Sorry guys, but it's probably not a glass prism

dkloi · Sep 5, 2007

David Kilpatrick said:
David Kilpatrick said:

Well, there is one way of decreasing moire and not losing too much
resolution and that is by a frequency dependent optical low pass
filter. If the red and blue components can be smeared more than
green, then chroma spatial resolution is sacrificed in return for
better moire resistance but luminance resolution would be less
affected. It'll come down to actually testing the cameras to see just
how well they have designed the AA filter.

Click to expand...

Minolta already had that patented under another use. It was used in
the Portrayer Set P, a layer-coated soft focus filter which used
selective scattering of green light to soften defects in skin tones
without losing the definition created by red and blue. A
diffraction-based, thin film deposition, portrait filter.

Neat.

David Kilpatrick said:
Reverse the concept, leave green intact and scatter red and blue, and
you have their AA filter, I guess.

Matsushita have patented it under US5280388 unfortunately :-(. I thought I was going to make my millions.

Abstract of US5280388

A phase grating optical low-pass filter having wavelength selectivity consisting of a plurality of layers which are the same in refractive index at a certain wavelength and different from one another in refractive index dispersion, and has a configuration for generating a phase difference at a boundary of the layers.

David Kilpatrick said:
OCLI eat yer heart out!

Optical Coating Laboratory Inc?

Cheers,
Daniel.

Alpha Doug · Sep 5, 2007

David,

You and I both know that this stuff gets written by non technical wordsmiths, who try to make the technical jargon sound yummy. I suspect (but can't prove yet) that what they mean is that the coating on the prism is "internal" in the sense of being "inside" the camera, rather then "external". I know that sounds a little silly, but I've worked with these advertizing guys before. I believe the camera has a nice bright pentaprism viewfinder (what else would Minolta guys use?) and they coated it to reduce reflexions. Just as with a lens surface.

But.....you could be right that it is mirrors in the shape of a pentaprism.
--
Only my opinion. It's worth what you paid for it. Your mileage may vary! ;-}
http://www.dougwigton.com/

cgarrard · Sep 5, 2007

Great post pic.

Ahh, what a sigh of relief this camera is for this forum... for now.

Carl
--
I'm here: http://www.carlgarrardphotography.com

b shaw · Sep 5, 2007

Daniel,

Based on a little homework, I made a guess back in early March regarding the EXMOR CMOS chips, that is, the chip incorporates a Grating Light Valve (GLV). Although, Sony doesn't come out and say the CMOS is a MEMs device with a GLV, they allude to it in their CX-NEWS articles. (For more background, I copied the post and link below)

So my questions for you are:

Do these new chips have a MEMs structure / GLV on them? If so, are they including a mechanical shutter in the A700 only because of their concern that photographers may be reluctant to adapt to the new technology?

If these chips don't have it, is Sony going to use the GLV concept just for video cams? Or, am I way off base altogether?

At your convenience, please advise.

b shaw

http://forums.dpreview.com/forums/read.asp?forum=1037&message=22308053

b shaw said:
GiorgioPM... yes, you are right, it's an "electronic shutter" More
precisely, a kind of a Micro Electro-Mechanical shutter called a
Grating Light Valve.

Importantly this really fast GLV shutter easily allows 60 fps, with
proper design allows live-view from sensor, and with proper DSP
allows noise reduction.

WARNING - GEEK-O-RAMA BELOW ..... "Danger, Will Robinson, Danger"
This following is kinda' technical, only read this if you want more
background info on how the new chip probably works.
A few years ago Sony introduced a Grating Light Valve (GLV) for
things like a HDTV operating at a 60 Hz refresh rate, etc. They
licensed this technology from Silicon Light Machines. (SLM)

At the time, they hinted that "this functionality as a new platform
technology to I/O components and all types of sensors." ....hmmmm...
maybe CMOS imaging sensors.

