Pages: 1 ... 4 5 [6] 7 8 9   Go Down

Author Topic: A Biased Evaluation of The Differences...  (Read 129251 times)

Nemo

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 276
A Biased Evaluation of The Differences...
« Reply #100 on: January 19, 2008, 04:52:58 am »

Quote
Having now compared a few 5D and D3 RAW images, it is my view that the noise advantage of the D3 is nowhere near 2 stops. I've got more images to compare, that differ by a 1/3rd of a stop, 1/2 a stop and 2/3rds of a stop etc., so at this stage I'd hesitate to put a figure on the D3 noise advantage. It's probably about 1/2 to 2/3rds of a stop,
[a href=\"index.php?act=findpost&pid=168116\"][{POST_SNAPBACK}][/a]

Your analysis is very interesting Ray.

It isn't surprising you can see an improvement in noise performance in the D3, compared to the 5D. On one hand, there are two years of technical progress (hardware, software). On the other hand, the base level ISO of the D3 is a full stop higher. One stop of difference is predictable. Lets see if Canon updates the 5D now (keeping the resolution).
« Last Edit: January 19, 2008, 04:54:04 am by Nemo »
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
A Biased Evaluation of The Differences...
« Reply #101 on: January 19, 2008, 05:12:20 am »

Quote
It isn't surprising you can see an improvement in noise performance in the D3, compared to the 5D. On one hand, there are two years of technical progress (hardware, software). On the other hand, the base level ISO of the D3 is a full stop higher. One stop of difference is predictable. Lets see if Canon updates the 5D now (keeping the resolution).
[a href=\"index.php?act=findpost&pid=168117\"][{POST_SNAPBACK}][/a]

The really meaningful comparison is not comparing the ISO numbers but the exposures. A full stop improvement in noise means that noise in a D3 image at f8 and 1/50th should be no worse than noise in the 5D image at f8 and 1/25th.

I doubt that the improvement is quite that good.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
A Biased Evaluation of The Differences...
« Reply #102 on: January 19, 2008, 05:44:55 am »

Well, I think these crops below indicate that the the D3 noise advantage compared with the 5D is definitely less than 1 stop. The 2 images are unsharpened, unfiltered and generally unprocessed apart from EC adjustment in ACR. Both images are slightly underexposed in relation to an ETTR at their respective ISO settings of ISO 3200 for the 5D and ISO 6400 for the D3.

[attachment=4725:attachment]

The 5D is f8 at 1/25th and the D3 f8 at 1/50th.
« Last Edit: January 19, 2008, 05:46:51 am by Ray »
Logged

Streetshooter

  • Full Member
  • ***
  • Offline Offline
  • Posts: 134
A Biased Evaluation of The Differences...
« Reply #103 on: January 19, 2008, 06:07:58 am »

Quote
The really meaningful comparison is not comparing the ISO numbers but the exposures. A full stop improvement in noise means that noise in a D3 image at f8 and 1/50th should be no worse than noise in the 5D image at f8 and 1/25th.

I doubt that the improvement is quite that good.
[a href=\"index.php?act=findpost&pid=168118\"][{POST_SNAPBACK}][/a]


Ray, what are you trying to prove ?  I think you are arguing with yourself and everybody else has left the room !  Michael has stated along with numerous other busy pros that the new D3 has the edge in high iso shooting now. They use their cameras day in and day out shooting thousands of images, so their testimonials are enough for me. Most of them are die hard Canon guys too.

There really is nothing to more that can be said.

Pete
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
A Biased Evaluation of The Differences...
« Reply #104 on: January 19, 2008, 06:37:50 am »

Quote
Ray, what are you trying to prove ?  I think you are arguing with yourself and everybody else has left the room !  Michael has stated along with numerous other busy pros that the new D3 has the edge in high iso shooting now. They use their cameras day in and day out shooting thousands of images, so their testimonials are enough for me. Most of them are die hard Canon guys too.

There really is nothing to more that can be said.

