Ad
Ad
Ad
Pages: « 1 ... 18 19 [20] 21 »   Bottom of Page
Print
Author Topic: Is the difference of DR on MFDB vs 35mm dslr discernible on print?  (Read 49488 times)
ThierryH
Sr. Member
****
Offline Offline

Posts: 409


« Reply #380 on: September 27, 2009, 08:12:44 AM »
ReplyReply

Dear "Professional",

The term "FAST" lens is commonly understood by photographers as describing a lens allowing for large/wide maximal aperture, respectively opening of the diaphragm. The larger the maximal opening the faster the lens in terms of exposure time in relation to the maximal aperture, which are always interdependent factors.

Best regards,
Thierry

Quote from: Professional
Thank you very much, now you explained it better, but saying fast in general can't explain much, i am a noob and don't understand quickly.  
Logged

JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #381 on: September 27, 2009, 10:46:20 AM »
ReplyReply

Quote
It seems that image quality of the Canon 5DII and the 1DsIII is quite similar and I would suggest that the Nikon 3DX is also similar.
I think a fair number of people might disagree with that suggestion, especially if the topic is dynamic range.
Logged

Christopher
Sr. Member
****
Offline Offline

Posts: 944


WWW
« Reply #382 on: September 27, 2009, 02:31:06 PM »
ReplyReply

Quote from: JeffKohn
I think a fair number of people might disagree with that suggestion, especially if the topic is dynamic range.

The same fair amount of people who, as far as I know, have never shown any REAL raw prove for that, just rambling on with DOx numbers. (Could be that I missed it)
Logged

JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #383 on: September 27, 2009, 03:43:51 PM »
ReplyReply

Quote from: Christopher
The same fair amount of people who, as far as I know, have never shown any REAL raw prove for that, just rambling on with DOx numbers. (Could be that I missed it)
I take the DxO results for what they are, so I'm not sure what exactly you want me to prove. I'm pretty sure the guys at DxO know more about camera measurement  and dynamic range than myself or most others participating in this thread. Just because somebody doesn't like what the DxO results say doesn't mean they can be dismissed as irrelevant. Can you explain why you feel the DxO tests are wrong? Do you know of credible tests showing that the 5DII has as much or more dynamic range as the D3x at low ISO?
« Last Edit: September 27, 2009, 03:44:24 PM by JeffKohn » Logged

cmi
Sr. Member
****
Offline Offline

Posts: 491


« Reply #384 on: September 27, 2009, 03:54:04 PM »
ReplyReply

Jeff, it's about blind trust. I regard someones practical experience higher than numbers, even if this numbers come from a undoubtely very qualified team. In what position do I put me if I trust someone blindly? Even if the one I trust is RIGHT, it puts me in a very uncomfortable position. I do not KNOW anylonger from my own experience, so that degrades me. And if we trust in this discussion blindly in dx0 numbers, then that degrades the discussion.

My opinion.

Christian

Logged
JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #385 on: September 27, 2009, 03:58:36 PM »
ReplyReply

Quote from: Christian Miersch
Jeff, it's about blind trust. I regard someones practical experience higher than numbers, even if this numbers come from a undoubtely very qualified team. In what position do I put me if I trust someone blindly? Even if the one I trust is RIGHT, it puts me in a very uncomfortable position. I do not KNOW anylonger from my own experience, so that degrades me. And if we trust in this discussion blindly in dx0 numbers, then that degrades the discussion.
It's not about 'blind' trust. But you do have to decide who has credibility and who does not. You can't test everything for yourself; even if you think you have the knowledge and expertise (doubtful), who has the time? The DxO guys have more credibility in camera testing than random folks on the web who happen to not like what the DxO tests say about their brand of choice. So I think the burden of proof is on the folks who want to discount the DxO tests, not the other way around. Besides, it's not like DxO are the only ones to report that the D3x has very dynamic range.
Logged

ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7332


WWW
« Reply #386 on: September 27, 2009, 04:12:12 PM »
ReplyReply

Of course there is the question what DR means to us? It's defined as "well capacity" / "read noise" both measured in electron charges. Is it something we see in print? Well, maybe...

