No A78 this year according to SAR

[mvanrheenen]

And the big question is can they nail good, fast, accurate AF that can track action without a mirror?

And: will the body size be getting smaller or not?

For both I hope not...

[alphaomega]

Barry Fitzgerals wrote

And the big question is can they nail good, fast, accurate AF that can track action without a mirror?

If Sony go for mirrorless then maybe they are confident that the new combined phase detect/contrast detect system works (or can be made to work) well enough such as now used in my NEX-6. Presumably there will be new improved sensors incorporating phase detect pixels. Maybe the relative success (for Sony) of the NEX system versus SLT have encouraged them to put everyting into mirrorless. I don't see Sony going back to mirror. They will not make any dent in CAN/NIK. According to SAR Sony are not ditching A mount so it will be someting different from NEX maintaining the A mount sensor to lens distance (or maybe a hybrid mount that can take both A mount and E lenses. Who knows what Sony are up to now. If a hybrid mount in in the pipeline then any lens of repute can be fitted using suitable converters including speed boosters. The appeal will widen to just about every enthusiast and professional photographer.

[agorabasta]

Ditching SLT altogether requires that on-sensor PDAF of competitive efficiency is foregoing.

So I assume it's a very good news indeed; once it's promoted out of the 'rumour' status, that is.

[Greg Beetham]

The future of screw drive lenses isn’t looking too good I think, will Sony even include a screw-drive focus motor in the new mirror-less version of the A-mount?
Maybe the new version will have an all-electronic shutter as well thus getting rid of another mechanical source of trouble.
The accompanying next logical step is to get rid of the troublesome A-mount mechanical aperture actuator as well and make future lenses with an electronic aperture and that would most likely require the inclusion of a couple more contacts in the body and on lenses. The only drawback (for customers not for Sony) in doing that is that all previous A-mount lenses would need replacing for use on the new body, so they might hold off doing that for a while.
Or they could introduce the extra contacts (and a couple of compatible lenses) but keep the mechanical actuator for the next body duration or two for legacy lens users. (if the extra contacts arrive you know what the future holds)
Also IBIS could disappear at some point (the switch has disappeared already and they use electronic IBIS in some models for video) and be replaced with in lens stabilization like Nikon and Canon do, that is much more suitable for video than IBIS is so that’s the way to go for sure.
Exciting times ahead for the A-mount (RIP).
Greg

[bfitzgerald]

I think the key here is "backward compatibility" it's all very well taking out AF motor in bodies and aperture levers. But you stuff up that consistency across your range and possibly alienate your users.

I've never had a problem with screw drive lenses or mechanical apertures myself, some say not great for video but how often do I shoot video? I think that's part of the problem makers have had, they're placing their bets on video as the future. Even Canon with their STM lenses, we go back to those awful days of focus by wire rather than mechanical focus. Thanks, but no thanks been there done that and it sucked back in the 90's and it still sucks now!

As for Sony they have to tread with care here, this isn't Playstation land you have a very different market and user.
Camera industry as a whole I think the party is "over" compact sales have dived hugely, even the ILC market is down. Far as I can see things need to settle down a bit, less models, less frequent updates, more emphasis on the enthusiast sector and less on the p&s one. Sure I see folks using an ipad to take photos, but let's get real it's not exactly the choice of enthusiasts or serious shooters. So I expect to see a few makers bow out in the next few years.

If you told folks 20 years ago Kodak would be out of the photo business people would have laughed at you, but that's the reality of the modern day.
Part of the problem is over consumption and trying to re-sell the same thing to people. I think over time the market has become over saturated and people buy less/upgrade less often. Good enough is the phrase that strikes fear into any company, but it's something they have to deal with

You can only have so many smile shutters, huge pixel counts and sure 4k video will be the next thing they throw at us. Then 8k video will be in demand and the cycle continues on and on. I admit we're all consumers to a point. But my extended period of hassle resulted in something that hit me quite a bit. I was forced to use my older 6mp Km5d's for some time, and guess what for "most stuff" quirks and clunky or not..they're actually capable of doing "most" of what I need them for..ie taking photos!

[agorabasta]

I think the only logical solution is to make camera bodies with interchangeable mount modules. As long as the mount module is a standard part across the bodies, the cost of the system may stay reasonably low. Such mount modules may include an E, a full legacy A, a no-screwdrive A and then maybe some fully electronic 'post-A' type.

[Atgets_Apprentice]

I think the only logical solution is to make camera bodies with interchangeable mount modules. As long as the mount module is a standard part across the bodies, the cost of the system may stay reasonably low. Such mount modules may include an E, a full legacy A, a no-screwdrive A and then maybe some fully electronic 'post-A' type.

Haven't Ricoh already done this?

[alphaomega]

Anyway, SAR are back with another SR5 rumour regarding the ditching of SLT and doubts about the A mount.
http://www.sonyalpharumors.com/new-alph ... arly-2014/

After yesterday’s post about Sony removing the SLT tech form the APS-C series quite some people got scared that this also means the end of the Alpha Mount lines. I keep stressing you that this isn’t true and that the A-mount line will keep being developed. Three years ago Sony switched from Optical viewfinder to Electronic Viewfinder with translucent mirror. And next year we will see another switch (with some surprise?). The dead of SLT is just the dead of a feature..not the dead of the camera line

So what’s the goal of Sony? One of my best Sony sources that always shared correct info sent me a short sentence:

“The Alpha line is not going away. They are working on technology that will outshine the competitors for early release next year.”

