ISO discussion
Posted 2 years ago
Hi everyone, I would like to start a discussion about ISO and low light photography.

I have seen many wonderful shots, like those by Sasson and Andre (http://1x.com/v2/#/?action=profile&user=20146 , http://1x.com/v2/#/?action=profile&user=14976)
They have some beautiful results even with high ISO.

I have to say every time I use ISO more than 400 I get lots of noise and I do have the D90 nikon which supposedly has a high iso performance. So I must be doing something wrong.

Are some lenses better with high ISO?

How can you still maintain an image with good technical quality even with high ISO?

Do you have to use noise-removal software? I heard such software often result in very soft image.

What is the maximum ISO you can use without losing too much quality? to me it seems anything above 400 is problematic.

Suggestions, feedback, experiment all are more than welcomed.
 
Posted 2 years ago
I have the D90 too, used to have a D70s. The D90 is only marginally better at ISO noise. I almost never go above 800. I have read with much jealously about these magical cameras that can shoot at 3200 ISO with very little noise...

I don't think lenses make any difference.

I have Noise Ninja, and although I have read many times that it is very good, I have not spent enough time with it to get any for myself.
 
Uzay 
Posted 2 years ago
Faddoush, links direct me to the main page, is there something wwrong with me or the links?
About ISO performance, i use D700 and prime lenses with 1.4, 1.8, 2, 2.8 max to get good results.
But as far as i observe if you don't knopw how to use the light properly none of the above are important.
I rarely use noise removal programs at least none of my published works in 1X has noise removal.
I also don't go above 800.
 
Posted 2 years ago
I'm lucky to have one of those "magical cameras", a D700. Up to 1600, no problemo. I even use 3200 quit often, with good results - both my photographs "Bees Nest" & "Barajas" where shot at 3200! I don't use any special software package to remove noise - only some tweaks in LR. I even sometimes use 6400 for indoor portraits of my daughter - noisy, but I do find it more grain like, once converted to B/W. Above that (for indoor use) it's not usuble anymore imo.

I tested the 5D mark II, and must say I found it nearly as good as the D700/D3. 3200 is certainly usuble, just a bit noisier. Above that I didn't really tested this camera.

Two considerations:
- exposure is critical at high iso (above 1600) - a bit of and my pictures tend to be very difficult to recover - more than at lower iso;
- I think, but never really tested this - high iso is probably more usefull in good light than poor light. Might sound a bit stupid to have high iso in good lighting, but think of a combination of small aperture & high shutter speed, for macro, sports & action photography...

About lenses... don't really know. But it seems logical to me that better glass gives better results.

 
Posted 2 years ago
I know Nic Duncan often pushes the ISO in bright circumstances, with beautiful results : http://1x.com/v2/#member/9920/nic-duncan/
 
Uzay 
Posted 2 years ago
KoPi wrote
I know Nic Duncan often pushes the ISO in bright circumstances, with beautiful results : http://1x.com/v2/#member/9920/nic-duncan/


What a pity that no EXIF data in his photos, i wonder if he shots RAW or JPEG,


 
Posted 2 years ago
hey faddoush :) happy you started it :)
from my little experience i believe that performance really depends on the quality and distribution of light. low light is great and very rewarding...
shadowed areas create more noise so low light with good light distribution works better for me....
i use up to iso 1250 with canon 50d and the 24-70 2.8 ... exposure time can be low at 1/20 or 1/30 but sometimes 1/80 or 1/100 is okay
sometimes you lose an image because either you moved or the person :) sometimes it works out... i find the the more challenging the light is the better result is obtained,,,,

Andre is the real master of low light

by the way for pp i process a bit in lightroom and continue on cs4 ... no noise removal just levels curves contrast sharpning...
if someone can reccomand a good way of reducing noise i will be happy :)

also i am interested to know the relation between exposure time iso and quality.....

 
Posted 2 years ago
What a pity that no EXIF data in his photos, i wonder if he shots RAW or JPEG

Almost sure she (better don't call her a he lol) shoots RAW Uzay. Just drop her a message and ask; she's always very helpfull (not overly active on 1x, but I know she drops in every now and then)
 
Posted 2 years ago
Very well said Michael! I agree completely.
 
Posted 2 years ago
Interesting topic indeed,

My personal experience is: OK up to 800 and mixed results above, depending on subject matter, light conditions etc...
I use a FujiS5, which is supposed to be good at that, and have superior dynamic over most sensor up to 800ISO.

