48mm x 48mm Sensor

[eronald]

So Sony and Canon can do 24x36mm, but only in the way that anyone can by buying (or outsourcing to) a suitable stepper, such as some AMSL models: by stitching. The same is probably true for Dalsa, Kodak, Tower Semiconductor in Israel (which fabbed the original sensor for the Kodak 14/n), the English foundry that fabbed the sensors for the Kodak SLR/N and SLR/C, and whoever fabs the sensor for the Nikon D3. Matsushita also used stitching to make the roughly 18x27mm CCD for the original Canon 1D.

In summary, there are many foundries capable of stitching and thus fabbing a 24x36mm or larger sensor, including third party foundries, so that a DSLR maker does not need to do it in house, but apparently no one can do it without stitching, which keeps costs high.
[a href=\"index.php?act=findpost&pid=186758\"][{POST_SNAPBACK}][/a]

My impression (rumor) is that Canon can *now* do 24x36 without stitching. It's quite possible that they designed their own one-off equipment for this, they have their own fabs and claim to have a process line optimized for the sensor CMOS chips which is better than what their competitors would get from a foundry. Canon appear dead frightened of Sony at the moment, and the reason might be another rumor that Sony too can do 24x36 without stitching. Maybe someone with close ties to the Japanese industry can tell us what's going on.


PS. I shall write up the Poisson stuff later for those who are interested.

Edmund

[mcfoto]

QUOTE(EPd @ Apr 3 2008, 08:20 PM)
No, Dalsa specifically makes large sensors for photographical use. When you buy their sensors you can choose between three different qualities. {snip} C-grade is near-outtake and I suspect these are the sensors used by Mamiya for their ZD back.
*

Hi
Having been a happy owner of the ZD for 2 years the ZD uses the same Dalsa chip that is in the Leaf Aptus 22.  http://www.dalsa.com/news/news.asp?itemID=165
Denis

[canmiya]

QUOTE(EPd @ Apr 3 2008, 08:20 PM)
No, Dalsa specifically makes large sensors for photographical use. When you buy their sensors you can choose between three different qualities. {snip} C-grade is near-outtake and I suspect these are the sensors used by Mamiya for their ZD back.
*

Hi
Having been a happy owner of the ZD for 2 years the ZD uses the same Dalsa chip that is in the Leaf Aptus 22.  http://www.dalsa.com/news/news.asp?itemID=165
Denis
[a href=\"index.php?act=findpost&pid=187078\"][{POST_SNAPBACK}][/a]

denis,
your information is consistent with the conversation i had with the dalsa tech person last year when i was choosing between the the aptus 22, zd and e22.....

[BJL]

My impression (rumor) is that Canon can *now* do 24x36 without stitching.
[a href=\"index.php?act=findpost&pid=187070\"][{POST_SNAPBACK}][/a]

Alright, I will file that rumor/impression/belief with the one that Dalsa has a design for a 48x48mm sensor suitable for use in a DMF camera, waiting for a customer.

But since no official statements and publications of Canon and Dalsa that I know of indicate such a thing, and in Canon's case two white-papers offer the opposing idea that the need for stitching is imposed upon Canon by the 26x33mm maximum field size of current steppers, I will remain skeptical until I see corroborating information, because in particular rumors that correspond strong wishes or fears should be treated with particular skepticism. (I do mean skepticism, as in "show me why I should believe that", not cynicism as in " that's not true and you are a fool for believing it".)


The price and spec's of the much anticipated replacement for the 5D, and any related Canon white-paper, might be a key source of information. The price of the forthcoming Sony 24x36mm flagship model could also tell something about how low the price of a camera using a stitched 24x36mm sensor can be.

... another rumor that Sony too can do 24x36 without stitching. Maybe someone with close ties to the Japanese industry can tell us what's going on.
[a href=\"index.php?act=findpost&pid=187070\"][{POST_SNAPBACK}][/a]

Is Sony close enough to the Japanese industry? Sony has recently said that it needs to use stitching to make the sensor for its forthcoming flagship, citing the same 26x33mm restriction as Canon has. So this seems to be another false rumor.

[BJL]

Given that Kodak is currently the principal sensor supplier for both the Hasselblad and Mamiya systems (including Phase One backs), Dalsa is left with the Hy6 as its main DMF platform
[a href=\"index.php?act=findpost&pid=186572\"][{POST_SNAPBACK}][/a]

I should clarify one thing here. I believe that Mamiya uses (previous generation) Dalsa sensors in its ZD bodies and backs, but I suspect that those will be phased out in favor of Phase One backs, which use Kodak sensors. So the Mamiya/Phase One alliance is potentially further reducing Dalsa's access to 645 systems, pushing it more towards doing what it can with the 56x56mm based Hy6 system.

