo. So, some of the “plus” of the added pixels is taken out by the “minus” of diffraction.
While it is still somewhat controversial, try the advanced calculator at the Cambridge Color site. It actually has proven to be very close to what I see in actual print testing. Note that there a lot of options you have to select–part of the controversy. DOF involves picking a theory. Human eye limit of perception is a (mostly) agreed upon theory. Diffraction is a fact. Putting print size and viewing distance into the calculator is a critical variable. Putting facts and theories and variables together in a calculator is bound to cause discussion ;~).
What is this pixel race all about?! 16MP gives a generous print @300DPI for A3. Now tell me who is printing larger on a regular base? 5% of us? If more resolution is said to be be better, why not talking about better colors, better noise, better dynamic range? If those high res sensors would be built for special purposes, then OK, but we do not get low res sensors anymore that improve on other tech aspects as mentioned. And as a sidenote, if only Sony would improve their cameras as a whole, say better intuitive handling, then this would be something. I feel like Sony would use its customers to test their sensors before they sell these sensors to the real camera makers.
You might be right about that one. With multiple systems, I fail to see a significant difference in what I shoot. I recently downloaded a ton of RX1R MKII RAW files and guess what- I could care less. I seriously was not all that impressed. What took me back was the excellent color and overal dynamic range but then again, I can get that elsewhere for a fraction of the price. If Roger over at lensrentals- who has literally played with every camera there is- can choose an M43 system as a main set-up or an “inferior” 6D over a D810 or otherwise- well, that tells me something.
Exactly. Those who know the business tell me about excellent pixel quality from m43, that the lenses matter more, paper characteristics for printing, know how of post processing etc. These specs numbers are marketing driven meant to feed GAS. The camera producers avoid to give us the ‘perfect camera’ or they could not sell new stuff every now and then. I saw a new professional print 80x120cm (31x47inch) made from a pic from the Sony A900 / ZA 16-35mm that was just perfect, even from close. That tells me that sensor improvement over the last 7 years is almost irrelevant for real photography. We just need to learn how to use our stuff. The differences we see in reviews are real but on laboratory level barely showing in reality or can be solved with a clever work around, not worth the money.
Not just megapixels – but lenses too. We just don’t need more resolution or more sharpness because our output media – paper prints and web images – can’t make any use of it. We’re just in an arms race, throwing money at technology.
Better low-light performance still makes a difference and that’s where I’d like to see the R&D money spent.
Since Sony Devices (and soon Sony Semiconductor Solutions) and Sony Imaging are not the same, I have always thought of Sony cameras as display for the sensors. The camera around the sensor doesn’t seem too important 😉
Where’s A7000?
More slowly cam and higher noise for nothing… since i got FF APS-C res seems so bad.
More MP equals less noise!
Lenses are the issue. Sony needs to get out a better line up of APSC only lenses because frankly- Fuji and even M43 are kicking your assss right now.
Oh, that’s right….you want me to buy FE lenses for my E mount camera. Read my lips….
HOMIE DON’T PLAY THAT!
Sony is taking the same route at CaNikon did with APS-C DSLR lenses. They will provide the standard zooms and will rely on the FF lenses for most of the fast primes. I would not expect this to change.
Yeah cuz that massive 35 1.4 FE makes even more sense on an APS-C body.
I belive that Sony have a 35mm F/1.8 OSS for APS e mount for those wanting a smaller , lighter cheaper option.
Commlite: X-mount to E-mount adapter plz!
impossible, X-mount flange is shorter and narrower
If its really 36mp then UHS-II support pls
QXD+SD
*XQD
Sounds like a good news if the rumour turns out to be true. I don’t need to carry telephoto lenses on a trip anymore.
I’ll believe when I see it, so far for the past year or so the A6000 successor is just a fairy tale.
BSI and no IBIS would have been disappointing on the A7RII, then i wonder why not adding both on A6000 successor? A 1500$ high end APS-C E mount camera with both features would be a good evolution of A6000
No one expected BSI on the A7rII nor any full-frame camera dude. So it wouldn’t have been a disappointment, it was rather a good surprise, so stop the wonders and just wait.
