RAW files: 1Ds3 and Phase P30+

[eronald]

What makes you think I do?

http://theory.uchicago.edu/~ejm/pix/20d/Tr...MG_6318-web.jpg
[a href=\"index.php?act=findpost&pid=203039\"][{POST_SNAPBACK}][/a]

The sort of tame bird we like our kids to feed from their hands
They had some tame birds of this sort flying around at Photokina, fortunately none of them decided to order some food to go

Edmund

[Panopeeper]

Here is a RAW file which has no photographic merit. It's simply a shot which includes data from slightly clipping highlights to clipping shadows. I was outside today and was curious about how the P25+ would handle this ridiculous range. If it's not useful for your purposes I can upload something different.

Dale,

this image is the perfect demonstration of what I was saying to 203. The dynamic range of the scenery was huge. The exposure was perfect: 1/3 stop more and lots of clipping occurs.

The factual highlight clipping was negligable - only 115 pixels were saturated (these are reflecting particles in the granit).

The area under the large rocks is underexposed (down to black clipping), but very fine details can be extracted from the 10th and partly from the 11th stop.

Anyway, this image does not exhibit the electric fault; good for you (though I still don't know the reason - it is not impossible, that this is coming and going, like a contact error).

There is another interesting thing to observe on this image, combined with that of Guy's: the stitching of the sensor parts. This is, what is called - incorrectly - "centerfold". It is in the center of the P45+ I have seen (five different backs), but Phase One seems to have made some smart improvent (I guess the P25+ is newer than the P45+): pieces of different sizes are stitched together. For example yours consists of three vertical strips, 1932, 640 pixels and the rest (I can't make out the horizontal stitching). Guy's copy consists of 646, 2526 and 920 pixel wide strips. This would mean better utilization of the chips (this should make the back cheaper!).

[Guy Mancuso]

Dale,

this image is the perfect demonstration of what I was saying to 203. The dynamic range of the scenery was huge. The exposure was perfect: 1/3 stop more and lots of clipping occurs.

The factual highlight clipping was negligable - only 115 pixels were saturated (these are reflecting particles in the granit).

The area under the large rocks is underexposed (down to black clipping), but very fine details can be extracted from the 10th and partly from the 11th stop.

Anyway, this image does not exhibit the electric fault; good for you (though I still don't know the reason - it is not impossible, that this is coming and going, like a contact error).

There is another interesting thing to observe on this image, combined with that of Guy's: the stitching of the sensor parts. This is, what is called - incorrectly - "centerfold". It is in the center of the P45+ I have seen (five different backs), but Phase One seems to have made some smart improvent (I guess the P25+ is newer than the P45+): pieces of different sizes are stitched together. For example yours consists of three vertical strips, 1932, 640 pixels and the rest (I can't make out the horizontal stitching). Guy's copy consists of 646, 2526 and 920 pixel wide strips. This would mean better utilization of the chips (this should make the back cheaper!).
[a href=\"index.php?act=findpost&pid=203083\"][{POST_SNAPBACK}][/a]

Gabor maybe have me confused with someone else. I have the same back as Dale P25 plus,mine was just purchased new. I did demo the the P30 plus though

[Panopeeper]

Gabor maybe have me confused with someone else. I have the same back as Dale P25 plus,mine was just purchased new. I did demo the the P30 plus though

Guy, you posted a set with different ISOs, of your bathroom with a GretagMacbeth color checker, shot with a P25+. I have not seen any *raw* images of you by a P30+. Did you post some? Where?

I have not been clear enough, but I mentioned the P45+ only as contrast to the P25+ with the sensor stitching.

[Dale Allyn]

Gabor,

I'm glad that the file worked for the purposes here.

Thanks for the info regarding the sensor configurations as well.

Dale

Dale,

this image is the perfect demonstration of what I was saying to 203. The dynamic range of the scenery was huge. The exposure was perfect: 1/3 stop more and lots of clipping occurs.

The factual highlight clipping was negligable - only 115 pixels were saturated (these are reflecting particles in the granit).

The area under the large rocks is underexposed (down to black clipping), but very fine details can be extracted from the 10th and partly from the 11th stop.

Anyway, this image does not exhibit the electric fault; good for you (though I still don't know the reason - it is not impossible, that this is coming and going, like a contact error).

