Why doesn't medium format have high ISOs?

[Deleted-pending]

is irrevelant, unless the bud is photojournalist who only posts thumbnails on a website or newspaper.

 

[UponQuai]

sorry, D3S is a Pro camera

 

[Tyrone Wellhung]

That means the D3s will have about 1.6x the pixel sensor area, and hence better ISO characteristics.

How do you make that out?

 

[ejmartin]

As far as I understand, medium format does not have high ISO sensitivity? (Is this true?) If so, why is that? Shouldn't medium format's super large pixels mean even lower noise?

Pixel size is not the dominant issue, rather it's CCD vs CMOS sensor design. Nikon and Canon's recent sensors have read noise in the neighborhood of 2 to 2.5 electrons per pixel at high ISO; MFDB sensors have more on the order of 20 electrons or more.

--
emil
--



http://theory.uchicago.edu/~ejm/pix/20d/

 

[Billx08]

That means the D3s will have about 1.6x the pixel sensor area, and hence better ISO characteristics.

How do you make that out?

Such an innocent appearing, very short question. What HorsePix doesn't realize that it's a trap, and when sprung he'll be inundated with reams and reams of esoteric verbiage that goes round and round and never leads anywhere that's even remotely useful.

 

[lightmagic]

FX and for FM the sensor is even bigger. If high ISO are not available, they can't do it for one reason or another. But they--the makers-- are not dumb. High ISO are big selling item. Just look a the success of D3! Maybe the makers of MF should get help from Nikon or Canon. Even better, Nikon should sell MF.

--
The Lightmagician
Sun is my eye
Winds my breaths
Sky my open Mind.
http://www.lightmagical.com

 

[Deleted-pending]

I am speaking about the user, not the camera, so I repeat myself :

Using a Pro camera at 12 400 iso is irrevelant, unless the bud is photojournalist who only posts thumbnails on a website or newspaper !

I really doubt that any pro would sell any big enlargment with noisy p&s quality shots. The only purpose of extreme iso is photojournalism, magazines & thumbnails.

 

[Joseph S Wisniewski]

I'm going to take a wild stab at this :

Ouch! What did "this" ever do to you, to deserve stabbing.

(sorry, in a humorous mood today).

It's probably because since the beginning of time, the medium format has been used in controlled (or landscape) settings with plenty of light (or time, with landscape) to nail a shot.

Actually, that's not at all true.

Way back, at the beginning of time (say up to 1940), large format was used for darn near anything that had to look good. People shot weddings on it. Reporters carried the classic "Speed Graphic", a folding 4x5 with rangefinder focusing that could shoot 6 frames from a sheet film magazine surprisingly fast.

Medium format was an amateur format, people got a box "Brownie" or a Kodak folding camera (I have an autographic 3A in my collection) for the family snaps. Yup, snaps, on 6x9 MF.

Eventually, as film quality increased and customer expectations decreased, medium format became the mainstream of journalism and event photographers. You had folding 645, 6x6, 6x7, and 6x9 rangefinders, non-folding rangefinders like the infamous "Koni-Omega", journalists who liked the TLR (and interchangeable lens TLRs to accommodate a wider range of "missions", and MF SLRs like the Hasselblad.

35mm and "tweener" formats like 828, 127, and instamatic captured the consumer market.

Film kept getting better, and soon people were actually shooting wedding and serious work on 35. The "revolution" came in 1961, when Vietnam got hot, and American and European journalists hit the place with MF gear, and saw what the Japanese were doing with the new Nikon F.

Next thing you know, Blads and Konis are littering the beeches, and the entire field of wartime journalism has converted to 35mm.

Did you know that Oskar Barnack developed the Leica specifically for landscape shooters? People who needed to travel light, but still get good results. Even with 1913 film, a 35mm landscape wasn't that bad. The high resolution "microfilm" type films were surprising. Sure, some were ISO 6, but the detail was stunning. It surprised him that it caught on with dilettantes, self styled "street shooters" and documenters of "life". But he adapted.