No, but he does have a point that it would have been a bit dissapointing if the A7II had IBIS and the A7RII did not…
I said BSI, not IBIS.
I’m more interested in 4k and higher fps than 36mp.
BSI, IBIS, Stacked sensor and 4k, that will put that camera above the A7II.
If true, this 36mp APS-C sensor creates a problem. While Sony/Zeiss has released some good FE lenses, it has not done much with APS-C E-mount lenses.
Sony better develop a much better kit lens with this 36mp sensor as the current kit lens for the 24mp cameras leaves much to be desired in terms of its optical performance when compared to the dSLR kit lens. Using the same kit lens for 24mp cameras on the new 36mp camera will make a bad situation even worse.
Many of Sony’s APS-C E-mount zoom lenses will likely also need replacement if these are to perform decently with this 36mp sensor. The better Sony E-mount primes (24mm f/1.8, 35mm f/1.8, 50mm f/1.8) will likely still perform well.
I absolutely agree. It is no smart move from Sony to bring such High MP camera to their APSC E-Mount system. The Zeiss and Sigma lenses might deliver but Sonys lenses are not up to the task. Maybe the 50mm F1.8 but even the 35mm F1.8 is SOFT until F2.8 on a A6000 already.
Sony would have to bring a lot of R&D back to the APSC department. The former NEX system is lacking big time compared to Fuji and is even behind compared to Samsung. There are no F2.8 zooms. There are no F1.4 primes. I can’t imagine to use the gigantic Zeiss FE 35mm F1.4 on a lightweight camera as the A6000 compared to the tiny Fuji 35mm F1.4. There are a lot of important focal length not covered by primes yet.
Sonys APSC offer is still seriously lacking and even with such a new High MP offer I would skip Sonys Axxxx line and go with Fuji as long there are no serious new lens offerings.
Definitely. And more primes. Same problem as with FE ‘system’ only even more limited.
I am pretty sure the only native APS-C E-mount lenses I own that would have a chance of resolving even close to the 36MP are the Sony 35, Sony 50, and Rokinon 12, and those would have to be stopped down to at least f/4. The Sigma 19 and Zeiss 16-70 can barely deliver the goods on my A6000 now, and forget about the 18-55 and 55-210 zooms.
hohoho
actually will be 82MP… come on ! for such a base/entry model more then 20-24MP is total overkill and will lower it’s own sells *for sure*.
i want lenses (f4 zooms mostly), not megapixles !
A6000/6100 is the high end. There’s also the A3000/3500 and A5100, and the A3000 was as base as you could ever want. A3500 is looking like a mild refresh.
The A77 is high-end, not the A6000. You see it in the price point. All high-end APS-C cameras are double the price of the A6000. Sure it’s the highest Sony MILC camera now, but there’s a lot of room to add more value (just take weather sealing, full metal construction and IBIS from the A77 line).
the f4 zooms already exist…
not the ones i want:
FE 24-105/4 OSS
FE 100-300/4 OSS
BSI is nice… But I’d rather have is used to generate such high quality 24mp files rather than increase resolution to 36mp.
I’m wondering whether it will get IBIS and the a7rii type autofocus. Would love to see phase detect AF support for adapted lenses and continuous eye AF.
PDAF for adapted lenses is a given, specially if they manage to make a stacked sensor.
36 mp downsampled to 24 mp will be better than native 24 mp
Even if true… Before you can down sample– you will have 36mp files clogging up your buffer faster. You will have slower transfer of files to your computer. More lag in post processing . None of this is worth the extra resolution for me. I know several people who switched from the d810 the the d750, just because the larger files were clogging up their workflow.
“One more thing: A source told me the A6xxx replacement camera got delayed twice last year simply because there was no “serious” competition.”
No. This is simply not believable. This is purely propaganda and your source ate into it. If the A6000 had retained a high price, I might have believed it, but it didn’t.