Pete
[a href=\"index.php?act=findpost&pid=168122\"][{POST_SNAPBACK}][/a]

I also use my camera day in and day out and shoot thousands of images. I've shot so many images on this trip my 5D's mirror fell off. I'm not trying to prove anything. I'm trying to find specific information which I've been unable to find on the internet. Everyone seems happy to go along with the mantra that the D3 has a very substantial noise advantage over the current Canon ff DSLRs. Michael has stated in this thread that it's about 2 stops better than anything else. This is not the same as saying the D3 has the edge in high ISO shooting, is it?

Finding such claims difficult to believe, I've checked it out for myself. I'm sharing the results with you. You're quite entitled to ignore the evidence before your eyes and believe any testimonials you like.
Logged

michael

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5084
A Biased Evaluation of The Differences...
« Reply #105 on: January 19, 2008, 08:20:32 am »

Wait a second.

What I said (or at least meant) is that the D3's high ISO capability is at least 2 stops better than anything else, not that there is inherently anything superior about the D3 at anything else other than the top ISO settings.

And, I would add that simply cranking the exposure slider in raw processing is emphatically not the same as having either a true ISO setting or even a boosted ISO setting in the camera.

As for comparisons done by me, or by others, there's no point in arguing. We each see what we see. Owning a 1Ds MKIII, a 5D and a D3, and having done enough comparisons for now my own purposes, I stand by my assertion. The difference is clearly visible, and enough of my colleagues have now also seen these comparisons that I feel comfortable with my conclusion.

I will NOT be publishing anything on this because I have no desire to be hung, drawn and quartered in the town square. Even with lots of charts and graphs with lines and arrows on the back, I'd still be called to task, and have no patience for this, at least in this instance.

Michael
Logged

James R Russell

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 992
    • http://www.russellrutherford.com/
A Biased Evaluation of The Differences...
« Reply #106 on: January 19, 2008, 09:06:14 am »

Quote
Wait a second.

What I said (or at least meant) is that the D3's high ISO capability is at least 2 stops better than anything else, not that there is inherently anything superior about the D3 at anything else other than the top ISO settings.

And, I would add that simply cranking the exposure slider in raw processing is emphatically not the same as having either a true ISO setting or even a boosted ISO setting in the camera.

As for comparisons done by me, or by others, there's no point in arguing. We each see what we see. Owning a 1Ds MKIII, a 5D and a D3, and having done enough comparisons for now my own purposes, I stand by my assertion. The difference is clearly visible, and enough of my colleagues have now also seen these comparisons that I feel comfortable with my conclusion.

I will NOT be publishing anything on this because I have no desire to be hung, drawn and quartered in the town square. Even with lots of charts and graphs with lines and arrows on the back, I'd still be called to task, and have no patience for this, at least in this instance.

Michael
[a href=\"index.php?act=findpost&pid=168137\"][{POST_SNAPBACK}][/a]

I've shot the D3 next to the 1ds MKII the 1dMKIIN in daylight, subdued light, window light, mixed light, stadium light, tungsten light, tungsten with flash fill, models, celebs, atheltes, real people, non real people (yes there are non real people)  not one but thousands of frames and frame per frame, regardless of processor, regardless of exposure the D3 gives me 2 added stops and at high iso the detail is more defined.

Now this doesn't mean it's my favorite camera, only camera, or camera I chose to use for everything  because it's not, but at high iso, nothing I've seen is close.

In fact it allowed me to shoot something in studio with 1k tungstens that we just could not have done with any other camera I've owned and I can't comment on the circumstance but it is one of the few instances where the client noticed and the camera actually saved the day.

James Russell
Logged

Graeme Nattress

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 584
    • http://www.nattress.com
A Biased Evaluation of The Differences...
« Reply #107 on: January 19, 2008, 09:06:29 am »

What we're seeing is camera / software / computer as system, rather than camera = image. And I think we're seeing two different raw philosophies - raw =  raw and raw = corrected raw.