For a technical discussion on how this GLV works... see this paper
"Overview and applications of Grating Light Valve based optical write
engines for high-speed digital imaging" from the boys at SLM.
http://www.siliconlight.com/webpdf/SPIE%20Jan04%20SLM%20WrtEng%205348-5%20.pdf

Also, part of the SLM patent says this...Light diffracted from the
pixel units of the GLV array will be directed through the aperture
... to generate an image that can be viewed by the eye of an
observer, input to a camera, or projected onto a screen. The image
will be full color and can either be static for a fixed or selectable
duration, or dynamic in that it changes with time and can even be a
video-type image.
http://www.freepatentsonline.com/6219015.html

With all of this in mind, review sony's CX-NEWS articles on High
Speed Imaging CMOS devices and you'll see how all of this ties
together.
http://www.sony.net/Products/SC-HP/cx_news/vol47/index.html
http://www.sony.net/Products/SC-HP/cx_news/vol43/index.html
Although, they don't come out and say the CMOS is based on a GLV,
they allude to it in their sidebars and related articles.

Lastly, they have just trademarked the name EXMOR... remember MOR
usually stands for 'Model Order Reduction' which is part of the way
someone would model the performance of a MEMS device.

Note that Sony's EXMOR trademark in the US was designated to apply
to: "CMOS sensors; video cameras equipped with CMOS sensors; video
cameras; digital still cameras equipped with CMOS sensors; digital
still cameras; mobile phones equipped with CMOS sensors; mobile
phones"

So, bottom line...
yes 60 Hz which translates to 60 fps,
and yes it may be possible to have a live view, from the sensor.
Moreover, adding some DSP to combine to sequential images can reduce
noise etc....

IF this is how their doing it, and
IF they can build an APS-C sized sensor, and
IF they bundle this into an Alpha body,
All I can say is "WOW",
This really will be industry shifting when they release it.

b shaw

PS. All of this is my guess based on publicly available information.
I don't have an NDA with them, and I may be wrong anyway.

Cliff WYS · Sep 5, 2007

There is mention of low dispersion glass in second sentence.

http://www.sony.jp/products/Consumer/dslr/products/body/DSLR-A700/feat03.html
Halfway down the page.

クリアで明るいファインダーを実現するため、光学ガラスペンタプリズムを搭載。接眼光学系を高屈折率低分散ガラスで構成し、ARコーティングを施すことでファインダー内のゴーストも低減。収差を抑えた大きく見やすいファインダーを実現しています。フォーカシングスクリーンには、ピントの山がつかみやすいスフェリカルアキュートマットを採用し、MF時も快適なピント合わせができます。

Crude babel fish translation:

Because the bright finder is actualized with clearing, loading the optical glass pentaprism. High refractive index it forms ocular optical system low with the dispersed glass, it decreases also the ghost inside the finder by the fact that AR coating is administered. The largely easy to see finder where aberration was held down is actualized. The mountain of focus adopts the clamp easy スフェリカルアキュートマット to the focusing screen, at the time of MF the comfortable focus adjusting is possible.

--
Cliff

PhotoTraveler · Sep 5, 2007

This settles it for good.

Prismatic · Sep 5, 2007

Phil's Review (actually Simon's, who works for Phil), shows on the specs page the following line:

Viewfinder * • Eye-level fixed optical glass pentaprism

So I hope this puts the matter to rest
--
Prismatic (formerly: kodakrom)

David Kilpatrick said:
Because the leaked specifications clearly state it has a new INTERNAL
COATING. That mean's its mirror prism, with improved reflective
surfaces. You can not coat glass internally.

Now of course they could have got the wording wrong, but the other
specs which used the 'roof mirror pentaprism' term (Sony's language
for a porroprism) made it unlikely it would be a solid glass prism,
and this one just backs that up.

So don't get annoyed and disappointed if they turn out to have made a
brighter, better mirror prism and not fitted a glass prism.