The images below were shot jpeg, all underexposed -2/3EV without noise cleaning plug-in:


1600


1600


1250

All are in B+W, which helps a lot because noise ends up looking not unlike grain.
...But in color it is a much more a problem:


1600

I believe the problem with high ISO is in mid-low and intermediate f-stops (intermediate zone if you prefer), because they will get very noisy if you crank up the contrast whereas very bright and very dark areas will be more dense, with fewer noise. In other words it works better with rather contrasted scene/subject, at least this is my impression. Also sometimes it may be good to have spot metering instead of matrix with high ISO - again from personal experience

 
Posted 2 years ago
I'm presuming Faddoush you mean shots with ambient light, no flash etc.

The points made by Sasson and Jacques are good.

I would just emphasise that correct exposure is the main ingredient for success of any photograph, no matter the ISO used. So, especially in low light conditions, good practice is to use a lightmeter. The camera meter uses reflected light, which probably doesn't indicate the best settings for low light shooting. The histogram can be misleading. And yes, high contrast scenes will have less midtone where the worst noise occurs. Very fast lenses make it easier to get correct exposure in low light, and long lenses accentuate camera shake and movement blur in the slower shutter speeds typically needed for low light shots. In regard to camera shake image stabilisation/vibration reduction helps. Or a tripod. Longer exposure times can induce more noise. And again yes, it's better to use the whole ISO values than the intermediate ones.

I have a Canon 40D, and I did some shots in RAW recently in a mall in dim light without lighting assistance. Noise at ISO 1600 was very apparent in skin at 100%, but negligible elsewhere. I processed the images with CaptureOne5Pro with default NR and sharpening in preconversion, and in the first step after conversion I ran Nik Dfine2 noise analysis and accepted the recommended NR. At the end of PP I used onOne Phototools2.5 'Edge Sharpening Only' for final sharpening. The result was very negligible noise throughout the image at 100% and very satisfactory sharpness.

brose

http://www.brosepix.com
 
Posted 2 years ago
In addition, in PP, bracketed exposures can be useful, or some method which generates a relatively overexposed version, which can be blended with the normal, since there is less noise in overexposed images.

brose

http://www.brosepix.com

 
Posted 2 years ago
in principle, assuming that we compare sensors of similar technical characteristics, noise depends on photo-site photo-gathering area and ISO, the larger the photo-site area and the lower the ISO the less noise (that is say why cameras with full-frame sensors have less noise compared with cameras with smaller sensors at the same ISO, note that sometimes it is better to have less pixels, as that for a fixed sensor size means greater photo-site photo-sensitive area), by increasing ISO one in fact amplifies the signal and this means that one also amplifies the noise (which by the way is always present), the lower the light the greater the noise to signal ratio and by setting higher ISO (amplifying the signal) the nose becomes prominent, in practice on high ISOs noise is seen only in portions of images with low light, say if you shoot with high ISO in good light you should get less noise than if you shoot in low light, but the best results should be achieved shooting in good light with ISO as low as possible (say Canon EOS 5D, even it has full-frame sensor, has ISO setting of 50 for the best quality), APS-size sensor has its limitation noise-wise, some camera manufacturers try to deal with noise 'artificially' by using image processing even before outputting RAW image, I personally do not like that approach, as it is in my opinion fake and in most cases unsuccessful

nevertheless, my impression is that noise is regarded as 'artistic' in art world and likely on 1X too, so shooting with technically inferior camera should give you more chance of producing an image which will be praised in art world and possibly on 1X ;)
 
Posted 2 years ago
... one more thing Faddoush. I don't know the D90 but make sure that there are no camera settings/parameters that will bring you more noise.

For example if you have a sharpness parameter on board I think it should not be set too high, because additional, software-made sharpness will cause additional noise and rather ugly one. Same for contrast I think. I also believe that such parameters only apply to .jpg as opposed to RAW but I am not sure...
 
Posted 2 years ago
ok so a bit more thoughts thanks to the remarks

all i say are just assumptions from my use..... i find distance a factor as well... being too close can be a problem and also using wide lens helps as well...
it all relates to amount of light reaching the sensor
color images are harder to deal with since there is color noise so bw looks better low light high iso
keeping color is a real challenge especially when there are mixed sources of light

i suggest you try using a tripod when photographing people .... it can help especially if you are working with a model.....
the benefits of low light its the atmosphere
structures with contrasts look good

will think about some more stuff :)

 
Posted 2 years ago
WOW so many things to try here, thank you all for your feedback!!!