Now, noise reduction is a very powerful and useful technique, and doubly powerful on raw data compared to RGB data, if you know what you're doing. I think with what Nikon are doing, and this is speculation as I obviously don't have a circuit diagram, is clever RAW noise reduction, especially on chroma. If Nikon have a very good technique, and looking at the images they probably do, then it certainly makes sense for them to allow you to do it in RAW which gives them a better JPEG, and keeps the files compatible, with a noise advantage, with normal raw conversion apps that lack advanced noise reduction.

In photography it's the end result that counts, but knowing how that end result is achieved can educate your shooting and help you get the best out of your images. That's the beauty of photography - it's technical and artistic, and ignore either of those aspects and your images suffer.

I think it's over-simplistic to look at the D3 and say it's image advantage is all sensor, or all image processing software on it's raw data. It's more complex than that, and that's why we get long threads like this...

Graeme
Logged

Nemo

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 276
A Biased Evaluation of The Differences...
« Reply #108 on: January 19, 2008, 09:11:48 am »

Quote
The really meaningful comparison is not comparing the ISO numbers but the exposures. A full stop improvement in noise means that noise in a D3 image at f8 and 1/50th should be no worse than noise in the 5D image at f8 and 1/25th.

I doubt that the improvement is quite that good.
[a href=\"index.php?act=findpost&pid=168118\"][{POST_SNAPBACK}][/a]

Well, I don't fully agree on this.

Cameras have different signal-to-noise ratios. Keeping the signal at the same level (identical exposure) you can compare noise. Noise depends on many variables. Read noise, thermal noise, reset noise...

It is difficult to make a meaningful comparison. Too many variables at play, so too easy to argue against any conclusion. Several comparisons under different practical circunstances is the best way to follow. Experienced photographers are an aid here. I understand Michael's comment about not publishing (at this moment) any comparison. His opinion is valuable for me. The Nikon D3 presents a serious improvement on any other camera on the market now, due to better signal handling (quantum efficiency, microlenses, on sensor filters?) and better noise reduction (hardware or software), and this is that counts. It is good news for photographers. Sony and Canon are forced now to match. The 1Ds Mark III is a different beast (higher pixel count, lower base ISO), but the sucessor of the 5D might do it, but this camera is an all-around product aimed at a different public. The 1D Mark III is Canon's professional reportage camera, and the D3 is Nikon's answer to the 1DMIII treat. Maybe we must wait for the 1D Mark IV...
Logged

Mort54

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 590
    • http://
A Biased Evaluation of The Differences...
« Reply #109 on: January 19, 2008, 10:01:07 am »

Quote
Well, I'm glad you find me entertaining.
Hi Ray. I didn't say I found you entertaining - I said I found your assertion that you weren't being biased on this amusing. You seem like a very smart person, and it isn't my intent to demean you in any way. I'm sure I could learn a lot from you.

Quote
There's a strong tradition on this site of promoting the benefits of RAW capture. All of us with a bit of experience are aware that RAW capture gives us the potential to get the best results.
Yes, I shoot almost totally RAW myself. I'm not sure what this has to do with any of this, however. The D3's advantages in noise aren't unique to its JPEGs. Many of those commenting on this haven't qualified their statements to being only about JPEGs.

Quote
I'm not a fan boy in any respect. If you read my posts, I've praised the capabilities of the D3 in this thread. If you shoot jpegs, the benefits are obvious and clear.
Same comment as above.

Quote
Do you not believe your eyes, Mort54? Would you rather believe the opinions of experts unsupported by RAW comparisons?
Of course. I don't go by the comments of others. I'm simply using the comments of others to butress my arguments. You aren't the only person doing comparisons, Ray. To suggest that only yours are valid is an interesting position to take. I've done my own comparisons and I'm content with my position based on those comparisions. All of the comments by others are simply icing on the cake.