There is another interesting thing to observe on this image, combined with that of Guy's: the stitching of the sensor parts. This is, what is called - incorrectly - "centerfold". It is in the center of the P45+ I have seen (five different backs), but Phase One seems to have made some smart improvent (I guess the P25+ is newer than the P45+): pieces of different sizes are stitched together. For example yours consists of three vertical strips, 1932, 640 pixels and the rest (I can't make out the horizontal stitching). Guy's copy consists of 646, 2526 and 920 pixel wide strips. This would mean better utilization of the chips (this should make the back cheaper!).
[a href=\"index.php?act=findpost&pid=203083\"][{POST_SNAPBACK}][/a]

[Guy Mancuso]

Guy, you posted a set with different ISOs, of your bathroom with a GretagMacbeth color checker, shot with a P25+. I have not seen any *raw* images of you by a P30+. Did you post some? Where?

I have not been clear enough, but I mentioned the P45+ only as contrast to the P25+ with the sensor stitching.
[a href=\"index.php?act=findpost&pid=203097\"][{POST_SNAPBACK}][/a]

I posted some raws i think with the P30 plus from when I was in San Juan playing with it but that was some time ago not sure were they are.  But yes you are correct the bathroom shots are with the P25 plus which i am very pleased with BTW with the higher ISO's. ISO 400 is really nice and 800  is nothing to sneeze at. I'm still playing around with the higher ISO stuff to find my happy spot but so far pretty good.

[Panopeeper]

the bathroom shots are with the P25 plus which i am very pleased with BTW with the higher ISO's. ISO 400 is really nice and 800  is nothing to sneeze at

You mean "underexposed by 1 2/3 EV and by 2 2/3 EV", right? (According to those images.)  ISO 800 of that camera means "undermetering by 2 EV".

I'm still playing around with the higher ISO stuff to find my happy spot but so far pretty good

If one can believe those shots of the bathroom, I mean if the exposure was according to the metering based on the ISO, unbiased, then ISO 200 should be the only one ever selected.

Correction: based on that serie, it should be 100. (See details in a post further down.)

[Dale Allyn]

If one can believe those shots of the bathroom, I mean if the exposure was according to the metering based on the ISO, unbiased, then ISO 200 should be the only one ever selected.
[a href=\"index.php?act=findpost&pid=203164\"][{POST_SNAPBACK}][/a]

Gabor, are you suggesting that ISO 200 is the true "native" ISO of the P25+ back? And further suggesting that one gains nothing in terms of image quality by shooting at ISO 50 or ISO 100?

I need to do more "testing", but so far I have preferred the look of the ISO 100 and ISO 50 shots over the ISO 200 shots. This may be purely coincidental, as my ISO increases have been to accommodate reduced light conditions or when I wanted a slightly higher shutter speed. I have not shot the same scene on each ISO setting while manually manipulating exposure time to balance for the ISO steps.

Frankly, I've simply been shooting at the lowest ISO suitable for the available lighting, that is, depending on the aperture or exposure time needed for the desired results.

Edit to add: I see that you, Gabor, edited your post to correct your remark. You are suggesting ISO 100 to be base or "native" on the P25+ back. Thanks for the correction.

[Guy Mancuso]

You mean "underexposed by 1 2/3 EV and by 2 2/3 EV", right? (According to those images.)  ISO 800 of that camera means "undermetering by 2 EV".
If one can believe those shots of the bathroom, I mean if the exposure was according to the metering based on the ISO, unbiased, then ISO 200 should be the only one ever selected.

Correction: based on that serie, it should be 100. (See details in a post further down.)
[a href=\"index.php?act=findpost&pid=203164\"][{POST_SNAPBACK}][/a]

They were replaced and uploaded , please let's stay up to date on this stuff. You make it sound like I cheated somehow, sure you want to say something like that. Let's be careful how we word things please . The first set was underexposed and my reshoot was exposed on the money.

[Panopeeper]

are you suggesting that ISO 200 is the true "native" ISO of the P25+ back?

I am suggesting, that there is no "native" ISO, because there is only one. The ISO selection changes the metering, and the question is, which ISO yields the best exposure (i.e. utilizing the dynamic range and minimizing the noise).

Guy posted a set with ISO 100, 200, 400 and 800, with successively reduced exposure. From those, the ISO 100 shot is perfectly exposed (1/3 EV down from the right edge); however, the question is, how good that metering reflected the entire scenery, i.e. how reliable the camera's metering is in that particular setting.

On the other hand, when I see ACR's automatic adjustments, ISO 200 seems to be the basis. Is ACR reliable in this relation? Did they make respective tests? I doubt it, because ACR applies the same adjustment to P45+ images as well, which do not need to be adjusted, for it has different ISOs.

And further suggesting that one gains nothing in terms of image quality by shooting at ISO 50 or ISO 100?