 
Posted 2 years ago
Just an update, I tried shooting with high iso even with good light source following some of the remarks given here, I got some really interesting results.

Since we are talking about ISO, when do you guys prefer to shoot with very low ISO?
 
Posted 2 years ago
From what I've understood, it is best to shoot at base ISO. For most recent Nikon DSLR's it is 200.

Going up gives, as you already know, noise.

Going below, will introduce clipping. This means that whites will tend to be blown out quicker.

I've never used ISO below the base one. I suppose one would use it in very bright situations or to achieve long exposures.
 
Posted 2 years ago
Faddoush wrote
when do you guys prefer to shoot with very low ISO?

Allways: Shoot at lowest ISO possible (I normally shoot at 100, but might go up to 400). Use a Tripod to make it happen!
Lars :-)


 
Posted 2 years ago
I have the D90 and the D700.

My experience is:

The D90 up to iso 800 with right exposure = No probles with noise. Sometimes iso 1600 will work fine.

The D700 up to iso 3200 with right exposure = No problems with noise. Sometimes iso 4-5000 will work fine.

No experience, but the new Nikon D3s should be iso 12.800 with no noise problems - should work absolute fine.

I do not think lenses have something to do with that, even better lenses always is better, it is the camera.
 
JBA 
Posted 2 years ago
So at high ISO settings, the exposure is critical. Does that mean you go exactly by the light meter or is it best to under/over expose slightly? I have a D90 and I have tried high ISO without much success. Even if noise is not too bad, the colours get more saturated and contrasted.
A fast lens is going to help, but only because you won't need such high ISO settings. . .
Jon
 
Posted 2 years ago
I very rarely shoot anything other than base ISO of 200 on my D300 simply because I want the best quality, I always use a tripod and I prefer slowish shutter speeds for my type of images. I do though occasionally use ISO 100 or upto say 800 if I need to for creative effects or require a faster shutter speed. I honestly do not notice any difference in my printed images at any of these ISO settings, testament to how good modern digital cameras are.

JP
 
Posted 2 years ago
JBA wrote
Does that mean you go exactly by the light meter or is it best to under/over expose slightly?

I never go by what the D300 light meter wants me to expose for Jon, I quickly learnt I get awful images that are overexposed and washed out. I exclusively spot meter in manual mode then take a reading -0.5 to -0.7 EV from the centre line on the bar graph in the viewfinder. I do this for various areas of the scene, foreground, water reflections, sky etc, 3 or 4 places with varying amounts of light. I do this to gauge the differences in exposure then make corrections with ND grads to balance the overall exposure to within what the sensor can comfortably handle. I then rely on the histogram and disregard the light meter as it usually ends up off scale once filters are fitted and my spot zone is used for focusing after that.

I use this technique for all ISO settings I may use ie from 100 to 800/1000.

JP
 
Posted 2 years ago
I personally keep getting more "loose" in terms of noise in a picture, i mean film had a lot more grain/noise at ASA 400 than we have today at ISO 800-1600 on majority of more recent cameras and i almost always give extra noise to my photos in Photoshop anyway :)

I kinda like the mood noise/grain gives to images, of course the noise a D60/D90 (which i have both) gives in no where near to the "quality" we get on a D700/D3 or the amazing D3s. On those DSLRs the noise is looks a lot more like the Film grain.

Sure sometimes its better to shoot with the lowest noise possible since it can be alter added as Post but i wont refrain taking a picture even with 3200 ISO if i want it, sometimes it can work others it wont.

Noise/Grain is harder on the eyes on color photos for sure, but for B&W not having it kind feels that something is missing in my opinion :)


 
Posted 2 years ago
If one wishes to avoid noise (grain) than one should shoot with ISO as low as possible. If noise is an issue then one may consider using a camera with full-frame or larger sensor (and smaller number of pixels).

Sometimes I deliberately use my compact camera (with 'tiny' sensor) in low light in order to get 'natural grain', which I sometimes find appealing (usually in darker B&W photos).

