Apple buys camera-technology company LinX

And most Apple "innovation" has come from outside, the GUI came as a gift from Xerox.

There were some giant failures, the years of the awful Mighty Mouse. And that was Jobs. ITunes, the software, is still pretty far behind in sound quality. There was the whole iPhone antenna issue, and it was several models.

More accurately, and this is after the passing of Jobs, Apple is abandoning support for pro user software like Final Cut Pro and Aperture.

And the latest Mac Pro really doesn't compete with say a high end Windows work station from HP, but the Mac is only $10,000 at most, while the HP can cost $20,000 pretty easily. But the HP is a much more capable piece of hardware. (And the HP DreamColor monitor just won technical Oscar.) No, I don't work for HP, but I sure think the DreamColor monitor on my laptop is excellent.

Apple is driving customers away from its laptops, but gluing in the battery, and making the GPU and RAM integral. This started under Jobs.

Apr 18, 2015permalink

Apple bought the GUI rights, but they then developed a lot of what ended up in the OS, like drop down menus, double-clicking to launch files, overlapping windows, and the whole desktop metaphor. They didn't actually use any of Xerox's code.

Jobs bought SoundJam and turned it into iTunes.

I don't think I've ever had a Mac Laptop that you could upgrade the GPU.

Apr 20, 2015*permalink

darn:

Sure Xerox was paid by Apple. Could be I guess.

Apr 20, 2015permalink

Xerox got Apple stock out of the deal.

Apr 20, 2015permalink

darngood,

Okay. Not exactly payment. But real.

Apr 20, 2015permalink

focalphotography

Easy way to avoid a patent dispute buy the competition and destroy them

Apr 17, 2015*permalink

Lenscraft

How is 'give $20 million' equal to 'destroy'? And where does this article mention a patent dispute?

Apr 17, 2015permalink

falconeyes

I am long predicting smartphones featuring lens arrays.

I did even predict Apple to be first to market and this news is a hint it will happen sooner than later.

Large enough lens arrays (like 5x5) can rival full frame image quality.

Lens/sensor array cameras are actually MORE sophisticated (not less) than lightfield cameras because they don't suffer from the severe diffraction-bound resolution limit which plenoptics does.

Apr 17, 2015*permalink

TrojMacReady

The HTC One M8 already featured a dual lens setup (second unit was mostly for depth information).

Only the execution (mainly software part) was lacking.

Apr 17, 2015permalink

falconeyes

Dual lens isn't an array. 3x3 is what I would consider the smallest meaningful array. It would be able to create images with 1/3 the nominal aperture, aka 1/3 the crop factor.

My point is phone cameras will have outstanding image quality some day. I did not talk about gimmicks for kids like depth extraction, background blurring etc.

Even (limited) refocusing in post is meaningful, not as an effect but in order to optimize focus accuracy after the shot is taken and can be analyzed.

Apr 17, 2015*permalink

TrojMacReady

I don't see a 3x3 array coming to iPhones, for the simple fact that it would take too much space on the limited form factor. Not to mention the fact that Apple cares a lot about margins...

" Phones impose limits on depth, not width"

Not true either. They're closely related. Many phones have their cameras in the area above or under the screen, because of depth constraints even for the most shallow imaging units. Especially with phones still becoming thinner. The alternative being a pucker found on phones like the S6. Again, I don't see Apple fitting an array in the top or bottom part. Or when more shallow, practically the whole back. Too much cost at the expense of their holy margins anyway. 9 modules instead of 1 can easily be a loss of >10% margin on the cost price.

And their iPads have never had the top mobile photography hardware.

BTW, you did mention lightfield cameras. Which are all about " gimmicks for kids like depth extraction" after the fact.

Apr 17, 2015*permalink

falconeyes

That's the point exactly. An 3x3 or 5x5 array would take zero space. Phones impose limits on depth, not width which is why phones will feature camera arrays. Its the only way they can progress w/o compromising their form factor. Think it thru ...

Moreover, camera modules are cheapo, and 9 or 25 of them still are ... Apple was never shy of expensive technology if it serves a purpose for the usage experience.

Apr 17, 2015*permalink

So what is the future of photography? Mirrorless, DSLR or iPhone?

Apr 17, 2015permalink

Two of those options have much bigger sensors, and always will, and the bigger sensors disperse heat per pixel better.

Two of those options are in bodies that can take interchangeable lenses.

Two of those options can readily usually do shallow depth of field, with the right lens.

So even if we get cold electricity tomorrow...

But in all seriousness, the answer to the silly question is all three, but for different purposes and ends. (And the iPhone, though it doesn't shoot raw, stands for smartphone here.)