David

--
Prismatic (formerly: kodakrom)

dkloi · Sep 6, 2007

b shaw said:
Based on a little homework, I made a guess back in early March
regarding the EXMOR CMOS chips, that is, the chip incorporates a
Grating Light Valve (GLV). Although, Sony doesn't come out and say
the CMOS is a MEMs device with a GLV, they allude to it in their
CX-NEWS articles. (For more background, I copied the post and link
below)

I actually looked into these things a few years back for possible use in a quantum optics experiment. Wasn't quite what we were looking for but it was an interesting use of MEMS.

b shaw said:
So my questions for you are:
Do these new chips have a MEMs structure / GLV on them? If so, are
they including a mechanical shutter in the A700 only because of their
concern that photographers may be reluctant to adapt to the new
technology?

GLVs work in reflection, not transmission so it'd be hard to incorporate it as a shutter on top of a sensor. You could use in an optical system where it was a reflective element which could be turned off. In this way, it is similar in effect to the TI DLP micromirror MEMS devices, though it can operate at higher speeds due to the mechanism of operation (a difrractive grating elements only needs to move fractions of the wavelength of light compared to swinging a mirror).

b shaw said:
If these chips don't have it, is Sony going to use the GLV concept
just for video cams? Or, am I way off base altogether?

I can see it being used for projection purposes, though I'm not so sure about how best to incorporate it into a sensor.

b shaw said:
At your convenience, please advise.

No problem at all since I'm already familiar with them, just haven't followed them in a few years so it an excuse to catch up.

Cheers,
Daniel.

David Kilpatrick · Sep 6, 2007

Alpha Doug said:
But.....you could be right that it is mirrors in the shape of a
pentaprism.

No, it's definitely solid glass - I've seen multiple websites today and downloaded as much info as possible. There's some very mixed up wording generally about many aspects of the camera.

For example, no two Sony sites seem to agree on how many raw shots it can take in sequence - 17, 18, 19 or ImageResource's tests which apparently have it not managing any more than 13!

David

David Kilpatrick · Sep 6, 2007

dkloi said:
Optical Coating Laboratory Inc?

Yes, about an hour's drive from your university Daniel - Tayside based. All the facilities you need for testing new coating types. Make you a hot mirror for any cutoff. They made me a couple of really useful IR filters for the early Leaf Lumina back in 1995, free of charge for the sake of the experiment.

David

David Kilpatrick · Sep 6, 2007

Cliff WYS said:
There is mention of low dispersion glass in second sentence.

That's for the eyepiece optics - the magnifying glass bit. Looks as if has helped them get 3mm more eye relief than the best of the new Nikon and Canon offerings (at 22mm)

David

Barry Fitzgerald · Sep 6, 2007

A-300 ??
Pentamirror?
No increase in flash synch
! 4000 sec shutter.

I dunno but its time to get shot of that sony insider! they have been feeding your bad info!
--

Clint is on holiday! Soon to return! ;-)

Alpha Doug · Sep 6, 2007

Hey David,

Considering that the Sony has "detection on 11 beaches", it would be hard to imagine initial marketing hype translated from the Japanese to be very accurate!

Marketing, Marketing, Marketing.

On the other hand, I am ecstatic! This is exactly what I have been waiting for to upgrade my 7D! I am really impressed at how well they listened, and how well they have implemented so many improvements. I wish I had the money immediately, but it will have to wait for my tax return in February. Bummer!

--
Only my opinion. It's worth what you paid for it. Your mileage may vary! ;-}
http://www.dougwigton.com/

David Kilpatrick · Sep 6, 2007

Barry Fitzgerald said:
A-300 ??
Pentamirror?
No increase in flash synch
! 4000 sec shutter.

I dunno but its time to get shot of that sony insider! they have been
feeding your bad info!