He is a top source and in one year from now we will be able to see and touch the new Alpha Camera Generation. Trust SAR and that source…he always told me the truth! Spread the word!

So I am "spreading the word". Perhaps someone close to Sony have been instructed to send this message to quell any speculation that their A mount lenses may become obsolete.

[agorabasta]

Haven't Ricoh already done this?

They did, in a way; and it was not too good. Sony must try much better to succeed with that.

[classiccameras]

Only a guess, but I suspect Sony will drop the SLT system on future A mount DSLR's and use NEX technology for all cameras. I say this because NEX tech has a future road map, SLT is doubtful.

[agorabasta]

Having a large mirror chamber is a good thing. It improves telecentricity.
But the most important thing is that they may put there something else, like the hot mirror; and more importantly, that hot mirror may be in the same slanted position as the current SLT mirror. This way the reflected IR is removed from the optical path. And then the IR absorption filter may also be glued onto the hot mirror, and would completely eliminate the problems with the sensor toppings internal reflections. And Sony actually applied for a patent describing the IR filter placed on the SLT mirror.
And then, having that IR filter in place of SLT mirror makes it possible to have the sensor sitting in an effectively sealed compartment.

Then there's another possibility - the mirror chamber doesn't have to be filled with air (empty), it may be filled with some dense optical medium having all the IR filter parts placed inside of chunk of solid matter or in some liquid. That could serve two purposes - push the diffraction limit farther, and also it would make the former mirror chamber depth smaller, thus helping to make the system thinner.

So there's a lot of opportunities they may think of; the question remains if they are going to use any of those.

[bfitzgerald]

Wondering how they are going to "outshine" competitors next year? All seems rather vague and like I said bar a sensor revolution, can't really see anything else on the cards that can do this. We're still waiting for a properly updated flash system!

It could hurt resale values on the A99 though, not that I expect Sony have sold loads of them. I have felt for some time that the idea of a fixed mirror in the light path, wasn't something that would go down well with buyers (or some of them)

[Dr. Harout]

Having a large mirror chamber is a good thing. It improves telecentricity.
But the most important thing is that they may put there something else, like the hot mirror; and more importantly, that hot mirror may be in the same slanted position as the current SLT mirror. This way the reflected IR is removed from the optical path. And then the IR absorption filter may also be glued onto the hot mirror, and would completely eliminate the problems with the sensor toppings internal reflections. And Sony actually applied for a patent describing the IR filter placed on the SLT mirror.
And then, having that IR filter in place of SLT mirror makes it possible to have the sensor sitting in an effectively sealed compartment.

Then there's another possibility - the mirror chamber doesn't have to be filled with air (empty), it may be filled with some dense optical medium having all the IR filter parts placed inside of chunk of solid matter or in some liquid. That could serve two purposes - push the diffraction limit farther, and also it would make the former mirror chamber depth smaller, thus helping to make the system thinner.

So there's a lot of opportunities they may think of; the question remains if they are going to use any of those.

This answers many of my questions. Thanks.
I always thought that a greater flange (register distance) has its own benefits, of which (probably) design and construction of lenses. I might be wrong. Frankly that's an intuition...

[Greg Beetham]

I was thinking much the same about the empty space agorabasta, camera designers abhor empty spaces, they’d feel compelled to fill it full of something but I think sealing the sensor in might not be good unless they can get rid of the mechanical shutter as that’s inclined to shed particles of stuff, some of which can end up on the sensor, and maybe they will…ditch the mechanical shutter, who knows.
I was reading some stuff about sensor design and they were saying the wiring grid overlay was some kind of IR interference filter in its own right, but that one had me baffled, unless it runs at a temperature that’s less than the rest of the sensor maybe, (absorption) it wouldn’t be small enough to act as a band-pass filter surely? They are making sensor components on a µm scale not an nm scale correct?
So that means they would have to rely on filters for blocking ‘near’ IR but that would tend to trap the sensor produced IR on the sensor side of the filter where it would cascade…unless the hot filter radiates the sensor produced IR, and one of the other filter layers does interference (null) at the visible IR wavelength, that might make some sense.
Greg
Ps I think the SLT was going to do that too Barry, maybe the next advance will apply the finishing touch.

[agorabasta]

I was reading some stuff about sensor design and they were saying the wiring grid overlay was some kind of IR interference filter in its own right, but that one had me baffled, unless it runs at a temperature that’s less than the rest of the sensor maybe, (absorption) it wouldn’t be small enough to act as a band-pass filter surely? They are making sensor components on a µm scale not an nm scale correct?

The sensor sees its own IR anyway, but that IR is the low-K heat radiation and the silicon is not sensitive to that.

But the wiring reflects visible light that passes back through the IR absorption layer and then reflects from hot mirror back onto the sensor. That hot mirror appears purple to the eye, and so it appears to the sensor too, and that's how we get the purple fringing and the blue-purple ghosting going into the edges/corners. Using amorphous copper for the wiring improves the things, as the copper gets blacker the more amorphous it is; but the silicon itself gets quite reflective at strong exposure.

The sensor wiring feature size is well into the nm scale, modern sensors are made with 120-90nm processes. Canon is the only extant dinosaur with their 500nm process.