It seem to me that there isn't such a thing as 'photographic photograph' :) (allusion to expression 'photographic painting'), and that 'unreal' (what is reality anyway ;) ) in photography is regarded as 'valuable' (possibly as 'real' looks common, simple, imperfect, etc.), though different people may like different reality 'distortions', some may demand hi-end photo equipment while some may find low-end photo equipment better suiting their taste, my taste could be best summarized as "Nullum simile est idem" and requires higher-end photo equipment and photo processes (in principle it hinges on fine details like pianoforte 'details' in music or hi-fi approach in audio recording and reproduction).
 
Posted 2 years ago
Faddoush wrote
What is the maximum ISO you can use without losing too much quality? to me it seems anything above 400 is problematic.

It is my impression that ISO 400 on my D200 corresponds very well to ISO 6400 on my new D700 and it is the maximum for me. It is not a scientific study but my personal experience. I do not know where D90 is in comparison.
 
Posted 2 years ago
Here follows an extract from one of my posts posted on a photographic forum (can't remember which one and when it was written) which may be of some interest for 1X users:

I bought E-1 as my estimate was that it was the cheapest professional dSLR camera on the market at the time (looked to me as the best value for money). Nevertheless, I think E-1 is an excellent camera, certainly very well built, if not one of the best builds of all times, it is very easy to handle, ok it is not so fast, has not so many AF points, but it looks as a solid camera. I think that professional Canon and Nikon dSLR are better (although for non-digital lenses they have problems with lens optical resolution and telecentricy), but they are more expensive. Olympus E-1 with Zuiko Digital lenses was the first digital system designed from scratch, do not know, say, if Hasselblad has it designed all for digital (they should not lack lens optical resolution as they use larger format sensors, but what about telecentricy), I guess they should have done that.

As an owner of Olympus E-1 dSLR I cannot refrain from mentioning that the top two scorers (according to info a couple of months old) in Amateur Photographer ever (if that means anything), with 95% scored, are: Canon EOS MK II and Olympus E-1.

The main concern with Four Thirds is a smaller sensor, I think, and that is what it was mostly criticized for. What now follows is analysis of how much photosite area affects the quality of the image, i.e. nose.

For the following cameras:

* Hasselblad H2D-39: LF, 49.0x36.7mm, 39.0MP
* Canon 1Ds MkII: FF, 36.0x24.0mm, 17.2MP
* Nikon D2H: APSN, 23.3x15.5mm, 4.26MP
* Olympus E-1: 4/3", 18.0x13.5mm, 5.5MP
* Canon 10D: APSC, 22.7x15.1mm, 6.3MP
* Olympus E-20: 2/3", 8.8x6.6mm, 4.95MP
* Canon G5: 1/1.8", 7.18x5.32mm, 5.3MP

I generated the following list which compares sensor quality taking into account only photosite areas:

* Canon 1Ds MkII vs Hasselblad H2D-39: x = 0.1 stops
* Hasselblad H2D-39 vs Olympus E-1: x = 0.1 stops
* Olympus E-1 vs Canon G5: x = 2.6 stops
* Canon 10D vs Olympus E-1: x = 0.3 stops
* Canon 1Ds MkII vs Olympus E-1: x = 0.2 stops
* Canon 10D vs Canon 1Ds MkII: x = 0.1 stops
* Olympus E-1 vs Olympus E-20: x = 1.9 stops
* Olympus E-20 vs Canon G5: x = 0.7 stops
* Nikon D2H vs Canon 10D: x = 0.6 stops
* Nikon D2H vs Canon 1Ds MkII: x = 0.8 stops
* Nikon D2H vs Olympus E-1: x = 0.9 stops

How to interpret the above list? Say, Olympus E-1 vs Canon G5: x = 2.6 stops, says that one would get the same nose level if one shoots with Canon G5 set on ISO 50 and Olympus E-1 set on ISO between 200 and 400 (2.6 stops difference), etc. The calculation takes into account only photosite areas, it is obvious that Olympus E-1 produces better results on ISO 400 than Canon G5 on ISO 50, but it can be used to compare cameras according to sensor size, i.e. photosite area.

It is interesting to mention that according to photosite area criteria Canon 1Ds MkII is "better" than Hasselblad H2D-39 for 0.1 stops and that Hasselblad H2D-39 is better than Olympus E-1 for "only" 0.1 stops.