Hence, the camera systems have not been optimized, or even really concerned with, high ISO shooting. It would be ridiculous to do so, because one assumes that a photographer toting a $30,000+ camera around is going to have the cash neccessary to light up his own studio, or a sturdy tripod from which to shoot his lovely landscape.

Actually, that's a pretty flawed argument. LF photographers often shoot that "lovely landscape" at f64 or 128 (remember "group 64?") and MF shooters do it frequently at f22, 32, or 45 (even though that's well below the diffraction limit on even 6x9. MF shooters don't pay much attention to the science of photography. If they did, the MF industry would have collapsed even earlier than it actually did).

At f32, with 50 speed film, you're down at 1/12 sec in daylight. That's often too slow for landscapes. Even a mild breeze blurs the trees at that speed. Want a wave to look like a wave? Shoot it above 1/125.

The studio is just as bad. I have a mix of 600 and 1200 W-S strobes, with one 3200. It's easy to shoot f8-16 and ISO 200 FX with that combination. The power needed for f16-32 ISO 50 MF is insane. Ever worked with four 4800 W-S packs? Try to avoid it, it's scary. And your subject will think they've been nuked.

Perhaps there are other internal reasons, but some of the Sony guys (here and other forums) have an idea that high ISO also results in lessened color depth/richness/whatever,

Actually, that's only true when you're past the decent operating range of the sensor. Sensors near their base ISO work well. And if the base ISO is 200, instead of 50, you've just picked up a 2 stop advantage in flash power or shutter speed.

which is why low ISO Sony A850 shots are awesome

The 850 and 900 are "broken". Seriously, at least two engineers have torn them down to determine why Nikon D3X, with the same sensor, gets two stops better low light capability.

and the "color of Sony" is always mentioned as a plus for the system

No, it's not. Sorry.

(and nope, you cannot exactly match it with editing - that's because you're working with less color to begin with).

Of course you can match it with proper profiling. Your "less color to begin with" argument is fabricated.

The big tradeoff is noise performance that isn't very good.

So, in summary, my from-the-rear-reply

: High ISO performance is a moot point to MFDBs and would actually hinder their overall performance.

There's no such thing as too much capability.

--
Rahon Klavanian 1912-2008.

Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.

Ciao! Joseph

http://www.swissarmyfork.com

 

[HorsePix]

Such an innocent appearing, very short question. What HorsePix doesn't realize that it's a trap, and when sprung he'll be inundated with reams and reams of esoteric verbiage that goes round and round and never leads anywhere that's even remotely useful.

And I'm not a math tutor, so I'll leave the derivation of the pixel area as an exercise for the student There were enough details in the original post to figure out how I came up with the number. Now, if I'm wrong, and Hassey is able to somehow stick 3.25x the pixels in only 2x the area while increasing the sensor area, that would be quite something!
--Michael

 

[Joseph S Wisniewski]

As far as I understand, medium format does not have high ISO sensitivity? (Is this true?) If so, why is that? Shouldn't medium format's super large pixels mean even lower noise?

Yes, given equal sensor technology.

The problem is that the sensor tech between MF and 35mm systems is not equal. It's not even close.

The MF market is very small, and until the arrival of the Pentax 645, was shrinking at a nice, steady 30% per year, going back to the film days. 2008 saw the market at 6000 units, total, between Blad, P1, and Leaf. That's with seven different sensors, so the market for any given MF sensor is about 1000 units. Canon sells more than 1000 units of the 5D II each day (based on serial number tracking). And the sensor tech is shared with their APS bodies. That means that Nikon, Canon, etc. deal with annual volumes 1000x that of MF manufacturers.

So, who gets the good tech? Canon's got their own labs. Nikon's in pretty deep with Sony Semiconductor. Panasonic has their own labs.

Blad and P1 deal with the only two companies in the whole world who will talk to you if you come to them and say "I want 1000 units of a custom sensor", DALSA and Kodak.