Let's see the meaning of ISO selection on different cameras, on the histograms, which show the effect of the exposure.

First, Canon 40D;

40D ISO 200

40D ISO 400

40D ISO 800

40D ISO 1600

The histograms show, that *roughly* identical image data has been delivered with increasing ISO setting, although the shutter was 1/20s, 1/40s, 1/80s and 1/160s. The increased ISO gain "made up" for the reduced exposure. Of course this "substitute exposure" is not a 100% substitute, i.e. the noise will be higher with the lower exposure.

Now, let's see what is happening if there is no ISO implemented on that level; following are the histograms from Guy's shots:

P25+ ISO 100

P25+ ISO 200

P25+ ISO 400

P25+ ISO 800

Here too the shutter has been halved with each stop increase of the ISO; however the pixel data got halved as well (the thicker white bars under the histograms mark one stop, the thinner bars 1/3 stop).

The higher ISO setting did not make up for the lower exposure. In other words, the higher ISO means metering as if there was a higher ISO implemented, but the camera does not implement it, it has to be compensated for in the raw processing.

This is equaivalent to the "ISO by pushing the exposure in raw conversion".

What is the result? If you pick a certain spot on the image, its brightness (in terms of pixel values) goes down with each ISO increase. Something, which has been in the sixth stop (from the very right edge) at ISO 100 will be in the range of the ninths stop with ISO 800. Accordingly, the noise will be higher and the details less.

There is nothing wrong with it, but it is important to understand: the higher ISO is in no way a substitute for higher exposure. Therefor my suggestion: always shoot with that ISO, which yields a close to ETTR exposure. If you can't afford the required exposure (moving subject, low light), then you will see, that you are underexposing, but you get the same as if you had increased the ISO.

Downsides:

- the display on the LCD will be darker,

- you have to increase the intensity in raw processing.

Of course you may shoot with higher ISO *when you realise, that you can't go with the required exposure*; the resulting image is the same without the downsides. But, again: be aware, that this is an "emergency measure", just like I am shooting with ISO 800 and 1600 only in "emergency cases".

[Panopeeper]

You make it sound like I cheated somehow, sure you want to say something like that. Let's be careful how we word things please . The first set was underexposed and my reshoot was exposed on the money.

Please read my above post to Dale carefully, and check out the histograms.

[Guy Mancuso]

Please read my above post to Dale carefully, and check out the histograms.
[{POST_SNAPBACK}][/a]

The latest versions are here. [a href=\"http://idisk.mac.com/guymancuso-Public?view=web]http://idisk.mac.com/guymancuso-Public?view=web[/url]
Noise test 2

Whatever results you get really as i said before are meaningless to me . All shooters care about is if we set it at ISO 800 do you get nice results. How we get there or how it works on the back end is somewhat meaningless because we don't shoot the science of it we look at final results.

[sergio]

I really don't see the 3D effect with MF. I think that is plain fantasy. Maybe 20,000K makes people see things...

The very real thing for me about MF is the VIEWFINDER. It lets me see, compose and shoot in a way that really sets in a mood to shoot and create. Makes me want to make photographs.

I really don't care that much about the pixels nowadays. I got over the angst of having uber IQ. I care about how I feel when I shoot. If a camera is great to use, then you'll probably end up with better images from that camera than from another that isn't, regardless of IQ.

That is the sole reason I prefer MF over 35mm. I feel that 35mm full frame viewfinders are small, and uncomfortable to use, no matter how good they can be.

[thsinar]

No Guy, I don't think Gabor is suggestion any cheat. All he is saying is even if you have set it to ISO 800, all it was is an under-exposure of the files, not a real ISO 800.

Thierry

They were replaced and uploaded , please let's stay up to date on this stuff. You make it sound like I cheated somehow, sure you want to say something like that. Let's be careful how we word things please . The first set was underexposed and my reshoot was exposed on the money.
[a href=\"index.php?act=findpost&pid=203187\"][{POST_SNAPBACK}][/a]

[Murray Fredericks]

The latest versions are here. http://idisk.mac.com/guymancuso-Public?view=web
Noise test 2

Whatever results you get really as i said before are meaningless to me . All shooters care about is if we set it at ISO 800 do you get nice results. How we get there or how it works on the back end is somewhat meaningless because we don't shoot the science of it we look at final results.
[a href=\"index.php?act=findpost&pid=203223\"][{POST_SNAPBACK}][/a]

Guy, on one hand you are right, but if you are working with uncontrollable situations where you rely on natural light (such as in architectural photography) you are always making 'informed' decisions about what will be usable/acceptable to the client in extreme  circumstances.

