Rumor of a new Hasselblad 100 megapixels back

[synn]

What about a 100MP Leaf back?  I've always liked Leaf files better.

I am sure that's coming. Wasn't that long after the IQ 250 that the Credo 50 was released, right?

[torger]

As per the 50c CMOS despite it's resolution the chip is different from the others 50Mp from Phase and Pentax, also the resolution is different,  this because hasselblad have a different aproach on color filters and a different blind pixel mask that permit a more precision in noise cancelling / long exposure.

This sounds very interesting. It would be interesting to know what chip customizations Sony offers. Different blind masks (that is the frame around that contains dark pixels used for calibration of black level) seems like a very basic thing, IR filter glass is basic too of course, but if they're really customizing the CFA response that would be most interesting.

I've never had the opportunity to compare side by side shots with colorcheckers of the different 50MP cameras, if I had I could report something about the hardware color differences. For now I have just assumed that they are small or even equal.

Anyway as stated before, Leaf and Hasselblad 100MP backs are almost 100% certain so any rumors/leaks are not really necessary :-). Less certain but quite likely is that there will be a CFV 100MP back, and unlikely but most interesting would be if Pentax would make a body with this sensor.

[Theodoros]

I've never had the opportunity to compare side by side shots with colorcheckers of the different 50MP cameras, if I had I could report something about the hardware color differences. For now I have just assumed that they are small or even equal.

Anders, EIZO monitors are all automatically calibrated (there is a written agreement/colaboration between the two firms) as to accurately present the color of Hasselblad sensors (despite the technology -CMOS or CCD- of the sensor) and Hasselblad developed further as well as simplified the color calibration method of (again all) their backs as to be 100% in line with the (famous for color accuracy) EIZO monitors... http://www.hasselblad.com/products/phocus-2-9-colour-calibration/  the result is amazing... One doesn't have to use Cmos or CCD to see it, color of Hasselblad backs is really TOP quality for accuracy and neutrality and its the same with all their backs despite technology, age or sensor size... One could claim during the past years that Hasselblad (or Imacon) backs where (close) second to Sinarbacks for color accuracy... It's not the case any more... they now are at least on par if not on (close) top.

That said, if one have compared a CF-39 to a P-45+ in the past, the difference for color accuracy and neutrality (in favor to Hasselblad) was both clear and obvious, the P-45+ in comparison was "translating" color in a manner that maybe pleasing or not to ones taste, but it certainly was not neutral or accurate. Clearly, color accuracy or neutrality was never P1's leading field despite being to the likes of customers.

[BAB]

Another rumor of the new arrival of the Hasselblad 100c is arrived from my dealer,
They said the developement of the new camera is at advanced stage.

As per the 50c CMOS despite it's resolution the chip is different from the others 50Mp from Phase and Pentax, also the resolution is different,  this because hasselblad have a different aproach on color filters and a different blind pixel mask that permit a more precision in noise cancelling / long exposure.
They are waiting the new 100Mp CMOS Made with their specs to switch between prototype Phase to the market.

There is no other news, they said the new camera would be annunced and presented on Photokina'16
That's all.

I was right with you until you mentioned Photokina '16 this date would certainly be 6 months to late for a meaningful release of a new product, going against all reasoning. After an ambious campaign selling all remains backs in stock to clear they way for a new inventory of a competitive product line why would any company then wait 6 months to release a new offering?

[mi-fu]

I also believe this rumor is true.

The problem of the new 100MP sensor to Hasselblad is: LENS.

The more modern HCD line, I believe, is not quite enough to cover the 100MP sensor's size. With the HC series, only the 50 and 120 got new designs. the 150N seems to be a coating improvement only. I have the 50II, absolutely fantastic. But Hass will need to improve their well-acclaimed 100mm. Color fringe is serious.

Like Torger said, I think Hass at least needs to improve the screen. True focus actually is still very fine. I like the ergonomics of Hass. It is solid, professional, proven-technology. Touch screen or not, I don't really care. (maybe it is more useful for tech cam users).

If there would be a 100MP CFV, like the double of the current 50C price, many people will be happy. But I doubt it would happen.

All in all, the 100MP H won't be that cheap either. Probably not as pricy as Phase. I can hardly believe it will be lower than 30k USD. (still much lower than Phase though).

[Wayne Fox]

Another rumor of the new arrival of the Hasselblad 100c is arrived from my dealer,
They said the developement of the new camera is at advanced stage.

As per the 50c CMOS despite it's resolution the chip is different from the others 50Mp from Phase and Pentax, also the resolution is different,  this because hasselblad have a different aproach on color filters and a different blind pixel mask that permit a more precision in noise cancelling / long exposure.
They are waiting the new 100Mp CMOS Made with their specs to switch between prototype Phase to the market.

There is no other news, they said the new camera would be annunced and presented on Photokina'16
That's all.

Interesting statements.  any sources to back it up?

[ErikKaffehr]

Hi,

I would suggest that almost all HC lenses are very good. Much better than the older Zeiss lenses.

But, to make use of those lenses a very careful workflow is needed, including optimal apertures, mirror lock up and dead accurate focusing.

I have shot a few of my Hasselblad V lenses on a 4.5 micron sensor and they work just fine. Perhaps they don't deliver the crispness of the HR digitars, but they can definitively produce a very decent image on a 100 MP MF sensor.

Best regards
Erik

I also believe this rumor is true.

The problem of the new 100MP sensor to Hasselblad is: LENS.