Blad and P1 don't have the budget to develop new sensors, or to pay people like DALSA and Kodak to actually develop something new for them. If you look at the size of the MF market, they're each doing about 3000 units/year, mostly of the $18,000 entry level 33x44mm backs. The distributors for small quantity European goods are gougers, if half that money gets back to Blad or P1, I'd be amazed. So they've got about $27M each/year. Older European companies, 6% R&D reinvestment is actually on the high side, but we'll be generous: $1.6m per year. That's not the sensor budget, that's the whole R&D budget: lenses, camera bodies, backs, software, and sensors. If sensors get $300k, that would be amazing.

All that kind of money gets you is Kodak or DALSA taking a "standard cell" that they've created for a different customer, typically aerospace, and laying them down for 33x44mm or 36x48mm sensors.

There's two things wrong with aerospace tech.

• The aerospace timetables are insanely long. A sensor is specified in 2010 for a launch in 2018. The technology is supposed be stable for that timeframe, at the cost of performance.
• Aerospace customers don't let you spin civilian apps from the latest tech. God only knows what's actually in the labs at Kodak and DALSA today. Blad and P1 are getting declassified 2002 tech.

Literally, when you sit the MF backs down next to a FF Nikon or Canon and start measuring them, instead of listening to MF fan boys saying "everyone knows MF is better", you find that MF has lower dynamic range than FF sensors that operate at 3 stops better speed, and poorer color accuracy.

Fuji tried to change the game around 2005, they built their own MF sensor using the same "Super CCD" that they used in the S3 and S5 cameras. Wide dynamic range, good high ISO performance, state of the art color. And they built that sensor in a slightly larger size than P1 or Blad were using at the time, and gave the back nice features like one-touch power sensor rotation from horizontal to vertical format.

Blad, who was doing about $25M/year business in lenses and H bodies with Fuji in 2005, freaked, and Fuji management killed the Fuji MF back. It's a pity, because I've got sample pics, and the thing was good.

The interesting thing about the Pentax 645 is that it's a potential game changer. It's more ergonomic than anything Blad or P1 have. It's got the same processor as a Pentax K, about 3x the power of a Blad or P1. It's got an AF system that looks more like 2005 than the 1985 systems in Blad or P1 (seriously, capability and speed wise, they're just like a Nikon 8008 or Canon EOS 630), and metering that's also not stuck in the previous millennium.

It's got the potential to substantially outsell Blad and P1 put together. And Pentax is tied closely to two Asian sensor suppliers. If the thing takes off, we could see them asking Samsung or Sony to spin a MF sized version of a 2010 sensor.

wizfaq MF

--
Rahon Klavanian 1912-2008.

Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.

Ciao! Joseph

http://www.swissarmyfork.com

 

[SNRatio]

As for pixel size, smaller is only better when we go for the highest possible resolution. Before Canon passed into mini-pixel-land,

Being in 'mini-pixel-land' doesn't seem to have hurt the performance of Canon's camera. Your statement 'as for pixel size, smaller is only better when we go for the highest possible resolution' is quite wrong. Shrinking pixels is quite a good way of improving most sensor characteristics, which is probably why Canon keeps on doing it, their control over production makes it easy, and to their advantage, to steadily improve their sensors, and shrinking the pixels is the straightforward way to do it. There is a grain of truth in the notion that larger pixels are better for low light situations. Ar any technology node, making the pixel larger than normal compared with the read transistor trades base DR for high ISO DR, but that is a design decision, and certainly doesn't represent a general trend that bigger pixels are better.

there was a lot of talk about full well capacity. It is still an important concept, simply because you need at least 2^16 electrons to reach a dynamic range of 16 stops (16 bits RAW), in practice quite a bit more, because of read noise.

Well, no. If it is pixel level DR that you are talking about (this isn't clear) then it's defined by the full capacity (not necessarily the full well capacity in a MOS sensor) divided by the read noise. The read noise can potentially be a fraction of an electron, so the full capacity needn't, in theory, be close to 64k electrons. In any case, having very high per-pixel DR's is not a great engineering strategy, since it requires read electronics with a bigger DR, which can be tricky to engineer and expensive. Better to subdivide into smaller pixels each with lower DR, putting less demand on the read electronics. After all, aggregated together, those pixels will achieve a DR the same as the original.