Quote
I never said that. Why are you misquoting me? I've said that it appears the D3 is applying more in-camera noise reduction than the 5D, especially with regard to chroma noise which can, however, be removed in software without degrading resolution.
I didn't quote you Ray, I paraphrased you :-) Your thesis (as I read it) is that the D3's magical noise properties are due largely to the fact (your assertion, not mine) that Nikon does noise reduction on their RAWs, and that Canon doesn't. I'm simply pointing out the obvious - that you have yet to support your thesis (your Noise Ninja comments don't cut it - I also see noise reduction on my D3 RAWs when I apply NN and other post processed NR techniques). I, on the other hand, did provide a reference to Nikon's D3 manual that stated that NR isn't applied to it's RAWs until ISO 2000 and higher (this isn't totally convincing, I agree, but it's more than you've offered up in support of your assertions). In any event, I'm not sure why I'm arguing this point (probably just to be annoying :-), since all I really care about are the results, regardless of how they are achieved. Maybe we're in violent agreement on this.

Quote
....but I have said, and have also demonstrated, that such objectionable chroma noise in the 5D image can be easily removed with Noise Ninja.
Yes, you have.

Quote
Nonsense! All camera companies need to be kept in the game. If they are not kept in the game, they're out of the game or taken over by other companies like Minolta was. All these companies have marketing departments and do research on the customers needs and wants.
Your assertion implied (to me, at least) that the hubub over D3 noise advances was mostly marketing hype. I think it goes way beyond that.

Quote
It's my experience that most DSLR owners I meet on my travels shoot in jpeg mode. I bet most journalists do too. I think Nikon are appealing to that customer base with the D3. It seems there are only a few fanatics on LL who consistently shoot in RAW mode and are prepared to spend time processing and manipulating their favourite images. Good results out of the box seem the preferred option for most.
I fall into the RAW is better camp as well. Unless I'm shooting family snapshots in good light, I have my D3 set to shoot NEFs. So I am very familiar with examining RAW files, and evaluating RAW files, and processing RAW files. And I know noise, and the lack thereof, from long personal experience with Nikon cameras. I've seen the bad, and now I'm seeing the good. The biggest ding on Nikon bodies has been their higher noise at high ISOs, and that ding was largely deserved (tho highly overrated by the Canon community, in my opinion). But again, you seem to be the only person declaring that the D3's low noise advantages are mostly a JPEG thing. That's certainly not what I'm seeing.

Anyway, hopefully we can get on more positive ground and agree that the D3 is a huge advance for Nikon in the noise department (your comments suggest you agree with this statement), and in fact sets new standards in the noise department (however they achieve this). Is it the be all, end all. Certainly not - in fact, for what I like to shoot, the 1DsIII is certainly a better choice (my P45+ is an even better choice, but that's a whole other subject), but since I'm invested too heavily in Nikon glass, that's not going to happen. I'm patiently awaiting the D3X. Untill then, I'm happily shooting my D3 (and my P45+).

Regards,
Mort.
« Last Edit: January 19, 2008, 10:16:34 am by Mort54 »
Logged
I Reject Your Reality And Substitute My

John Sheehy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 838
A Biased Evaluation of The Differences...
« Reply #110 on: January 19, 2008, 10:29:13 am »

Quote
Those who are saying that the D3 is 2 stops better are most probably not just talking about noise, they are talking about overall image quality, meaning mix of detail retention, color saturation, ability to post-process the image, look of the noise,...
[a href=\"index.php?act=findpost&pid=168091\"][{POST_SNAPBACK}][/a]

That sounds like a lot of superstition about what makes an image; most of the things you mention are directly related to noise.

There are so many variables available for people to ignore when making comparisons.  People are easily fooled.  Most people don't realize that small prints and monitors give an unfair advantage to images with less pixels, and they attribute that to "better image quality because of lower pixel density", which isn't true at all.  The methods that most programs use to reduce the image for viewing on the screen preserve individual pixel noise to a small or large degree, and the image with the most pixel resolution loses resolution the most, and gains the most image-level noise.  Same can happen when going to a small print.  The way printer drivers have traditionally worked is nonsense.  Intermediate binning (to 300 PPI or 360 PPI, for example) is BS, and ruins images.  It ruins them more, once again, for the higher resolution images, especially at small sizes.  We are living in an age where our capture technology resolution surpasses our display technology resolution by a huge margin.

