[M1ke A]

Had a very quick and utterly unscientific play with the demo unit in Park cameras yesterday.

I only had my Tokina 10-17 fisheye with me but hey.

Shot as jpgs due to lack of raw converter. No PP other than cropping to 1500px.

ISO 100
http://www.pbase.com/m1ke_a/image/10...0/original.jpg

ISO 800
http://www.pbase.com/m1ke_a/image/10...1/original.jpg

ISO3200
http://www.pbase.com/m1ke_a/image/10...2/original.jpg

IS6400
http://www.pbase.com/m1ke_a/image/10...3/original.jpg

ISO 6400 100% crop


Obviously the Tok isn't the first word in image quality but what do we think?

[mr cat]

Canon Digital Learning Center - Sample EOS 5D Mark II Video: Reverie

[VinnyP]

A lot depends on how much light there was? High ISo perversely does worse in the situations that you need it most ie low light. The 100% crop looks pretty good especially given the high pixel count. If it was low light then that is very good. If there was plenty of light I'd think it was a little noisier than say a Nikon D300.

[ccsharry]

for 21MP, this is absolutely stunning performance!!

[Dave]

Originally Posted by VinnyP

A lot depends on how much light there was? High ISo perversely does worse in the situations that you need it most ie low light. The 100% crop looks pretty good especially given the high pixel count. If it was low light then that is very good. If there was plenty of light I'd think it was a little noisier than say a Nikon D300.

Would agree with this, as well lit D3 examples I have seen are more impressive. I expect the high ISO capabilities of 5D MKII & D3/D700 to be so close as not to make a difference in the real world

[Steve]

Its also very hard to judge objectively a camera's capability using a fisheye lens as the optics

[VinnyP]

Originally Posted by Dave

Would agree with this, as well lit D3 examples I have seen are more impressive. I expect the high ISO capabilities of 5D MKII & D3/D700 to be so close as not to make a difference in the real world

I suspect that may be stretching things a bit. The early tests show the D3 having around 1.5 stops advantage over the 5D MkII at high ISO but at nearly double the pixel density that is to be expected.

[M1ke A]

Originally Posted by VinnyP
A lot depends on how much light there was? High ISo perversely does worse in the situations that you need it most ie low light. The 100% crop looks pretty good especially given the high pixel count. If it was low light then that is very good. If there was plenty of light I'd think it was a little noisier than say a Nikon D300.

Would agree with this, as well lit D3 examples I have seen are more impressive. I expect the high ISO capabilities of 5D MKII & D3/D700 to be so close as not to make a difference in the real world

Sorry guys but am I being really dumb? If the light was 'good' why would you crank the ISO?

I can understand higher ISO for freezing action etc but if you don't have to, why would you?

[Charlotte]

Originally Posted by mr cat

Canon Digital Learning Center - Sample EOS 5D Mark II Video: Reverie

Love the credits saying "Cast (In order of appearance)" ... even though there are only two of them

[VinnyP]

Originally Posted by M1ke A

Sorry guys but am I being really dumb? If the light was 'good' why would you crank the ISO?

I can understand higher ISO for freezing action etc but if you don't have to, why would you?

In most cases you wouldn't which is the issue for me, a lot of high ISO tests are shot in good light so you get ISO 6400 at 1/500 and F8. These tend to look good at web resolution on a lot of cameras. But the same camera at 6400 and 1/15 at F1.4 will suddenly have a whole lot of noise and that's where you are more likely to be shooting at 6400 plus.

[Steve]

Which is another reason why most web comparisons are useless. Unless the user has tested the equipment they are comparing at the same time using the exact same settings and in a totally controlled situation then the results are simply not worth much.