For those who would like to conduct the calculations on their own, here is the Mathematica code:

ClearAll[a, b, ap, bp, na, nb, MP];
Solve[{a/b == na/nb, na*nb == MP}, {na, nb}]

ClearAll[x, Ap1, Ap2, a1, b1, ap1, bp1, na1, nb1, MP1, a2, b2, ap2, bp2, na2, nb2, MP2];
a1 = 49.0; b1 = 36.7; MP1 = 39.0*10^9; (*H2D39*)
na1 = Sqrt[a1/b1*MP1];
nb1 = Sqrt[b1/a1*MP1];
ap1 = a1/na1;
bp1 = b1/nb1;
Ap1 = ap1*bp1;
a2 = 18.0; b2 = 13.5; MP2 = 5.5*10^9; (*E1*)
na2 = Sqrt[a2/b2*MP2];
nb2 = Sqrt[b2/a2*MP2];
ap2 = a2/na2;
bp2 = b2/nb2;
Ap2 = ap2*bp2;
Log[Ap1/Ap2]/Log[2]

a1 = 18.0; b1 = 13.5; MP1 = 5.5*10^9; (*E1*)
a2 = 18.0; b2 = 13.5; MP2 = 5.5*10^9; (*E1*)

a1 = 7.18; b1 = 5.32; MP1 = 5.3*10^9; (*G5*)
a2 = 7.18; b2 = 5.32; MP2 = 5.3*10^9; (*G5*)

a1 = 22.7; b1 = 15.1; MP1 = 6.3*10^9; (*10D*)
a2 = 22.7; b2 = 15.1; MP2 = 6.3*10^9; (*10D*)

a1 = 36.0; b1 = 24.0; MP1 = 17.2*10^9; (*1Ds*)
a2 = 36.0; b2 = 24.0; MP2 = 17.2*10^9; (*1Ds*)

a1 = 8.8; b1 = 6.6; MP1 = 4.95*10^9; (*E20*)
a2 = 8.8; b2 = 6.6; MP2 = 4.95*10^9; (*E20*)

a1 = 23.3; b1 = 15.5; MP1 = 4.26*10^9; (*D2H*)
a2 = 23.3; b2 = 15.5; MP2 = 4.26*10^9; (*D2H*)

a1 = 18.0; b1 = 13.5; MP1 = 8.15*10^9; (*E300*)
a2 = 18.0; b2 = 13.5; MP2 = 8.15*10^9; (*E300*)

a1 = 49.0; b1 = 36.7; MP1 = 39.0*10^9; (*H2D39*)
a2 = 49.0; b2 = 36.7; MP2 = 39.0*10^9; (*H2D39*)

 
Posted 2 years ago
Robert Jurjevic wrote
Here follows an extract from one of my posts posted on a photographic forum (can't remember which one and when it was written) which may be of some interest for 1X users:
I bought E-1 as my estimate was that it was the cheapest professional dSLR camera on the market at the time (looked to me as the best value for money). Nevertheless, I think E-1 is an excellent camera, certainly very well built, if not one of the best builds of all times, it is very easy to handle, ok it is not so fast, has not so many AF points, but it looks as a solid camera. I think that professional Canon and Nikon dSLR are better (although for non-digital lenses they have problems with lens optical resolution and telecentricy), but they are more expensive. Olympus E-1 with Zuiko Digital lenses was the first digital system designed from scratch, do not know, say, if Hasselblad has it designed all for digital (they should not lack lens optical resolution as they use larger format sensors, but what about telecentricy), I guess they should have done that.

As an owner of Olympus E-1 dSLR I cannot refrain from mentioning that the top two scorers (according to info a couple of months old) in Amateur Photographer ever (if that means anything), with 95% scored, are: Canon EOS MK II and Olympus E-1.

The main concern with Four Thirds is a smaller sensor, I think, and that is what it was mostly criticized for. What now follows is analysis of how much photosite area affects the quality of the image, i.e. nose.

For the following cameras:

* Hasselblad H2D-39: LF, 49.0x36.7mm, 39.0MP
* Canon 1Ds MkII: FF, 36.0x24.0mm, 17.2MP
* Nikon D2H: APSN, 23.3x15.5mm, 4.26MP
* Olympus E-1: 4/3", 18.0x13.5mm, 5.5MP
* Canon 10D: APSC, 22.7x15.1mm, 6.3MP
* Olympus E-20: 2/3", 8.8x6.6mm, 4.95MP
* Canon G5: 1/1.8", 7.18x5.32mm, 5.3MP

I generated the following list which compares sensor quality taking into account only photosite areas:

* Canon 1Ds MkII vs Hasselblad H2D-39: x = 0.1 stops
* Hasselblad H2D-39 vs Olympus E-1: x = 0.1 stops
* Olympus E-1 vs Canon G5: x = 2.6 stops
* Canon 10D vs Olympus E-1: x = 0.3 stops
* Canon 1Ds MkII vs Olympus E-1: x = 0.2 stops
* Canon 10D vs Canon 1Ds MkII: x = 0.1 stops
* Olympus E-1 vs Olympus E-20: x = 1.9 stops
* Olympus E-20 vs Canon G5: x = 0.7 stops
* Nikon D2H vs Canon 10D: x = 0.6 stops
* Nikon D2H vs Canon 1Ds MkII: x = 0.8 stops
* Nikon D2H vs Olympus E-1: x = 0.9 stops

How to interpret the above list? Say, Olympus E-1 vs Canon G5: x = 2.6 stops, says that one would get the same nose level if one shoots with Canon G5 set on ISO 50 and Olympus E-1 set on ISO between 200 and 400 (2.6 stops difference), etc. The calculation takes into account only photosite areas, it is obvious that Olympus E-1 produces better results on ISO 400 than Canon G5 on ISO 50, but it can be used to compare cameras according to sensor size, i.e. photosite area.

It is interesting to mention that according to photosite area criteria Canon 1Ds MkII is "better" than Hasselblad H2D-39 for 0.1 stops and that Hasselblad H2D-39 is better than Olympus E-1 for "only" 0.1 stops.

For those who would like to conduct the calculations on their own, here is the Mathematica code:

ClearAll[a, b, ap, bp, na, nb, MP];
Solve[{a/b == na/nb, na*nb == MP}, {na, nb}]

ClearAll[x, Ap1, Ap2, a1, b1, ap1, bp1, na1, nb1, MP1, a2, b2, ap2, bp2, na2, nb2, MP2];
a1 = 49.0; b1 = 36.7; MP1 = 39.0*10^9; (*H2D39*)
na1 = Sqrt[a1/b1*MP1];
nb1 = Sqrt[b1/a1*MP1];
ap1 = a1/na1;
bp1 = b1/nb1;
Ap1 = ap1*bp1;
a2 = 18.0; b2 = 13.5; MP2 = 5.5*10^9; (*E1*)
na2 = Sqrt[a2/b2*MP2];
nb2 = Sqrt[b2/a2*MP2];
ap2 = a2/na2;
bp2 = b2/nb2;
Ap2 = ap2*bp2;
Log[Ap1/Ap2]/Log[2]

a1 = 18.0; b1 = 13.5; MP1 = 5.5*10^9; (*E1*)
a2 = 18.0; b2 = 13.5; MP2 = 5.5*10^9; (*E1*)

a1 = 7.18; b1 = 5.32; MP1 = 5.3*10^9; (*G5*)
a2 = 7.18; b2 = 5.32; MP2 = 5.3*10^9; (*G5*)

a1 = 22.7; b1 = 15.1; MP1 = 6.3*10^9; (*10D*)
a2 = 22.7; b2 = 15.1; MP2 = 6.3*10^9; (*10D*)

a1 = 36.0; b1 = 24.0; MP1 = 17.2*10^9; (*1Ds*)
a2 = 36.0; b2 = 24.0; MP2 = 17.2*10^9; (*1Ds*)

a1 = 8.8; b1 = 6.6; MP1 = 4.95*10^9; (*E20*)
a2 = 8.8; b2 = 6.6; MP2 = 4.95*10^9; (*E20*)

a1 = 23.3; b1 = 15.5; MP1 = 4.26*10^9; (*D2H*)
a2 = 23.3; b2 = 15.5; MP2 = 4.26*10^9; (*D2H*)

a1 = 18.0; b1 = 13.5; MP1 = 8.15*10^9; (*E300*)
a2 = 18.0; b2 = 13.5; MP2 = 8.15*10^9; (*E300*)

a1 = 49.0; b1 = 36.7; MP1 = 39.0*10^9; (*H2D39*)
a2 = 49.0; b2 = 36.7; MP2 = 39.0*10^9; (*H2D39*)

Now I have a headache...
 
Niels Christian Wulff  Book editor
Posted 2 years ago
Robert Jurjevic wrote
For those who would like to conduct the calculations on their own, here is the Mathematica code:

ClearAll[a, b, ap, bp, na, nb, MP];
Solve[{a/b == na/nb, na*nb == MP}, {na, nb}]

ClearAll[x, Ap1, Ap2, a1, b1, ap1, bp1, na1, nb1, MP1, a2, b2, ap2, bp2, na2, nb2, MP2];
a1 = 49.0; b1 = 36.7; MP1 = 39.0*10^9; (*H2D39*)
na1 = Sqrt[a1/b1*MP1];
nb1 = Sqrt[b1/a1*MP1];
ap1 = a1/na1;
bp1 = b1/nb1;
Ap1 = ap1*bp1;
a2 = 18.0; b2 = 13.5; MP2 = 5.5*10^9; (*E1*)
na2 = Sqrt[a2/b2*MP2];
nb2 = Sqrt[b2/a2*MP2];
ap2 = a2/na2;
bp2 = b2/nb2;
Ap2 = ap2*bp2;
Log[Ap1/Ap2]/Log[2]

a1 = 18.0; b1 = 13.5; MP1 = 5.5*10^9; (*E1*)
a2 = 18.0; b2 = 13.5; MP2 = 5.5*10^9; (*E1*)

a1 = 7.18; b1 = 5.32; MP1 = 5.3*10^9; (*G5*)
a2 = 7.18; b2 = 5.32; MP2 = 5.3*10^9; (*G5*)

a1 = 22.7; b1 = 15.1; MP1 = 6.3*10^9; (*10D*)
a2 = 22.7; b2 = 15.1; MP2 = 6.3*10^9; (*10D*)

a1 = 36.0; b1 = 24.0; MP1 = 17.2*10^9; (*1Ds*)
a2 = 36.0; b2 = 24.0; MP2 = 17.2*10^9; (*1Ds*)

a1 = 8.8; b1 = 6.6; MP1 = 4.95*10^9; (*E20*)
a2 = 8.8; b2 = 6.6; MP2 = 4.95*10^9; (*E20*)

a1 = 23.3; b1 = 15.5; MP1 = 4.26*10^9; (*D2H*)
a2 = 23.3; b2 = 15.5; MP2 = 4.26*10^9; (*D2H*)

a1 = 18.0; b1 = 13.5; MP1 = 8.15*10^9; (*E300*)
a2 = 18.0; b2 = 13.5; MP2 = 8.15*10^9; (*E300*)

a1 = 49.0; b1 = 36.7; MP1 = 39.0*10^9; (*H2D39*)
a2 = 49.0; b2 = 36.7; MP2 = 39.0*10^9; (*H2D39*)



Cool down Mr. Spock

/Wulff
 
Posted 2 years ago
@ Robert Jurjevic

Some nice comments here, Robert. I enjoyed your analogy of some photographic styles-appreciation with fine detail in music. While your mathematical description of the effect of the pixel density of sensors is not immediately accessible to me, I think it is a worthwhile note to a discussion of noise.

brose

http://www.brosepix.com
 
JBA 
Posted 2 years ago
Robert Jurjevic wrote
It seem to me that there isn't such a thing as 'photographic photograph' :) (allusion to expression 'photographic painting'), and that 'unreal' (what is reality anyway ;) ) in photography is regarded as 'valuable' (possibly as 'real' looks common, simple, imperfect, etc.), though different people may like different reality 'distortions', some may demand hi-end photo equipment while some may find low-end photo equipment better suiting their taste, my taste could be best summarized as "Nullum simile est idem" and requires higher-end photo equipment and photo processes (in principle it hinges on fine details like pianoforte 'details' in music or hi-fi approach in audio recording and reproduction).

Nice point, though I think a lot of photography is at least in part 'about photography' or what you may call 'photographic seeing', whether the photographer is aware of that or not. About what a camera does to what it photographs, to 'reality' if you like. About, in the immortal words of Garry Winogrand, 'what something looks like when photographed'. So I guess there may be such a thing a 'Photographic photograph'. . . ;-)
Cameras lie ;-)
Jon
 
Posted 2 years ago
JBA wrote
So I guess there may be such a thing a 'Photographic photograph'

Yes, definitely. And to further comment what Robert says about high end vs low end gear I don't think there is such a breakdown between "truthful" photograph on one side and altered/fake reality (fantasy if you prefer) on the other side. It is more different degrees of alteration because as Jon points out camera lies ... On the end of the day there are different visions, different point of view (from photographers). Furthermore the tension between what looks/seems truthful in the first place and how reality is altered anyhow (subtly or not) is not the least interesting aspect of photography (euphemism).