Best regards
Erik

Quote from: JeffKohn
It's not about 'blind' trust. But you do have to decide who has credibility and who does not. You can't test everything for yourself; even if you think you have the knowledge and expertise (doubtful), who has the time? The DxO guys have more credibility in camera testing than random folks on the web who happen to not like what the DxO tests say about their brand of choice. So I think the burden of proof is on the folks who want to discount the DxO tests, not the other way around. Besides, it's not like DxO are the only ones to report that the D3x has very dynamic range.
Logged

Professional
Sr. Member
****
Offline Offline

Posts: 306


« Reply #387 on: September 27, 2009, 04:49:42 PM »
ReplyReply

Quote from: ThierryH
Dear "Professional",

The term "FAST" lens is commonly understood by photographers as describing a lens allowing for large/wide maximal aperture, respectively opening of the diaphragm. The larger the maximal opening the faster the lens in terms of exposure time in relation to the maximal aperture, which are always interdependent factors.

Best regards,
Thierry

Thank you very much!

Just i am not a photographer, let's say i am a camera shooter maybe, a hobbyist or whatever but not an true real photographer so simple as that, so for that reason i can understand fast in term of focus or in term of aperture or something else, but i always make it as in term of focus before i make it as a term of aperture value, so apologize my ignorant.

Best Regards,
Tareq
Logged
cmi
Sr. Member
****
Offline Offline

Posts: 491


« Reply #388 on: September 27, 2009, 04:56:36 PM »
ReplyReply

Quote from: JeffKohn
It's not about 'blind' trust. But you do have to decide who has credibility and who does not. You can't test everything for yourself; even if you think you have the knowledge and expertise (doubtful), who has the time? The DxO guys have more credibility in camera testing than random folks on the web who happen to not like what the DxO tests say about their brand of choice. So I think the burden of proof is on the folks who want to discount the DxO tests, not the other way around. Besides, it's not like DxO are the only ones to report that the D3x has very dynamic range.

I dont know anything about high end cameras, but I talk about a general position, about the way I think.

And so I say: I rather trust my own thinking than someone others. I consider it, I try to incooperate it, but I don't take it blindly. And I mistrust opinions wich take it "blindly".


Cheers

Christian
« Last Edit: September 27, 2009, 05:08:30 PM by Christian Miersch » Logged
ThierryH
Sr. Member
****
Offline Offline

Posts: 409


« Reply #389 on: September 27, 2009, 05:02:53 PM »
ReplyReply

Nothing to apologize, Tarek, you are welcome.

Thierry

Quote from: Professional
Thank you very much!

... so apologize my ignorant.

Tareq
Logged

BernardLanguillier
Sr. Member
****
Online Online

Posts: 7900



WWW
« Reply #390 on: September 27, 2009, 07:03:42 PM »
ReplyReply

Quote from: Jack Flesher
Doug is right -- there is an awful lot of sanctimonious argument on this question from armchair "experts" that have NEVER even picked up an MF camera...  So buddyjanes, here is my post, now let's see yours from your DSLR that proves me so "wrong" -- or is this another one of YOUR typical meager and unsatisfactory armchair arguments?

Hello Jack,

Regarding the armchair comment, you might want to give a look at this. Whether you like the pictures or nor, whether you appreciate the technical value of the images or not... I hope that you will at least acknowledge the fact that 20,000 feet vertical in one week is hard to do from an armchair...

Regarding your 1ds3 clipping issue, have you at all considered the possibility that not all DSLR might be equal?

Cheers,
Bernard
« Last Edit: September 27, 2009, 07:11:41 PM by BernardLanguillier » Logged

A few images online here!
rethmeier
Sr. Member
****
Offline Offline

Posts: 779


WWW
« Reply #391 on: September 27, 2009, 08:07:54 PM »
ReplyReply

Having used the Canon 1DsMkIII and now the D3x,I know which has the better sensor,it is the so called overpriced D3x.
I also can't wait for Nikon to start using the new kid on the block,the new Sony  CCD SuperHAD II sensor with 34.8 MP.
No AA filter etc.