Unless, Barry, the Japanese site with translated info for the 'A300' is actually referring to a genuine A300. Some of the most plausible information said 1/4000th etc. Anything I have written has a source, none of it has been invented.

You may yet see an Alpha 300 (or some other number) which replaces the 100. It might even use the new sensor.

David

b shaw · Sep 6, 2007

Daniel,

Thank you,

I think the MEMS used as a GLV was a red herring for me. It's been more than a decade since I was fab'ing GaAs FETs - so I assumed there were new fabrication tricks that Sony could employ. I won't tell you what tricks I imagined, because you, as a well respected physicist, might hurt yourself from laughing too hard at my ideas. So, you see, I'm just watching out for your well-being. ;-)

Anyway, after re-reading their material, I have another question. Do you think, they are using a MEMS structure on the back side of the sensor, as part of their parallel column A/D converter? Perhaps it's part of a resonator - perhaps, that is, in part, how they are obtaining their 60 to 300 Hz.

I hope I'm not over staying my welcome, but here are the reasons why I keep re-looking the MEMS issue:
1) the name, Ex MOR, and

2) they discussed MEMS when talking about their high speed chip, at least during the initial development, and
3) I'm geekier than I let on and this stuff still fascinates me.

b shaw

dkloi said:
dkloi said:

At your convenience, please advise.

Click to expand...

No problem at all since I'm already familiar with them, just haven't
followed them in a few years so it an excuse to catch up.

Cheers,
Daniel.

dkloi · Sep 6, 2007

b shaw said:
I think the MEMS used as a GLV was a red herring for me. It's been
more than a decade since I was fab'ing GaAs FETs - so I assumed there
were new fabrication tricks that Sony could employ. I won't tell you
what tricks I imagined, because you, as a well respected physicist,
might hurt yourself from laughing too hard at my ideas. So, you
see, I'm just watching out for your well-being. ;-)

Well, I'm not a semiconductor device physicist so I probably won't laugh too hard. I do get to talk to real condensed matter experimentalists and I'm sure they chuckle a bit at my suggestions at times. I'm lucky in that my job allows me to wander all over the map, from semiconductors, superconductors, quantum optics, ion traps, optical lattices, Bose-Einstein condensates, lasers, microwaves, NMR, ESR, buckyballs and nanotubes, though my expertise is mostly top level rather than at the nitty gritty, wrong end of a cryostat range of things

.

b shaw said:
Anyway, after re-reading their material, I have another question. Do
you think, they are using a MEMS structure on the back side of the
sensor, as part of their parallel column A/D converter? Perhaps it's
part of a resonator - perhaps, that is, in part, how they are
obtaining their 60 to 300 Hz.

From all indications, it's just standard CMOS circuitry. From the CX News 43 article, I think they're using a 75MHz input frequency to the chip, internally up-converted to 300 MHz which then drives the DAC and ADC counters. Don't know what the timings are for the IMX021, can't wait to get my hands on the application sheet.

b shaw said:
I hope I'm not over staying my welcome, but here are the reasons why
I keep re-looking the MEMS issue:
1) the name, Ex MOR, and
2) they discussed MEMS when talking about their high speed chip, at
least during the initial development, and
3) I'm geekier than I let on and this stuff still fascinates me.

I prefer these geeky discussions over stupid ones like, "The CF card on the 7D is backwards compared to all other cameras, therefore it's rubbish and I'll never buy a KM product again" type of discussions (mooted as a criticism of the 7D when it was released AFAIR).

I must have missed the discussion of MEMS with reference to parallel column ADC. Do you have an exact reference? I really can't imagine what function they could implement which would be specific to parallel column ADC. I can think of ways of using microfabrication techniques for optimisation of microlens/filter/AA assemblies, perhaps even a tunable filter array.

The only MEMS I can be sure is in the A700 are the angular rate sensors for SSS.

Cheers,
Daniel.