[eronald]

Alright, I will file that rumor/impression/belief with the one that Dalsa has a design for a 48x48mm sensor suitable for use in a DMF camera, waiting for a customer.
[a href=\"index.php?act=findpost&pid=187105\"][{POST_SNAPBACK}][/a]

I don't think there's anything particularly hard about the design here, as they already have modular sensor setup. Of course that's just conjecture on my part, the last and only time I designed a chip was in the eighties. But even then, you just designed the base cells and then got the automation to iterate and lay the whole thing out.

Edmund

[BJL]

I don't think there's anything particularly hard about the design here ...
[a href=\"index.php?act=findpost&pid=187112\"][{POST_SNAPBACK}][/a]

I will not disagree: scaling up sensor designs is probably relatively easy, apart from scaling up the fabrication procedure, in particular the stitching process needed. Commercial viability is probably mostly a matter of the extra cost in relation to the demand for the squarer images and/or the avoidance of camera or back rotation for verticals.

[BJNY]

Has anyone compared the Aptus 65S to a P30+ to see how the image quality and high i.s.o. performance is helped or hurt by the microlenses?

I ask because Steve H's i.s.o. 800 photo of the cat with the H3DII-31 is very impressive

[Ray]

But since no official statements and publications of Canon and Dalsa that I know of indicate such a thing, and in Canon's case two white-papers offer the opposing idea that the need for stitching is imposed upon Canon by the 26x33mm maximum field size of current steppers, I will remain skeptical until I see corroborating information, because in particular rumors that correspond strong wishes or fears should be treated with particular skepticism. (I do mean skepticism, as in "show me why I should believe that", not cynicism as in " that's not true and you are a fool for believing it".)

[a href=\"index.php?act=findpost&pid=187105\"][{POST_SNAPBACK}][/a]

BJL,
Having now read with a clear head your exellent explanation of the reason why the number of reject chips increases exponentially with the area of the chip, there might seem to be little economic incentive to invest in the development of a stepper which can accommodate a larger chip in the one process. Any savings in economy flowing from such a development are bound to be at least partially (and probably significantly) offset by the hugely greater number of rejects.

If there is a significant cost savings to be made by applying the lithographic process to the full chip size, one might wonder why Canon have not already given us a new format, 26mmx33mm, which has the benefit of being closer to the 4/3rds format, has very nearly the same area as 24x36 and very nearly the same diagonal.

Most 35mm lenses should be able to accommodate such a format. Vignetting in the corners would be no more of a problem because the diagonal of a 26x33 format is just a millimetre or so less than the standard 35mm format.

I would think that any development of a larger stepper would have to occur in tandem with a significant reduction in the number of rejects resulting from improvements in other areas of the chip making process.

[BJL]

... one might wonder why Canon have not already given us a new format, 26mmx33mm
[a href=\"index.php?act=findpost&pid=187134\"][{POST_SNAPBACK}][/a]

Ray,

I have wondered about the same thing! In fact I speculated that Nikon might try some such approach to produce a sensor of roughly the same area as 24x36mm and potentially usable with the same lenses as it has about the same diagonal.

Here are some possible problems
1) mirror and shutter assemblies would have to be redesigned for the extra 2mm of height
2) some high end lenses might have tight 24x36mm anti-flare baffles, also a bad fit for 26mm frame height.
3) that 26x33mm region must include some "non-imaging" parts of the sensor: the non-imaging  pixels around edge, and the A/D convertors on a CMOS sensor. (Especially now that Sony is putting an A/D convertor atop each column of pixels on its CMOS sensors, for higher frame rates and such.)

For all those reasons, limiting frame height to the current 24mm might be far more cost effective, so I proposed 24x32 or 24x30.

But item (3) might require trimming more. Pessimistically, it might be that Canon stayed with exactly the same 18.7x28.1mm mm in the 1DMkIII sensor as in the MkII because this is up against the height or width restrictions of the 26x33mm stepper field size. Canon has in fact implied this. The gap is 7.3mm vertically (26-18.7) and 4.9mm horizontally, so it might be that about 4.9mm is needed in each direction and the width is already maxed out. That would allow only a small vertical growth to about 21.1x28.1mm.

