IQ250 14 stops DR

[KevinA]

At all iso settings! How? I've not seen anything else keep it's dr when the iso goes up to 6400.

[torger]

At all iso settings! How? I've not seen anything else keep it's dr when the iso goes up to 6400.

Of course it's not at all ISO settings. It's at base ISO 100. MF marketing and truth is not always very well combined.

[KevinA]

Of course it's not at all ISO settings. It's at base ISO 100. MF marketing and truth is not always very well combined.

It's on Youtube so must be true 

[ErikKaffehr]

+1

On the other hand, there is a lot of good information from folks like Steve Hendrix, Dave Peterson and knowledgeable posters.

Erik

Of course it's not at all ISO settings. It's at base ISO 100. MF marketing and truth is not always very well combined.

[Doug Peterson]

I'd much rather share raw files and hands on experience than talk numbers.

We have a ton of raw files. I'm racing to properly name, organize, and upload them. Hope to have a lot to share tomorrow.

[KevinA]

Not that I'm in the market for one. But isn't the sensor size a bit of a let down, does it not play into the hands of the 35mm brigade.
If the physical size of the sensor is what gives MF the look, making it smaller kind of dilutes the reason for shooting MF?

[Doug Peterson]

Not that I'm in the market for one. But isn't the sensor size a bit of a let down, does it not play into the hands of the 35mm brigade.
If the physical size of the sensor is what gives MF the look, making it smaller kind of dilutes the reason for shooting MF?

We have a lot of happy P30+, P21+, IQ140, Aptus II 8, Aptus II 6 shooters. All of which are the same size sensor as the IQ250.

[torger]

Not that I'm in the market for one. But isn't the sensor size a bit of a let down, does it not play into the hands of the 35mm brigade.
If the physical size of the sensor is what gives MF the look, making it smaller kind of dilutes the reason for shooting MF?

Sure a full-frame sensor would have been a nicer start in terms of impressing us, but probably the 44x33mm format was a necessity to start with so Sony could sell the same sensor to many manufacturers and get a better business case. The 54x41mm format is not as all-around. 44x33mm has been around for a while, and while I do agree that it is a bit smallish, many seem happy with it. Many factors contribute to a look, sensor size is just one part of it. I tend to think that the meaning of sensor size in terms of look contributor is a bit exaggerated actually, but with a camera system designed for full-frame like the 645DF it surely is nice to actually have that.

[Ken Doo]

I agree with Torger.  Imo, while sensor size is significant, digital capture tends to magnify things (for lack of a better term) and the relationship to film capture size isn't really a one to one relationship.  Medium format digital (actually all formats) really needs to be considered in its entirety. In other words, its the complete system working together.  Taking one number/figure (such as from DXO) really doesn't do much when looking at the entire equation and creating an image.  To me, the most overlooked consideration is how much photographic enjoyment/satisfaction a particular camera system provides the photographer.

  ken

[Telecaster]

Not that I'm in the market for one. But isn't the sensor size a bit of a let down, does it not play into the hands of the 35mm brigade.
If the physical size of the sensor is what gives MF the look, making it smaller kind of dilutes the reason for shooting MF?

I'm not into the bragging rights game. For me MF is far more about the approach I take with it—deliberative, careful, precise, on a tripod—than the sensor size. I like the 4:3 aspect ratio too. The higher spatial resolution is a plus mainly for its tonal benefits...more samples of a subject equates to finer gradation. I'm seeing this same benefit with Sony's A7r, BTW, but I'm using that camera in a more spontaneous way, as befits its small size, so it's not a replacement for my Pentax.

-Dave-

[synn]

We have a lot of happy P30+, P21+, IQ140, Aptus II 8, Aptus II 6 shooters. All of which are the same size sensor as the IQ250.

+1

I really wish people would stop looking at spec sheets and actually shoot with 44x33 backs to see the "Look" first hand.

As far as DR is concerned, why single out MF for marketing speak? The almighty D800's DR at anything past 800 ISO is nothing special either.



Casual shot at 6400 ISO. I cried when I pushed the shadows.
(I rarely shoot anything serious above 800 ISO, but that's just me).

