Oh dear....5D3 vs D800

yakim_peled1 · April 21, 2012

Life happens today, not in the future. The cameras that we have today are amazingly capable. Their only weakness is that they depend on humans to carry them around and activate them, humans who sometimes would rather long for what they don't have than employ and enjoy what's in their hands.

</blockquote>

Hummmm........ Could the same rational be employed to other aspects in life as well? I mean, should I stop dreaming about younger women and be satisfied with the one I live with for the past 21 years? Nahhhh....... It can't be right.

:-)

Happy shooting,

Yakim.

david_k · April 22, 2012

I'd have to agree that the leap in pricing is quite unjustified. £2999 for the MK iii in the UK is far too much.

The same goes for Canon's new (and upcoming) lenses (24-70 & 200-400) which are extraordinarily expensive.

paulie_smith1 · April 22, 2012

"So does my 1978 Chevy Chevette, my 8-track tape player and my Betamax video recorder."

I can't find anyone to repair the 8 track player in my 74 Saab Sonnet.

On the other hand, for ultimate image quality the D800 is a better deal.

I still like 8x10 contact prints from the 50's vintage Deardorffs I use - tho the glass is much newer.

And I still sell prints taken with a Canon G3 from time to time.

g dan mitchell · April 22, 2012

Paulie Smith wrote something about:

"... ultimate image quality..."

</blockquote>

"Utimate" image quality? Really?

Dan

robert_turner5 · April 22, 2012

I truly don't get the abundance of conspiracy theory about the pricing of this camera.

For US customers, the 5dMarkII retailed at $3,000 in November 2008. Since that time, purely as a function of currency fluctuation, the US dollar has lost 22% of it's value against the Yen (99.00 11/2008; 81.14 3/31/12). $3,000 at MarkII introduction now costs $3,660 at MarkIII introduction. And this is for a camera that is arguably more competitive with its contemporaries than the MkII was at introduction.

Corporate greed? Sure it exists, but it doesn't have anything to do with the price in terms of reality. Consider how quickly major retailers immediately sold out their initial allocations of thousands of MkIIIs, which reportedly were shipped in large quantities.

Real reasons may lie in US monetary policy, Quantitative Easement, or increase in debt, and are beyond the scope of this forum. ;-)

Landrum Kelly · April 22, 2012

<blockquote>The cameras that we have today are amazingly capable. Their only weakness is that they depend on humans to carry them around and activate them, humans who sometimes would rather long for what they don't have than employ and enjoy what's in their hands.</blockquote>

Meanwhile, Dan, you have the 5D II, the 5D III, and the D800 on order:

http://www.photo.net/canon-eos-digital-camera-forum/00aCnS

From that eminence you tell the rest of us to be content with what we have.

Let them eat cake, eh, Dan?

--Lannie

dan_south · April 22, 2012

If you convert dollars to Japanese yen, the new camera doesn't cost more. It costs LESS than its predecessor.

Also, the 5DIII is manufactured in Japan. The D800 is manufactured in Thailand where labor costs are lower.

Lannie, well, I sold my car last year, so I had a few extra bucks. How much did you pay for your car? Probably more than I've ever paid for a camera. ;-)

Yakim, no no! Please keep dreaming about younger women.

rishij · April 22, 2012

The DR range is odd. It only appears at low ISO settings. However at both low and high ISO settings the SNR of the 5D3 and D800 are similar. That would suggest shadow noise should be similar, which would normally make DR similar.

...

DxOMark don't seem to explain exactly where the extra DR at low ISO is coming from, so I guess we'll have to wait for more explicit and transparent tests to figure that one out.

</blockquote>

I think Ilkka already explained this pretty well, but to recap: it'd seem that Canon's read noise due to 'dirty' electronics is what leads to the low DR. If DR is determined by the # of stops between the signal where SNR=1 & the saturation signal, then lower read noise will automatically increase DR... by a significant amount as well.