Although this would give the 4:3 that I and many MF enthusiasts prefer to 3:2, there would still be a noticeable vertical crop of 1.14x and horizontal crop of 1.28x from the 24x36mm frame of current 35mm format lenses, so not helping much with recovery of full FOV performance from wide angle lenses and the wide end of standard zooms.


So I suspect that any sensor format currently possible without stitching would still impose a noticeable wide angle crop on 35mm format lenses, and so every DSLR maker prefers to match "sub 35mm" sensors to their existing smaller format lenses: DX, 4/3, EF-S, etc. (Except the 1D series, where wide-angle is arguably a lower priority. But I expect the 1D to go 24x36mm eventually.)

[BJNY]

See page 21 of F&H's e-magazine
http://www.franke-heidecke.net/files/image...n_Poetry_EN.pdf
stating Hy6 as Full frame 6x6 medium format camera with the option of using future 48x48mm sensors.

[BJL]

See page 21 of F&H's e-magazine .... stating Hy6 as Full frame 6x6 medium format camera with the option of using future 48x48mm sensors.
[a href=\"index.php?act=findpost&pid=195135\"][{POST_SNAPBACK}][/a]

That seems only to be making, in an ambiguous way, the obvious point that:
IF in the future sensors in 48x48mm format and suitable for MF photography become available
THEN these 6x6 based cameras will be able to use them, while 645-based cameras will not.
I very much doubt that that F&H has suddenly become the first company to announce that such sensors are in fact coming, before any sensor maker like Kodak or Dalsa has announced such a product.

Square sensors continue to be offered, but in recent years the only new models have been monochrome designs adapted to other uses like X-ray equipment.


To be cynical, the Hy6 partners would of course like potential customers to believe that 48x48mm DMF backs are likely to come, and thus to prefer the Hy6 over 645-based alternatives, but this sort of talk about unannounced future larger square sensors could well be just marketing-driven obfuscation.

[thsinar]

Dear BJL,

I don't know where and when Sinar (or Leaf, if I can allow myself speaking for them), or even F&H have started a hype about this bigger 48x48 sensor coming: all these companies did was to mention the readiness in their respective Hy6/AFi brochures, which is a fact. I for myself have mentioned it once.

None of these company has started a talk about it, none of it has "played" on this to lure potential customers, at least not Sinar and myself.

I do not see here any "marketing-driven obfuscation" and wonder why talks/rumors/discussions by photographers in threads here and elsewhere are turned into being a company's marketing communication.

Best regards,
Thierry

To be cynical, the Hy6 partners would of course like potential customers to believe that 48x48mm DMF backs are likely to come, and thus to prefer the Hy6 over 645-based alternatives, but this sort of talk about unannounced future larger square sensors could well be just marketing-driven obfuscation.
[a href=\"index.php?act=findpost&pid=195248\"][{POST_SNAPBACK}][/a]

[James R Russell]

Dear BJL,

I don't know where and when Sinar (or Leaf, if I can allow myself speaking for them), or even F&H have started a hype about this bigger 48x48 sensor coming: all these companies did was to mention the readiness in their respective Hy6/AFi brochures, which is a fact. I for myself have mentioned it once.

None of these company has started a talk about it, none of it has "played" on this to lure potential customers, at least not Sinar and myself.

I do not see here any "marketing-driven obfuscation" and wonder why talks/rumors/discussions by photographers in threads here and elsewhere are turned into being a company's marketing communication.

Best regards,
Thierry
[a href=\"index.php?act=findpost&pid=195332\"][{POST_SNAPBACK}][/a]

Regardless, I've yet to see an article on the HY6 that doesn't mention that it's ready for larger square sensors, so that information is coming from somewhere and those thoughts get transfered from reps, sales people, photographers and it continues.

I've seen survey's asking about square sensors and obviously someone is either planning to do it, or probably trying to decide if there is a market for it.

I doubt seriously if all those quotes could be attributed to Phase or Hasselblad and call it just a mention, or a sales point, but let's face it there is no reason to mention anything unless it's to boost sales.

The thing is, in the digital world, few of us that work with these cameras, softwares, backs, and computers really believe anything until it's on the street and ready to use, usually with all the bugs worked out.

Many of us here were early adopters and though learning and knowledge is power a lot of us did a great deal of free beta testing for the manufacturers, hoping they would get the product right, so the mention of a 48mm square sensor doesn't make me want to stand in line to be the first, but then again I'm not much for standing in line for anything.

