the image sensor with the built-in of phase detection range finder

Nikon Rumors Forum » Nikon DSLR

the image sensor with the built-in of phase detection range finder

(21 posts) (6 voices)

Tags:

PeterZheng

member
Joined: Mar '11
Posts: 16
offline

Since 2007 later, the built-in of phase detection range finder method will take build-up within image sensor. As of July 2010 in Japanese published patents, which Nikon has nearly 20 patents published, Canon has nearly 25 patents, Sony has nearly 15 patents of publication, and Panasonic also has a few of patents. However, Pentax is no patent in this one side. These are also R&D for the image sensor.

In a hereditary way, an AF sub-mirror is added to a master-mirror's backside, that is source from the Pentax's patents, (registered in 1973-1979). Pentax ME-F was first applied example, released 1980 and sold into market 1981, while the ME-F was taking CD-AF analog circuit. Today's the master-mirror way still be a faraway origin came from the Pentax's patent, in 1956.

In July 2010, Fujifilm announced FinePix F300EXR camera, the phase difference detection range finder successfully achieved the built-in method on the image sensor. However, Fujifilm did not publish about the source of technology and that who is the sensor's manufacturer.

Perhaps, in the Nikon V1 its sensor manufacturer is Fujifilm, and not Sony? "we engineered and developed this sensor inside Nikon. But for the production side, that is done by our partner", Suzuki Masahiro said. (Mr. Suzuki is the General Manager, Research & Development Department, Development Headquarters, Imaging Company, and managed the team that created the new products).

I wish to see the built-in method of phase detection range finder will be into APS-c and 35mm sensors, but the OVF should not disappear at the DSLR camera, does it only remove the AF sub-mirror and complicated AF module.

In the DSLR, this built-in method will cut down costs and heightens AF reliability, can avoid the traditional AF module's potential accuracy error, (the original two optical planes are merged into a single optical plane), no longer need AF module's mechanical adjustment procedure. The Live-view mode to be AF faster.

Also, if the Coolpix P7100's successor is a 2/3-inch sensor with the built-in of phase detection range finder would be exciting.

Nikon, Publication JP 2010129783, Date: 2010-06-10
http://patent.astamuse.com/ja/published/JP/No/2010129783

Sony, Publication JP 2010-78707 (P2010-78707A) Date: 2010-04-08
http://www.ekouhou.net/%E6%92%AE%E5%83%8F%E7%B4%A0%E5%AD%90%E3%81%8A%E3%82%88%E3%81%B3%E6%92%AE%E5%83%8F%E8%A3%85%E7%BD%AE/disp-A,2010-78707.html

Canon, Publication US2010165176 (A1) Publication date: 2010-07-01
http://appft1.uspto.gov/netacgi/nph-Parser?Sect1=PTO2&Sect2=HITOFF&u=%2Fnetahtml%2FPTO%2Fsearch-adv.html&r=1&f=G&l=50&d=PG01&p=1&S1=20100165176&OS=20100165176&RS=20100165176

Posted 1 year ago #
El_Pickerel

preferred member
Joined: Jul '11
Posts: 581
offline

Correct me if I'm wrong, but if you had the AF sensor built into the image sensor and lost the separate AF module, wouldn't you get no autofocus unless the image sensor was exposed? Wouldn't that make an OVF nearly useless?

Now if you did this to make a full-frame / APS-c compact camera to be a competitor to the likes of a Leica, that would be pretty cool. Or if Nikon was to up their competitiveness in the video market and have a camera with a normal AF array for stills and one built into the sensor for video / live view, that would also be cool for those who would like to use Nikon for video and shut up all those Canuts.

Posted 1 year ago #
Testing123

preferred member
Joined: Oct '10
Posts: 232
offline

El_Pickerel said:
Correct me if I'm wrong, but if you had the AF sensor built into the image sensor and lost the separate AF module, wouldn't you get no autofocus unless the image sensor was exposed?

Correct

Wouldn't that make an OVF nearly useless?

No.
EVF would require power to the full sensor as opposed to just the AF segments. Power = heat and until active cooling is a part of cameras heat will be slow to dissipate.

The effects of heat can be mitigated and minimized but a hot sensor will always have a worse SNR than a cooler sensor, regardless of the downstream technology.

If you add cooling to a EVF camera to remove (some of) said heat you can add cooling to an OVF camera. The OVF will always have a SNR (and thus ISO) advantage.

Be that ISO advantage worth the hassle of an OVF (though we haven't gotten into the whole list of positives and negatives) is left for the individual to decide.

Also a highres EVF is still more expensive than a OVF if we're talking a basic penta-mirror system.