The more modern HCD line, I believe, is not quite enough to cover the 100MP sensor's size. With the HC series, only the 50 and 120 got new designs. the 150N seems to be a coating improvement only. I have the 50II, absolutely fantastic. But Hass will need to improve their well-acclaimed 100mm. Color fringe is serious.

Like Torger said, I think Hass at least needs to improve the screen. True focus actually is still very fine. I like the ergonomics of Hass. It is solid, professional, proven-technology. Touch screen or not, I don't really care. (maybe it is more useful for tech cam users).

If there would be a 100MP CFV, like the double of the current 50C price, many people will be happy. But I doubt it would happen.

All in all, the 100MP H won't be that cheap either. Probably not as pricy as Phase. I can hardly believe it will be lower than 30k USD. (still much lower than Phase though).

[landscapephoto]

It would be interesting to know what chip customizations Sony offers. Different blind masks (that is the frame around that contains dark pixels used for calibration of black level) seems like a very basic thing, IR filter glass is basic too of course, but if they're really customizing the CFA response that would be most interesting.

All sensor manufacturers can customise the CFA response. You will find that the primaries are different between the same chip used in different brands.

[eronald]

All sensor manufacturers can customise the CFA response. You will find that the primaries are different between the same chip used in different brands.

Data, please

[Wayne Fox]

All sensor manufacturers can customise the CFA response. You will find that the primaries are different between the same chip used in different brands.

Even a low volume run like these  MF chips?  seems customizing for each manufacturer would make it hard to produce any kind of a volume that would make it attractive for Sony to produce.

I could see sony doing it for Canon or Nikon or themselves, but not sure MF makers are moving enough product.

[ErikKaffehr]

Hi,

Sensors don't have primaries, but they do have a spectral response. Monitors and colour spaces have primaries.

My guess is that sensor makers use a limited number of compounds but can choose the balance between the channels.

Best regards
Erik

All sensor manufacturers can customise the CFA response. You will find that the primaries are different between the same chip used in different brands.

[landscapephoto]

All sensor manufacturers can customise the CFA response. You will find that the primaries are different between the same chip used in different brands.

An example of data (old, but it is the only one about MF I could find fast):

P 45+
R raw   3.19   -1.76   -0.43
G raw   -0.15   1.5   -0.35
B raw   0.18   -0.98   1.8

H3DII-39
R raw   3.19   -1.83   -0.37
G raw   -0.21   1.64   -0.43
B raw   0.18   -0.98   1.8

[ErikKaffehr]

Hi,

That is the compromise matrix that converts from RGB channels to internal colour space.

The two backs you mention use the both Kodak sensors and the compromise matrices are pretty similar.

Best regards
Erik

An example of data (old, but it is the only one about MF I could find fast):

P 45+
R raw   3.19   -1.76   -0.43
G raw   -0.15   1.5   -0.35
B raw   0.18   -0.98   1.8

H3DII-39
R raw   3.19   -1.83   -0.37
G raw   -0.21   1.64   -0.43
B raw   0.18   -0.98   1.8

[landscapephoto]

The two backs you mention use the both Kodak sensors and the compromise matrices are pretty similar.

The matrix is different, which means the color filters are different, although the two cameras use the same sensor.

[bjanes]

The matrix is different, which means the color filters are different, although the two cameras use the same sensor.

That is not necessarily correct. For a non Luther-Ives sensor, creating of a matrix is a compromise to minimize metameric error. Optimizing for one color can adversely affect the coefficients for another color. Also the matrix for different illuminants is different. See Jim Kasson for a discussion.

Bill

[landscapephoto]

That is not necessarily correct. For a non Luther-Ives sensor, creating of a matrix is a compromise to minimize metameric error. Optimizing for one color can adversely affect the coefficients for another color. Also the matrix for different illuminants is different. See Jim Kasson for a discussion.

ISO standard 17321 defines relatively strictly how the matrix is derived from the illuminants. I don't see how the same filters fed through the same ISO 17321 procedure would give different values for the matrix.

[ErikKaffehr]

Hi Bill,

The article you refer to is a pretty good one!

To me the compromise matrices for the P45+ and Hasselblad back are pretty similar, no really large differences. It used to be my understanding that Phase One backs are individually calibrated. If you compare matrices they may convert to a different colour space. There are many unknowns.

It is my understanding that Phase One mounts their own IR-filters. IR-filters obviously affect the compromise matrices.

So, I don't think conclusions can be drawn by just looking at  the compromise matrices. To many unknowns…

Best regards
Erik

That is not necessarily correct. For a non Luther-Ives sensor, creating of a matrix is a compromise to minimize metameric error. Optimizing for one color can adversely affect the coefficients for another color. Also the matrix for different illuminants is different. See Jim Kasson for a discussion.

Bill

[landscapephoto]

To me the compromise matrices for the P45+ and Hasselblad back are pretty similar, no really large differences. It used to be my understanding that Phase One backs are individually calibrated. If you compare matrices they may convert to a different colour space. There are many unknowns.

There aren't any unknown in the links I have cited. Everything is defined in ISO standard 17321. The output colour space is defined and the same for the two matrices. The illuminants are defined. The procedure to measure the data and estimate the matrix is defined. I don't understand what you are talking about.

[bjanes]

ISO standard 17321 defines relatively strictly how the matrix is derived from the illuminants. I don't see how the same filters fed through the same ISO 17321 procedure would give different values for the matrix.

What causes you to think that the matrices you quote were made to ISO 17321 specs. Whoever made the profiles could derive their matrices.

Bill

[landscapephoto]

What causes you to think that the matrices you quote were made to ISO 17321 specs.

The text in front of them in the links I gave is "Color matrix as defined in ISO standard 17321".