Apples & Oranges - nearly

[Sheetshooter]

Last night I whiled away some time shooting a little still-life motif on my office desk just by the available light. I shot the same thing with the D70s (200 ISO) and the EOS 5D (100 ISO). Witht he 5D I set the colour temperature to the metered rating and with the D70s I shot in 'Incandecent' with a +3 setting. In both shots the colour rendering was quite good - although a tad different.

Now, I had found during my investigations, that some folk advise using a blue conversion filter when shooting in tungsten because otherwise it is necessary for the processing algorithms to amplify the BLUE channel considerably and, of course, it is generally the BLUE channel that noise finds most appealing as an abode. Needless to say I didn't frig around with filters last night but I might return to the test a little later with filters.

I shot at f/11 with the 45mm lens on the Nikon and the 50mm lens on the Canon. As you'd expect there was a slight difference in depth of field. But the most glaring difference was in the noise. The Nikon was a shocker. Lumpy like a tapioca pudding in the low values. The Canon was as smooth as the proverbial infantile tusch.

Cheers,

[big pix]

SS........ any chance of seeing some images.......

[Glen]

Sheetshooter, I too would love to see the comparisons if you think it would show the differences as jpg's. Were both processed using the same program or each firms proprietry software?

[Sheetshooter]

Both are 100% images from the BLUE channel

Canon





Nikon




I shall do a subseqquent test using conversion filters and see if that makes a difference but real work has to be done now - the sun is out and I am off to Bayview for a pool shoot.

Cheers,

[stubbsy]

OK To my uneducated brain it would seem a more valid comparison would be ISO 200 on both.

[Sheetshooter]

Peter,

I shall do that in time but my primary concern was to get an indication of who things would work with my given standard practices. I don't ever envisage shooting anything that is not at 100 ISO on the Canon or 200 ISO (minimum) on the Nikon.

Both images were processed through Photoshop and Bridge - it was a pretty level playing field.

In my wanderings with both cameras thus far I have to say that I have been fairly underwhelmed by the performance of the Nikon - even as a walk-around happy snap thing. Keep an eye on the For Sale forum .....

[Glen]

Sheetshooter, that is a pretty amazing comparison, than you very much for it. Must say I am surprised.

[gstark]

SS,

Thanx for posting this. I see you bought an Epson printer then?

While the noise factor here is quite startling, we're also seeing some other aspects of the differences in what I suspect are the in-camera processing algorithms that Canon and Nikon use.

The Nikon image appears (to me) to be significantly sharper than the Canon's, and this is exactly what I expect, because I believe that in order to gain the lower noise characteristics that the Canon has, some of the sharpness is sacrificed to the smoothing algorithms.

I'm not for a moment suggesting that one method is any better than ther other, but simply that they are different, and it then comes down to personal preferences as to what one prefers.

Certainly a very valid and useful comparison; thanx again SS for doing this.

[LOZ]

SS Seen that the D70 is such a crappy camera @ 200 ISO I would be only to pleased to help you and take it off your hands for $500 cash. pick up today. .

[Oneputt]

I think that the noise issue (which I have stuck myself) is the only downside to what otherwise is a brilliant camera for the money.

[pippin88]

SS: Interesting. Did you adjust exposure at all in PS?

[Nnnnsic]

So you're comparing noise from a one and a half year old sub-2000 dollar camera (which has a sensor damn similar to the older D100's) and a two-month old 5000-dollar camera (with a brand new sensor)?

Phew.

Glad this topic was called "Apples & Oranges" because I'd be confused as to what the point is.

[PiroStitch]

Just out of interest, did you shoot in jpg or in RAW? I've never seen a D70 that noisy at ISO200. No i'm not trying to be overly defensive, I'd just like to know if it was shot in Jpg.

[Glen]

Piro, it is the blue channel only

[PiroStitch]

Piro, it is the blue channel only

Whoops Thanks Glen Shows you how much i read of that first post

[Sheetshooter]

The images were captured in RAW. Of course these are dumbed down JPEGS saved for web and so the finer details of comparison are a little lost.

Gary, what you perceive as a difference in sharpness is in fact an example of the depth of field - although I shot with the Nikkor 45mm and the Canon 50 mm obviously the range altered to retain similar cropping. Keep in mind that this is part one of a series of tests I shall be conducting in order to determine the parameters within which I can operate in the field. Similar framing is a pivotal criterion which is why the range changed.

Next stage of the test is to shoot a similar set-up (the same set-up??) both with colour correction by a filter and colour correction by the camera sytem and see how that affects this noise issue.

Today saw calibrating - or at least getting a handle on - the dynamic range of my work system as it relates to my normal modes of exposure control. I also tested how far I can shift the 24mm in various configurations before mechanical and optical vignetting become an issue.

The metering data actually benfits using the TS-E lens because once shift is applied the TTL meter is unablke to offer a reliable reading.

And yes, you are right Gary. I opted for the Epson R2400 printer, although what you see in the test shot is actually a 4990 scanner for the 4x5s. (Learned today that the purchase of the scanner untitlkes me to some other Epson printer F.O.C.)

Cheers,

[gstark]

Gary, what you perceive as a difference in sharpness is in fact an example of the depth of field

Ok, that's fair enough too; thanx. What was the actual (equivalent, probably) focus point?

And yes, you are right Gary. I opted for the Epson R2400 printer, although what you see in the test shot is actually a 4990 scanner for the 4x5s.

Perhaps, but I was looking at the two roll paper holders that are also visible in the images. That's where I was making my sharpness observation, hence my question regarding the focus point.

[Sheetshooter]

Of course, how silly of me .... it's been a long day in the sun. Terrified out oif my wits too by a burglar alarm and siren that went off when I entered an unattended home to access their pool.

The point of focus was the core-clip of the roll holder on the left. I would add that the Canon was focussed automatically and of course the 45mm P lens was focussed manually with the 'assist light'. Let me make clear that I don't think either lens is a slouch and one must remember that my sole purpose in choosing a D70s was to be able to use that splendid little classic Tessar design lens. It certainly is no slouch!!

I eventually located the carton with the old Nikon F2 in it over the weekend and so I am keen to see what happens with my 105 1:2.5 with this digital camera also - although fully manual exposure (which is possibly even preferable if the truth be known.) I also bought a 28mm AF for doing 'favour' shots for local bludgers. What are people's views on that lens, by the way? The jury is out for me on it.

Cheers,

[MCWB]

I also bought a 28mm AF for doing 'favour' shots for local bludgers.

The f/2.8 or the f/1.4?

[Sheetshooter]

The f/2.8. As I said it is only to keep the bludgers at bay. I used to use an Olympus MJU II for this but times today are a' changing.

[gstark]

The point of focus was the core-clip of the roll holder on the left.

Ok .... that section of the Nikon image still looks sharper to me.

I'm happy that DoF accounts for the difference on the other roll holder, but I still think that we're also seeing the results of some image softening due to the Canon noise reduction algorithms.

I agree, btw, that the noise evident in the Nikon image is way too high, but I think that a large part of what we're seeing is simply down to the fundamental differences in how the engineers want the image to turn out.

one must remember that my sole purpose in choosing a D70s was to be able to use that splendid little classic Tessar design lens. It certainly is no slouch!!

Leigh is certainly head over heels with his.

[MCWB]

Given what Gary's postulating, maybe it might be worth whacking the blue channel of the Nikon image through an anti-noise program (Neat Image or similar) and see if you can get a similar noise level, then check the sharpness?