If you, Phase, Leaf, or Hasselblad want's to get a professional's attention, then first start working on better previews and  lcd's (not just bigger), open platforms that give us the choice to maximize our investment, faster lenses, in some cases wider lenses, easier workflow, lower cost storage, bug free software, high isos, or here's a novel thought how about a native 400 iso camera back that goes to a clean 1600 iso, that has build in nd filters to drop the iso down to 25, 50, or 100, without loss of quality.

As far as square goes, this is just personal opinion, but I loathe the thought of a square format camera.  I don't shoot many album covers (are there still real album covers?), I don't hang a lot of stuff in galleries and I really don't want an intended horizontial shot of a face to be framed at the waist, anymore than I want a vertical ad to have the leg's of c-stands and flags in the frame.

JR

[thsinar]

Hi James,

Regardless, I've yet to see an article on the HY6 that doesn't mention that it's ready for larger square sensors, so that information is coming from somewhere and those thoughts get transfered from reps, sales people, photographers and it continues.

I've seen survey's asking about square sensors and obviously someone is either planning to do it, or probably trying to decide if there is a market for it.

I doubt seriously if all those quotes could be attributed to Phase or Hasselblad and call it just a mention, or a sales point, but let's face it there is no reason to mention anything unless it's to boost sales.

The thing is, in the digital world, few of us that work with these cameras, softwares, backs, and computers really believe anything until it's on the street and ready to use, usually with all the bugs worked out.
[{POST_SNAPBACK}][/a]

I have not seen any specific article other than the official (Sinar) brochure mentioning this purposely and with insistance. THAT information, like you call it, is a FACT: this camera is simply ready for the future, whatever this future is or will be. It is certainly not a shame or being disrespectful to mention it with ONE single sentence in a brochure, is it?

Nobody I know from "my" distributors, nor our sales people is ever mentioning it, so far I could hear and experience up to now. I also doubt you could "accuse" Sinar, in the past and present, of  purposely having created rumors about new products, of having spoken of things which do not exist or were not planed or well advanced in its development.

Unlike you James, I am with the belief that the subject of a bigger sensor has been debated and discussed here by many, but none of the mentioned manufacturers.

If you, Phase, Leaf, or Hasselblad want's to get a professional's attention, then first start working on better previews and  lcd's (not just bigger), open platforms that give us the choice to maximize our investment, faster lenses, in some cases wider lenses, easier workflow, lower cost storage, bug free software, high isos, or here's a novel thought how about a native 400 iso camera back that goes to a clean 1600 iso, that has build in nd filters to drop the iso down to 25, 50, or 100, without loss of quality.
[a href=\"index.php?act=findpost&pid=195366\"][{POST_SNAPBACK}][/a]

- Please have a look at our new display: it is not only bigger, but improved in terms of contrast, color rendition, brightness.

- the Sinarbacks are open to all technically possible platforms, the opposite, means all platforms are not open to the Sinarbacks.

As for high ISO,  I have mentioned it 4 times already here: go to the following link and check out the capability of the new eMotion 75 with new sensor board:

[a href=\"http://forum.getdpi.com/forum/showthread.php?t=1683&page=6]http://forum.getdpi.com/forum/showthread.php?t=1683&page=6[/url]

Best regards,
Thierry

[James R Russell]

Hi James,
I have not seen any specific article other than the official (Sinar) brochure mentioning this purposely and with insistance. THAT information, like you call it, is a FACT: this camera is simply ready for the future, whatever this future is or will be. It is certainly not a shame or being disrespectful to mention it with ONE single sentence in a brochure, is it?

Nobody I know from "my" distributors, nor our sales people is ever mentioning it, so far I could hear and experience up to now. I also doubt you could "accuse" Sinar, in the past and present, of  purposely having created rumors about new products, of having spoken of things which do not exist or were not planed or well advanced in its development.

Unlike you James, I am with the belief that the subject of a bigger sensor has been debated and discussed here by many, but none of the mentioned manufacturers.
- Please have a look at our new display: it is not only bigger, but improved in terms of contrast, color rendition, brightness.

- the Sinarbacks are open to all technically possible platforms, the opposite, means all platforms are not open to the Sinarbacks.