And at 4 steps above base ISO, a full exposure generates only 1/16 of the electrons at base ISO. So when the MF sensors operate at about the same pixel size as FF sensors, they simply lack the capacity to record nuances at the pixel level way above base ISO.

Generally, CCD's have higher photoelectron saturation densities and quantum efficiencies than MOS sensors, so you'd expect, if the pixels were the same size, for the MF sensors to be working with more electrons than the 135 ones. The real problem is that the read noise of CCD sensors is much higher than that for MOS sensors. An MF sensor might have a read noise ten times that of a Canon 1DIV. It is the high read noise which limits the DR as photon counts fall.

Well, we get another lecture in sensor theology here.

By that reasoning, with read noise at, say 1/8^=2^-3 electron, we only need 32=2^5 electrons to achieve 8 stops DR!

And, as variances are additive, the variance from pooling, say 4 pixels, with read noise A is 4A. Unless the read noise of each pixel is less that 1/4 of that of a 4x bigger pixel, the total S/N-ratio will become smaller by pooling this way, resulting in less dynamic range recoverable. And, everything else being equal and random allowed to be random, this seems to be what we observe. BUT, variances won't be fully additive unless the engineers allow them to be.............

It is only for the Poisson shot noise, with variance equal to expected value, and distibutions reproducing additively, that we have this wonderful property of full additivity. Which means that by pooling, we can get the S/N-ratio higher "at no cost".

I have heard exactly this same theological exegesis before, from another nick, with the same Canon-blessings. Seemingly no capability or willingness at all to acknowledge that ALL sensor development strategies have their strengths and weaknesses, and that many of the problems with Canon's approach are widely acknowledged. In a very authoritative way, but rather void of documentation, we are told how things REALLY are. So we better not check for ourselves!

I feel this is kind of trolling, and that it defeats much of the intention of such forums. It may be a very conscious strategy from Canon, in that they have some large problems with their approach which they at present don't want to have openly and penetratingly discussed, and rather try to turn sensor discussions into theology and name-calling.

I'll just pick one example here, it is typical and very central to the theological argumentation:
" Shrinking pixels is quite a good way of improving most sensor characteristics"

True or false? For almost all practical purposes, measurements indicate that it is, at best, only partially true. It IS true for resolution, and when read noise is low, there is little sense in capturing electrons way beyond the bit depth of the RAW file. But, as soon as we enter high ISO territory, even the sampling strategies may become problematic. 6 steps above base ISO, if a full exposure at base is 32000 electrons, we have only 500 left. If we say that gives 9 stops of DR, halfway down the tonal curve we are working with 30-60 electrons. That may, for example, affect the color precision, as the Bayer filter doesn't allow us to simply pool the closest pixels. We may have to go down to a far lower resolution to get acceptable quality.

And the point is, that high ISO performance IS relevant for many applications, so any changes affecting that performance may be problematic. The "smaller is better" theology is also contradicted by the performance of the D3s sensor.

 

[Joseph S Wisniewski]

maljo

But that's not the real issue.

Rayman said his images weren't at 100%. Well, that's the neatest thing about digital imaging, the more you downsize the images, the better the high ISO looks. A Nikon D3, at web size 1024x768, about 1/4 size, looks insanely clean at ISO 125,000.

Print the images, side by side, at a decent size. That's all you need to do to see whether MF's more pixels and allegedly "better pixels" actually makes a difference. I've done this, MF and FF, on the same shoot.

I'm convinced that the fact that more MF shooters and image customers don't try side-by-side comparisons is the only thing that keeps Blad and P1 in business.

--
Rahon Klavanian 1912-2008.

Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.

Ciao! Joseph

http://www.swissarmyfork.com

 

[spectra]

They are always intelligent, well researched and thought provoking...

