125 replies to this topic

[Rocha]

Don't get me wrong, its looks like a really nice camera with a lot of advances.
But for a camera with that price tag ($3299) It’s looks odd that there are some important features(at list to me) that aren't there but they are in my 'entry level' camera that cost 1/4 of the D5.

<{POST_SNAPBACK}>

I am glad you understand this, lots of people want the newer camera only because it is the new kid on the block, and usually those people don't pay attention to what they need. If the features that are important to you are in an entry level camera, you should by all means get that!

To me, the most important features are the autofocus, build quality or ergonomics (I do a lot of topside too), image quality (color) and resolution, in this order. My personal opinion is that the D2x is better than the 5D in all (don't know about image quality yet) of those features except for resolution (but that's only in paper, I don't know how much of a difference you would see between 12.4 and 12.8 megapixels). Since they have very similar resolution, I think the difference between the D2x and 5D is equivalent to the difference between the film Nikons F5 and F90 (both have the same res. film), so, even with the price difference I would still get the D2x if I was buying today. Now, if Nikon didn't have the 10.5 fisheye I would think twice about the Canon.

[herbko]

To me, the most important features are the autofocus, build quality or ergonomics (I do a lot of topside too), image quality (color) and resolution, in this order. My personal opinion is that the D2x is better than the 5D in all (don't know about image quality yet) of those features except for resolution (but that's only in paper, I don't know how much of a difference you would see between 12.4 and 12.8 megapixels). Since they have very similar resolution, I think the difference between the D2x and 5D is equivalent to the difference between the film Nikons F5 and F90 (both have the same res. film), so, even with the price difference I would still get the D2x if I was buying today. Now, if Nikon didn't have the 10.5 fisheye I would think twice about the Canon.

<{POST_SNAPBACK}>

The most important difference is signal-to-noise and dynamic range. The difference between 12.4 and 12.8 Mpixels is negligible, and the detail resolution will depend on the anti-alias filter which trades off resolution/sharpness for less artifacts and fringing. The sensors on the 5D are more than twice the area of the ones on the D2x. If Sony and Canon are equal at sensor design and processing, and I'm skeptical that Sony has caught up to Canon this quickly, the 5D sensor records more than twice as much light for the same exposure. ISO 200 is better than ISO 100 for the D2x. This in some sense makes all your lenses one stop better. If you consider that it typically cost about twice as much to buy any particular lens that is one stop better, it's a big advantage.

[Alex_Mustard]

This thread has been a weird one - pictures of birds and camera bashing! I must be reading DPReview

----

Some news on housings I got on the email from Lee Peterson:

"Nexus will show their Canon EOS 1Ds Mark II housing at DEMA in Las Vegas
Booth # 2513 Toshi Kozawa is also working on the Canon 12 MP EOS 5D."

Alex

[Kasey]

As far as I know 1:1 still means the image on the sensor is the same size as the subject. I haven't seen any lens makers using a fudge factor on it.

<{POST_SNAPBACK}>

Herb - 1:1 traditionally means that the image on the 35mm fram is the same size as the subject, so the APS crop is actually larger than 1:1. This leaves the question whether the EF-S 60mm lens is 1:1 on the EF-S sensor, or really greater than 1:1.

Unfortunately I don't know the answer.

My real question is - why would canon build this lens? Why make it EF-S instead of a FF lens? I must be missing something here - cost? size? It doesn't seem much cheaper than a FF variety would be, is it smaller?

[herbko]

Herb - 1:1 traditionally means that the image on the 35mm fram is the same size as the subject, so the APS crop is actually larger than 1:1.  This leaves the question whether the EF-S 60mm lens is 1:1 on the EF-S sensor, or really greater than 1:1. 

Unfortunately I don't know the answer.

My real question is - why would canon build this lens?  Why make it EF-S instead of a FF lens?  I must be missing something here - cost?  size?  It doesn't seem much cheaper than a FF variety would be, is it smaller?

<{POST_SNAPBACK}>

Hi Kasey, I don't know the answer to either questions. Have you seen lens makers make reference to 1:1 for anything other than 35mm format? My bet would be that the EF-S 60mm is really 1:1. That is the image is the same size as the subject at maximum magification, but I don't really know. Your second question is even more puzzling , considering that the 60mm cost the about the same as the EF 100mm full frame, which is an excellent lens.

My pet peave is the opposite that of Dru. I think it is confusing to put in fudge factors for specs that has clear physical meaning, like the focal length of a lens and 1:1 magnification.

