Ad
Ad
Ad
Pages: « 1 2 3 [4] 5 6 7 »   Bottom of Page
Print
Author Topic: D800E -Truly Amazing Low-Light Performance  (Read 23475 times)
ErikKaffehr
Sr. Member
****
Offline Offline

Posts: 7915


WWW
« Reply #60 on: November 20, 2012, 03:35:34 PM »
ReplyReply

Hi,

I would just say that it is really nice to have a first class camera at a reasonable price. The D800 is something like 3000 $US below the D4, for many of saving 3000$US is quite significant. So if the D800 offers D4 like high ISO quality is not just about mesurbation but about saving money. The other side of the coin that the D800 also gives you 36 MPixels.

Yes, you can do your own test. Just get a rental on all pieces, read the fine manuals, set up a few test subjects and do some well executed test shots, avoid mistakes and learn to process your files optimally and than do an educated choice.

Best regards
Erik
Logged

Slobodan Blagojevic
Sr. Member
****
Offline Offline

Posts: 6190


When everybody thinks the same... nobody thinks.


WWW
« Reply #61 on: November 20, 2012, 03:48:56 PM »
ReplyReply

... Jeremy... To read that as my 'looking out for' anyone is beyond absurd....

... Slobodan...For crying out loud read what's actually written. 

... h, no that's not what I'm saying...

Seems like a lot of people do not know how to read what you are saying.

Could it be that the problem is in what you are saying and how?

Logged

Slobodan

Flickr
500px
RFPhotography
Guest
« Reply #62 on: November 20, 2012, 04:21:47 PM »
ReplyReply

Truthfully, no I don't think so.  I think it's more a case of people with an agenda twisting or inferring meanings to suit that agenda.
Logged
jeremypayne
Guest
« Reply #63 on: November 20, 2012, 06:43:47 PM »
ReplyReply

Truthfully, no I don't think so.  I think it's more a case of people with an agenda twisting or inferring meanings to suit that agenda.

Is it still paranoia if they REALLY ARE out to get you?

Thanks for the laugh of the week, Bob.  Truly.

Logged
BernardLanguillier
Sr. Member
****
Offline Offline

Posts: 8359



WWW
« Reply #64 on: November 20, 2012, 10:26:40 PM »
ReplyReply

I think it's more a case of people with an agenda twisting or inferring meanings to suit that agenda.

Aaahhhh... so you do not have any agenda? I am glad this is cleared now.

I must have been confused by the many posts you have written about the lack of actual benefits of the technologies embedded in cameras not bearing a Canon logo, in particular the D800.  Wink

Cheers,
Bernard
« Last Edit: November 20, 2012, 10:33:35 PM by BernardLanguillier » Logged

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

Posts: 1712


« Reply #65 on: November 21, 2012, 02:23:31 AM »
ReplyReply

Truthfully, no I don't think so.  I think it's more a case of people with an agenda twisting or inferring meanings to suit that agenda.
Since the quote seems to refer to me, Slobodan and Jeremy: What is my agenda? In what way am I twisting your meanings? Please provide support for your accusations.

In the mean time, perhaps you can try to explain what you really think, because your posts leave me confused. You seem to be saying that:
 Measurements can sometimes be poorly carried out or misinterpreted


It is hard to disagree on that. Given that we all agree that a given measurement _can_ be of little use for a given problem, what is the point in repeating that ad infinitum?

...Good data or bad data?  But it doesn't need to be a doctoral dissertation on the subject to do that.
Are you saying that doing measurements and interpreting them is simple? I tend to disagree: there is a reason why we educate PhDs: doing measurements of some kind that are repeatable, applicable outside of the lab and sufficiently novel that the results actually add to the accumulated human knowledge is quite hard.

I believe that I have given you examples where misguided experts/reviewers have given bad advice based on their home-brew "down-to-earth" testing methology. In other words: while a PhD may not be needed to do measurements and analyze their results, there are plenty of examples of good practitioners (without a PhD) giving poor advice.
As far as 'better' results with a D4 or D800, better is a wildly subjective term.  What is 'better' in one instance for one person may not be for another.  
I don't think it takes a full understanding of, for example, Emil's treatise on noise to really understand what noise in a digital image is and that less is generally better.

I don't think it takes an engineering degree to realise that, for the most part, today's cameras are better than ones that came before in terms of noise performance/characteristics and dynamic range, among other things.
So which is it... Are some cameras "better" or is it all subjective?