--
Spectras Photos:

http://www.flickr.com/photos/spectrapix/sets/72157594150842281/detail/?page=1

 

[Joseph S Wisniewski]

They are always intelligent, well researched and thought provoking...

Thanks.

Be careful, it's dangerous to feed something as large as my ego.



--
Rahon Klavanian 1912-2008.

Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.

Ciao! Joseph

http://www.swissarmyfork.com

 

[Ed Ingold]

There are fundamental differences between medium format sensors and small format sensors. The former is designed to have the highest image quality, while the latter place emphasis on sensitivity and speed of transfer. Raw data seems to be lacking from the camera and back manufactures, but this is what I've gleaned from more basic sources.

Medium format and astronomical CCDs have what is called "full frame" design. This makes efficient use of the available real estate at the expense of transfer speed. Data is transferred from cell to cell in a bucket-brigade manner, to the edge of the sensor where it is read into memory.

Most small format CCDs have an "interline" design, where every other row of cells is masked off from light. Data is transferred to an adjacent "dark" cell, which is read in the bucket-brigade method while the original cell can be receiving a new image. Variations on this technique may block off the sensor in a different manner, but the principal is the same - only half of the sensor is available for imaging.

Small format digital is nearly completely converted to CMOS technology, which is faster and cheaper than CCD, and runs cooler (i.e., less thermal noise). Much of the image processing can occur in the sensor itself, improving speed even further. Both CCD and CMOS sensors use micro-lens technology to offset the loss of efficiency of the interline structure.

The difference between MFD and a DSLR results is significant too. I have a D3 and an Hasselblad CFV-16 back. The CFV has outstanding dynamic range, both in resisting blowout and especially in shadow detail. The color is accurate and consistent, but not necessarily exciting, whereas the the D3 color is usually more saturated. The dynamic range of the CFV appears to be in the 12-13 stop range, consistent with other MFD backs. The dynamic range of the D3 is in the 7-9 stop range, consistent with tests performed by DPReview and others. The resolution of the CFV is about 50% better than that of the D3 when cropped to a 3:2 aspect ratio, yielding about the same number of active pixels. The difference is due to both the absence of an anti-aliasing filter in the CFV, and larger pixels which make better use of the lens' resolution.

In practical terms, I wouldn't bother taking the Hasselblad to shoot a play or concert (or football game, if I were inclined to attend). I prefer the Hasselblad for more "contemplative" photography, including landscapes and closeups, for reasons not limited to the potential image quality.

Medium format digital is not necessarily ISO-challenged. Phase One offers ISO 800, and up to 1600 via "binning" technology wherein the outputs of four adjacent cells are read as one. While the CFV works best at ISO 50, the results at ISO 400 are pretty good and getting better as Adobe and Hasselblad RAW conversion software improves.

 

[Joseph S Wisniewski]

There are fundamental differences between medium format sensors and small format sensors. The former is designed to have the highest image quality, while the latter place emphasis on sensitivity and speed of transfer.

That is totally incorrect, but since you go into detail, I'll dismantle the erroneous assumptions as we come to them.

Raw data seems to be lacking from the camera and back manufactures,

That is true.

but this is what I've gleaned from more basic sources.

Please cite some of those "basic sources".

Medium format and astronomical CCDs have what is called "full frame" design. This makes efficient use of the available real estate at the expense of transfer speed.

Incorrect, or at least incomplete.

While "full frame transfer" CCDs often have fill factors (ratio of active light gathering area to total cell area) higher than either "interline transfer CCDs" or CMOS or NMOS sensors, many CMOS designs have fill factors as high as typical full frame transfer CCDs.

But the designs: FFT CCD, Interline CCD, CMOS, and NMOS, have little to do with the transfer speed. CMOS sensors typically have the slowest transfer speeds of all.

Transfer speed is a function of three things:

• the switching circuitry used.
• the clock speed of the sensor in a particular application (some applications run the sensor well under max speed, in order to improve noise).
• and the number of readout channels.

The number of channels (more information on that in a minute) is the key determining factor.