Then there's noise reduction.  It seems that most people think that low noise means seeing low noise in the final output, with no regard for what is lost in the process, as if the very purpose of photographs was to not show obvious signs of noise.  I don't think this is wrong for the purposes of an individual photograph, but the facts should be clear about what the camera itself is actually capturing, regarding noise vs detail.  When you heavily reduce chromatic noise in PP (post-RAW, not post conversion), you can get away with more saturation, as saturation would increase the visibility of the chromatic noise.  What you usually get is a very low-frequency chromatic noise, with soft blotches covering hundreds of pixels each, something you would probably not see at all if you did no noise reduction at all, and your monitor could display your 12 or 22 million pixels all at once; much like what you get when you look at an image that looks noisy at 100% on the monitor, and then go stand 12 feet from the monitor.  Edges are retained, while 2-dimensional pixel noise dissolves away (1-dimensional or "banding noise" is a bit more persistent, but that tends to be much lower than it used to be in the newest cameras).
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
A Biased Evaluation of The Differences...
« Reply #111 on: January 19, 2008, 11:19:17 am »

Quote
What I said (or at least meant) is that the D3's high ISO capability is at least 2 stops better than anything else, not that there is inherently anything superior about the D3 at anything else other than the top ISO settings.

Michael,
That's what I meant by 'superior' too. The D3 is being marketed as, and being supported by many professionals as a camera with break-through low noise technology. It might be break-through for Nikon, but all indications from my comparisons of the RAW files from both cameras is, it's only a marginal improvement on the 5D in respect of low noise at high ISO in RAW mode.

However, I'm in a quandary here. I've botched the tests with regard to FoV. I'm going to put the blame on the accuracy of the naming of lens focal lengths for this. The 5D RAW when converted to 16 bit tiff is 72.8MB. The D3 RAW when converted is 69MB.

Reducing the 5D file size to 69MB should result in equal size images at the same enlargement. However, because of the mismatch of FoVs, I have to reduce the 5D file to 62MB. I'm throwing away over 10MB of 5D image information in the comparisons.

Should this more or less compensate for the mismatch? I certainly can't be sure, so I'm not sure I should continue posting comparisons demonstrating that the noise advantage of the D3 compared with the 5D is of the order of 1/4 to 1/3 of a stop and with a bit of judicious out-camera noise reduction maybe even zero.

I'll see if I can persuade Nik's (Thailand Co. Ltd) on Monday to hire out their demo D3 to me for a day, but I'm doubtful they'd agree.

Quote
As for comparisons done by me, or by others, there's no point in arguing. We each see what we see. Owning a 1Ds MKIII, a 5D and a D3, and having done enough comparisons for now my own purposes, I stand by my assertion. The difference is clearly visible, and enough of my colleagues have now also seen these comparisons that I feel comfortable with my conclusion.

Michael, I don't think we are arguing about what we are seeing. We're not arguing about, 'this print or image on monitor has more (or less) noise that that image'. We're arguing about methodology. I would argue that choice of f stop and shutter speed is fundamental in the taking of a shot. How the camera responds to that choice is the crucial issue. ISO settings can be a smoke screen. The mere fact that the D3 actually has an ISO 25,600 setting is a wow! factor that I believe is misleading. It's main use is for jpeg shooters.

Quote
And, I would add that simply cranking the exposure slider in raw processing is emphatically not the same as having either a true ISO setting or even a boosted ISO setting in the camera.

I agree that it's not the same as having a true ISO setting. But a boosted ISO setting? There's a limit to the amount of personal checking of statements by others one has time to make. I've accepted that ISO 3200 on the 5D is a boosted ISO which is essentially the same as ISO 1600 underexposed by one stop.