[Ken R]

Not that I'm in the market for one. But isn't the sensor size a bit of a let down, does it not play into the hands of the 35mm brigade.
If the physical size of the sensor is what gives MF the look, making it smaller kind of dilutes the reason for shooting MF?

I used a 645D for a few weeks (it also has a 44x33mm sensor) and I did notice the shallower DOF even at mid apertures when using lenses with similar angle of view than my 35mm DSLRs and shooting similar subjects at equal distances. The size of the chip is large enough that it does have that MF look. Yes the IQ160 that I own is even stronger in that regard but even with the 44x33mm chip you will see a significant difference compared to 35mm DSLRs. The 4:3 aspect ratio also helps. If you can, try it yourself (DEMO or rent) and see if it is for you.

[NikoJorj]

At all iso settings! How?

Because they have implemented ISO-as-metadata?
Do note that DR shifts towards HL with ISO setting though (at high ISO, you got much more DR in the highlights than in the shadows), if that is the case.

[eronald]

One can take these hard-DR claims of 14 stops.

2. Lose a couple of stops at the bottom to noise, defects, unbalanced readouts, pattern noise due to stitching and other issues
1. Lose a stop at the top to non-linear effects

You now have a "real DR" of 11 stops. Of which
1 Lose a stop to balance lighting
1 Lose a stop to ISO to get decent shutterspeed

So by now you're at 9 "real" stops, you need 8 for an image, so you have a chance to get a couple of stops of shadow lift with some sensor noise, or to go 1 1/2 stops into the highlights with some tint changes due to non linearity-one of these will work well, the other a bit less, depending on exposure.  

As anyone knows, 10 bits is the absolute minimum for a digital camera, 12 is usable and 14 is good - not extraordinary.

This back has a breakthrough feature -Live View- and decent quality. It's a hex wrench to add to the screwdrivers which are already in the toolkit.

And yes, the sensor has another breakthrough feature, probably: Price. But Phase is keeping this detail under wraps, while Pentax is publicizing it


Edmund

[Ken R]

Someone should make a dynamic range test scene and test a few setups.

My ultimate real "test" for the dynamic range of cameras is just about any architectural interior space with windows on one side only during a bright sunny day with a stunning view. Try to get perfect detail and color outside (including detail in white clouds) and inside (the entire space even 30 feet from the windows) with one exposure and no supplemental lighting.

[Theodoros]

One can take these hard-DR claims of 14 stops.

2. Lose a couple of stops at the bottom to noise, defects, unbalanced readouts, pattern noise due to stitching and other issues
1. Lose a stop at the top to non-linear effects

You now have a "real DR" of 11 stops. Of which
1 Lose a stop to balance lighting
1 Lose a stop to ISO to get decent shutterspeed

So by now you're at 9 "real" stops, you need 8 for an image, so you have a chance to get a couple of stops of shadow lift with some sensor noise, or to go 1 1/2 stops into the highlights with some tint changes due to non linearity-one of these will work well, the other a bit less, depending on exposure.  

As anyone knows, 10 bits is the absolute minimum for a digital camera, 12 is usable and 14 is good - not extraordinary.

This back has a breakthrough feature -Live View- and decent quality. It's a hex wrench to add to the screwdrivers which are already in the toolkit.

And yes, the sensor has another breakthrough feature, probably: Price. But Phase is keeping this detail under wraps, while Pentax is publicizing it


Edmund

Edmund, all the above refer to DSLRs… Usable vs. captured DR is much more with CCD MF if compared with DSLRs... we just have to wait and see if this new Cmos sensor behaves like a DSLR or like MF or is some ware in-between the two… Certainly, 14 stops refers to captured DR... usable DR will be less than that and it will be different between photographers too (as with any other camera) depending on each individual's idea of "correct" development.

[eronald]

Edmund, all the above refer to DSLRs… Usable vs. captured DR is much more with CCD MF if compared with DSLRs... we just have to wait and see if this new Cmos sensor behaves like a DSLR or like MF or is some ware in-between the two… Certainly, 14 stops refers to captured DR... usable DR will be less than that and it will be different between photographers too (as with any other camera) depending on each individual's idea of "correct" development.