The A6000 was being overtaken on a technical level by the Samsung NX1 sensor. It even retreated to incredibly low price points which is why it became such an incredible value for the money.
There WAS a strong competition, but Sony used the pricing on the A6000 to give it an edge to gain more marketshare and stop any momentum for the other cameras. But lowering the price on the A6000 could have been done in conjunction the release of a successor camera without issues, if that successor camera was worthy of the price tier above the A6000.
I’m suspecting that Sony simply got overwhelmed by Samsung’s sensor, and had to remake a new sensor. I’m guessing bringing in a “high end” APS-C sensor to the market that is inferior to the competitors would be bad for their company identity, so they had to redo it.
Despite all the advantages of the Samsung 28 mpx BSI sensor over the 24 mpx CMOS from Sony, I think that Sony just don’t give a shit about the moves of Samsung in the mirrorless department.
Between a NX1 which looked to much like a reflex to be a real competitor of the A6000 and a very expansive NX500 without EVF, the conceptual mistakes of Samsung cameras make imho your point irrelevant. A very slight advantage in one department (sensor) doesn’t make a modest challenger a brand that you are really scared of when you’re the leader…
If competition there was, it was more Fuji XE2/XT1/XT10… But if you’re wrong about the competitor you’re right about the price of the A6000. It’s its incredible low price which make it a no brainer. At 800-900 it would be more compared to the Fujis.
In the end, competition is behind most corporate product decisions. It’s pretty clear that the a6000 price has been dropping.
But all their MILC cameras dropped by the same margin of 25%. And name some competition for the A7-line!
Indeed, the competition comes from the lower A5100 and the upper the A7 and any entry-level DSLR. Fuji is the only significant competitor in the mirrorless APS-C game, larger though but with a matured system and many choices.
I think the A7 cameras will adjust their prices because they compete with Nikon on a technical level. For most people out there, an A7ii would be an excellent buy in comparison to D750 for example.
But the A7 line at least had the excuse that they would be replaced by successors.
It sure is believable. Look at Amazon, one of the largest retailers in the world.
Look at their top-selling mirrorless camera for months – it was the A6000. It is only recently that the A6000 is no longer marked as being the best-selling camera in category. (Edit: Choose the 16-50mm kit and it is still the #1 best seller in the category…) In fact the top 9 bundles are all A6000 kits – EOS M3 bare body comes in at #10. http://www.amazon.com/gp/bestsellers/electronics/7242008011/
In the Mirrorless Cameras category, NX3000 comes in at #6. Despite its price, the A7RII is top-10, NX1 isn’t even close to the top 10 – http://www.amazon.com/Best-Sellers-Electronics-Mirrorless-Cameras/zgbs/electronics/3109924011/
Argue the technical merits of the Samsung all you want, the market says that it was not “serious competition” – the A6000 outsold it by a huge margin.
Do you think Sony enjoyed lowering the price on the A6000?
No they hated it, every company hates being forced to lower their prices by the competition. One thing is the A6000 succeeded, but you can bet that Sony would have wanted it to go for higher prices point. It simply didn’t, because the competition was overwhelming.
They only lowered prices very recently, on a unit they’ve had TWO ENTIRE YEARS of brisk sales (it was a top seller even before the price drop and has been for a long time) to amortize NRE across.
Once you take into account NRE amortization and production yield improvements/cost reductions (you can be sure they’ve been making minor tweaks to get prices down) and reduced component costs (paying off the sensor production line’s capital costs), I wouldn’t be surprised if they’re still making a healthy amount of profit per unit even at the reduced price.
The A6000 had a few sales campaigns during last year’s black Friday as well.
Nobody is doubting the camera is very profitable for Sony. But the price war happened because there was strong competition. Sony simply won the price war, but that’s a different story.
I find this “serious competition” statement unbelievable as well. But if you limit the competition to mirrorless cameras with APS-C sensor only and a price tag under $600 – then it is true.