Barry Fitzgerald · Sep 6, 2007

David Kilpatrick said:
Unless, Barry, the Japanese site with translated info for the 'A300'
is actually referring to a genuine A300. Some of the most plausible
information said 1/4000th etc. Anything I have written has a source,
none of it has been invented.

You may yet see an Alpha 300 (or some other number) which replaces
the 100. It might even use the new sensor.

David

You know I forgot about that! Oh well ooops!

You are forgiven ;-)

--

Clint is on holiday! Soon to return! ;-)

b shaw · Sep 6, 2007

dkloi said:
From all indications, it's just standard CMOS circuitry. From the CX
News 43 article, I think they're using a 75MHz input frequency to the
chip, internally up-converted to 300 MHz which then drives the DAC
and ADC counters. Don't know what the timings are for the IMX021,
can't wait to get my hands on the application sheet.

My bad. I think I must have assumed the "Sideview" of the CX News 43 was somehow related to the Main article. I assumed it was some part of the DSP. Ultimately, with that bad assumption, I researched all the way to a bad conclusion.

However, my original hare-brained idea is not so hare-brained after all. Check out what these rocket scientists are doing. (That wasn't meant to be sarcastic, they work at NASA!) It's not a GLV, but a micro-shutter. http://www.nsti.org/publ/MSM2002/229.pdf

Although that application is complete different, I was imagining something with substantially less torsion, with Micro-ribbons flexing to a side over a micro-lens, the length of the ribbon rolled about it's longest dimension. Each end of the ribbon is attached above a micro-lens at an angle. More ribbons are necessary than if they were attached flat above each micro-lens. Upon deflection the ribbon would flex away, exposing the lens ... gosh, it would be easier to whiteboard this ...

dkloi said:
I must have missed the discussion of MEMS with reference to parallel
column ADC. Do you have an exact reference? I really can't imagine
what function they could implement which would be specific to
parallel column ADC. I can think of ways of using microfabrication
techniques for optimisation of microlens/filter/AA assemblies,
perhaps even a tunable filter array.

Seems like MEMS, and now NEMS, is still growing in applications.

dkloi said:
The only MEMS I can be sure is in the A700 are the angular rate
sensors for SSS.

I understand the micro-accelerometer application. I wonder what they use for the context awareness of the LCD menu - portrait or landscape. See bottom of page.

Thanx for your considerations, no need to reply. We should probably let this MEMS discussion pass away quiet and peacefully.

So, um.... check out the Popphoto images, taken with the A700, hand-held, inside a dark cathedral!

http://www.popphoto.com/gallery.aspx?section_id=26&section_prefix=cameras&webtrends_section=cameras&article_id=4570&window_id=1&gallery_id=1082&page_number=1&seq=1&cnt=1&slide=on

b shaw

Anastigmat · Sep 6, 2007

David Kilpatrick said:
Bengeo said:

David,

This site http://a700.ic.cz/ has screen dumps of the comparisons
screens that I saw earlier when they were live and they say Mirror
for the 100 and Pentaprism for the 700.

Click to expand...

It certainly does - so why do the specs say 'new internal coating'? A
glass prism can only have an external coating.

That is true. It is not possible to paint the inside surfaces of a solid piece of glass.

David Kilpatrick said:
Unless they mean a new
multilayer coating between the glass and the mirror deposition (the
prism doesn't actually need mirror coating to function, but most
prisms can be improved by mirrorizing the surfaces).

David

It could be a mistake in the translation or in the authorship of the specs. To err is human.

Sorry guys, but it's probably not a glass prism

Senior Member

Veteran Member

cgarrard

Guest

Leading Member

Active member

Forum Pro

Forum Enthusiast

Senior Member

Veteran Member

Veteran Member

Veteran Member

Barry Fitzgerald

Guest

Veteran Member

Veteran Member

Leading Member

Senior Member

Barry Fitzgerald

Guest

Leading Member

Forum Pro

Similar threads

Keyboard shortcuts