Cheers,
Willem.
Logged

Willem Rethmeier
www.willemrethmeier.com
Sydney Australia
JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #392 on: September 27, 2009, 09:41:26 PM »
ReplyReply

Quote from: Christian Miersch
And so I say: I rather trust my own thinking than someone others. I consider it, I try to incooperate it, but I don't take it blindly. And I mistrust opinions wich take it "blindly".
I'm sorry, but you're building a straw-man argument. I never said anything about taking anything blindly. DxO Labs has a wealth of information available about what they test, how they test it, and how to interpret the results. Start here, here, or here.
Logged

JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #393 on: September 27, 2009, 09:49:53 PM »
ReplyReply

Quote from: ErikKaffehr
Of course there is the question what DR means to us? It's defined as "well capacity" / "read noise" both measured in electron charges. Is it something we see in print? Well, maybe...
I don't disagree with you. No single test can tell you everything. I said as much in one of my first posts in this thread. More recently I just took exception to Christopher's assertion that the D3x's image quality is the same as the 5D2/1Ds3, which he offered no proof of despite criticizing others who supposedly "have never shown any REAL raw prove for that..."
« Last Edit: September 27, 2009, 09:53:10 PM by JeffKohn » Logged

Ray
Sr. Member
****
Offline Offline

Posts: 8878


« Reply #394 on: September 27, 2009, 09:58:38 PM »
ReplyReply

Quote from: Christian Miersch
I dont know anything about high end cameras, but I talk about a general position, about the way I think.

And so I say: I rather trust my own thinking than someone others. I consider it, I try to incooperate it, but I don't take it blindly. And I mistrust opinions wich take it "blindly".

Christian,
Whilst I agree with the principle in general that one should attempt to think for oneself as far as possible and not blindly trust the opinions of others simply because they may present themselves as being qualified in a particular area, I have to agree with Jeff that it doesn't really make much sense to mistrust the highly qualified opinions from an organisation like DXO which present the results of their rigorous, scientific testing in a clear, precise and consistent manner..... unless you have some justified suspicion that DXO have got it wrong. Since you claim to know nothing about high-end cameras, one wonders where your suspicion that DXO might be wrong comes from.

You should bear in mind that DXO do not do such testing for a bit of fun, or merely for academic interest. They are in the business of producing a very sophisticated RAW converter which not only attempts to get the very best from each proprietary RAW file, in terms of DR, SNR, tonal range etc, but also attempts to correct a variety of lens defects through add-on modules. If their testing methodology were not sound, consistent and reliable, it's difficult to imagine how the results of their testing could be useful for their product.

It seems clear to me that certain owners of very expensive MFDB equipment see red at the mere hint that a DSLR might produce at some fundamental level, at base ISO, some property of image quality which is actually superior to their DB. Such photographers might subsequently dismiss the DXOMark results as being out of touch with reality before they've taken the trouble to work out for themselves what the DXO results are actually saying.

Having referred to the DXOMark charts quite frequently since they were made available to the public, I see only one instance of a DSLR outperforming a DB in some aspect of image quality at base ISO, and that's the Nikon D3X with regard to Dynamic Range. In all other aspects of image quality, that DXOMark test (SNR, tonal range, color sensitivity), the DB has superior performance at equal image/print size, at base ISO.

At higher than base ISO, it's a different story. Even the most ardent DB fan boys would have to admit that the DSLR is either as good or better at high ISO in every parameter except perhaps resolution.

I quote from the DXO website:

"This website presents a large set of measurement data built over time in the testing laboratories at DxO Labs. Our imaging experts have developed a thorough understanding of the technologies and methods involved in measuring the parameters of digital camera image quality. Indeed, the strength of DxO Labs’ industry-leading image quality evaluation solution, DxO Analyzer, lies in its precisely-described test protocols in tandem with strict control of all physical parameters that might influence measurements.

In keeping with accepted scientific protocols, all measurements can be repeated independently under the same bias-free conditions. This ensures that DxO Labs' measurements and its DxOMark scale are objective and reliable metrics to help photographers evaluate digital camera image quality performance."
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8878


« Reply #395 on: September 27, 2009, 10:07:41 PM »
ReplyReply

Quote from: ErikKaffehr
Of course there is the question what DR means to us? It's defined as "well capacity" / "read noise" both measured in electron charges. Is it something we see in print? Well, maybe...

Erik,
I thought we'd already established earlier that DR defines the range of brightness levels that the camera can record with a single exposure. Of course one doesn't 'see' that full DR if the scene being photographed has a brightness range which is less than the DR capability of the camera, just as we do not see the full resolution capability of the camera if the scene being photographed is devoid of  fine detail or texture.