[herbko]

This thread has been a weird one - pictures of birds and camera bashing! I must be reading DPReview

Alex

<{POST_SNAPBACK}>

I'm sorry if you consider my post bashing, and I can't do much about pictures of birds. It is after all a thread on the introduction of the Canon 5D, and comparisions between it and the D2x were first made by others, which I don't think is inappropriate either. More generally the question of FF vs cropped sensors will be raised now that there is a FF camera priced below a cropped sensor camera. I'm making the case for one advantage of FF. Of course, there are other considerations as many have pointed out.

[Phil Rudin]

Regarding 1:1, it has nothing to do with 35 mm, it means life size. If you use a 6 x 7 camera with a 1:1 lens at life size then the image is the size of the film (i.e 6 x7) not the size of 35 mm. An 8" x 10" neg at 1:1 would be an image 8" x 10". An image on the Olympus 4/3 format is the size of the sencer at 1:1. The Oly 50 mm macro is a 1:2 lens but produces an image which is life size for 35 MM because the chip has what you would call a 2x crop factor. The image also has twice the DOF because it is a 50 and not a 100 mm.

Phil Rudin

[Rocha]

I know, I know it is all my fault, I started the camera bashing I think comparisons between these two are going to be inevitable, either here, on dpreview or anywhere because of the reasons that Herb mentioned (FF cheaper than cropped), because of the close pixel count and because this will be a great underwater camera (just like the D70, dRebel, 20D, etc). Anyways, who hasn't compared the D70 and the 20D or dRebels here at wetpixel?

Now, Herb pointed out advantages of the FF that I didn't know. Thanks for that, I am sure it will be useful for anybody that shoots in high iso. I am more interested in low iso, the most appealing feature of the 5D for me so far is iso 50. Now, if the pixels in the D2x are smaller, there are more pixels per unit area, does this translate into higher resolution?

[Phil Rudin]

Herb I don't get how a sensor that is twice as large gives you twice as much light, I.E. 100 to 200 iso. If I use 100 film in a 35 mm camera with the lens set to say F8 at 250 I will get the same exposer as using a 6 x 7 camera with 100 film and F8 at 250th. The 6 x 7 film is four times larger than the 35 mm but it dosen't increse the exposure at all. am I missing somthing?

Phil

[herbko]

Herb I don't get how a sensor that is twice as large gives you twice as much light, I.E. 100 to 200 iso. If I use 100 film in a 35 mm camera with the lens set to say F8 at 250 I will get the same exposer as using a 6 x 7 camera with 100 film and F8 at 250th. The 6 x 7 film is four times larger than the 35 mm but it dosen't increse the exposure at all. am I missing somthing?

Phil

<{POST_SNAPBACK}>

The first point is simple. For a given exposure, a FF chip will catch 2.25 (1.5x1.5) times more light than a 1.5x cropped chip. In the cases that we're look at the pixel count is about the same so each sensor on the FF catches 2.25x more light than the ones on the 1.5x cropped chip. There's an efficency factor in that a little area is taken up by transistors and wires needed for the amplifier and read out and partly made up for by micro lens which basically takes the light that would otherwise fall on this area onto the sensor. The basic scaling should be pretty close.

The ISO does not change with this scaling. I should have been more clear on that on the last post. The lowest ISO depends on how much exposure it takes to fill the storage capacity of the sensor which is independent of area: the bigger sensor catches light at twice the rate but is twice as big so will fill up at the same rate as the small sensor. What is generally true given the same processing technology and circuit design is that the noise level does not depend on sensor size. So if you correctly expose a shot with the small sensor at ISO 100 and take the same shot with the big sensor at 1/2 the exposure and bump that up to ISO 200. You'll still have a cleaner image with the bigger sensor given the assumptions above. In that sense the bigger sensor is more than one stop better.

To briefly address the point Rocha broght up about shoot always at the lowest ISO and higher ISO performance does not matter. There are cases, like shooting a sunburst shot where some shadow details are underexposed due to the need to keep the sun from being overexposed, where it does matter. These details can be brought out with curves in photoshop, and the quality of these details depend on how well the sensor performs at low light.

[Phil Rudin]

Thanks Herb, I understand your point but only as it regards like sensors of diffrent sizes. The first FF sensor was the 6MP Contax and the images at 400 were worst than my 5 mp Oly E-1 at 800 which has a chip about one fourth the size of the Contax chip.

Phil

[herbko]

Thanks Herb, I understand your point but only as it regards like sensors of diffrent sizes. The first FF sensor was the 6MP Contax and the images at 400 were worst than my 5 mp Oly E-1 at 800 which has a chip about one fourth the size of the Contax chip.

Phil

<{POST_SNAPBACK}>

Yes. It only strictly applies to sensors of the same design and process just scaled up. I don't have any details on the relative merits of the Canon vs Sony processes and circuit design. So if they were to make a FF D3x with the same process and design, you can expect one stop better signal-to-noise.

Your experience with the Contax surprises me, but maybe that's why they are out of this business.

[Phil Rudin]

It is sad, I have a Contax N-1 (35 mm) and the glass is wonderful.

Go to Pentax and look at the smcp-FA 645 120 mm F4. This is a macro lens that shots life size, 1:1 in 645. Life size image on 645 film or sensor. E-ray does the same thing on film over 12 inches wide. Image is still life size.

My Oly E-1 shots life size, 1:1 (the 4/3 chip size or about one- fourth 35 mm size) with an extension tube. In 35 mm that would be called 2:1.

Phil

[Kasey]

Regarding 1:1, it has nothing to do with 35 mm, it means life size. If you use a 6 x 7 camera with a 1:1 lens at life size then the image is the size of the film (i.e 6 x7) not the size of 35 mm. An 8" x 10" neg at 1:1 would be an image 8" x 10". An image on the Olympus 4/3 format is the size of the sencer at 1:1. The Oly 50 mm macro is a 1:2 lens but produces an image which is life size for 35 MM because the chip has what you would call a 2x crop factor. The image also has twice the DOF because it is a 50 and not a 100 mm.

Phil Rudin

<{POST_SNAPBACK}>

I suspect that the 1:1 claim of the Canon 60mm lens is based on the 35mm format, even though it mounts only to EF-S camera, thus yielding greater than 1:1. Otherwise the degree of magnification would be less than the 1:1 of the 100 macro (which IS based on 35mm), and offer even further confusion to the lens lineup!

[Giles]

add my 2 cents .. i would rather have a 1d mkII ....... that killer speed and pro body ... the Ds may have resolution and full frame but its price is out of my league ... and besides the 1d is still a massive choice for pros in the world

[whitey]

5D - it's all about resolution and FULL FRAME. Full frame, in particular, is a Very Good Thing. Many non-Canon shooters seem to have forgotten this point, but it's actually sort of nice to shoot your 17mm lens at 17mm again etc. And I believe that a bigger sensor is a better thing - I'd like a 6*9 sensor rather than a 35mm size sensor if possible!

So anyway, that's what your paying for with your 5D. It's certainly not cheap considering the build quality - you're getting a 'consumer' body and autofocus. But there's been a lot of demand on the forums for full frame at an 'affordable' price, and this is Canon's response.

I still think it's a pretty good underwater camera, as long as it can be made to fit inside existing housing moldings (unlike the 1 -series, which are substantially bigger and thus seem deprived of sensibly priced housing solutions).

Topside I'm not entirely convinced that it's better value than a 1Ds, but that's probably just because I'm bitter that my 1Ds is now devalued to the point where's it's worth about the same as my Yashica Electro (that's not very much, BTW).

Another point to consider - with some of the cheaper housing systems, there have been concerns re: corner sharpness on crop sensor cameras. Full frame potentially makes corner sharpness much more of an issue.

I will resist the temptation to add more bird pictures to this thread. I still reserve the right to post brick wall photos, however. <_<

[ssra30]

Personally for me full frame vs cropped sensor is not really a big issue. More important issue is how well it is executed. Nikon already took care of the wide angle disadvantage with 12-24mm and 10.5mm DX lens.
Full frame, at equal pixel count to cropped sensor (all else being equal) theoretically will have advantage in digital noise level but if Nikon can manage somehow to keep the noise level down to an acceptable level, especially when compared to the full frame counter part then I don't see cropped sensor as a disadvantage beside having to do a little math when trying to figure out the lenses.
The trade off of higher noise level especially at high ISO vs problem with seeing deficiency in the periphery of full frame glass is not a dealbreaker for me as far as cropped sensor is concerned. If Nikon decides to stay with cropped sensor for their next 18-20+ mps camera, I have no problem with that as long as they can somehow make the sensor works and keep the noise down. If they can't then they probably be better off going full frame.

[herbko]

Now, if the pixels in the D2x are smaller, there are more pixels per unit area, does this translate into higher resolution?

<{POST_SNAPBACK}>

If you fill the frame with the subject on both a D2x and a 5D there should be very little differences in most cases. Of course we'll have wait until we see a detailed review of the 5D to compare, but I would be a little surprised if there's much of a difference.

At small apertures, the smaller, higher density, sensor will be limited by diffraction more than the larger sensor. It's a little complicated to sort out but the larger sensor has an advantage on macro subjects where you shoot to get the same print size and depth of field.

[Jolly]

I suspect that the 1:1 claim of the Canon 60mm lens is based on the 35mm format, even though it mounts only to EF-S camera, thus yielding greater than 1:1.  Otherwise the degree of magnification would be less than the 1:1 of the 100 macro (which IS based on 35mm), and offer even further confusion to the lens lineup!

<{POST_SNAPBACK}>

I am not sure if I understood you properly. A lens doing 1:1 is recording the subject 1:1 and has so far nothing to do with relation to certain sensor/film sizes. The only thing which is true: if you shoot a small subject 1:1 which fills the smaller sensor (APS-C) entirely, you would require a 1,5:1 (Nikon) or 1,6:1 (Canon) lens in order to fill the full frame sensor in the same way with the same subject.

Personally for me full frame vs cropped sensor is not really a big issue. More important issue is how well it is executed. Nikon already took care of the wide angle disadvantage with 12-24mm and 10.5mm DX lens.

<{POST_SNAPBACK}>

I still know some film shooters who really like their fast primes (in which they have invested a lot of money) a lot. Some of them consider switching to digital finally, but not all of them would be happy to exchange several f2.8 primes (14mm, 20mm, etc.) for those few and slow APS-C 10/11/12-20/22/24mm zooms. Add the price for a new APS-C fisheye and the 5D is already a quiet good deal. I don't say these zooms aren't good, but some people who would spend the money for cameras like the 5D might love to continue using their fast glass. From one point of view I would even prefer APS-C wideangle zooms because the have a lot more DOF than a full frame lens with the same FOV and therefore perform better behind dome ports in this respect. But I can not confirm that FF wideangles deliver poorer corners in general. I think we have been spoilt by FF lenses on cropped sensors where corners weren't critical because the lenses had been designed to cover a larger image circle. But APS-C lenses have shown the same effect now as they “just” cover the used image circle, like FF lenses with slides or FF sensors. I can not see the Canon EF-S 10-22mm beating all full frame L glass wideangle zooms and fast primes (17-40mm f4, 16-35mm f2.8, 20mm f2.8, 24mm f2.8, and so on) to replace them with just one slow zoom. I am sure Nikon shooters had more wideangle choices when shooting slides than “just” one equivalent 18-36mm f4 (12-24 on APS-C). I am sure Nikon will add more good glass to their DX line but for the time being, it’s not the worst thing if you switch to digital and simply continue to use your existing glass which is not only available now, but already owned.

Don't get me wrong, its looks like a really nice camera with a lot of advances.
But for a camera with that price tag ($3299) It’s looks odd that there are some important features(at list to me) that aren't there but they are in my 'entry level' camera that cost 1/4 of the D5.

<{POST_SNAPBACK}>

With film bodies it was simple, the next more expensive camera was better in most every aspect (frame rate, build quality, etc.). Resolution and sensor size was the same with all cameras: film. With digital we have to face different models being different kinds of compromise between the old days’ criteria and the digital part of the camera (sensor size / resolution). 5D’s priority is nothing but FF and resolution. So I would not say it’s odd that it doesn’t feature certain things just because of the high price. A 5D with Nikon D2X body and features would not be available for 3000 bucks, the 5D sensor is much more expensive than the D2X sensor. On the other hand a studio photographer with no requests to speed but resolution could say the Nikon D2H is so overpriced, just 4 megapixels for the same price like the 5D? Different Cameras for different users with different preferences, hourses for courses. Judging from the very first responses, it seems that the 5D could be a quiet attractive compromise for Canon underwater shooters.

Julian

[Simon K.]

Hi Simon,

Do you think that the 5D will fit in your S&S D60 housing - with a few mods? This might be the cheapest option and you wouldn't have to wait for the new housing?

Alex

<{POST_SNAPBACK}>

Interesting Idea. The S&S Housing is pretty big (What i first disliked but now like) so it is a possibility. I will do a Measuring on the Weekend when I have the Housing Back from the Suitecase of my Girlfriend who is still in Spain (had to put som weight to her suitcases)

But probably you have to change most Buttons and i don't know if this is

- much less expensive than a new housing
- Good for the future tightness of the housing

Anyone did something like this in tha past?