Apr 17, 2015*permalink

Thanatham Piriyakarnjanakul

@HowaboutRAW, totally agree with you at this moment. It is the same when people said Mirrorless is the future of photography.
EVF still not real time, AF tracking still not on par, except sensor.
Photography is not about sensor alone. Is one of the key factors.
Photography is an art, graphic, information, moment, memory.
Thanks

Apr 17, 2015permalink

Phone camera still no optical zoom or interchangeable lens.

Apr 17, 2015*permalink

Joe Ogiba

I can't image future cameras having a flipping mirror box. After using my full frame A7r that is much smaller than my APS-C DSLRs I will never again purchase a camera with a mirror. I also enjoy getting stills from my GH4 4K videos.

Apr 17, 2015permalink

Joe,

And that A7R is much louder that all APSC DSLRs. (I guess this will change.)

DSLRs are going to be around for a long time. But EVFs will improve, so DSLRs will make up less of a percentage of consumer camera sales.

Apr 17, 2015permalink

KW P,

No, but better sensors and lenses, and processors sure help.

Apr 17, 2015permalink

No body know about the future. Get the camera the meet your need now. Can I get the future camera without mirror to cover the football match, now?

Apr 17, 2015*permalink

KW,

Yes, the EM1, and the NX1 would be excellent for shooting a football match today.

And the GH2, XT1 and A7II would be good too.

So no need to wait for a "future", today is good.

Apr 17, 2015permalink

In the field Pro use DSLR, outside the filed more people use smart phone, presently. Let wait for the future.

Apr 17, 2015permalink

KW:

There are a lot of cameras between the D4S and the smarphone, several of which can shoot a football match well.

Then there are studio cameras, many of which are DSLRs.

No smarphone takes interchangeable lenses, few smartphones have optical zoom, and few shoot raw.
The point being that there are big reasons to consider non-smartphone cameras that aren't $5000+ bodies.

Apr 17, 2015permalink

Both Samsung and Panasonic have smartphones with 1" sensors, so once the iPhone, for example, moves form 2/3 to 1" it will be a more viable camera for people who care about photography.

Apr 20, 2015permalink

"The acquisition would indicate that Apple is planning to do more of the imaging hardware development in-house. "
No, it doesen't. LinX is a pure software company that previously have bought camera modules from chinese sunny optical. The software is usable with other camera modules as well. Apple will likely choose camera modules that suits them best regardless of what LinX previously used.

Apr 17, 2015permalink

jacosta

In my non expert opinion, combining an RGB sensor with a BW sensor should increase low light performance (by a factor of 3) as well as resolution.
In terms of resolution, the BW sensor has no antialias filter and no demosaicing is involved and therefore provides a much sharper BW image; furthermore it allows to recover per pixel color information, thus also increasing color resolution (by a factor of 2?)
Any expert opinions?

Apr 17, 2015permalink

I'm not an expert but i agree on most of that. Low light performance will improve by a factor of about 2, not 3. Thats because the color filters in the bayer array doesent filter out 2 out of 3 color ranges. It filters out one color channel almost completly, one almost not at all, and the third just half way. Demosaicing algoritms perform better that way. Real quantum efficiency is normaly around 50%.

A monocrome sensor would improve that to around 90% compared to a sensor with bayer filter.

AA filters are not necessary when the lens is doing that job by diffraction limitation or other imperfections.

Apr 17, 2015permalink

jacosta

Thanks for the explanation SimenO1 !

Apr 17, 2015permalink

jacosta

Incidentally, the Huawei Honor 6 Plus comes with a dual 8Mpx camera setup, to achieve better performance and 13Mpix shots. I wish there was more info on it .

Apr 22, 2015permalink

ThatCamFan

Facebook would of bought it probably for a billion dollars as usual lol

Apr 17, 2015permalink

dialstatic

My guess: iPhone 8 will have a 3D camera :-)

Apr 17, 2015permalink

zodiacfml

This is just a sign that Apple is dead serious with smartphone photography....killing the entry level P&S of Canikon, we even saw a patent for three sensor, original Foveon style image capture.

Have we heard from Canikon innovating the entry level segment outside from increasing MP counts?

Next thing we'll know, smartphones would eat into mid level P&S too.

Apr 17, 2015permalink

http://www.androidauthority.com/samsung-galaxy-s6-sony-imx240-camera-sensor-588726/

Prime lens P&S are a very small part of the P&S market, and Apple is far from alone in the phone market. LinX are not alone in the multi camera module software market. I think P&S will survive the prime camera phones by emphasizing more on the zoom capability.

But the P&S makers have a job making the cameras more user friendly. Online instant sharing services and so on.

Apr 17, 2015permalink

Menneisyys

"This is just a sign that Apple is dead serious with smartphone photography...."

Too bad they haven't produced anything outstanding so far. They have never been the BEST in smartphone photography. Heck, not even mainstream smartphone photography like the Galaxy Note 4; that is, not niche devices like the 1020 or the 808.