What's interesting about the DXO tests comparing the D3X with the P65+ is that, at probably any print size, the D3X will record as much (or more) detail in the deepest shadows as the P65 can, but not if the deepest shadows are at a brightness level equivalent to 18% gray, which is the level DXO uses to measure SNR. The P65+ still retains a noise advantage over the D3X in the mid-tones and lower mid-tones.

Owners of DBs who have made their own comparisons with a D3X, are probably seeing these slightly higher noise levels in the mid-tones and lower mid-tones of the D3X shot (at equal image/print size) and are therefore concluding that the DXO results must be wrong. In fact, the DXO results show that at an average scene reflectance of 18% gray, the P65+ has 4dB better SNR than the D3X. That's more than a whole stop better.

The fact that the D3X can have as much as 2/3rds of a stop lower noise in the deepest shadows, which may often be clipped to black for esthetic reasons when processing the image, is probably of little concern to the user of an MFDB system in the studio, but could be of concern, for example, to a landscape photographer shooting a waterfall in a dense rainforest, who wants to retain some detail in the surrounding, dense undergrowth of the forest.
Logged
Panopeeper
Sr. Member
****
Offline Offline

Posts: 1805


« Reply #396 on: September 27, 2009, 11:37:10 PM »
ReplyReply

Quote from: Ray
What's interesting about the DXO tests comparing the D3X with the P65+ is that
Ray,

for the case that you want to know what is interesting in that comparison:

Quote
not if the deepest shadows are at a brightness level equivalent to 18% gray
18% grey has the intensity 18%. This is about 2.5 stops from clipping, i.e. in the middle of the third stop of the DR. Do you know a DSLR, which captures the third stop of the DR as "the deepest shadow"?

Quote
The P65+ still retains a noise advantage over the D3X in the mid-tones and lower mid-tones
From what did you infer this?

1. The noise of the D3X is shown as lower at mid-level (18%), except in print, i.e. if the P65+ image is downresed.

2. DXO indicates, that the best DR of the D3X (with ISO 75) is 0.5 EV greater than that of the P65+ (with ISO 44). This relates to the deepest shadows, where SNR=1 is.

Quote
In fact, the DXO results show that at an average scene reflectance of 18% gray, the P65+ has 4dB better SNR than the D3X. That's more than a whole stop better.
1. Again, this is the printed (downresed) version. Of course, the P65+ is better.

2. This decibel has nothing to do with stop. The noise of the D3X with ISO 100, measured on an 18% intensive patch is 1.1%, which corresponds to SNR=90.  The printed P65+ image would show SNR=226 or about 0.44% noise.

This noise level is totally irrelevant. DXO's 18% SNR is an absolutely useless measurement, it is for those, who don't know its meaning.

Quote
The fact that the D3X can have as much as 2/3rds of a stop lower noise in the deepest shadows, which may often be clipped to black for esthetic reasons when processing the image, is probably of little concern to the user of an MFDB system in the studio, but could be of concern, for example, to a landscape photographer shooting a waterfall in a dense rainforest, who wants to retain some detail in the surrounding, dense undergrowth of the forest.
1. Again, the noise is not 2/3 of a stop lower or higher or whatever. "Stop" is not the measurement of the noise.

There is one sense to qualify noise with stop: compared two not equal cameras, with the same ISO, the exposure of one must be a certain amount higher/lower than that of the other to achieve the same level of noise. This is just what DXO's DR tab shows: the same noise, namely SNR=1, will be observable with the D3X on 2/3 EV darker patches than with the P65+. Not the noise is 2/3 EV lower but the intensity can be 2/3 EV lower to get the same noise.

2. If someone cuts off the deepest shadows by black point, then one does not have any reason to talk about dynamic range.

----------------------------------------------------------------------------------------------------------

To the others: the above should not be interpreted as any position for or against any camera in the comparison. I do not advocate any measurement accept my own ones (and I don't have any usable raw images from the P65+ and not enough from the D3X).
Logged

Gabor
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7332


WWW
« Reply #397 on: September 27, 2009, 11:51:51 PM »
ReplyReply

Ray,

I like your explanation. What I really wanted to say was that some photographers find that their MFDBs have better DR than their D3X and blame DxO. In my view the DxO measurement is based on standards and is probable correct. It could well be that the DR in the standards is something different from that MFDB owners see.

