The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Canon 5Ds/5DsR

ShooterSteve

New member
Despite the maybe possible advantages of a Sony 50MP version camera, the most important thing is the quality lens lineup. And there Canon wins hands down WRT IQ and number of high quality L glass in every focal length and for all special purposes.

So my choice would be clear .... Canon! Even if DR is maybe not as good as an upcoming 50MP Sony sensor :cool:
I've been very unimpressed with most of my wide to normal Canon L glass compared with Leica R's and Zeiss. Even the 5Dmk3 brought out optical deficiencies with my 16-35 2.8, 17-40, former 23-70 2.8, 35 1.4, etc. The TS-E's and the new 24-70 2.8 are superb in my experience. The L long lenses have been great though. But 50MP with no AA filter is going to be a new challenge for sure. Even my Leica R's on my Nikon D800E shows slight issues that I never noticed on my 5Dmark3 images from the same lenses. I think we would have to expect new optical designs and tolerances for such high MP sensors. With prices to match.
 

turtle

New member
The best part about releasing a 50MP camera (for Canon) is all the upgraded Mk II lenses they will sell. And think of all the people who will eagerly await the next round of super lenses.

Releasing a 50MP camera will create a whole bunch of problems, which Canon will be only too happy to profit from solving. I am not being critical here - fair play to them - but we need to recognise the road in front of us if we go down it.

Some interesting insights being raised regarding aliasing and false colour/detail. I do, however, disagree that DR is all about pushing files to the max just because. I think it comes down to what we shoot and it is perhaps because I come from B&W film, where I am used to being able to do pretty well anything I want in terms of contrast ranges (with a lot of hard work, sadly), that I need digital to be able to give me that if I am to continue shooting in the way visual instinct now drives me. In general use, I would say less that 1% of my images runs into DR issues with 11.7 stops (banding more often). If I am shooting London at night, that rises to about 15-20% (including banding in there with the DR issue), depending on the circumstances of each shoot. I can't stand HDR imagery and so my DR aspirations are totally unrelated to achieving that (IMHO appalling) end result.

The aspiration for more DR is nothing more than a reflection of what people used to do with B&W negative film. Nobody noticed or cared to comment then, because it was taken for granted as integral to the process, if you really wanted to pull it out. Few people did, but for them it was important, just as it is now. One only has to remember St Ansel talking about using water bath development of some of his famous negs.

If Canon gets the banding issue fixed in this body it will at least be progress.

I've been very unimpressed with most of my wide to normal Canon L glass compared with Leica R's and Zeiss. Even the 5Dmk3 brought out optical deficiencies with my 16-35 2.8, 17-40, former 23-70 2.8, 35 1.4, etc. The TS-E's and the new 24-70 2.8 are superb in my experience. The L long lenses have been great though. But 50MP with no AA filter is going to be a new challenge for sure. Even my Leica R's on my Nikon D800E shows slight issues that I never noticed on my 5Dmark3 images from the same lenses. I think we would have to expect new optical designs and tolerances for such high MP sensors. With prices to match.
 

fotografz

Well-known member
I guess it would be possible to make a 50% cancelling, but I don't think that's how they've done it, you wouldn't get that cutting feeling when peeping if they leave some AA-effect in.

It's important to note that a camera AA filter is quite weak too, it *is* possible to get moire with AA filtered cameras too, it's just a whole lot harder. So the manufacturer chooses some threshold where you get away with almost all moire and aliasing in almost all situations, but not not more than that so you keep almost all real resolution.

Then we get the interesting question - if the AA filter does not solve the problem 100%, how much worse is it then to be without AA filter? I cannot answer that question really so I'd like to test. How sharp lenses are and how effective the microlenses on the sensor are will have an effect. If there was such a thing as an aliasing-problem-scale I'd think that being with AA filter would be 95% problem free, and without would be 30%, ie there is quite a large difference, but say it would be 70 vs 60 percent than one maybe think the R would be worth it...