Posted 1 year ago #
PeterZheng

member
Joined: Mar '11
Posts: 16
offline

In hereditary way, an AF sub-mirror is added to a master-mirror's backside, but the OVF is available, the master-mirror is translucent at the AF area. The built-in PD-AF method is in the DSLR camera still has translucent master-mirror, while a difference is that the mechanical shutter was opening state (or half opening) when AF running, but the exposure time is an usual way to run the shutter. This is DSLR camera, not FSLR camera, does not leak films.

Posted 1 year ago #
Testing123

preferred member
Joined: Oct '10
Posts: 232
offline

PeterZheng said:
The built-in of PD-AF method still has a translucent master-mirror, a difference is that the mechanical shutter was with opening state (or half opening) when running AF,

The camera in question does not have said master-mirror, though the technique could be applied, as you implied (through removal of the sub-mirror) to SLR cameras to give one OVF and PD-AF on-sensor. This would reduce mechanical complexity, part cost, and manufacturing difficulty (alignment) as well as allow higher frame rates (as the AF system would not be obscured during image acquisition and the mirror assembly would be much simpler and more cheaply capable of high-G operation).

I'll bet my last dollar the D3200 has only on-sensor AF. Haven't seen enough details to know if betting on higher models is a safe bet yet.

Posted 1 year ago #
El_Pickerel

preferred member
Joined: Jul '11
Posts: 581
offline

Okay. So let me get this straight.

The mirror is translucent in the areas that light would have to pass through to the AF points. The shutter is open for autofocusing. That is how you would have AF on the image sensor while still being able to have an OVF that covers the entire frame. Is this correct?

Posted 1 year ago #
Testing123

preferred member
Joined: Oct '10
Posts: 232
offline

That is correct.

As it works now with SLRs, the mirror is already translucent in the areas that light would have to pass through for the AF points. There is a second mirror behind the primary one (primary mirror bounces light UP to the OVF) which bounces light DOWN to the AF sensors.

If the AF sensors are on-sensor one would simply remove this second mirror and open the mechanical shutter (if any).

Posted 1 year ago #
PeterZheng

member
Joined: Mar '11
Posts: 16
offline

El_Pickerel said:
Okay. So let me get this straight.

The mirror is translucent in the areas that light would have to pass through to the AF points. The shutter is open for autofocusing. That is how you would have AF on the image sensor while still being able to have an OVF that covers the entire frame. Is this correct?

did you know actual existing DSLR cameras about the basic structure of the AF and OVF?

Posted 1 year ago #
PeterZheng

member
Joined: Mar '11
Posts: 16
offline

Testing123 said:
That is correct.

As it works now with SLRs, the mirror is already translucent in the areas that light would have to pass through for the AF points. There is a second mirror behind the primary one (primary mirror bounces light UP to the OVF) which bounces light DOWN to the AF sensors.

If the AF sensors are on-sensor one would simply remove this second mirror and open the mechanical shutter (if any).

Thanks for the correct explication

Posted 1 year ago #
El_Pickerel

preferred member
Joined: Jul '11
Posts: 581
offline

PeterZheng said:
did you know actual existing DSLR cameras about the basic structure of the AF and OVF?

I've only even had DSLRs since January. I haven't stopped to look extensively into the internals save for specifications and what those mean. A little slack would be sort of nice. My engineering mind wants to figure out how things work but doesn't always have the time to.

Posted 1 year ago #
PeterZheng

member
Joined: Mar '11
Posts: 16
offline

http://fotkidepo.ru/photo/143879/15383I8E2H3JCr5/bwQaWbOUZA/543095.jpg

In this URL provides an existing DSLR camera's section plane showed, makes you can see the master-mirror and sub-mirror. The top of mirror is optical viewfinder, and the bottom is AF module.

This AF formation was originally designed for the FSLR cameras during the films era. In today, most of the DSLR cameras cannot interchange the image sensor, nor leaks films, so it ought to remove the AF sub-mirror, and takes the PD-AF range finder to be built-in method on the image sensor. (Unless the image sensor is interchangeable way, then it could be an exception).

In the picture, maybe you already saw the AF module is complicated configuration, and it also needs the mechanical position adjustment procedure in the production line, its costs and method are not perfect result if with AF built-in method compared.

Posted 1 year ago #
donaldejose

preferred member
Joined: Mar '11
Posts: 1,043
offline

How does the phase detection built into the chip work to focus the lens when the shutter curtain is still closed before you take the photo?

Posted 1 year ago #
Testing123

preferred member
Joined: Oct '10
Posts: 232
offline

donaldejose said:
How does the phase detection built into the chip work to focus the lens when the shutter curtain is still closed before you take the photo?