ISO 3200 on the 5D underexposed one stop is therefore equivalent to a boosted ISO 6400 (ignoring discrepancies in ISO accuracy). The fact that Canon has not provided a boosted ISO 6400 setting on the 5D should not deter users from providing their own ISO 6400 setting by simply underexposing ISO 3200 by one stop.

There's a psychological element to this. The fact that Nikon have provided an ISO 12,800 and 25,600 option is going to fool some people into believing that it's of real and fundamental benefit. My contention would be it's of real benefit only for in-camera processing of jpeg images and for the appearance of camera LCD previews which will look correctly exposed rather than very underexposed.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
A Biased Evaluation of The Differences...
« Reply #112 on: January 19, 2008, 11:34:29 am »

Quote
I've shot the D3 next to the 1ds MKII the 1dMKIIN in daylight, subdued light, window light, mixed light, stadium light, tungsten light, tungsten with flash fill, models, celebs, atheltes, real people, non real people (yes there are non real people)  not one but thousands of frames and frame per frame, regardless of processor, regardless of exposure the D3 gives me 2 added stops and at high iso the detail is more defined.
[a href=\"index.php?act=findpost&pid=168142\"][{POST_SNAPBACK}][/a]

Wow! There's something very odd going on here. This warrants further investigation. My tests show something in the order of 1/3rd of a stop.

One point which puzzles me. How can the D3 give you a 2 stop advantage regardless of exposure? It's all about exposure (at a given f stop) isn't it?
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
A Biased Evaluation of The Differences...
« Reply #113 on: January 19, 2008, 11:43:32 am »

Quote
In photography it's the end result that counts, but knowing how that end result is achieved can educate your shooting and help you get the best out of your images. That's the beauty of photography - it's technical and artistic, and ignore either of those aspects and your images suffer.
[a href=\"index.php?act=findpost&pid=168143\"][{POST_SNAPBACK}][/a]

My sentiments exactly! Well put!  
Logged

Graeme Nattress

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 584
    • http://www.nattress.com
A Biased Evaluation of The Differences...
« Reply #114 on: January 19, 2008, 12:02:31 pm »

The way digital cameras work means that they no over-exposure capabilities as they hard clip, and dynamic range is defined by how far you can go into the shadows and still either pull out detail, or what your tolerance to noise is. This is not helped by patterns in the noise, or subjective differences in the texture of noise and the different definitions of dynamic range that individuals use.

To say a camera has a stop advantage in noise, to me, means that it has a stop more dynamic range. Dynamic range in notoriously tricky to judge, and I've found it cannot be properly measured without very controlled tests and calibrated test targets in an environment that eliminates stray light. To properly examine the full DR of these cameras that now are going to 14bit AtoD, we are at the limits of what a single backlit step wedge test can deliver, and we have to be super careful to eliminate any stray light  that would make blacks not black enough.

Then we need to define what we're comparing when cameras have different size and density sensors. That alone makes exact comparisons tricky.

What tests can do, however, is help us know the limits of the technology, what we can get away with, what we can not. With noise, what software in post is useful, which is not, what is better done in camera or in RAW data or RGB data. What can be important shooting RAW is to learn that benefits in raw processing algorithm, something I'm doing on a daily basis, can vastly alter or improve how the final image looks. It can mean we can re-visit an old image and develop it differently and get results that make it usable, when once it was not. Or we might find out that no amount of clever processing can rescue it, and if we intend to do more of that kind of shooting, we need a camera with a lower noise sensor.

Graeme
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
A Biased Evaluation of The Differences...
« Reply #115 on: January 19, 2008, 12:52:07 pm »

Quote
To say a camera has a stop advantage in noise, to me, means that it has a stop more dynamic range. Dynamic range in notoriously tricky to judge, and I've found it cannot be properly measured without very controlled tests and calibrated test targets in an environment that eliminates stray light. To properly examine the full DR of these cameras that now are going to 14bit AtoD, we are at the limits of what a single backlit step wedge test can deliver, and we have to be super careful to eliminate any stray light  that would make blacks not black enough.
[a href=\"index.php?act=findpost&pid=168190\"][{POST_SNAPBACK}][/a]

Graeme,
Isn't there a distinction to be made between devising an objective standard for DR that applies to all cameras, and comparing the DR of one camera with another in relation to a specific scene?