Apr 17, 2015*permalink

JakeB

Cheap, plastic Samsung crap?

No, thanks.

Apr 17, 2015permalink

Menneisyys

jakeb, you haven't ever seen a Note4, I presume.

Apr 17, 2015permalink

Coyote_Cody

Funny I posted about this yesterday in Canon M forum w/ zero responses? Not many inquiring minds??

A 'clever' solution to the thickness issue, not useful for larger sensors though, unless space is the big issue.

Likely easier to do very fast multiple exposures and then blend them via h/w algorithms (like HDR but for 'normal' stills) for both jpg or raw.

Hope the engineers who came up with this got big bucks in apple buyout !! Good work, big daddy buys, big bucks!!

Now Samsung must solve the same problem - no issue for them of course, they certainly have ALL the necessary skills, apple likely did not, buying is easier!!

Apr 17, 2015permalink

Peiasdf

Samsung is using SONY sensors again after the in-house ISOCELL sensor totally bombed in Galaxy S5.

Apr 17, 2015permalink

jennyrae

load of nonsense.

Apr 17, 2015permalink

Boerseuntjie

Read it and weep, you are full of nonsense jennyrae.
http://www.gsmarena.com/samsung_galaxy_s6_has_a_sony_imx240_camera_sensor-news-11392.php

http://www.phonearena.com/news/Samsung-Galaxy-S6-and-Galaxy-S6-edge-use-the-Sony-IMX240-camera-sensor_id66700

Apr 17, 2015*permalink

DVT80111

$20M is just pocket change. This company does not have a lot of technology. Maybe just a few patents.

Apr 17, 2015permalink

beenthere

More info on LinX:
http://image-sensors-world.blogspot.com/2015/04/apple-acquires-array-camera-developer.html

Apr 17, 2015permalink

Trying to figure how this is any different from Pelican Imaging and their 2x2 and 4x4 array cameras.

Linx and Pelican both have too low "fill factors", most of their array is the space between the microcameras: wasting pretty much all the potential of the idea. They are also "comb filtered", which makes image recombination incredibly difficult. They definitely can't fulfill the claims of matching the performance of a single lens the size of their arrays.

My prediction: this is going to be as unnewsworthy as Apple's PrimeSense purchase.

Apr 17, 2015*permalink

Pelican is "comb filtered", but i dont think LinX are. They seem to use standard imaging modules and have only 2-4 simultaneous focusing depths. Corephotonics seems to use the LinX approach without "comb filtering".

Multi modules are about much more then adding sensor area. Its also about maximizing sensor area within thikness constrains, video frame rate multiplying, depth sensing, HDR, and syntetic aperture resolution techniques. 2-4 modules are of course not the end of this development.

And light gathering area is more about front lens coverage area then sensor area. Front lenses may one day be covering as good as microlenses are doing today over pixels.

Apr 17, 2015permalink

WaltFrench

Think of the Very Large Array of multiple radio telescopes that are now state-of-the-art. Separate sensors covering a wider radius, so able—with some not-actually-horrible computations—to get the equivalent of a big sensor/lens w/o the depth that others here aptly note.

Combining the separate images is a bit trickier than for radio telescopes due to the much more challenging depth of field issues. Note that clever as Lytros was, the images never had the fine resolution that people increasingly expect from even snapshot cameras.

Apr 17, 2015permalink

Radiotelescope arrays operate by correlated sampling throughout the array: phase is important in synthetic aperture imaging.

You'd need to build the camera array out of Kinect 2 style gated sensors to pull that off in an array camera. Optical telescope arrays like ROTSE don't correlate phase, they orient the scopes to different headings to simultaneously capture wider areas.

> Note that clever as Lytros was,

Lytro wasn't actually "clever". Array cameras are more clever, but they suffer from the same problem Lytro did, dramatic undersampling.

Apr 17, 2015permalink

Does LinX make sensors?

Apr 17, 2015permalink

Nope.

Apr 17, 2015permalink

They buy complete sensor modules from chinese sunny optical. Now they will probably be using whatever camera modules Apple decide.

Apr 17, 2015*permalink

iAPX

"leading the way to DSLR performance in slim handsets" lol!

Maybe it's as good as dslr from 12years ago or more?!? lol!

Apr 17, 2015permalink

Not likely, more like a highend DSLR from about 1999.

Apr 17, 2015permalink

Mister Roboto

Not even a chance with microscopic sensor of iPhone. How the heck can it match DSLR output? Not even better than a $50 camera nowadays.

Apr 17, 2015permalink

Sensors are irrelevant, what drives system performance is the size of the aperture at given coverage angle. That's basic information theory.