.
Yes, I do think that at times people can get caught up in the details and lose good pictures as a result.  It's also true; however, that technically perfect pictures aren't always the best either.  
You seem to be argueing against a strawman here. Everyone (?) agrees that technical issues are only one component of the final image.
Contrary to what your comment may seem to indicate, I'm not opposed to lab tests; and said as much in my previous remarks.  
That is good to hear.
Quote
I read them and take the information into account when I'm evaluating a new camera or lens.  But my point is there's more to it than just the lab numbers and that an over-reliance on the lab numbers alone isn't a good approach.  
So who among the thread-contributers is it that does this, in your view? Or are you discussing hypothetical problems?

-h
« Last Edit: November 21, 2012, 02:38:39 AM by hjulenissen » Logged
jeremypayne
Guest
« Reply #66 on: November 21, 2012, 06:13:35 AM »
ReplyReply

I think it's more a case of people with an agenda twisting or inferring meanings to suit that agenda.

Yes, I, too would like to know what my secret agenda is.

It would be much easier to make the correct inferences to suit that agenda if I knew what it was ... please, do tell!
Logged
RFPhotography
Guest
« Reply #67 on: November 21, 2012, 06:17:49 AM »
ReplyReply


I must have been confused by the many posts you have written about the lack of actual benefits of the technologies embedded in cameras not bearing a Canon logo, in particular the D800.  Wink

Cheers,
Bernard


What?  First, I don't shoot Canon.  Used to.  Don't any longer.  So pumping up Canon isn't something I'd really be interested in doing.  Second, I've never said that the D800 was a bad camera.  Never.  What I have said is that it isn't the camera I wanted from Nikon as a successor to the D700.  Despite that, since Nikon isn't, apparently, going to build the camera I (and a lot of other people) want, I did break down and buy one.  Much to my own chagrin.  Roll Eyes
Logged
Ray
Sr. Member
****
Offline Offline

Posts: 8939


« Reply #68 on: November 21, 2012, 06:26:09 AM »
ReplyReply

Bobfisher,
I really can't understand your concerns here. Have you got some sort antagonism against graphs? Do you suffer from graphobia (or should that be graphphobia)?  Grin

In my view the incessant arguing and bickering over differences in the various  aspects of  a camera's comparative technical performance, as often seen at sites such as dpreview, is due to a lack of appreciation of the significance of the sorts of measurements conducted by DXO.

However, I do understand that for anyone who is not familiar with the units used in the graphs, such data may not be at all illuminating. There's a basic requirement, which is surely not unreasonable, for a person to familiarize himself with the significance of the units used, before they become meaningful. And DXOMark, in their various articles and descriptions of the performance aspects they have measured, do provide explanations of what the units mean and how significant the differences in values may be.

For example, DR is measured in Exposure Values, abbreviated as EV, and a difference in one EV is equivalent to a difference in one F stop of exposure, all else remaining the same. DXOMark also mention that a difference of 0.5EV (or 1/2 a stop) will probably be noticeable, implying that differences of less than 1/2 a stop may not be noticeable.

This is the sort of information which I find useful when making a decision whether or not to buy a new camera. Without such reliable measurements, I might be tempted to make a decision based upon hyperbole and misinformation which can abound in the absence of reliable measurements, each news source just repeating the misinformation or biased view already perpetrated.

For example, to extend Erik's analogy, we already have comments in this thread that cameras such as the Nikon D4 and Canon 1DX have better high-ISO performance than the D800. Do you really not want to know how much better and over what range of high ISOs the D4 is better? The DXOMark graphs will give you a reliable indication of what to expect.

Perhaps you rarely shoot above ISO 400 with your D800 (if you have one) because of noticeable image degradation, but you've heard that the D4 has amazing high-ISO performance, so you buy the camera hoping that at last you can get clean images at high ISO.

How are you going to feel if you later discover the truth, that the D4 only has better DR than the D800 above ISO 800, in other words, at ISOs which you never use because they are too noisy? A 1/2 stop improvement in DR at ISO 1600 is still worse than DR at ISO 800 using either camera, and significantly worse than DR at ISO 400 with the D800.

How are you going to feel when you discover that any improvement in high-ISO noise from the D4 is only apparent in the deep shadows by a barely noticeable degree, and that SNR over most of the tonal range at high ISOs is no better than that of the the D800, nor is Color Sensitivity any better?
Logged
RFPhotography
Guest
« Reply #69 on: November 21, 2012, 07:30:43 AM »
ReplyReply

Ray, your response is so completely ridiculous and childish that it deserves no more response than that.
Logged
Slobodan Blagojevic
Sr. Member
****
Offline Offline

Posts: 6190


When everybody thinks the same... nobody thinks.