If we presume that the viewer is applying her own visual standard of acceptable shadow detail, then it's easy for any individual viewer to determine that image A has a higher dynamic range than image B, and probably most people with average or normal vision would agree with the comparison. I find it improbable that most people could raise a difference of opinion as to image A being more, or less, noisy than image B, assuming such viewers were familiar with the concept of visual noise and could recognise it, and assuming also we're not talking about extreme pixel-peeping scenarios.
Logged

Graeme Nattress

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 584
    • http://www.nattress.com
A Biased Evaluation of The Differences...
« Reply #116 on: January 19, 2008, 02:42:02 pm »

Well, that's still tricky as you never quite know  how the camera is mapping light to code values. Say one sensor is 1 stop more sensitive than another - then equal exposure settings could produce clipped highlights on one, and not on the other. Say that how the RAW data is developed into a JPEG either in camera, or to a TIFF in your RAW converter uses a different curve for different cameras, and maybe the cameras treat black point data differently, you could get fooled by just comparing images.

That's why a standardized target makes life easier, especially if you can access the raw data without a curve applied, then you can measure between clipping (where ever it is) and what you tolerate in the noise, where ever that might be, and declare the dynamic range as the difference between the two. As long as you're consistent in what you call clipping and what you call too noisy, it works well. Doing a log / log plot of values also helps  as then you can visually see the deviation from linearity as you head into the noise floor.

Graeme
Logged

Sunesha

  • Full Member
  • ***
  • Offline Offline
  • Posts: 103
    • http://sunesha.se
A Biased Evaluation of The Differences...
« Reply #117 on: January 19, 2008, 03:21:36 pm »

I have a friend that works as Private Dective, the big reason he bought a Nikon D3 is the silly 25,6 k ISO. That he tells me does his "spying" business a great deal. He say that as long you can be sure that you can see who it is on the photo it is a moneyshoot for his "spy" business.

He used to use a 5D because its great iso performance. But he loves to have those silly settings as that stops blur, blur is more often his enemy than noise. As most off his photos done by night in avaible light. To use a flash would probaly be unwise in his kind off work  

I think he does very interesting type off photography.

This is not my opinions, i just repeat what he told me. That he has very new abilities to make moneyshoots. Just to bring the discussion, just because you think it un-useble some other can make money off function you dont need.

For me as a more a "normal" type off photographer the D3 brings Nikon to complety new level off photography that wasnt possible just 6 months.

I always watched the Canon as they both make fullframe cameras, but my lens collection that I am very attached too and the NikonĀ“s approach how you control your camera has always made me stay with the Nikon brand.

Actually my personal reason to use Nikon is that controls are much easier use, that is more important for me than marginal Image quality differences which has becam

What I find interesting as digital cameras evolve is how much less image qualtity doesnt matter. As most digital SLRs has a very small difference in Image quality nowdays. I think manufactures shall make more and more smarter access to the functions off our cameras. After all I work with the tool many hours and it is interface is what makes to take the shoot. I think that biggest diffrence in "technical" quality is made in photoshop. As we have great tools to correct "technical" flaws today.

The only thing that ticks me off is that Nikon is so so stupid that mirrorlockup can be used with selftimer. That would save me to use a remote-control. I am so so stupid that often forgets my remote control at home.

I think people are to obsessed with image quality that they overlook most others functions that we deal with everytime we take a photo.

My dream camera is fast used camera with accessible smart design off the buttons. Make it almost waterproof. I think we are close to that with the current nikon cameras. I am no genius but every design news that make take shoot easier and faster can make me a better photographer.

What I like with Luminous Landscapes "reviews" is that they review how it is to use the camera rather than measure the camera.