Samsung NX1 is better than A6000 except its larger body size, NX500 beats A5100 and Fuji XT1 / 10 is another serious competition.
Sony – how about producing some lenses for E – APS-C mount – Sony A6000 has been one of their top sellers because it was priced around $800 with kit lens and offered a nice sensor – EFV and good AF and was a nice improvement over the NEX6 – but now with supposedly 4K and 36MP sensor etc I cannot see the price stay around $800 – I am guessing $1299 – as they say there was no real competition for it in 2015 – but seriously Sony – it is always the same – tons of bodies and few lenses and most lenses are all expensive Zeiss branded lenses with few G branded also expensive – no rumors about any new E APS-C mount lenses – I assume they are expecting folks to buy Zeiss FE mount lenses from now on – for top top dollars –
Absolutely not buying any Zeiss lenses. Ever. Just waaay overpriced bling IMHO.
Nobody force you to get Sony camera. Simply get other brand that is more affordable to you.
I do have other brands – Fuji X100 X100s XT1 – Olympus EM1 Em5 – Lumix GX7 GX8 GM1 Gm5 GX1 LX100 and Nikon APS-C & FF DSLR’s – Canon G3X G5X G7X –
but I do have the Sony NEX5n NEX7 A6000 A7 RX100 RX100m4 –
Why people can’t understand that while new APS-C lenses would be nice the FE line up is top priority for now?
New APS-C only would mean newer designs to improve what they already have like their kit lens. While on FE there’s a lot of big holes on focal length range.
I think many if not most of us understand that perfectly. We’re just telling Sony what we want, and hoping to influence their direction. They need to hear from APS-C users and quit smoking so much of their own FF hype.
Surely some FE lenses are just as good a choice for APS for different reasons. I have the excellent FE 55mm F/1.8 it would make a superb 82.5mm portrait lens on APS. There are also the Zeiss touit lenses. I am pretty certain that their FE cameras will have far better profit margins than their APS models hence their interest in FF.
I hope this is not true. At 24 Mpix most of the lens line-up already struggle and the noise pattern is not very pleasing. Improve the AF instead so it can live to its hype. Import genious things like continuous Eye AF from the A7 series. Improve high ISO performance, even if it means downgrading to 16-20Mpix. In-camera level gauges, etc….
Simply not true. I tried all my APS-C lenses on A7RII and they are as sharp as the sharpest FF lenses.
Only SEL 16 2.8 was as abysmal as it was on NEX7.
Sigma 60 2.8 is sharper and with better colors than Sony FE 55 1.8.
Dude, APS-C 24MP sensor has higer pixel density on the sensor than FF 42MP.
Why in the world you think APS-C lenses work worst on FF sensors?!
Feels too small in my hands. Much prefer the larger size cameras.
Completely personal. I feel that the bigger cameras are too heavy, and the A6000 is perfect. I like shooting it more than my D750 despite the gap in performance. Hell, my D750 is barely heavier feeling than my friend’s A7R II. Height and width aren’t too far off, either.
Lol
Is the sensor stabilised? That’s what I’ve been waiting for, not an increase in resolution and a likely increase in noise…
A7RII has 42MP, A7II has 24. A7RII still has better ISO performance than the A7II!
But yes, stabilisation would be nice
36mp feel today like 24mp when Nex7 was released
Please refresh my memory
What were the terms of the ” I told you so bet””
If this 36mp bsi rumor is true who wins?
it’s a draw
It will be good to see what the final sensor is gonna be, as just like last time this will make way into A77 mark3
It’s interesting to read comments on this topic, and on the 100 mp ‘medium format’ from Phase One. I’m surprised at how many photographers are saying they’re simply not that interested in more pixels, and would rather see improvements in noise and color depth. And I agree. At the end of the process, we still have paper prints and LCD screens as our only output media. We’re past the point of diminishing returns for resolution, and all I’m seeing for myself would be bigger files and slower processing..