I am sure Ferrari drivers can point their cars downslope to make'em go faster, but as an engineer I only compute the top speed on the flats

All of us here love photography, and are fascinated by the tools of the art. Most of us in this forum are incredibly privileged to have used a wide variety of both MF and 35mm equipment, and many of us have engineering qualifications; some of us (not me obviously) are also very talented artists; everyone here can take a decent picture with a Holga if warranted, and can recognize true quality.

I think when the hype has settled we will start to place the IQ250 in context. For me this thing is, until proven otherwise, a big D800 back end that mostly gets be clipped to a legacy Mamiya 645 front-end and legacy lenses; and that's what the pictures look like. I don't believe in the whole magically transcending the sum of the parts, at least when it comes to mechanical contrivances - although I am always fascinated by the limitless superiority of woman over man, with so little genetic difference

Edmund

PS. The Leica S whatever its faults is a custom-designed body with custom designed lenses for the smallish MF format; it may be better positioned to squeeze all the juice out of the lemon - if super-hi-rez lenses are needed on the D800, then I don't see why any old Mamiya lens should magically get the best out of this comparatively small sensor.

[ErikKaffehr]

Hi,

Much of the discussion is nonsense. Increasing ISO is just underexposure. On some cameras it is possible to eak out some more detail in the shadows, by amplifying the signal before AD conversion, Canon does this.

You can expose to protect highlights, but that is done best at base ISO.

Now when you underexpose you can do two things. Either you can tag the raw data with ISO number and do nothing, probably the correct way, or you can multiply the raw data by ISO/base ISO that makes raw files look good but is otherwise a bad thing, IMHO.

If you underexpose you will loose shadow detail, and DR will be reduced and noise will go up. It could be argued that you can protect extreme highlights by underexposure, but you can do that by shooting at base ISO and just get a decent histogram.

This stuff about MFDBs and DSLRs is just marketing BS. The IQ 250 uses a Sony sensor with Sony ADC:s, every bit of the information in that raw file is coming from Sony.

Phase one can do some tricks, like increasing integration times and add an excellent processing pipeline, but they cannot circumvent physics.

Best regards
Erik

Edmund, all the above refer to DSLRs… Usable vs. captured DR is much more with CCD MF if compared with DSLRs... we just have to wait and see if this new Cmos sensor behaves like a DSLR or like MF or is some ware in-between the two… Certainly, 14 stops refers to captured DR... usable DR will be less than that and it will be different between photographers too (as with any other camera) depending on each individual's idea of "correct" development.

[eronald]

Erik,

 Careful, with such blunt speech you will offend the artistic sensibilities of our artist friends - I wonder whether Picasso and friends had impassioned discussions about the magical quality of canvas and paints

Edmund

Hi,

Much of the discussion is nonsense. Increasing ISO is just underexposure. On some cameras it is possible to eak out some more detail in the shadows, by amplifying the signal before AD conversion, Canon does this.

You can expose to protect highlights, but that is done best at base ISO.

Now when you underexpose you can do two things. Either you can tag the raw data with ISO number and do nothing, probably the correct way, or you can multiply the raw data by ISO/base ISO that makes raw files look good but is otherwise a bad thing, IMHO.

If you underexpose you will loose shadow detail, and DR will be reduced and noise will go up. It could be argued that you can protect extreme highlights by underexposure, but you can do that by shooting at base ISO and just get a decent histogram.

This stuff about MFDBs and DSLRs is just marketing BS. The IQ 250 uses a Sony sensor with Sony ADC:s, every bit of the information in that raw file is coming from Sony.

Phase one can do some tricks, like increasing integration times and add an excellent processing pipeline, but they cannot circumvent physics.

Best regards
Erik

[Paul2660]

If you underexpose you will loose shadow detail, and DR will be reduced and noise will go up. It could be argued that you can protect extreme highlights by underexposure, but you can do that by shooting at base ISO and just get a decent histogram.

With the Sony chips at base ISO and at least 2 stops up in ISO you have a much better range of shadow detail when you underexpose for highlights.  I see this all the time with the D800 files.  Do you see it with the A99?  The trade off seems that when pushed for higher ISO the D800 seems to fine till 3200 past that and the files fall apart pretty quick.  Once the raw files from the IQ250 become more available it will be most interesting to see what the ISO range of 400 to 3200 looks like at full resolution.

Paul C