WWW
« Reply #70 on: November 21, 2012, 07:41:43 AM »
ReplyReply

I admit my responses so far are so completely ridiculous and childish that they deserve no more response than that.

Oooooops!

 Did I just misquote/misread what you wrote, again? My bad Wink
Logged

Slobodan

Flickr
500px
jeremypayne
Guest
« Reply #71 on: November 21, 2012, 07:48:23 AM »
ReplyReply

Ray, your response is so completely ridiculous and childish that it deserves no more response than that.

"Calling all secret agendas!  We have a live one in Sector 9!"

You are dropping 'ef bombs and conspiracy theories ... And Ray is being ridiculous and childish?
Logged
RFPhotography
Guest
« Reply #72 on: November 21, 2012, 08:00:35 AM »
ReplyReply

"Calling all secret agendas!  We have a live one in Sector 9!"

You are dropping 'ef bombs and conspiracy theories ... And Ray is being ridiculous and childish?

Yeah, you deride me for cursing after doing it yourself.  What's that word I'm looking for.... hy.... hyp-something.... hypocritical, that's it. 
Logged
BJL
Sr. Member
****
Offline Offline

Posts: 5171


« Reply #73 on: November 21, 2012, 08:05:54 AM »
ReplyReply

... 18% SNR is what you probably would have on facial tone or deep blue sky. These are typically surfaces you would like to have smooth tonality.

With a wide dynamic range it is well possible that we never utilize the low end. Shadow detail is normally kept pretty dark and would be even more compressed in printing.
Thanks for that comment Erik. Although there are situations where extreme DR is important (like scenes of unusually high Subject Brightness Range or forgetting to set the ISO speed appropriately when photographing moths at dusk), I would say that the vast majority of photography involves not much more that the traditional "normal" SBR of about three stops either side of mid-tones. Indeed the ASA/ISO film speed scale incorporated a similar idea, giving a speed that is the maximum Exposure Index that keeps the "noise floor" of the film about four stops below normal mid-tone placement.

With a typical mid-tone placement at about 12-18% of maximum diffuse luminosity, that puts the lower end of the commonly interesting luminosity levels at about 1-2%. So I would say that for the purpose of _most_ low light/high ISO speed photography, what matters most to the visible noise levels in the final displayed image is the SNR at various levels in this range, say down to 1%, or over the range of about seven stops down from highlight placement, or down to four stops below mid-tone placement. I would be far more interested in measurements and photographic examples of "performance over a given luminosity range" like this than I am in measurements of an extreme like "DR down to 1:1 SNR". Such quantification of extremes is often of dubious direct practical relevance, in the way that extinction resolution is less informative than MTF50.

The 18% SNR data are a useful start, and for good modern sensors like the D800(E) these seem to be dominated by photon shot noise up to very high ISO speeds, but I would expect sensor noise to become relevant about three stops lower if one also looks at noise down to some moderate-to-deep shadow level.
« Last Edit: November 21, 2012, 09:37:03 AM by BJL » Logged
jeremypayne
Guest
« Reply #74 on: November 21, 2012, 08:11:25 AM »
ReplyReply

Yeah, you deride me for cursing after doing it yourself.  What's that word I'm looking for.... hy.... hyp-something.... hypocritical, that's it. 

I'm not deriding you for cursing.

I'm making fun of you for other stuff ... Like cursing and claiming secret agendas while calling Ray childish ... or proclaiming that the rest of us aren't capable of using scientific data properly ...

That's not mocking you for cursing ... that's mocking you for ... let's just call it unintentional irony.
Logged
RFPhotography
Guest
« Reply #75 on: November 21, 2012, 08:24:50 AM »
ReplyReply

or proclaiming that the rest of us aren't capable of using scientific data properly ...



That comment is a perfect example of what I mentioned earlier.  Nowhere, nowhere did I say that people weren't able to properly use data.  There is absolutely no possible way that inference can logically or factually be gleaned from any remarks I've made. 

That is a fabrication of your own, personal insecurities and I take no responsibility for those.  That's something you'll have to deal with on yourself.

I'm done with this thread.
Logged
bjanes
Sr. Member
****
Offline Offline

Posts: 2866



« Reply #76 on: November 21, 2012, 09:37:42 AM »
ReplyReply

With a typical mid-tone placement at about 12-18% of maximum diffuse luminosity, that puts the lower end of the commonly interesting luminosity levels at about 1-2%. So I would say that for the purpose of _most_ low light/high ISO speed photography, what matters most to the visible noise levels in the final displayed image is the SNR at various levels in this range, from down 1%, or over the range of 6-7 stops down from highlight placement, or down to 3-4 stops below mid-tone placement. I would be far more interested in measurements and photographic examples of "performance over a given luminosity range" like this than I am in measurements of an extreme like "DR down to 1:1 SNR". Such quantification of extremes is often of dubious direct practical relevance, in the way that extinction resolution is less informative than MTF50.