I off course like some measure-bating. But when focus has shift from how it is use the camera from discussing image noise I stop reading the review.

A camera I cant take shoots with my gloves on easy, is a unuasble camera for me, it doesnt matter if takes the best photos in world. But if spend a day without gloves in a freezing climate I dont enjoy taking photos.

But we all have diffrence needs, thats why believe we have so many choises
Logged
Daniel Sunebring, Malmoe, Sweden
Homepag

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
A Biased Evaluation of The Differences...
« Reply #118 on: January 19, 2008, 06:11:07 pm »

Quote
In other words, there's a big sales push to keep Nikon in the game and maintain the pressure on Canon.

Over all image quality cannot be measured in f stops, Bernard. Post processing capability is all there in ACR.... detail enhancement, vibrancy, color saturation, temperature and tint, sharpening etc.
[a href=\"index.php?act=findpost&pid=168099\"][{POST_SNAPBACK}][/a]

Ray,

Well, you'll probably agree with me that the reason why some cameras deliver less saturated high ISO RAW files is that there is just more entropy on those files that do not enable them to make the image look good while maintaining a high level of saturation? In other words there is less color information in the file. Pushing the saturation back in ACR results in comb like histogram, and you know what that means as well as I do... the amount of post-processing doable from then on is very limited.

I believe that this can be measured in f stops, just like the rest, but I did personnally not compare the D3 that I own with the Canons in realworld situations. All I am saying is that at least MR, and others as well who have tried both cameras claim they see a significant difference that we assessed as being 2 stops.

Now, frankly speaking, the D3 is more than good enough at high ISO for my needs, and I do intend to devote more time to this topic.

Cheers,
Bernard

John Sheehy

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 838
A Biased Evaluation of The Differences...
« Reply #119 on: January 19, 2008, 11:22:29 pm »

Quote
Ray,

Well, you'll probably agree with me that the reason why some cameras deliver less saturated high ISO RAW files is that there is just more entropy on those files that do not enable them to make the image look good while maintaining a high level of saturation? In other words there is less color information in the file.

Noise and the quality of the CFA filters are the only things that are going to affect color.  There are no saturation pixies in the cameras.

Quote
Pushing the saturation back in ACR results in comb like histogram, and you know what that means as well as I do... the amount of post-processing doable from then on is very limited.

The more saturated "default" output is probably just a different default, internally, even though the sliders are the same.  Converters like ACR do not treat RAWs from all cameras the same.  They would lose customers if the conversions were too much different from the camera JPEGs, or the manufacturer's converter.

Quote
I believe that this can be measured in f stops, just like the rest, but I did personnally not compare the D3 that I own with the Canons in realworld situations. All I am saying is that at least MR, and others as well who have tried both cameras claim they see a significant difference that we assessed as being 2 stops.

But they have never seen the 1Dsmk3 RAWs converted the same way as the Nikon RAWs, or visa-versa.  That is why I say that someone should take exactly the same shot (with a tripod-mounted lens, if possible), with the same lens, same Av and Tv values, and provide the RAWs so that they can be converted homogenously.  Then the naked truth about the cameras' noise will be displayed.  Then people can start asking the right questions.  If the saturation levels look very different, then that avenue can be explored, but IME, both Canon and Nikon RAWs need about the same amount of saturation boost when converted homogenously, to look natural.

Quote
Now, frankly speaking, the D3 is more than good enough at high ISO for my needs, and I do intend to devote more time to this topic.
[a href=\"index.php?act=findpost&pid=168258\"][{POST_SNAPBACK}][/a]

The D3 is the second best low-light DSLR, assuming the 1Dsmk3 is the best.  No one is saying that its low-light performance isn't excellent.  It just isn't the Canon-Killer that it is made out to be, unless you're shooting for someone who values the lack of visible noise over anything and must hand them the CF cards right out of the camera, because Canon has always had a policy of very low noise reduction in their conversions.
Logged
Pages: 1 ... 4 5 [6] 7 8 9   Go Up