Data is transferred from cell to cell in a bucket-brigade manner, to the edge of the sensor where it is read into memory.

The data can be read from as many points on the edge as are desired. Those are called "readout channels". Generally, the readout of a multiple channel sensor is structured so that every N'th column is allocated to a channel, a 2 channel chip might have columns 1, 3, 5, 7 all part of one chain, columns 2, 4, 6, 8 part of the next.

Canon uses very large channel counts on their CMOS sensors, and clocks them at quite leisurely rates. 8 channel readout, at 10 MHz clock, gets 80 megapixels/sec off the chip.

Sony uses literally thousands of parallel readout channels, clocked at very slow rates. The CMOS sensor used in Sony A850, A900, and Nikon D3X has 6,000 parallel channels, clocked (depending on the camera) at rates as low as 20kHz (not MHz).

CCDs tend to use very low channel counts, because they are often used in applications emphasizing high dynamic range, and 16 bir A/D converters are large, expensive, and difficult to keep matched (resulting in "stripes" in the image). Current digital backs use 1 or 2 readout channels, clocked at rates from 15MHz (2 channel 33mp backs) to 25MHz (1 channel 22mp backs, 2 channel 40-60mp backs).

So, the fastest CMOS is restrained to rates about half CCDs, and the slowest CMOS is literally 1/1000 the rate of CCD.

There have been CCDs with higher channel counts. Digital cinema CCDs often use anywhere from 12 to 64 channels, with individual clocks in the 15-20 MHz range, for aggregate data rates as high as 800 megapixels/second.

Scientific CCDs built for high speed imaging (explosions, etc) can have thousands of channels at clocks to 15 MHz, for aggregates in the billions or tens of billions of pixels/sec.

Most small format CCDs have an "interline" design, where every other row of cells is masked off from light.

True, for point and shoot cameras. The Nikon D1X and D100, and all Sony and Pentax models to use CCD sensors did not use interline CCDs.

Data is transferred to an adjacent "dark" cell, which is read in the bucket-brigade method while the original cell can be receiving a new image. Variations on this technique may block off the sensor in a different manner, but the principal is the same - only half of the sensor is available for imaging.

Small format digital is nearly completely converted to CMOS technology, which is faster and cheaper than CCD, and runs cooler (i.e., less thermal noise).

Actually, that's a common myth. CCDs operate in the 100mW range, and with radiating areas over 1cm2, do not heat appreciably. The heat sources in a digital camera are the main processor (or processors) and the LCD backlight, both of which are commonly in the 1-3W range. Both CMOS and CCD sensors are subject to equal thermal input from the processor and backlight.

Much of the image processing can occur in the sensor itself,

Except that it doesn't.

The IC fabrication techniques needed for CMOS sensors are still large geometry, deep etch and implantation, and do not lend themselves well to integration with the sensor. A/D converters on chip is about the limit.

improving speed even further.

It's the highly integrated CMOS chips that have the lowest speeds of all. It's the large number of parallel channels that gives them the high throughput.

Both CCD and CMOS sensors use micro-lens technology to offset the loss of efficiency of the interline structure.

Actually, many FFT CCDs on MF digital backs also use microlenses. This started happening when pixel counts hit 30MP.

(to be continued)

--
Rahon Klavanian 1912-2008.

Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.

Ciao! Joseph

http://www.swissarmyfork.com

 

[Joseph S Wisniewski]

The difference between MFD and a DSLR results is significant too. I have a D3 and an Hasselblad CFV-16 back. The CFV has outstanding dynamic range, both in resisting blowout and especially in shadow detail.

I've tested about four different MF digital backs. On the same scene as a D3, all were inferior in dynamic range.

The color is accurate and consistent,

Again incorrect. The MF sensors have very old color filter formulas, with spectral responses that are not as colorimetric (mathematically mappable to human eye responses) as modern Sony and Canon sensors.

but not necessarily exciting, whereas the the D3 color is usually more saturated.