Actually, here's a great example of a controlled test by Fred Miranda that shows just how clean D800 shadows are vs. 5DIII (Canon users: seriously brace yourself):

http://www.fredmiranda.com/5DIII-D800/index_controlled-tests.html

As to your 2nd point: I fully agree. It irks me that the full testing protocols are often not enumerated on DXO's site... I feel they offer equations here & there to give you the feeling that everything's valid, but their methodology should be shared so it can be peer reviewed like any other scientific paper. I'm confused as to what they're actually doing when calculating DR. Are they doing something similar to a step wedge test? If, for example, they're just evaluating max/min signals in white(blown-out)/black(clipped) RAW files, those numbers would be fallible to non-linear signal processing or sensor response.

That being said, their intent appears to be noble & there appears to be internal consistency between cameras (one thing people need to remember when calling foul at DR results showing the D800 better than MF sensors is that, no matter how much highlight headroom you have in MF sensors, if the read noise is poor, that will dramatically decrease DR). But I'd still like to see their full methodology, & I still have trouble accepting that the 'normalized' DR goes up from 13.2EV to 14.4EV just from normalization. If it approached 14EV (14-bit ADC), that'd be easier to swallow. But even if read noise = 1ADU after 'downsizing', you'd max out at 14EV... can you drop below 1ADU read noise? And if you did, the signal at which SNR would then also be <1 ADU... which is not possible in a digital file. So that leaves me confused.

I find the Stouffer transmission wedge test to be pretty valid for comparing cameras. You're feeding each camera the same input, then looking at where you've clipped highlights vs. where SNR =1 in the shadows. I've carried out a Stouffer transmission wedge test on the 5DIII, & will compare it to my friend's D7000 later tonight (and D800 when he gets it). Though Fred Miranda's tests in the link above seem pretty conclusive, I do want to control for differences in highlight headroom between the cameras, as that would affect the DR result.

Cheers,

Rishi

Landrum Kelly · April 22, 2012

Lannie, well, I sold my car last year, so I had a few extra bucks. How much did you pay for your car? Probably more than I've ever paid for a camera. ;-)

</blockquote>

I'm glad you have your priorities in order, Dan. My newest car is a 1995 Honda Civic, but, hey, that's how I manage to buy cameras!

You knew that I was kidding, of course. As if I were the one making do with one camera and two lenses. . . .

--Lannie

g dan mitchell · April 22, 2012

Landrum wrote:

From that eminence you tell the rest of us to be content with what we have.

Let them eat cake, eh, Dan?

</blockquote>

Classic case of changing the subject, combined with ad hominem, which are often the strategies of a person who might feel more comfortable dissing the messenger than discussing the message. Rather than impugning the person's character, can we stick to the issue? Please?

I like Dan's concise and clear statement of the issue, namely the risk and distraction that can come with obsessing over small specification differences, exaggerated to the point that they take on importance far beyond any real effect that they will have on photographs, coupled with perhaps some personal anger and frustration over gear and the companies that make it - when the equipment is already far and away competent enough to allow one to make astonishingly great photography in both aesthetic and technical terms.

It would be interesting to see your response to the point he was actually making.

Take care,

(Another) Dan

g dan mitchell · April 22, 2012

Robert Turner wrote:

"I truly don't get the abundance of conspiracy theory about the pricing of this camera."

</blockquote>

Robert, you obviously haven't spent enough time in photography gear discussion forums yet! ;-)

Dan

robert_turner5 · April 22, 2012

Robert, you obviously haven't spent enough time in photography gear discussion forums yet! ;-)

Point well made, Dan...

I was under the impression Nikon/Canon threads were always reasonable and devoid of hyperbole. ;-)

On another note, I have a question about the Fred Miranda tests referenced before (to you, Dan, or anyone else who may want to chime in).