Pelican and Linx cover a large area with small cameras, which, from an information theory standpoint, has the potential to match a lens with that total covered area. The problem is that P and L don't sample enough of "created" aperture's actual area for this to work well, and they don't sample it with enough points. Linx literature shows 2x2. Pelican talks of 4x4.

Apr 17, 2015permalink

Mister Roboto

Irrelevant? How are you gonna cram that much information in a 1/3 sensor? Camera manufacturers are going forward with 1 inch going mainstream while Apple is going backward.

Apr 17, 2015permalink

Yes, totally irrelevant.

The problem with your "How are you gonna cram that much information in a 1/3 sensor?" question is that there isn't "a 1/3 sensor", there are (plural) up to sixteen sensors (plural, again) and that's just warming up.

Apr 17, 2015permalink

Peiasdf

@Mister Roboto
Why are you still trolling Apple articles? If you don't understand array processing you shouldn't comment on it. Now you just make yourself sounds like the ignorant fool you are.

Apr 17, 2015permalink

"Leading the way to" doesent mean iPhone 7 will get dslr quality. It means that in some distant future, maby 10 years from now, we might get dslr-like image quality from an iPhone using much more then four camera modules. Maby. But who knows what dlsrs are capable of in 10 years?

Apr 17, 2015permalink

Once Apple adopts a 1" sensor like Samsung and Panasonic have, iPhones will start taking sales from the enthusiast compacts.

Apr 20, 2015permalink

[OFF-TOPIC]
@Lars (or anyone else at DPReview) - Few weeks ago I've looked up for any info about a successor for the Lumia 1020 and, apparently, Microsoft will abandon the "big sensor" route because it was a niche market and Microsoft realised that it'd be better for them to do just like everyone else.

Do you have any info on that?

Apr 17, 2015permalink

Peiasdf

That'd be sad. The camera and unique UI is the only reason I am interested in a Window phone.

Apr 17, 2015permalink

Lars Rehm

I have no hot info on this but a while ago I spoke to a Lumia engineer who said that a lot of customer feedback they got for the 1020 was that the camera was too slow which of course has at least partially to do with the large sensor.

Apr 17, 2015permalink

@ Lars - Thanks for your reply! :)

As for the Lumia people's comment: well... yeah! If you use a single-core processor (808) when everyone else is using dual-core and then use dual-core processors (1020) when everyone else is using quad-core --and some flirting with octa-cores, of course the speed of your phone will be lacking!

But nowadays, with quad-core processors and RAM getting cheaper, I think that a 1030 wouldn't be so slow as the current 1020.

I could seriously consider the Panasonic CM1 if it wasn't for the Android...

Apr 17, 2015permalink

Menneisyys

"I have no hot info on this but a while ago I spoke to a Lumia engineer who said that a lot of customer feedback they got for the 1020 was that the camera was too slow which of course has at least partially to do with the large sensor."

High pixel count sensors can also be very quickly (and energy-efficiently) processed / handled with proper engineering - see the 808. It's what engineering was missing in the pretty much rushed 1020 - they simply didn't have the time to implement DSP purely in hardware, unlike with the 808.

Apr 17, 2015permalink

Lars Rehm

all a matter of cost though, at the end of the day manufacturers need to shift huge numbers of those devices to make the development worthwhile. Niche products that only appeal to a small number of users and are expensive will always be a difficult sell internally I think.

Apr 17, 2015permalink

"Niche products that only appeal to a small number of users and are expensive will always be a difficult sell internally I think."

Yes, absolutely!

But I still think that the major problem with the Pureviews 808 and 1020 wasn't being a niche product, but the hardware itself.

Like I said on my previous post, both had an inferior hardware compared to its peers. In the case of the 808 it was even worse because it came with Symbian, that was nearly dead already.
If it had come with Android (or even Windows Phone) I guess the acceptance would've been better.

I don't know, I think that MS should at least try one last time. Specially now that WP is maturing, Win10 is coming up, MS has more re$ource$ than Nokia had and can use a better hardware and so on.

Well, let's keep our fingers crossed :)

Apr 17, 2015*permalink

Tom May

All you lusting for the large imager might have another shot at it if Nokia builds and sells an Android device beginning in third quarter 2016.

In my opinion though, that boat has sailed and will be an even smaller niche 18 months from now. The future lies with folded optics, mems, lens arrays, and massive computation, all better suited to the thin form factor of smartphones.

Apr 20, 2015permalink

Mrrowe8

I am interested & It is the direction Photography is going the better the cell phone camera gets the greater the recording of the human experience is .. Much like Kodak did with the brownie and later instamatics

Apr 17, 2015permalink

And unless exposure happened to be perfect, the results from Instamatics weren't good, or it was one good shot out of 24.

Apr 17, 2015permalink

Mister Roboto

I would rather have a larger sensor at least 2/3 or 1 inch than invest on any lens which in the end will produce mediocre result.

Apr 16, 2015permalink