For the people complaining about the high MP, don’t forget that Sony have a good reason to increase the MP count even higher.
http://www.sonyalpharumors.com/sr3-sony-a7000-is-the-first-e-mount-camera-with-15-5-stops-on-sensor-hdr/
personally not excited for 36MP as I only print 20″x30″ max. IBIS, touch screen, weather sealed body, digital level gauge are more appealing to me.
That requires 54 mp…!!!
that really depends on your viewing distance.
Detailed comparison of the Sony A6000 and Nikon D750… some surprises..
http://www.disboards.com/threads/sony-a6000-vs-nikon-d750.3471961/
Why did you say the Sony 50mm/1.8 is an effictice 75mm/1.8 on FF?
It is a 75mm/2.7 with 2x higher ISO setting – and of coarse this explains the different DoF.
No… It doesn’t double the ISO. In terms of light gathering, it’s still 1.8, regardless of crop factor. In terms of depth of field equivalence, you are correct.
It would require double ISO, if you would use a 75mm/2.7 lens on a FF camera, to get the same result as with a 50/1.8 on a APS-C camera.
With same ISO, the FF picture would be darker, or you’d have to double the exposure time for same brightness.
That’s false. For light gathering purposes, the aperture doesn’t change — it’s still 1.8. As actually shown by my pictures– same settings generated same exposure, regardless of whether full frame or aps-c.
Try this out:
use your D750 with 85mm @ f2.8 (that is about the 75mm/f2.7 equivalence).
use your A6000 with 50mm @ f1.8.
Use on both cams the same shutter speed and the same ISO settings and compare. Then double the ISO setting on your D750 and repeat.
Tell me what you figured out. Tell me about noise and brigthness of the pictures. (We both know, the DoF is similar)
Done. Sorry my friend. You are incorrect. You don’t understand the exposure triangle. Where both cameras are set at the same ISO, same aperture, and same shutter speed, you get the same exposure. Where you set the a6000 at 1.8 and the d750 at 2.8, the a6000 results in an image that is one stop brighter.
In terms of crop factor, it affects aperture related depth if field. It does not change the exposure triangle.
Where you set the a6000 at 1.8 and the d750 at 2.8, the a6000 results in an image that is one stop brighter.
Of coarse this is true – and I think you agree that you get the same brightness on d750, if you double the ISO setting. And this is exactly what I said.
That goes to what I said from the very start. In terms of exposure, 50mm/1.8 on aps-c is equivalent to 75/1.8 on full frame. So now you’re withdrawing your disagreement and agreeing with my original statement?
You are translating 50mm to 75mm in order to have the same viewing angle. You are are using f1.8 on both cameras to have same exposure (and same ISO of coarse).
But you get a different DoF. And that is why 50mm 1.8 is not equivalent to 75mm 1.8.
Change to 75mm 2.7 and you have equivalent DoF, but you have different exposure.
Double the ISO setting, and you have the complete equivalent:
equivalent for 1.5x crop-factors means:
focal-length * 1.5
aperture * 1.5
ISO *1.5 * 1.5
If you do not all 3 changes together, it is not equivalent.
We are basically saying the same thing. I’m translating it as 75/1.8 with same ISO but different depth of field. You are translating it to 2.7 with increased ISO. Though the 1.5 is not accurate. If you go from 1.8 to 2.7, that’s about a full stop in ISO, so you need to double ISO.
But with different depth of field, it is not equivalent.
1.5 * 1.5 = 2.25, so it is about a full stop of ISO – that’s why I said, you have to double the ISO value in order to get the equivalent. Tony Northrup made a good video about that topic, but he used m43 and FF. Having a 2x crop-factor makes the math easier and he neglected the 2.25 inaccuracy problem very elegantly.
The funny thing about having the increased ISO setting to get the complete equivalent is, that the noise-level is also the same. Remeber: with the same sensor-tech and quality, a double sized sensor has 1 stop less ISO-noise.
So this 3-setting formula gives you an equivalent in terms of Field of View, Depth of Field, Brightness of Picture and ISO-noise level. And the exposure time is the same, too.