Fred says " It's unfair to downsize the D800 36MP file to 22MP in order to compare the advantage of higher resolution. Doing so, would throw away detail from the higher resolution file. Instead, I interpolated the 22MP file to 36MP. I believe this is a fair comparison because both files will be printed at the same paper size. "

Fine, but isn't it equally unfair to interpolate the 5d/III image? Isn't he essentially asking software to "create" missing data?

Granted, the color noise likely has nothing to do with interpolation, but I would think if he makes a decision in his tests so as not to be unfair to Nikon, he should more fully defend modifying the Canon's image in order to have the same sized "100%" image crops.

But, as usual, I could be missing something basic and essential...

rt

</blockquote>

Brad_ · April 22, 2012

>>> "Feeling good about your equipment is the most important aspect of photography"

I just came back from a 1,300 mile desert road trip in Nevada and California. <a

href="http://www.amazon.com/Canon-PowerShot-ELPH-100-HS/dp/B004J3V8UU/ref=sr_1_1?

ie=UTF8&qid=1335143708&sr=8-1">This camera</a> served me well for the trip. I should check and see what DxOMark has to

say about it, so I'll know where it slots in with other cameras, and how I should really feel about it.

g dan mitchell · April 22, 2012

Brad appropriately joked:

"I just came back from a 1,300 mile desert road trip in Nevada and California. <a href="http://www.amazon.com/dp/B004J3V8UU/?tag=nmphotonet-20" rel="nofollow" target="_blank">This camera</a> served me well for the trip. I should check and see what DxOMark has to say about it, so I'll know where it slots in with other cameras, and how I should really feel about it."

</blockquote>

+1

I did a 2000+ mile trip a few weeks ago, entirely for photographic purposes. I can report that I did not think once about how my camera stacks up against other cameras. I did think a lot about the potential for bringing back interesting and hopefully compelling photographs.

Do enough photography and gear is just gear. It is not an object of lust or a way of measuring personal worth. :-)

Dan

dan_south · April 22, 2012

First, to clear the air, Lannie and I were both just having some fun with each other.

Regarding pixels versus pricing, I'm sure that we'd all love to have more pixels at lower prices. However, the world has gone through some tough challenges, and businesses have to do what it takes to survive in this climate. We've just lost Kodak. Let's please not lose Canon and Nikon to bad business decisions.

Regarding the excellent points made by Brad and (the other) Dan, paralysis by analysis seems to be all the rage on photo forums these days. It's a real shame, because some brilliant technology has just come to market (5D3, D800, NEX-7). You'd think that people would be having so much fun making photos that they wouldn't have time to bicker about the other brand's specs.

If you're out there working hard to make the best pictures you can, you won't have time to obsess over pixel density, anti-aliasing filters, or the chromatic aberration bogie man.

Case in point: Luminous Landscape is featuring photos taken by an astronaut who piloted one of the last shuttle missions. It's some of the most spectacular photography that I've ever seen. What camera did he use? A Nikon D3S.

A typical reaction from a pixel and bargain obsessed gear head might go something like this: Hey, wait a minute! That can't be right. The D3S has only 12 MP. That's not enough for serious photography! NASA needs to purchase some D800's and fly that mission over again! Even worse, the D3S costs over $5000US. What a ripoff for only 12 MP!

Finally, with regard to the Fred Miranda 'controlled' test, IMO it's not acceptable to draw conclusions when comparing out of focus regions of photographs. Yet, so many Internet-published 'camera tests' do just that.

yakim_peled1 · April 23, 2012

Do enough photography and gear is just gear. It is not an object of lust or a way of measuring personal worth. :-)

</blockquote>

While I can fully agree with the second sentence I can not agree with the first. When I shoot I need the gear to be an extension of my eye and hand. For this reason I give the ergonomics of the camera at least as much importance as its IQ. A camera with great IQ but which is not comfortable to shoot with or if it lacks features I consider essential is of little use for me and I would rather spend my money on a different camera with a bit less IQ but which is more user friendly and have the other things I need.