The 18% SNR data are a useful start, and for good modern sensors like the D800(E) these seem to be dominated by photon shot noise up to very high ISO speeds, but I would expect sensor noise to become relevant about three stops lower if one also looks at noise down to some moderate-to-deep shadow level.

That makes sense to me. The engineering DR reported by DXO is less relevant in practical photography than is the photographic DR. I used Emil's method to determine the photographic DR at various noise floors for the D800 and the D4, using the DXO full SNR data. The DRs are normalized for the pixel count using the DXO method. As expected, the D800 has better DR at base ISO and the D800 is better at high ISO. The differences are less pronounced at higher noise floors due to the lesser contribution of read noise.

The DR advantage of the D4 at higher ISOs is minor, around 0.5 EV. IMHO, one should choose between these two cameras on factors other than DR.

Regards,

Bill

Logged
hjulenissen
Sr. Member
****
Offline Offline

Posts: 1712


« Reply #77 on: November 21, 2012, 09:59:37 AM »
ReplyReply

The 18% SNR data are a useful start, and for good modern sensors like the D800(E) these seem to be dominated by photon shot noise up to very high ISO speeds, but I would expect sensor noise to become relevant about three stops lower if one also looks at noise down to some moderate-to-deep shadow level.
My point earlier was that if the 18% SNR is dominated by shot-noise, I would guess that there is little to differentiate different cameras as long as one keeps the lense and sensor size variables constant?

So what an 18% SNR measurement may tell us is that having plenty of light, good large-aperture lenses and a large sensor is a good thing. Am I missing something?

-h
Logged
BJL
Sr. Member
****
Offline Offline

Posts: 5171


« Reply #78 on: November 21, 2012, 12:09:58 PM »
ReplyReply

My point earlier was that if the 18% SNR is dominated by shot-noise, I would guess that there is little to differentiate different cameras as long as one keeps the lense and sensor size variables constant?

So what an 18% SNR measurement may tell us is that having plenty of light, good large-aperture lenses and a large sensor is a good thing. Am I missing something?

There is for now at least the likelihood that 18% SNR is not dominated by shot noise at all ISO speeds for all current "interchangeable lens cameras" [ILCs], and that noise generated within the camera (everything except shot noise) starts to make a visible difference in moderately deep shadows even at just moderately high ISO speeds, where the 18% SNR is still dominated by photon shot noise. Even if the visibly relevant shadow regions are only relevant down to about "zone 3", two stops below mid-tones, this "camera noise" could become significant about two stops lower in the ISO speed scale.

But maybe sensor technology is approaching the point where 18% SNR is dictated by basic physics, not the limits of the technology. (Once Canon and the MF industry get their acts together!)

Apart from that, you are perhaps overlooking the cases (see my added emphasis above):
- when one does _not_ have plenty of light
- when the desire for a lightweight, mobile walking/hiking kit rules out large-aperture lenses as anoise-reduction strategy
- when the desire for substantial depth of field makes large apertures unusable.


P. S. Once a large enough lens delivers a certain effective aperture size (entrance pupil diameter) is achieved, sensor size does not matter to photon shot noise in light-limited, high ISO speed situations: a larger sensor with a lens of equal effectively aperture size used with equal exposure duration gives about the same photon signal and noise statistics, just at a higher f-stop and higher exposure index. Larger sensors only help with noise through:
- enabling the use of larger lenses with larger aperture diameters (a given aperture diameter being achievable at less low f-stops), and
- allowing the gathering more photons when using base ISO speed and "full exposure" of the sensor is possible.
« Last Edit: November 21, 2012, 12:18:06 PM by BJL » Logged
jeremypayne
Guest
« Reply #79 on: November 21, 2012, 01:24:08 PM »
ReplyReply


Bob: "Nowhere, nowhere did I say that people weren't able to properly use data."

In addition, too many people tout the hype of the test numbers without inquiry.  DxO Mark's headline test numbers for the D800/e are terrific.  But those numbers are massaged.  How many people actually realise that?  I'd venture it's a minority.

Hmmmm ... kinda seems like ya did ...

Gobble, gobble! Shocked Shocked
Logged
Pages: « 1 2 3 [4] 5 6 7 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad