Page 1 of 1

D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Sun Jun 20, 2010 4:16 pm
by gstark
As you know, We currently have one of each of these cameras to review.

Thanks to Nikon Au, who supplied us with a D3s and 70-200 lens, and thanks to Canon Au, for supplying us with a Mk IV, also with their latest 70-200 lens.

First impressions are important, and here both cameras shine. They both feel great in the hand, and both feel very tight and solid. Oddly, although they're both near enough to the same weight, the Canon feels somewhat lighter. Subjectively, the Canon seems to be the more responsive of the two, acquiring focus more readily.

The Nikon feels a little slower in this regard, but it's no slouch either, and its build quality is just tank-like. it seriously reminds me of the old school film Nikons, such is the heft of the D3s.

While things like ergonomics are often a personal matter, and especially when it comes to how a camera feels and handles in your hands, there are a couple of things that I think that both manufacturers need to pay attention to. For instance, it seems that almost every time Canon release a new model camera, they relocate the on/off switch. It might be on the back of the camera, or on the top deck, maybe to the left of the prism, or maybe not. The bottom line is that whenever I pick up a Canon body, I have to think about which camera am I using, and look for and find out where the this most basic of controls is.

By way of contrast, on picking up the Nikon, the power switch is in the same place on every one of their SLRs - surrounding the shutter release button. It's readily accessible without needing to remove your eye from the viewfinder, and switching between bodies requires no real retraining the hands in terms of locating many of the controls.

In a similar vein, one of the controls I use very frequently is the focus point selection dial, and on the Canon, this is a two handed exercise. Sorry guys, but no; you've got that wrong. This should be as simple as using only the command dial; the need for pushing buttons for this functionality is just plain wrong.

Nikon have yet to get their act together when it comes to image previewing though. I think that the manner by which the image zooms on each Nikon body is different: hold one button and turn a dial as on the D3s - which I think is wrong, or plus and minus buttons somewhere in the ether (wrong again), or maybe just as simple as the context sensitive buttons next to the rear lcd on the D300/D700 bodies, which, to me, seems to be the simple, logical, and most appropriate means of implementing this functionality.

In use, both cameras are simply pleasures to drive. They're both high performance vehicles, with fast frame rates, great AF systems, able to sustain long bursts of image making, and able to shoot in something that's close to darkness.

And while both cameras will let you go where no camera has gone before, the Nikon truly excels with its high ISO performance.First up, here's the test scene, Canon first. These shots are full frame, raw images, resized and converted to jpg only.

These were shot using one floodlight, through a beauty dish. ISO was 12800 +3EV (102000) and according to my handheld meter, correct exposure was f/13 and 1/200. So that's where we are. I recomposed the Nikon shots to take account of the crop factor on the Canon; although Canon's marketing people might want you to believe that these cameras are direct competitors, the fact is that the Nikon is a full frame camera, while the Canon is not. There are advantages to both formats, and we all are well aware of them. Suffice to acknowledge that this is a fundamental difference in the cameras, and move along.

Image
Image

I think that for non-critical application, either camera can produce acceptable results, but observe the lower section in the Canon's image. Note the purple tinge in the bottom rh corner. We observed this sort of thing across all of the Canon's high ISO images, and we don't know if this is common across the camera model, or just with the unit we have.

Given the exposure as set by the handheld meter, the Canon's images appear to be slightly over-exposed and verging upon burned out, in the higher brightness areas of this image. I would probably choose to pull back the exposure a little in terms of EV compensation on the Canon when using the camera in a more realistic environment. Unfortunately, this extra brightness in the image means that the blacks are not as rich as they should be,which, overall, makes the image look, to my eyes, a little "thin" and underexposed, but over-compensated with brightness adjustments.

The Nikon's image appears to be pretty close to spot on exposure wise, and this is confirmed in the histogram.

If your images are for a more critical application though, this becomes a Nikon show. The purple tinging seems to be an indication that the Canon's sensor isn't quite there yet. Here's a pair of 100% crops; one from each camera, and again, Canon first.

Image
Image

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Sun Jun 20, 2010 6:23 pm
by Matt. K
Just a matter of time before flashgun become obsolete? The low light performance of both cameras is astonishing....but the NIKON? It's night vision technology! Thanks for the heads up Gary.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Sun Jun 20, 2010 6:45 pm
by glamy
Hi Gary,
I do agree with your comment about the image preview on the D3s. Why did they change from the D2x? There was nothing wrong with it and it was much more practical to get out of the zoom (unless I am doing something wrong...).Makes you wonder about the process used to get to these changes :? There are two bits of plastic in the box, one is the cover for the flash shoe, do you know what the other one is for?

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Sun Jun 20, 2010 7:40 pm
by aim54x
WOW...FX supremacy!!! Looks like the Mk IV got blown out of the water in the HIGH ISO test.

Both are nice cameras though, and the ergonomics are very user specifics....

Thanks for the pics Gary!

I cant think of what the other piece of plastic for?

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Sun Jun 20, 2010 7:44 pm
by zafra52
Thanks for the very informative comparison. Yet, most of us
are already locked into a brand because of the substantial
investment on lenses and related paraphernalia. I believe the
choice is simpler for the starting photographer.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Sun Jun 20, 2010 8:13 pm
by glamy
zafra52 wrote:Thanks for the very informative comparison. Yet, most of us
are already locked into a brand because of the substantial
investment on lenses and related paraphernalia. I believe the
choice is simpler for the starting photographer.

True in a way but I was seriously looking at the 5D as an alternative. I could get a 5D II and a 24 tilt/shift and still have $1000.00 to spare to get to the D3s price. One may argue about the build quality and so on, but for what I do it would have been ok. In the end I stuck with Nikon because of the quality of the camera and other personal preferences.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Sun Jun 20, 2010 11:23 pm
by DaveB
An interesting comparison, thanks. But when making grand comparisons like this you need to be careful with some of your assumptions.

gstark wrote:In a similar vein, one of the controls I use very frequently is the focus point selection dial, and on the Canon, this is a two handed exercise. Sorry guys, but no; you've got that wrong. This should be as simple as using only the command dial; the need for pushing buttons for this functionality is just plain wrong.

It's configurable. I use the rear "joystick" for this on Canon bodies, and it only requires my thumb.

These shots are full frame, raw images, resized and converted to jpg only.

Ok, an important point then is which RAW converter did you use? ACR? If so, which version? Things improved in the high-ISO stakes with CS5/LR3, and not linearly (i.e. the gap between these cameras may be closer than with CS4/LR2). That's even without tweaking any noise reduction.

according to my handheld meter, correct exposure was f/13 and 1/200. So that's where we are.

[...]

Given the exposure as set by the handheld meter, the Canon's images appear to be slightly over-exposed and verging upon burned out, in the higher brightness areas of this image. I would probably choose to pull back the exposure a little in terms of EV compensation on the Canon when using the camera in a more realistic environment.

Probably not. In that case you would be using the Canon's internal meter, not the external handheld. For example there can be a slight difference in the actual ISO of each camera model, and the internal meter would presumably take that and other factors into account.

Use a grey card filling the frame in the same light, and with both cameras compare the histograms of images at the internally-metered settings (using appropriate metering patterns of course). This might eliminate confusion over what's causing the difference in brightness you've noticed.

Cheers

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Mon Jun 21, 2010 12:34 am
by Willy wombat
Interesting pics here Gary - love the low light technology becoming available. I just hope us poorer photographers get access to it in the next couple of years. :up:

I had a play with a D3 the other day myself.

The advantage of being able to shoot in auto ISO mode up to 1600 or 3200 iso without a care in the world is such a treat with these high end cameras .

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Mon Jun 21, 2010 12:50 am
by gstark
DaveB wrote:An interesting comparison, thanks. But when making grand comparisons like this you need to be careful with some of your assumptions.

gstark wrote:In a similar vein, one of the controls I use very frequently is the focus point selection dial, and on the Canon, this is a two handed exercise. Sorry guys, but no; you've got that wrong. This should be as simple as using only the command dial; the need for pushing buttons for this functionality is just plain wrong.

It's configurable. I use the rear "joystick" for this on Canon bodies, and it only requires my thumb.


Dave,

Thanks; please do tell. The Mk IV is with Leigh, and I don't think he has a manual.


These shots are full frame, raw images, resized and converted to jpg only.

Ok, an important point then is which RAW converter did you use? ACR? If so, which version? Things improved in the high-ISO stakes with CS5/LR3, and not linearly (i.e. the gap between these cameras may be closer than with CS4/LR2).



The latest, as downloaded from Adobe last evening. Prior to that download, I had no support - nor any need for support - for either of these bodies on my desktop system.

I deliberately avoided any NR; I just wanted to see the basic raw images produced by each camera.

Regardless of the processing engine and NR applied, I remain somewhat concerned over the purple tinges that appear in the bottom rh corner of these images. This was evident in all of the high ISO images from the Canon, but only along the bottom of the images. It is also evident in other high ISO images, not taken during last night's session.


according to my handheld meter, correct exposure was f/13 and 1/200. So that's where we are.

[...]

Given the exposure as set by the handheld meter, the Canon's images appear to be slightly over-exposed and verging upon burned out, in the higher brightness areas of this image. I would probably choose to pull back the exposure a little in terms of EV compensation on the Canon when using the camera in a more realistic environment.

Probably not. In that case you would be using the Canon's internal meter, not the external handheld. For example there can be a slight difference in the actual ISO of each camera model, and the internal meter would presumably take that and other factors into account.

Use a grey card filling the frame in the same light, and with both cameras compare the histograms of images at the internally-metered settings (using appropriate metering patterns of course). This might eliminate confusion over what's causing the difference in brightness you've noticed.


In this instance, I very much doubt it. As I said, I metered using the handheld meter; I neglected to mention that it was used in incident mode. The camera settings were manually set to those indicated by the light meter's readings: shutter speed of 1/200, aperture of f/13, and ISO to the 102000 setting, and the shots were then made. The in camera metering was neither relevant nor utilised.

Cheers.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Mon Jun 21, 2010 1:39 am
by surenj
Thanks Gary for this. It's good to see the results by someone who doesn't have any financial bias with the relevant companies. :wink:

gstark wrote:The in camera metering was neither relevant nor utilised.

I think one may underestimate the noise by the Canon if you have overexposed the shot inadvertently.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Mon Jun 21, 2010 2:44 am
by DaveB
gstark wrote:
DaveB wrote:
gstark wrote:In a similar vein, one of the controls I use very frequently is the focus point selection dial, and on the Canon, this is a two handed exercise. Sorry guys, but no; you've got that wrong. This should be as simple as using only the command dial; the need for pushing buttons for this functionality is just plain wrong.

It's configurable. I use the rear "joystick" for this on Canon bodies, and it only requires my thumb.

Thanks; please do tell. The Mk IV is with Leigh, and I don't think he has a manual.

Go to Canon's site and download the manual in PDF form. Towards the bottom of the "Guides and Manuals" section.
Have a look at C.Fn III-9. Even with this in the default setting (0), you'd press the AF-select button with your right thumb and then either the "multi-controller" (joystick) with the same thumb or the dials with your thumb/forefinger. With C.Fn III-9 set to 1, you can just tweak the joystick without having to press the AF-select button first.

an important point then is which RAW converter did you use? ACR? If so, which version? Things improved in the high-ISO stakes with CS5/LR3, and not linearly (i.e. the gap between these cameras may be closer than with CS4/LR2).

The latest, as downloaded from Adobe last evening.

The latest CS5 or LR3? Fine. The latest updates for CS4 or LR2 might not do either camera justice in this regard.

gstark wrote:Given the exposure as set by the handheld meter, the Canon's images appear to be slightly over-exposed and verging upon burned out, in the higher brightness areas of this image. I would probably choose to pull back the exposure a little in terms of EV compensation on the Canon when using the camera in a more realistic environment

DaveB wrote:Probably not. In that case you would be using the Canon's internal meter, not the external handheld. For example there can be a slight difference in the actual ISO of each camera model, and the internal meter would presumably take that and other factors into account.

gstark wrote:In this instance, I very much doubt it. As I said, I metered using the handheld meter; I neglected to mention that it was used in incident mode. The camera settings were manually set to those indicated by the light meter's readings: shutter speed of 1/200, aperture of f/13, and ISO to the 102000 setting, and the shots were then made. The in camera metering was neither relevant nor utilised.

The in-camera meter is relevant and should be used to test this. Note that the ISO of the camera is not "102000". It should be easy to do an experiment and compare the metered exposures for each camera at this "identical" (ha!) ISO setting to resolve this. Maybe your assumption was valid, but I suspect there's a slight error here.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Mon Jun 21, 2010 10:04 am
by ozimax
Gary, I would like to help you further test the Mk iv. Please courier to my address and I will return it (in about 2 years). Sincere thanks to Canon... :lol:

Seriously, even with the 5D Mkii, I rarely use flash these days. (I have kept my old 430EXi and sold my 580EX). With a nice prime lens, nice clean images are possible in very low light. I can only imagine how well one of these two pro bodies would work with, say, an F1.2 lens or similar?

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Mon Jun 21, 2010 10:11 am
by Mr Darcy
glamy wrote:There are two bits of plastic in the box, one is the cover for the flash shoe, do you know what the other one is for?

Other Nikons have a cover for the eyepiece so that stray light coming through that doesn't affect the metering when you are taking a photo remotely. My guess would be that htis is what it is.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Mon Jun 21, 2010 10:36 am
by photohiker
Gary,

Have you got a strong magnifier lens? My 5D had a strange purple cast in the same corner of the image at high ISOs (but somewhat lower than your test) :) When I handed it to my mate who is a camera mechanic for a clean he put it under his scope and found a filamentous strand of glue running over the corner of the sensor. I had cleaned the sensor myself several times and had not seen it, nor dislodged it. Haven't seen the purple since. I can only guess that it was catching some light and spraying it onto the sensor.

This might be unrelated, but worth looking for in any case.

I'm not sure what you are testing by the light metered exposures? Could you possibly do back-to-back camera metered vs external light metered shots? I'm with Dave on this one, as I think others have generally tested DSLR ISO's as not particularly accurate, but the in camera metering should know about that.

The Nikon certainly packs a hell of an image for that ISO!

Michael

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Mon Jun 21, 2010 9:04 pm
by aim54x
Mr Darcy wrote:
glamy wrote:There are two bits of plastic in the box, one is the cover for the flash shoe, do you know what the other one is for?

Other Nikons have a cover for the eyepiece so that stray light coming through that doesn't affect the metering when you are taking a photo remotely. My guess would be that htis is what it is.


I would have thought this, but the D700 and the D3 series have a built in curtain for that job....

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Mon Jun 21, 2010 10:08 pm
by wendellt
it still bothers me that nikon centred all the af points in the viewfinder on the d3 systems to mimic canons design
the d2x had focus points spread all around

i was pro nikon before the mark Iv came out but since canon fixed the focus point selector via joystick i think its a very close call since canon will undoubtably come up with better iso in their next model

as an interest what factors in canon's AF technology make the mark IV focus faster than the nikon system

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Tue Jun 22, 2010 12:09 am
by surenj
wendellt wrote:canon's AF technology

These three words shouldn't be used in the same sentence! :rotfl2: :rotfl2:

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Tue Jun 22, 2010 2:02 pm
by biggerry
I have no idea why I looked at this thread, I 'ain't ever gonna buy one these..but..

check this review out http://www.photographybay.com/2010/02/04/canon-1d-mark-iv-vs-nikon-d3s-iso-comparison/

it could be my imagination but the canon in this one has similiar purple fringing on the lower right...or it could just be that that corner is brighter (refer one of the last images)

there are a few reviews about and all seem to have the same plague of sh$t fights from nikon and canon fanboys :rotfl2:

http://www.neutralday.com/canon-eos-1d-mark-iv-vs-nikon-d3s-iso-comparison/

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Tue Jun 22, 2010 2:13 pm
by surenj
biggerry wrote:I have no idea why I looked at this thread, I 'ain't ever gonna buy one these..but..

I look at them because this will be the consumer level technology in a few years. Soon the 60D, d400 will have similar ISOs... exciting times especially for small flash strobism. We mite be able to get by with small/low powered flashes for the most part... :wink:

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Tue Jun 22, 2010 2:39 pm
by gstark
DaveB wrote:Go to Canon's site and download the manual in PDF form. Towards the bottom of the "Guides and Manuals" section.
Have a look at C.Fn III-9. Even with this in the default setting (0), you'd press the AF-select button with your right thumb and then either the "multi-controller" (joystick) with the same thumb or the dials with your thumb/forefinger. With C.Fn III-9 set to 1, you can just tweak the joystick without having to press the AF-select button first.


Thanks, Dave; will do.

an important point then is which RAW converter did you use? ACR? If so, which version? Things improved in the high-ISO stakes with CS5/LR3, and not linearly (i.e. the gap between these cameras may be closer than with CS4/LR2).

The latest, as downloaded from Adobe last evening.

The latest CS5 or LR3? Fine. The latest updates for CS4 or LR2 might not do either camera justice in this regard.


Would have been CS5.


The in-camera meter is relevant and should be used to test this.


If I'm testing the cameras' metering systems, then yes.

But I'm an old fart, and I wanted to do this in a way that made more sense to me: I had a constant light source, so I wanted to take the cameras' metering right out of the equation. With a constant light source, I have a constant EV, which directly translates, for any given ISO value, to a known set of shutter speed and aperture settings. Thus I can simply put the camera into the given range of settings (M) and go ahead and make my day. Er, images. :)

No different to how I might shoot in a real world setting today, using either digital or film.

Note that the ISO of the camera is not "102000". It should be easy to do an experiment and compare the metered exposures for each camera at this "identical" (ha!) ISO setting to resolve this. Maybe your assumption was valid, but I suspect there's a slight error here.