But you’re assuming one would want the same depth of field. There are many many situations where I want the widest aperture possible, especially if the lens is capable of it. So if I’m shooting with ff, I’ll use 85mm at 1.8. If shooting with aps-c, I’ll use the 50 at 1.8. I’m not stopping down and increasing the ISO.
I am assuming nothing – I just explained what the equivalent between FF and APS-C means.
Of coarse you don’t have any need to limit the capability of your D750 just in order to have only the A6000 50mm f1.8 equivalent.
In your blog you wrote in the 1st paragraph:
“Picking it up used, with the Sony 50/1.8. So also an effective focal length of 75/1.8.”
If you limit your statement only to the focal length of 75mm, it is true. Adding the f1.8 to 75mm turns the statement kinda wrong. The focal length statement is right – but why adding f1.8 to the statement? F1.8 does not change the focal length – so it is true what you write. But f1.8 tells also, that it equals a 75mm/f1.8 lens for FF. And this is the wrong part of your statement.
If you just write:
“Picking it up used, with the Sony 50/1.8. So also an effective focal length of 75.”
there is nothing to complain about. You would have made no statement about the equivalent aperture at all.
BTW: Did I really read it right – you are talking about stopping down your APS-C camera and increase the ISO. And that is what you don’t want to do?
My writing was confusing… I wouldn’t normally stop down the full frame. Some people make the argument that aps-c is just as good as full frame since you “have to” stop down the full frame to get the same depth of field. But far more often, I’m using among the widest possible aperture, whether full frame or aps-c. When I’m shooting low light, I’ll keep the ISO as low as possible and aperture as wide as possible to accomplish it. In individual portraits, I’ll shoot nearly as wide as possible.
Progress is measured in megapixels. At least by salesmen and consumers.
For my uses, the ideal resolution is about 24Mp for APS-C and 36Mp for FF. That is with current lenses and sensors.
At the same time I want there to be progress, so I must trust Sony to make good use of denser sensors.
Nikon has been using 36Mp APS-C sensors for a while and they work fine.
But I’d be happy with just better dynamic range and ISO performance. Hopefully, the resolution race will even out in the future and people will start thinking about more important imaging properties.
“Nikon has been using 36Mp APS-C”
No. Their 36 Mp is full frame.
No, the ideal resolution is 6 Mp. Or even better, 2 mp. That is with current lenses and sensors. Or even better, one single but very large pixel.
It would be one heck of a great pixel!
The A6000 is a sweet little camera for the price.
why would we need 36mp? we need better ISO…
36MP BSI is likely to outperform 24M non-BSI
That said, I’d prefer to see 24-28M BSI, which will have only slightly higher pixel density than the A7RII (approx. 18MP in APS-C crop area)
Shouldn’t 36 mp be reserved for the a7xxx model. I don’t see it happening, too much of a jump. Maybe a 28 mp be more plausible.
I was really hoping the new A6100 would be a sports camera with fast AF and fast FPS, If this is true I hope they have a function to reduce the resolution for higher frame rate.
I really hope they launch a new 16 50 f2.8 for APSC E Mount G Lens with this camera. That much resolution sensor begs high quality lenses!
If it has a collapsible design like the 16 50 pz, it would be even better!
“24 is too much, 36 is too much”
Hey, I have actually made a test. I took a cheap Industar 50-2 (50/3.5 4-element Tessar, bought WITH camera for $20) and made pairwise comparison shots of remote landscape placing lens on NEX-5N and Pentax Q. It is evident that NEX-5N 16 mp sensor does not extract all details from lens image. Note that the lens is a FF lens, not aps-c.
Unfortunately, I can’t do yet such experiment with E lenses, yet.