Testing123 said:
If the AF sensors are on-sensor one would simply remove this second mirror and open the mechanical shutter (if any).

Posted 1 year ago #
donaldejose

preferred member
Joined: Mar '11
Posts: 1,043
offline

I think you are correct. The shutter has to be opened every time you focus on a new object. So now you are causing a shutter actuation every time the lens focuses. That might be a ratio of something like 10 shutter actuations for each photo taken. Unless shutters using this system are built more robust a shutter rated for 150,000 actuations could exceed this number after just 15,000 photos. See the potential durability issue?

Posted 1 year ago #
Testing123

preferred member
Joined: Oct '10
Posts: 232
offline

Assuming there is still a need for a mechanical shutter.

I don't know the technology well enough to say why they are needed at all anymore. Perhaps someone else does.

Posted 1 year ago #
PeterZheng

member
Joined: Mar '11
Posts: 16
offline

about the mechanical shutter's drive way during the AF running, I think there are possibly three different solutions.

People ought to notice that an image sensor is behind the shutter, but is no longer films. For the DSLR camera, the initial position of mechanical shutter does not have to need the precondition in closed state. But the closed shutter can help to reduce the dust intrusion on the sensor module.

Method 1. the initial default position of shutter is fully opened, but the exposure period to run the shutter. This is the simplest way, but the shortcoming is also obvious, makes the dust would easier stain the image sensor module.

Method 2. when the AF beginning, the shutter will be opened and holds in 1s - 3s, unless the exposure started, would switch to an exposure state. The initial default state is with closed shutter. In today, most of the shutter durability are not a problem, like that.
Also, you may notice the Live-View mode (Nikon d7000 or other), once the release button has been taken a half pushed, then the mechanical shutter will be a full opened state.

Method 3. a better way, there is a need to innovate mechanical shutter solution at the DSLR camera. Said Method 1 shortcoming is dust troubles, however, in foreside of the shutter can set a simple sub-shutter (auxiliary blades), its running state is like the Method 2 said. But the master-shutter state is like Method 1, its initial shutter state is fully opened. The shutter durability and dust will be no longer issues.

In 1995 - 2000, In 35mm FSLR cameras had provided the 16:9 masked film mode, namely a set of auxiliary blades was in the foreside of shutter.

Posted 1 year ago #
jonnyapple

Goldfingers
Joined: May '09
Posts: 3,400
offline

PeterZheng said:
In 1995 - 2000, In 35mm FSLR cameras had provided the 16:9 masked film mode, namely a set of auxiliary blades was in the foreside of shutter.

I remember APS film cameras doing this and thinking it was a complete waste of the rest of the film's area.

I'm pretty sure method 1 is possible. The Sony NEX-5N I just got has an electronic front-curtain shutter option that is the default behavior (not electronically controlled, but electronic). On the NEX-5, the shutter had to be closed then opened again which hurt shutter release time and made for an ugly (IMO) shutter sound. The 5N is snappy when prefocused in this mode (they claim 0.02s shutter lag), and could benefit from on-sensor AF.

I should also mention that over a year of use with an NEX and changing the lens fairly often, I haven't had serious problems with dust. This is true even though the shutter isn't closed while changing lenses, or at least it isn't if you have the shutter release enabled without a lens attached—which I do so I can use a lens adapter for Nikon lenses. I think we've come a long way from the first digital sensors that seemed like dust magnets (I remember having to clean the D100's sensor pretty frequently).

Posted 1 year ago #
Drab

preferred member
Joined: Aug '10
Posts: 540
offline

I was under the impression APS accomplished 16:9 mode by tagging the negative, telling the printer to crop, not by actually cropping the negative itself. That would have involved mechanical complexity and therefore cost, two things anathema to the APS concept.

Posted 1 year ago #
PeterZheng

member
Joined: Mar '11
Posts: 16
offline

sorry, the advanced APS films did not relate to the 35mm FSLR camera's the 16:9 masked film mode.

Posted 1 year ago #
PeterZheng

member
Joined: Mar '11
Posts: 16
offline

Also, you may notice the Live-View mode (Nikon d7000 or other), once the release button has been taken a half pushed, then the mechanical shutter will be a full opened state. About the Method 2.

Posted 1 year ago #
jonnyapple

Goldfingers
Joined: May '09
Posts: 3,400
offline

I'm pretty sure that is how APS cameras let the printer know the image was panoramic, Drab. I wasn't being very clear, but there were some cameras that had a mask in the viewfinder that would show up when the camera was in panoramic mode to help with composition. To mask it on the film itself the stop would have to be placed at the image plane or the edges would be blurred.

Posted 1 year ago #