As for high ISO,  I have mentioned it 4 times already here: go to the following link and check out the capability of the new eMotion 75 with new sensor board:

http://forum.getdpi.com/forum/showthread.php?t=1683&page=6

Best regards,
Thierry
[a href=\"index.php?act=findpost&pid=195387\"][{POST_SNAPBACK}][/a]

theirry,

I suggest you get one of those rubber Ronald Regan masks and go into dealers that sell digital backs and start talking about "why a square format camera" and you'll hear a lot of "don't let this out but square sensors or coming".

Then buy one of those burner cell phones over on Canal street and start calling dealers worldwide asking the same questions.

You may have only mentioned it once, but it's out there.

In fact the rumor I keep hearing, (multiple times) and once again I qualify this as rumor, is that the reason Leaf didn't make a rotatable back for the HY6 is because they assumed square sensors were on the way.

Once again, don't kill the messanger, that's just what people have said.

Everybody gives blad a lot of grief for saying there camera is "full frame" but by the standards you just described, there is nothing wrong with explaining the possibilities as they see it.

As far as iso on the Sinar, don't have a clue but your still talking about using better electronics to help boost the iso, your not talking about new sensor technology.

What I mentioned was why not a native high iso sensor that has clip in nd filters between the back and the body.   That way you'd get real clean high iso and the ability to shoot low iso when needed?

As far as the lcd's and previews go, how about some screen shots of the new Sinar lcd's next to a dslr?


JR

[EricWHiss]

To each his own I guess.  I for one would buy a larger square sensor. I like the square format.

[thsinar]

James,

for once I agree to disagree with you: it does not happen that often.

I can assure you that we haven't (understand Sinar hasn't) "briefed" our dealers/distributors about new coming (square/bigger/larger/higher resolution, etc ...) sensors: it is not our style.
May be those dealers/distributors have been brainwashed by all the calls/talks?

I wish I could refer to some of the (numerous) PMs I have got asking me about this. They would all tell you that  not only did I not give any such indication which could lead to the belief that it will come, but I even asked those people not to bet on it.

I also sincerely doubt that such a single sentence in a brochure could alone lead sensed and mature people to the belief that it is on the way.

I have nothing against Blad speaking or claiming about "full frame", and have never blamed them for this (I would not allow myself to do so). But I have my idea and my thoughts about this, and if asked I can even speak it out.

I got your suggestion about a native 400 ISO: there are different approaches, I guess, to get this. For the time being, Sinar has made it possible to have a very "decent" ISO 800 file, with the existing sensors, and I won't speculate about what the future will be, since I don't know.

Best regards,
Thierry

theirry,

I suggest you get one of those rubber Ronald Regan masks and go into dealers that sell digital backs and start talking about "why a square format camera" and you'll hear a lot of "don't let this out but square sensors or coming".

Then buy one of those burner cell phones over on Canal street and start calling dealers worldwide asking the same questions.

You may have only mentioned it once, but it's out there.

Everybody gives blad a lot of grief for saying there camera is "full frame" but by the standards you just described, there is nothing wrong with explaining the possibilities as they see it.

As far as iso on the Sinar, don't have a clue but your still talking about using better electronics to help boost the iso, your not talking about new sensor technology.

What I mentioned was why not a native high iso sensor that has clip in nd filters between the back and the body.   That way you'd get real clean high iso and the ability to shoot low iso when needed?
JR
[a href=\"index.php?act=findpost&pid=195391\"][{POST_SNAPBACK}][/a]

[thsinar]

Would love to do it, James, but I don't have currently a dslr in my hands. If it happens that I cross one (which should not be that difficult!), I'll d it.

Thierry

theirry,

As far as the lcd's and previews go, how about some screen shots of the new Sinar lcd's next to a dslr?
JR
[a href=\"index.php?act=findpost&pid=195391\"][{POST_SNAPBACK}][/a]

[eronald]

As far as square goes, this is just personal opinion, but I loathe the thought of a square format camera.  I don't shoot many album covers (are there still real album covers?), I don't hang a lot of stuff in galleries and I really don't want an intended horizontial shot of a face to be framed at the waist, anymore than I want a vertical ad to have the leg's of c-stands and flags in the frame.

JR
[a href=\"index.php?act=findpost&pid=195366\"][{POST_SNAPBACK}][/a]

James,

 We're back to the discussion of which aspect ratio works best for what end.

 I find that a square camera tends to compose best when held perfectly horizontally while a rectangular SLR gives all these interesting handheld angles.

 It's interesting how aspect ratio-and angles work together.

 As female models get closer and closer to giant squids with those interminable matchstick legs, tentacle arms and no waist, it does get harder and harder to show all of them in a frame

Edmund