I was rounding the ISO ... At the pointy end of the ISO scale, and for discussion, rather than hard specifications, 102000 is near enough for me.

There may be errors, but I believe my logic is sound.

photohiker wrote: you got a strong magnifier lens? My 5D had a strange purple cast in the same corner of the image at high ISOs (but somewhat lower than your test) :) When I handed it to my mate who is a camera mechanic for a clean he put it under his scope and found a filamentous strand of glue running over the corner of the sensor. I had cleaned the sensor myself several times and had not seen it, nor dislodged it. Haven't seen the purple since. I can only guess that it was catching some light and spraying it onto the sensor.


Thanks Michael, interesting point. We shall look at this and see if we can come up with anything ./... bearing in mind though that the camera isn't ours.

wendellt wrote:it still bothers me that nikon centred all the af points in the viewfinder on the d3 systems to mimic canons design the d2x had focus points spread all around


Wendell, that's not why it's done this way. The reason is that the D3<whatever> (and D700) shares the AF system with the D300 bodies, which are crop cameras. As such, the AF system actually covers a smaller physical area, that physical area being related more closely to the whole of the crop cameras' frame size, and thus a much smaller (relatively speaking) area within the coverage frame of an FX bodied camera.

I agree that it's frustrating, and probably an accounting, rather than an engineering, decision on the part of Nikon. I'd certainly prefer a larger AF area as well.

biggerry wrote:it could be my imagination but the canon in this one has similiar purple fringing on the lower right...or it could just be that that corner is brighter (refer one of the last images)


Thanks for that, Gerry, and no, it's not your imagination. That seems to me to be pretty much the same issue that we are seeing here. Overall, I think that their images are showing much the same sort of outcomes that we're seeing here.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Tue Jun 22, 2010 7:45 pm
by glamy
[quote="wendellt"]it still bothers me that nikon centred all the af points in the viewfinder on the d3 systems to mimic canons design the d2x had focus points spread all around

I find that upsetting as well and I do not care about the ability to use DX lenses which I suppose would bring these focus points where we see them in the D2x.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Tue Jun 22, 2010 8:00 pm
by gstark
surenj wrote:Thanks Gary for this. It's good to see the results by someone who doesn't have any financial bias with the relevant companies. :wink:

gstark wrote:The in camera metering was neither relevant nor utilised.

I think one may underestimate the noise by the Canon if you have overexposed the shot inadvertently.


I don't think the image is overexposed, actually. To my eyes it's actually looking like a print made from an underexposed negative, so by that logic it's underexposed, but has then been (over) compensated in the processing with too much brightness.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Tue Jun 22, 2010 8:41 pm
by DaveB
gstark wrote:
wendellt wrote:it still bothers me that nikon centred all the af points in the viewfinder on the d3 systems to mimic canons design the d2x had focus points spread all around


Wendell, that's not why it's done this way. The reason is that the D3<whatever> (and D700) shares the AF system with the D300 bodies, which are crop cameras. As such, the AF system actually covers a smaller physical area, that physical area being related more closely to the whole of the crop cameras' frame size, and thus a much smaller (relatively speaking) area within the coverage frame of an FX bodied camera.

I agree that it's frustrating, and probably an accounting, rather than an engineering, decision on the part of Nikon. I'd certainly prefer a larger AF area as well.

Unfortunately I think there is a certain amount of engineering involved here. The further you get from the centre of the image circle the harder it is to have accurate phase-detect AF sensors. On the Canon FF cameras the sensors don't go a long way out towards the edges either.

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Tue Jun 22, 2010 9:37 pm
by surenj
gstark wrote:I don't think the image is overexposed, actually. To my eyes it's actually looking like a print made from an underexposed negative, so by that logic it's underexposed, but has then been (over) compensated in the processing with too much brightness.

It would be pretty easy to compare the histograms to find out whether the two pics are matching. If one is underexposed, then that would have MUCH more noise than a 'correctly' exposed or 'over'exposed picture.

gstark wrote:the Canon's images appear to be slightly over-exposed and verging upon burned out, in the higher brightness areas of this image

Hence I was pointing out that if it's TRULY overexposed (as you have stated here), then the noise would be UNDERestimated. IE the Canon would be even noisier when correctly exposed.

I am only using my camera as an example. There is loads of noise if I underexpose even at ISO 100. Maybe that rule doesn't apply to these? :?

Re: D3s Vs 1D Mk IV (Large file sizes in this thread)

PostPosted: Tue Jun 29, 2010 7:46 pm
by Glen
Thanks for the review Gary, amazing to see how far the high ISO technology has come :D