Case in point #1: About 2 years ago I owned the Sony A33. It had several features that I liked (e.g. revolving screen, AS, compactness and AF in video) but I couldn't make it work seamlessly for me. I thus sold it after a few months of use.

Case in point #2: I avoided the 5D and 5D2 and went with the 40D and 7D respectively. After playing a bit with the 5D3 though I came to a conclusion that it is a completely different beast (I think it's the mythical 3D we've been waiting for) and apart from the name has very little with its predecessors. Thus, I can certainly see myself owning one some day.

Happy shooting,

Yakim.

angkordave · April 23, 2012

So Canon is charging a whopping $500 more (list price) for improved AF and some modest tweaks here and there to upgrade to the 5D3.

</blockquote>

The differnce in price is significantly more here in SE Asia where the price of a 5D3 is $3500+ and the old 5D2 is $1900 . I have just purchased a 5D2 as opposed to a D700 after trying a 5D2 out for a day The 5D2 is a much better camera with video too. (the D800 is $3700 here at present and out of my price range).

Another point which bugged me about Nikon video is the awfull sound quality even with an external mic apparent;y the D800 and D4 has the same sound setup as D300s. A 5d2 has much better sound which can be controlled manually. It may not be so important to some here but HDVideo is an important consideration to me. Reading this forum, anyone would think that a 5D2 is outdated and obsolete. When I asked pros who use them and after my own testing; the 5D2 is an excellent camera and a relatively cheap entry to FF Digital. Especially for those like me who have to upgrade to FF lenses. Do I really need 36mp? If I can produce decent 90cm wide prints from a 6MP Fuji S2 Pro then I'll have no problems witf a 5D2.

g dan mitchell · April 23, 2012

Yakim wrote:

"While I can fully agree with the second sentence I can not agree with the first. When I shoot I need the gear to be an extension of my eye and hand. "

</blockquote>

I have to accept your first sentence, and I fully agree with the second. :-)

It gets at a different issue than the one being looked at here, but it is a very important point. (That "other issue" is the one about the small technical issues and their relative importance and, even more, the tendency of some to obsess about trivial differences and about the fetishisms that develop around gear.)

When I shoot, my ideal would be that the gear essentially becomes invisible. In other words, because I have done so much shooting with it and dealt with photographic issues frequently enough, the technical parts of the thought process will become so second nature that I don't struggle with them. In addition, the operation of my gear - how to set up the camera, which lenses might be right, an so forth - are familiar enough to me that they don't take mental energy or focus away from the more important aspects of making photographs.

This actually does loop back to Gear Upgrade Syndrome - the compulsion to feel obligated to upgrade or panic when any new "better" thing is announced. To my way of thinking, the intuitive skills to operate a camera are critical to making most good photography. By using a fine camera (and associated gear) for a number of years - let's take my 5D2 that I've used for over three years - I am able to function effectively and intuitively while shooting. If another camera comes along, especially if it is from a different manufacturer, it isn't enough for it to be a little bit better in a few ways. For me to consider "upgrading," it must provide truly compelling advantages that would convince me to relearn all of that important operational stuff. Every so often that happens, but not with every new bit of gear.

I think we're roughly on the same page on this one, Yakim.

Take care,

Dan

(Note: I rely on good gear that I use, appropriate to the types of photography I am doing, but I honestly don't think about the gear in terms other than what it does for my photography.)

ilkka_nissila · April 23, 2012

I still have trouble accepting that the 'normalized' DR goes up from 13.2EV to 14.4EV just from normalization. If it approached 14EV (14-bit ADC), that'd be easier to swallow. But even if read noise = 1ADU after 'downsizing', you'd max out at 14EV... can you drop below 1ADU read noise? And if you did, the signal at which SNR would then also be

The raw file contains the pixels with 13.2 EV dynamic range (at manufacturer's ISO 100). The file is first transferred from camera to PC and converted into a 16-bit TIFF and then is downsized to 8MP 16-bit TIFF. The camera hardware cannot produce these downsized images with enhanced DR (although recent Nikons do have 16-bit processing pathways the only reduced size image options are 8-bit per color). Here DX assume that the user shoots RAW and then processes the images on a computer using image formats that allow the high dynamic range sought.

As per whether even the average can go below 1 ADU, yes it can? The 4.5 pixels that are averaged to go from 36 to 8MP have 14*4.5 bits = 63 bits, plenty of room for improved DR.

yakim_peled1 · April 23, 2012

I think we're roughly on the same page on this one, Yakim.

</blockquote>

Yep. great minds think alike. :-)

So Canon is charging a whopping $500 more (list price) for improved AF and some modest tweaks here and there to upgrade to the 5D3.

</blockquote>

Yesterday I had a chance to play with the 5D3 and this short experience was enough to convince me that the changes are far from being modest. Yes, the improved AF system is the most noticeable change as the 5D2's AF system was very primitive (derived from the 20D) but the improved VF, better fps, better high ISO, added buttons, enhanced customability etc. all add up to a completely different beast. This is not a 5D3. This is the mythical 3D we've all been waiting for. As such it is well worth the added cost IMHO.

Happy shooting, Yakim.

Landrum Kelly · April 23, 2012

First, to clear the air, Lannie and I were both just having some fun with each other.

</blockquote>

Thank you, Dan. My original point (Apr 19, 2012; 08:00 p.m.) was simply that Canon is not and will not be standing still--not typically the kind of claim that generates a lot of controversy. I find it inconceivable that Canon will not try to match Nikon in the megapixel race, at least until such time that there is no profit in that. Nikon's latest crop sensor camera with about 25 megapixels indicates that Nikon still thinks that more megapixels sell--which is not to say that having more megapixels has very much to do with quality.

In the case of the D800 (not the crop sensor camera with 25 megapixels), I think that having more megapixels actually is about better quality--for a certain type of photography, at least. I am a bit more skeptical about packing 25 mp on a crop sensor, but we shall see. . . .

As for what DxOMark has to say, I can't say that I have ever been able to come to any firm conclusion based on its analyses. Some things can be meaningfully quantified, and some cannot. I recoil when someone tries to give a grade to something like top-of-the line cameras. I find Luminous Landscape and Dpreview.com more useful in general, and even with the latter I am not sure how useful its number ratings are. I did like its old method that was used to show side-by-side comparisons better than the present system, but it is all worth looking at, in my opinion--yes, it is pixel-peeping, and I confess to doing my fair share of that.

We do a fair amount of product comparisons on these threads, and I see no alternative. Most of us get pretty enthusiastic when we see a bigger leap forward than what we were expecting--especially when it occurs at a reasonable price. I remember how excited I was when Canon initially announced the 5D at about $3500. That sounds expensive in retrospect but it was a major breakthrough then. It was not so long before the price sank to $2500 or so, and perhaps the same thing can happen with the 5D III. Either way, I am sure that I would be delighted to have the 5D III, but the D800 or something like it has more appeal to me right now. Maybe if I used it, I would not be so enamored.

Then again, I am still playing around with an old Kodak DCS Pro SLR/n with about 13.5 megapixels--and the real reason is pixel-for-pixel sharpness:

http://www.photo.net/photodb/folder?folder_id=1028279

That kind of sharpness is what can happen when one leaves off an anti-aliasing filter. The Nikon D800 E is interesting for much the same reason. Why my obsession with resolution? It goes back to my adolescent love of good refactor telescopes. I am still in awe of high resolution Does high resolution have practical applications? You better believe it.