We need to do some excessive image manipulation to really utilize the DR of modern cameras, having both highlight and shadow detail with making the image flat, so a significant amount of processing is involved in making a printable image.

My impression is that MFDB owners see a difference but cannot define, many times they are more of artists than scientists, but call it DR or microcontrast. That said it seems that there are many MFDB photographers around who are both, some of them write articles on Luminous Landscape.

A related issue is that there are many photographers who like film, although it seems to be special cases where film resolves better than digital I think that it's proven beyond doubt that in the majority of cases digital outperforms film on technical grounds. Aesthetics may be different.

Best regards
Erik

Quote from: Ray
Erik,
I thought we'd already established earlier that DR defines the range of brightness levels that the camera can record with a single exposure. Of course one doesn't 'see' that full DR if the scene being photographed has a brightness range which is less than the DR capability of the camera, just as we do not see the full resolution capability of the camera if the scene being photographed is devoid of  fine detail or texture.

What's interesting about the DXO tests comparing the D3X with the P65+ is that, at probably any print size, the D3X will record as much (or more) detail in the deepest shadows as the P65 can, but not if the deepest shadows are at a brightness level equivalent to 18% gray, which is the level DXO uses to measure SNR. The P65+ still retains a noise advantage over the D3X in the mid-tones and lower mid-tones.

Owners of DBs who have made their own comparisons with a D3X, are probably seeing these slightly higher noise levels in the mid-tones and lower mid-tones of the D3X shot (at equal image/print size) and are therefore concluding that the DXO results must be wrong. In fact, the DXO results show that at an average scene reflectance of 18% gray, the P65+ has 4dB better SNR than the D3X. That's more than a whole stop better.

The fact that the D3X can have as much as 2/3rds of a stop lower noise in the deepest shadows, which may often be clipped to black for esthetic reasons when processing the image, is probably of little concern to the user of an MFDB system in the studio, but could be of concern, for example, to a landscape photographer shooting a waterfall in a dense rainforest, who wants to retain some detail in the surrounding, dense undergrowth of the forest.
Logged

cmi
Sr. Member
****
Offline Offline

Posts: 491


« Reply #398 on: September 28, 2009, 06:31:03 AM »
ReplyReply

Quote from: JeffKohn
I'm sorry, but you're building a straw-man argument. I never said anything about taking anything blindly. DxO Labs has a wealth of information available about what they test, how they test it, and how to interpret the results. Start here, here, or here.


Jeff, I can completely understand your position. Let me just say you are misunderstanding me. I have to make an important distinction: For your situation in judging cameras it is completely valid to trust on dx0, and you have all reason to do that, and for judging DSLRs (wich I wanna maybe buy, etc) I do exactly the same and use dxo. But this trust is no longer useful in a discussion which is about the actual FINDINGS of dx0! And correct me, but exactly the dx0 results are implicitely in question, because would we all trust in dx0 results, we would not need to make a discussion, it would be clear from the start! Right?

And since you cannot expect real working photographers to obey to the lab standarts of dx0, the relevant method here is own experience with two systems. No, I do not expect you to have this experience, but it does just mean, you, and also me, cannot take part in a meaningful discussion. Thats all.

Basically, every argument wich does rely on knowledge instead of trust and stand completely on own feets is valid.

Practical, result centered comparisations of any two cameras in production scenarios are valid.

If Gabor would argue why his experiences in checking out raw files are relevant in comparing real world scenarios with different cameras, if he would explain his method and how he arrive at conclusions, then fine, that is also valid. It just has to be relevant for real production.

But just citing dx0 with nothing but trust is not valid for a serious discussion. On the other hand, citing dxo along with explanations of WHY it is valid (other than trust) are ok again.

Maybe thats more clear.
Logged
Jeremy Payne
Guest
« Reply #399 on: September 28, 2009, 06:54:07 AM »
ReplyReply

Quote from: Christian Miersch
Maybe thats more clear.

Not at all.
Logged
Pages: « 1 ... 18 19 [20] 21 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad