Buying a DSLR - High ISO

[kmedina419]

He should mention that the lens really helps you shoot in low light if your with a kit lens and say your settings are 1/60 f5.6 1600 iso youll get noise bad (camera depending) if you get better class like a 2.8 your settings would switch to 1/250 f2.8 iso 1600.....to even get less noise then switch it 1/120 f2.8 iso 800..most all intro cameras do great at iso 800

[JanErikEdvartsen]

"Shells Orgasmatron Massage" LOL!

[roberthemsley2]

Quick Question

i need a camera for shooting in low light, so basically shooting at high iso's and getting clear images with low noise.

am i looking for a high megapixel camera in order to reduce noise? i was told that a 18 megapixel image with noise cropped down to a normal jpeg looks like it has less noise than a 12 megapixel cropped down?

so out of these cameras which is the best for low light?

nikon d90, canon 50d, canon 550d, ??

thanks in advance

[kmedina419]

@roberthemsley2 you need to get away from megapixels, you need a bigger sensor, the senor technology is what really makes your difference for high iso shooting

[71shoelover]

@kmedina419 So true. However I have a Nikon D90 and A Nikon D40 and I have to tell you that the litle Nikon D40 starts to fall to bits at ISO 1600 quite badly and at HI1 (ISO 3200) It is rubbish. The Nikon D90 is still great at ISO 1600

[famousPhotog]

In practical use, high resolution cameras such as the D3x are specialist cameras for those who are regularly shooting images at ISO 100 that will be printed at very large sizes, eg. 20x30 to 40x60. Using the D3x for anything else is just a waste of file handling time and hard drive space. Unfortunately, many people believe that a camera that produces files that are larger than needed for printing at modest sizes will somehow produce a superior result at modest sizes, which is simply not true.

[orestisdeepblue]

I don't think that anyone printing at "modest sizes" will buy D3x. And the fact that technology advances helps everybody use better cameras for less money. But you are free to shoot with a 4mp P&S if you feel is enough for your needs.

[famousPhotog]

The argument that there is superiority for the D3X at the final output print stage only applies for print sizes where the additional resolution is required. At all the smaller print sizes, 16x20 and smaller, the extra resolution of the D3x provides no resolution/detail advantage over the D3, and just causes the photographer to handle and store extra data. However, the superiority of the D3's larger sensels is apparent at all print sizes up to 16x20.

[famousPhotog]

Even though the D3x is a newer version of the D3, the D3 with its larger sensels provides superior dynamic range starting at ISO 200, superior tonal range, superior signal-to-noise, superior color sensitivity. The standard ISO range for the D3X is 100 - 1600, for the D3 is 200 - 6400, therefore, it is not surprising that the D3X outperforms the D3 at ISO 100. However, from ISO 200 upwards, the D3 is superior. So for the vast majority of the ISO range, the D3 is still superior.

[orestisdeepblue]

where do you get that facts ???

you are totally wrong

[famousPhotog]

For example, check DXOMark for camera comparison. Look at the individual graphs for each parameter. What is specifically wrong with what I said?

[orestisdeepblue]

The fact that you claim that D3 has better dynamic range overall because it has 1/3 advantage on high-iso's and you just pass the fact that at it's base iso D3x is 1.4ev better than D3.

High dynamic range is good to have at any iso speed. But where it counts most is in landscape and studio where you almost always use base-iso setting due to the controlled lightning conditions.

If small pixels where so bad for DR and High-iso capabilities then noone would pass the 5 or 10mp limit.

[famousPhotog]

In fact, the reason I am so unhappy about the Canon 7D, is because Canon elected to pack too damn many pixels on an APS-C sensor, thus compromising dynamic range forever. If the 7D had been a 10mp camera, it would have offered phenomenal performance AND image quality at its price point. Instead, we have a 7D camera with lots of exciting new features, but only a marginal improvement in IQ compared to the 40D with respect to dynamic range.

[71shoelover]

@famousPhotog They packed too many pixels on the 7D and the 550d for the "Mega Pixel Masterbators" who don't understand that more mega pixels DOES NOT mean better quality pictures. I have a little Nikon D40 and it kicks the pants off a Canon 550d for picture quality. WHY...Because it has a large sensor and only 6MP on the sensor..Therefore more colour is able to get in each pixel, The result is lovely punchy colourful pictures from the Nikon D40. and pants pictures From the Canon 550D.

[famousPhotog]

Furthermore, once dynamic range is compromised by over saturation of the sensels, there is no correcting for this later. No software or firmware update will overcome the loss of detail at the extreme limits of the dynamic range. The ONLY solution is bigger and/or deeper sensel wells, a hardware issue. At any given level of sensor technology, larger sensels will ALWAYS produce more dynamic range than smaller sensels, due to the photon saturation issue.

[famousPhotog]

Noise levels and dynamic range are two different parameters. Yes, you can minimize the noise coming from densely packed sensors by having less magnification upon output. However, the dynamic range limitation, especially at high sensitivity (high ISO), is related to the individual sensels propensity for becoming saturated. Larger sensels are more resistent than smaller sensels. Therefore, you will see fewer blown highlights and crushed shadows with larger sensels, than with smaller sensels.

[orestisdeepblue]

In iso6400 Nikon D3 has less than 1/3 of a stop more dynamic range than D3x. But in base-iso the D3x has more than 1.5stops higher dyn-range than D3. What would you choose ?

At any given level of sensor technology, larger sensels give more dynamic-range per sensel. Not per print

[canadianmaple09]

0:51 I would like to know more about the "Orgasmatron."

[orestisdeepblue]

Once again, wrong about the high-iso thing..

[transferases]

He is absolutely right... with newer sensor technology and if all else is same, less MP and therefore bigger pixels will equal better high ISO performance and Nikon bodies excel at this.

[orestisdeepblue]

That's absolutely wrong and a myth created -mostly- by dpreview. Larger pixels give better high-iso performance per pixel, not per print. That's why it's wrong to compare a 100% crop from a high-mp camera to a 100% from a low-mp camera. You are comparing un-equal magnifications that way

[transferases]

I understand what you are trying to say, because the larger MP photo will be much bigger at 100%. But even if you make the images the same size by reducing the magnification of the larger photo, the noise still is better on the photo with larger pixels.

[transferases]

This is ALWAYS the case when comparing canon and nikon. Take their 5d mark II with a full frame sensor and 21 MP and take nikon's D700 with full frame sensor and 12 MP.

[transferases]

If you take a photo say at iso 6400 framed exactly the same, the 21 MP image is a lot bigger and at 100% when compared the 21 mp will have a lot more noise than the 12 mp photo. If you reduce the size of the 21 MP photo to that of the 12 mp photo from nikon, the noise on the 21 mp photo will be less visible however I have found that the 12 mp photo from the D700 will still have better noise.

[orestisdeepblue]

Sorry man, you are wrong on that.

You can check DxOMark to confirm that 5DII, 1Ds3 , D3, D3x and D700 all have similar noise performance per print. The advantage of high-mp cameras is that they offer similar noise per print while capturing more detail.

You can't compare 1D4 with the others cause it's a different format. Same with APS-C and 4/3 cameras.

[transferases]

I just checked Dxomark, you can as well. If you go to the compare camera section you will see the nikon d3 and d700 have a much higher rated dynamic range, color depth, and low light iso then any of canon's FF.

Also, i know the 1d mk iv can't be compared to the D3S because it's a different format. That's exactly my point, but canon IS making it compete with the D3S and that is why it won't even compare. Canon's competitor to D3S is the 1d mk iv, that's why i was comparing.

[orestisdeepblue]

You probably did a wrong comparison again. Compare D3, D700 and D3x in high-iso and print, not on screen.

It's not a Nikon vs Canon thing i'm talking about. It's about the myth that high-mp cameras give more noise.

All 5 FF cameras from Canon and Nikon give the same high-iso quality. And it's sad that the "experts" on the matter don't even know why..!!

[km6xz]

You are claiming the high ISO performance of a 1dIV is similar to a D3s? Do you also sell used bridges and swamp land?

This high pixel count and density is causing a lot of working pros to switch for the observable obvious improvement in DR and noise.

A 7d for example, due to pixel density also has the honor of being unusable for serious landscape or studio work due to diffraction setting in at only f/9. Canon is going in the wrong direction in IQ in an effort to sell to px counters

[orestisdeepblue]

No , i never compared 1D4 to D3s.

I claim that ISO-performance has nothing to do with pixel density. What matters in high-iso performance is 3 things, the company that built the sensor, the electronics and how advanced they are and the physical size of the sensor. Pixel density has nothing to do with iso-performance when you compare 2 sensors from the same company, same chips and same sensor sizes.

If you are a Nikon fan, compare D3 to D3x for that matter at equal print sizes

[orestisdeepblue]

Oh.. and the diffraction is a whole different story..

Lets talk about iso here :)

[transferases]

Another comparison, canon's new 1d mk iv vs their D3S. Canon has a 1.3 x crop sensor with 16 mp, while nikon has a FF sensor with 12 mp. Obviously the image quality and high iso performance will be far better on the nikon, canon is loosing out on IQ, dynamic range, and high iso performance by making the pixels smaller AND stupidly using a smaller sensor as well.

[WhatDigitalCamera]

@orestisdeepblue

So we're all part of a photo industry conspiracy, led by dpReview, to con the world into thinking there's no correlation between pixel pitch and noise. And you're the lone voice of sanity fighting to expose the truth.

Whatever.

[orestisdeepblue]

I don't know if it's a conspiracy or just lack of interest from your part. If you consider the fact that you reply to a reader of yours in that way and you don't even bother to check if he is actually right about it..well, it's probably the second thing :)

Oh... one last thing..

I'm not the only voice.. DxOMark also says the same thing if you care to check :)

[WhatDigitalCamera]

Even Nikon themselves market the D3 for low noise and the D3s for high resolution. Are they wrong too? We're just going to have to agree to disagree on this.

[orestisdeepblue]

No.. they are not wrong.. they are good in marketing :)

I guess we'll agree to disagree, no hard feelings. But, just in case you are talking to a sane person and not just a crazy fan-boy of High-MP cameras, give it a try and check if its possible that DP-review is wrong and DxOMark and me (and lots of other people too) are right when we say that high pixel density is good for high-iso performance when you compare equal print sizes. Take care :)