I agree that more sensor resolution almost always gives more details. I guess what most people mean when they say “XY amount of MP is too much” is that with higher sensor resolution the shortcomings of lenses get more pronounced. The higher the resolution the bigger the difference in sharpness between center and corners, the more apparent aberrations show. You get a lot more details with more MP but you also need to take more care of the lens’ shortcomings. For some a uniform, low MP image with less details and faster processing in post is just more convenient. If given the chance to choose between low or high MP Sensor, I myself would always choose the high MP sensor however (given that the downscaled noise behaviour is comparable). Best regards and thanks for taking the time in making this comparison.
>low MP image with less details and faster processing
Any camera can downsample on-the-fly
Meanwhile sometimes people will want to use high pixel density as teleconverter.
Would be good to have a low-res RAW setting on Sony cameras, too.
Also, keep in mind that most RAW converters create 24M (approx. 6000×4000) RGB pixels from 24M single-color photosites – there’s some interpolation.
So there’s color benefit in the higher resolution if you can avoid the noise consequences, which BSI helps significantly with.
Indeed i think most are missing the point. If we assume that we won’t see many new aps-c lenses for e-mount what can sony do to make the old ones work better. Most of the primes now max out at about 13Mpix of sharpness and the zooms at about 7Mpix. If we assume they go 50% more pixels and maybe other tricks to make the current lenses work better (selectable anti-alias filter?) how many gains will we see on those numbers. A good guess would be between 16Mpix-19Mpix for primes and between 8Mpix and 10.5Mpix for the zooms. This would mean no new lenses and numbers that are still higher than most m43 results people are screaming about. Also opens the door on opening the power of say the FE70200 to act like a 300 to 400 range when cropped and still have a >11Mpix image. All this without any new lenses why would they not do it ? The a7 range showed us that more pixels doesn’t mean it has to be less in other areas. So if sony can add a new sensor have gains on their current lenses and little downsides. I hope sony is thinking this way and have tested this vs the a7 route that seems to be bigger lenses not that i am against that but we already have the a7 line 🙂
The Pentax Q and the NEX-5n don’t share the same sensor size so the image circle on the lens is not the same. The Pentax-Q’s image will be taken off the center part of the image circle, leaving the (generally less sharp) edges out.
So should we discard any improvements in center just because edges (which often don’t even contain subject) are worse? And there are other, sharper lenses…
Someone needs to make a 24mm F1.4/F1.8 for the A6100, and make it available at an affordable price – will sell like crazy.
zeiss already does.
Do you think price of a6000 will drop after announcing/shipping the new a6100?
I wonder if I should buy a6000 now or wait a little.
If the successor has a similar price tag, A6000 price drops, if the successor has a much higher price tag, then not.
Soon we can talk about why the a6100 wasn’t announced at CES, and then speculate on whether it might be at CP+. It’s a tradition.
bsi is a way to amp up the megapixelrace once more… Why doesn’t m43 have bsi-cmos yet?
Cause it isn’t released, yet. It’s just a matter of time.
cambridge in colour calc only works in full stops annoyingly, but im guessing its 36mp aps-c is diffraction limited at 6.3 ish
24mp is at around 7.1 is last bit before diffraction
43mp is diffraction limited at 5.6 , 42mp squeaks by
so not as bad as i thought it might be, but given we are really near everyday apatures itll have its limits.
dont know what size scales well to 4k or if it matters to sony, as they got 42mp as 1:1 apsc size 4k, didnt they?
Will they release a kit lens that can resolve the 36MP on APS-C is what I wonder. There will be a physics Nobel prize waiting for the designer. I decided to stick with my superb NEX-6 and wait for an affordable FF mirrorless camera from Sony.
I’ve been waiting for this to arrive WITH 5axis stabilisation during video mode…
If it’s not coming then for gods sake release a 12MP 6000s version please!
You can keep your high res and supply a low light APS-C sensor stabilised beast instead.
I would prefer 28-30 MP BSI. I sure hope previous rumors of IBIS and a weather-sealed body (and a weather-sealed Zeiss 16-70) prove to be true. I would rather have better low-light performance with fewer pixels.
How much improvements from BSI do we expect? (Looking at the RX100 it seems that improvements are small, but cost increases by a lot)