Back to how high/low ISO you shot I personally almost always shot at 400, goes up if required, and rarely goes down (200) even if possible, which is probably stupid (or better said out of laziness)

 
Posted 2 years ago
nice discussion Jon & jacques

@ jacques, if the light is good, ISO 400 is good

brose

http://www.brosepix.com
 
Posted 2 years ago
brose wrote
I enjoyed your analogy of some photographic styles-appreciation with fine detail in music.

Thanks, there were times when audio hi-fi was very important to me, ideas such as that only instruments giving a true 'sound' are acoustic, my "Nullum simile est idem." approach in art, etc., may all have something in common.

brose wrote
While your mathematical description of the effect of the pixel density of sensors is not immediately accessible to me, I think it is a worthwhile note to a discussion of noise.

Basically, it says that the larger the area of a photo-site the better the recorded light in each pixel... of course the quality of the light falling on each photo-site depends on the quality of lens... if compared to smaller sensors full-frame or larger format sensors have larger photo-site areas for the same number of pixels, but they require larger images which makes lens design more difficult and the lenses more expensive...

 
Posted 2 years ago
Hello John,

JBA wrote
About what a camera does to what it photographs, to 'reality' if you like. About, in the immortal words of Garry Winogrand, 'what something looks like when photographed'. So I guess there may be such a thing a 'Photographic photograph'. . . ;-)
Cameras lie ;-)

One could say that perfect hi-fi approach to photography would be to record the scene as our eye would see it, though this definition may be a bit restricted, as say almost any camera would have shallower DoF and significantly narrower dynamic range than our eye, and one could argue that there may exist a 'true' perception of reality independent of human receptory apparatus, but, as Bertrand Russell has pointed out, it would seem that all we are in contact with is 'sense data', not 'reality' itself... furthermore, there is a problem of who we are, i.e. what makes us a unique entities, say who is seeing an image, or who am I in me... another complexity arises with the fact that the picture does not exist in reality, all there is in front to you in photographic sense is a 'chaos' of electromagnetic radiation in certain wavelength range, the image is 'artificially' created by a lens which in principle has the same effect as a pin-hole, i.e. it 'dramatically' reduces the number of light beams which fall on a falt surface which we call image... again I, whoever may that be, am capable of perceiving the light spectrum in each 'pixel', which we call and 'feel' as colour, it would seem that all there is in 'reality' is light with its spectrum property, no colours at all... also it looks like we are used to motion pictures, we rarely see someone's face 'frozen', so 'frozen' elements from 'reality' may look to us unusual and rare... and finally it looks like it is impossible to prove the existence of 'reality'...

one could argue that 'photographic photograph' is a photo taken with a decent 'hi-fi -like' camera without any 'forbidden' post-processing which includes even unnecessary pre-RAW in-camera processing or any 'avoidable' lens errors, etc. ...nevertheless, what I would expect is that then one would seek for unusual 'natural' elements in the scene which would somehow 'mimic' what would one wished to achieve with image post-processing...

my personal artistic judgement looks to be some sort of percepto-emotional-rational process in me, whoever or whatever I am... i.e., I know very good which images I like but have no clue why

Regards,
 
Robert  Forum moderator
Posted 2 years ago
The ISO setting is more a philosophy for me and what i wane reach. The result for some camera is color noise which can be used as compositional element. Regularly i use at the day light ISO 400 which raises the dynamic and it works so much well. Even with my D200 at ISO 400 at the day time gives fine results. The resulting noise is also depend on the light or better how much light is available. There where is black of course the picture engine will create something which can be called noise but for such case a free hand speed light will help. ISO is not much a topic since the new Nikon top models are very good at low light.

Of course the quality of the lens also plays an important rule but not only for low light photography.

I make the experience that on my Nikon there is a kind of noise geometry. From ISO 200 till 1250 all works fine and the quality is very good. From ISO 1600 till 3200 some more color noise is there. The noise ware which is on board the camera makes the result sometimes not better and some detail sharpness get lost. So that is why i use the noise ware only on ISO 3200 and higher.

Robert
 
 
Compose a reply
You must sign in if you want to post a reply.
Fine Art Prints  -  Our books  -  Work with us  -  FAQ  -  About 1X
© 1X Innovations AB 2007-2011. All rights reserved.
 
 Stumble 1X