Both MF and FF can easily be profiled into reasonably accurate results, not "more saturated", etc.

The only difficulty is that, even properly profiled, a MF back will have greater problems with observer metamerism, especially in the reds.

The dynamic range of the CFV appears to be in the 12-13 stop range, consistent with other MFD backs. The dynamic range of the D3 is in the 7-9 stop range, consistent with tests performed by DPReview and others.

Apples to oranges.

Try resting them both by the same methodology. Anything done side by side (DXO, etc) has the MD come up short.

The resolution of the CFV is about 50% better than that of the D3 when cropped to a 3:2 aspect ratio, yielding about the same number of active pixels.

However, the CFV will have more aliasing. I've had more than one MF shot ruined by that.

The difference is due to both the absence of an anti-aliasing filter in the CFV, and larger pixels which make better use of the lens' resolution.

Only if you define "better" as "worse". The D3 is dealing with sensors that match the lens. The CFV is dealing with a large crop factor.

In practical terms, I wouldn't bother taking the Hasselblad to shoot a play or concert (or football game, if I were inclined to attend). I prefer the Hasselblad for more "contemplative" photography, including landscapes and closeups, for reasons not limited to the potential image quality.

Would the other reasons be the "forces you to slow down" mantra that is often heard from photographers lacking the discipline to slow themselves down?

Medium format digital is not necessarily ISO-challenged. Phase One offers ISO 800, and up to 1600 via "binning" technology wherein the outputs of four adjacent cells are read as one.

So, by reducing resolution to below that of a D3, it can approach being just 3 or 4 stops worse

While the CFV works best at ISO 50, the results at ISO 400 are pretty good and getting better as Adobe and Hasselblad RAW conversion software improves.

The results of everything, MF, FF, and APS, get better as the raw (lower case, it's a word, it doesn't stand for anything) conversion software improves. MF stays just as many steps behind...

--
Rahon Klavanian 1912-2008.

Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.

Ciao! Joseph

http://www.swissarmyfork.com

 

[coudet]

More often they say something like:

"MF cameras have incredible clarity"

"MF cameras give me incredible richness in my files, unlike anything I have ever seen"

And a bunch of other terms that dont actually mean anything.

Literally, when you sit the MF backs down next to a FF Nikon or Canon and start measuring them, instead of listening to MF fan boys saying "everyone knows MF is better", you find that MF has lower dynamic range than FF sensors that operate at 3 stops better speed, and poorer color accuracy.

 

[rayman 2]

maljo

But that's not the real issue.

Rayman said his images weren't at 100%. Well, that's the neatest thing about digital imaging, the more you downsize the images, the better the high ISO looks. A Nikon D3, at web size 1024x768, about 1/4 size, looks insanely clean at ISO 125,000.

Print the images, side by side, at a decent size. That's all you need to do to see whether MF's more pixels and allegedly "better pixels" actually makes a difference. I've done this, MF and FF, on the same shoot.

I'm convinced that the fact that more MF shooters and image customers don't try side-by-side comparisons is the only thing that keeps Blad and P1 in business.

--
Rahon Klavanian 1912-2008.

Armenian genocide survivor, amazing cook, scrabble master, and loving grandmother. You will be missed.

Ciao! Joseph

http://www.swissarmyfork.com

Joseph looks like you havent looked at the pictures carefully ... you would have seen that they are larger then 100 % not smaller.

i did it this way BECAUSE everything smaller looks OK on screen and I wanted to show the difference between the systems.
MF isnt bad up to 1600 iso if you choose the right back or camera.
I chose the 31 on purpose because of that.
But I´m with you that MF has lots less funds for R+D.
The format is still bigger and has some advantages.....
Peter

 

[RedFox88]

is irrevelant, unless the bud is photojournalist who only posts thumbnails on a website or newspaper.

Did you see in Popular Photography magazine, the d3s iso 12,800 image that was a 2 page spread taken at f6.3 and 1/30 of a second or so? No noise was visible on the paper. That's quite a bit bigger than thumbnails or low dpi newspaper print!