So many of my shots are produced in extremely low light and then pushed up as far as they can go...I'm sure many other photographers (particularly those who work with interiors) encounter these same situations.

For me, these tests can explain a lot (even if I arrived to the same conclusions by trial and error) and will influence decisions in real world situations and also what I recommend to my clients when being briefed on a job...


Murray

[rethmeier]

I agree with Murray,
to get the shot is what counts.
If I need to use my D3 at 6400 instead of my Hy6/75LV at 800, I'll be using the Nikon.
If  the image doesn't have to go beyond A3 I'm covered.
Still prefer the Hy6 as I like the viewfinder etc better.
Cheers,
Willem.

[hdomke]

Today I came across a magazine article that addresses the heart of the discussion on this thread: In the comparison of MF to 35mm, is it more than just pixel count?

The article is called "The Fall and Rise of Medium Format" by Simon Wakelin. It is in the July/August 2008 issue of Digital Photo Pro.

The bulk of the article was about MF manufacturers promoting bit depth  Here are quotes:

Jack Showalter, president of Hasselblad USA “Let’s face it, it’s not just pixel count. The reason you use a larger piece of silicon for medium format is to convert more photons to electrons and thus gain more data with deeper bit depth.”

“Dealing with true 16-bit color depth makes a difference” says Mark Rezzonico, vice president of Leaf America.

Jans H. Christiansen, marketing director of Phase One “…if you’re shooting in tricky lighting situations  and you want those subtle details, tones, shadows and highlights to show, this is where medium format really comes into its own.”

Is part of the appeal of MF the fact that the cameras are bigger and look more exclusive? Isn’t that what Hugh Milstein of Digital Fusion means when he says “The bottom line with medium format is that big glass makes a big impression… the guy with the bigger camera gets the better shot because it elicits an emotional response from the talent."

Personally, I found the argument about bit-depth unconvincing, but I think there is no doubt that MF cameras have a different presence in the studio.

[Panopeeper]

Let’s face it, it’s not just pixel count. The reason you use a larger piece of silicon for medium format is to convert more photons to electrons and thus gain more data with deeper bit depth

The sensel size  (linear, not area) of the 39 MPix H39 is a hopping 6.8micron.

For comparison:

Canon 5D: 8.2micron
Canon 1DMkIII: 7.4micron
Nikon D3: 11.3micron

There must be something else there.

[Snook]

Today I came across a magazine article that addresses the heart of the discussion on this thread: In the comparison of MF to 35mm, is it more than just pixel count?

The article is called "The Fall and Rise of Medium Format" by Simon Wakelin. It is in the July/August 2008 issue of Digital Photo Pro.

The bulk of the article was about MF manufacturers promoting bit depth  Here are quotes:

Jack Showalter, president of Hasselblad USA “Let’s face it, it’s not just pixel count. The reason you use a larger piece of silicon for medium format is to convert more photons to electrons and thus gain more data with deeper bit depth.”

“Dealing with true 16-bit color depth makes a difference” says Mark Rezzonico, vice president of Leaf America.

Jans H. Christiansen, marketing director of Phase One “…if you’re shooting in tricky lighting situations  and you want those subtle details, tones, shadows and highlights to show, this is where medium format really comes into its own.”

Is part of the appeal of MF the fact that the cameras are bigger and look more exclusive? Isn’t that what Hugh Milstein of Digital Fusion means when he says “The bottom line with medium format is that big glass makes a big impression… the guy with the bigger camera gets the better shot because it elicits an emotional response from the talent."

Personally, I found the argument about bit-depth unconvincing, but I think there is no doubt that MF cameras have a different presence in the studio.
[a href=\"index.php?act=findpost&pid=203888\"][{POST_SNAPBACK}][/a]

I have to disagree with you there as I see a big difference with Bit depth fro my P30 to my 1DsMII.
It is easily seen with a grey background... If you do any retouching you'll notice there is a big difference between my phase and Canon files.
I have not had a 1dsMIII file to play with, but do not think it is better than the P30's 16 bit.
The 16 Bit hold up a lot more abuse for sure.. even in just exposure.
I have heard there is not much of a difference to really upgrade to the 1DsMIII from the 1DsMII, But am I missing something?
Snook

[203]

But am I missing something?
Snook
[a href=\"index.php?act=findpost&pid=203902\"][{POST_SNAPBACK}][/a]

Yes. I will send you a couple RAW images (off-line) from model shoots if you want to check them out.