--Lannie

dogbert · April 23, 2012

I find myself at a bit of a cross roads with my Canon system; not because of the FF camera pricing, but because of the accessory pricing. The Canon lenses are simply moving out of my price range. $2K for a 24-70/2.8? Yikers! The new 600EX-RT flash (or whatever the model number is) is what. . .$600? For a single flash unit? Look. . .I make a good living. . .I can afford toys. . . .but this market has simply left me in the dust.

</blockquote>

+1

I make plenty of money and love photography but Canon's pricing of late is taking all the joy out of it for me. All I wanted was USM in their cheap legacy primes such as the 35 f2, and 24 f2.8 and given they were able to put aring USM in the $200 28-105 f3.5-4.5 I had high hopes. So what did Canon deliver? $800 primes. All I wanted was ring USM in the 70-300 IS, and Canon gave us a $1500 L version that weighs more than a housebrick.

I'm too stuck in the Canon system to change but over the last three our four years Canon has introduced very little of interest to me - the pricing is part of the problem - while Nikon has. And if I were starting today, I'd probably choose Nikon.

pierrearnaud · April 24, 2012

For all the speculation about Canon's marketing strategy, I think that the 5DIII will appeal to those who, like me, held off after the the 5DII came out, because they would have preferred improved AF as opposed to higher pixel count.

Regarding the DR difference between the 5DIII and D800, I'm quite ready to accept that Nikon has a serious lead, but I'm still wondering how this translates in practical terms. The best hands-on approach to this I have seen so far is shadow recovery examples such as :

http://www.fredmiranda.com/5DIII-D800/index_controlled-tests.html

as this translates numbers into real photographic meaning. However, I have an issue with this and other similar examples. One of my most common PP correction in LR3 is to push shadow recovery (fill-in) quite a bit on high contrast images. I do push the cursor halfway through the slider on occasions, sometimes even more, and never have had results remotely as bad as Fred Miranda's example, and this is with files shot with the lowly 5D classic.

Given that my camera has significantly lower DR than the 5DIII, I have to wonder if don't have to push correction beyond realistic use to reveal that kind of a difference between the 5DIII and the D800.

The other thing I am wondering is how the DR difference translates in terms of hilights, something I have not seen so far, and which might be more important than shadows. I remember playing with a D700 several years ago and comparing identical shots with my 5D classic. I loved the AF of the D700 but found some blooming artifacts in the blown out hilights which were more troublesome than those produced by the 5Dc.

Shouldn't we concentrate more on how sensors behave in the hilights, since no matter how much DR we have, there will always be situations where we face blown out hilights, such as night photography, for instance ?

dan_south · April 24, 2012

After reading the Fred Miranda report I'm thinking about canceling my D800 order. Live View and high quality TS lenses

are very important to me. I think I'll be sticking with the 5D3 for the foreseeable future.

rishij · April 24, 2012

I do push the cursor halfway through the slider on occasions, sometimes even more, and never have had results remotely as bad as Fred Miranda's example, and this is with files shot with the lowly 5D classic. Given that my camera has significantly lower DR than the 5DIII

</blockquote>

Not sure where you're getting your data from but if you look at DXO, the 5DIII actually has worse low ISO dynamic range than the 5D at the pixel level.

Just the other day I tried to do a +2 stop exposure on an accidentally underexposed (heavily backlit, before I dialed in EC) shot from my 5DIII and already saw banding.

I actually had similar experience with my 5D Classic, but my 5DII seemed to behave slightly better. I think that might be explained by inter-unit variability. All in all, the entire line of 5D sensors (maybe all Canon, I don't know, I'm only speaking for what I own, which is the entire 5D family) is unacceptably prone to banding & low ISO shadow noise when compared to (recent) competition (the D7000 & D800, for example).

-Rishi

Sign In

Oh dear....5D3 vs D800

Recommended Posts

Link to comment

Share on other sites

Top Posters In This Topic

Popular Days

Top Posters In This Topic

Popular Days

Posted Images

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Link to comment

Share on other sites

Create an account or sign in to comment

Create an account

Sign in