I do know that the microlens-less MFD sensors like 39 megapixels P45+ and indeed my H4D-50 combined with razor-sharp tech lenses will have more aliasing issues than the 5DsR, but without testing I don't know how large the difference is. Would I buy "in the blind" I'd surely go for the 5Ds though, for the principles behind it. 1-2% extra resolution is not worth the possible extra issues.

The other point is interesting, if those aliasing artifacts in fact is an important component of the elusive "MFD look". I don't know. I haven't really seen the "MFD look" and not really seen anything special about it. The older CCDs have both this pixel peep microcontrast thing but also quite high noise level so you get a certain "grain" or texture, while a clean CMOS with AA filter might look more "artificial" despite it's more exact in its capture. Possibly this is a "MFD look" thing, but again it's pixel peeping. Now when MFD has the superclean CMOS that aspect is going away.

I'm pretty allergic to pixelated prints, but that too is a matter of taste. I haven't yet had much experience of really large prints, but heard others experiment with adding some light high resolution noise on top of the upscaled image to get a pleasing texture up close on the finished print, and that's probably something I would look into too, but to make the upscale work well to start with the original picture must be suitably soft on the pixel level otherwise pixelation is very hard to get rid of. I don't like the look of fractal upscalers, creating more resolution than was captured is simply not a good idea. I've looked at a lot of upscaling algorithms and the basic "bicubic smoother" always gives the best looking result.

As noone really knows what the MFD look is it is tempting to mimic them as much as possible, and sure the 5DsR is more like them. But pixelpeeping an image to find the MFD look I think is cheating :)

The reason why MFD lacks AA filters is not fully clear to me, but if I may guess the reason is that the CCDs used, which are used for many other applications, did not come with AA filters in the standard product range so they would have to be developed specifically for photography probably at a steep cost, and then it was better to make it a feature. More resolution is easy to understand and see, while the advantages of a slightly softer image takes a bit longer time to explain... I think it became more obvious to me now when I've started with printmaking, you're always resampling the image to match the print size, that pixel peep sharpness becomes less useful and that aliasing becomes more of a pain.
Perhaps it would be of value to consult with a master print maker to get a handle on techniques they use to produce such lovely looking large prints? I vaguely recall reading some fairly complex articles on the subject from those dedicated to making large prints.

As a point of related interest, some older MFD CCD backs had a user removable AA filter.

- Marc
 

Paratom

Well-known member
They bring a new wide angle zoom and I assume they designed it so its up to the task.
We allready know the current 24-70/28II is a very good lens (better than Nikon counterpart) so maybe it is up to the task.
Than there are the Sigma 35 and 50 Art.
And Canons Tele lenses are very good as well.
So IMO chances are not that bad that there are quite some lenses which might perform fine on a 50MP FF sensor.
Do we need /want 50MP is another story. There are probably more people who want it than people who need it.
I am interested how color will look like.
 

torger

Active member
Perhaps it would be of value to consult with a master print maker to get a handle on techniques they use to produce such lovely looking large prints? I vaguely recall reading some fairly complex articles on the subject from those dedicated to making large prints.

As a point of related interest, some older MFD CCD backs had a user removable AA filter.
On common trick is to use large format film ;)

I've asked around but beyond bicubic smoother and fractal/vector upscalers I have not found that much, but there are indeed some home-cooked recipes to try out. Truth is that it's a quite narrow use case, few print at those very large sizes and nose the prints afterwards. I may look into cooking up some own methods. I think that if you make a crossover between bicubic and vector and mask with some light high frequency noise you could do something better. It will reduce the jaggies issue but it won't solve false colors, for that you need to do even more guess-tricks in the demosaicing in the raw conversion.

Anyway I know my view (well, I'm not exactly alone on this) on the merits of antialias is controversial and not so popular in the MFD camp (where all current sensor are without AA). For a long time being without AA was one of the claimed reasons why MFD had better image quality, and I say that it's the opposite, to get better image quality you should have an AA filter. Not popular.

Noone in high-end audio would think about removing the AA filter to get high frequency noise and claim it would be better quality. The eye is obviously easier to fool than the ear.

If someone is able to demonstrate that you can get better image quality on a file viewed on a distance by not having AA I will gladly change my view, but so far I haven't seen any such thing (ie it's only about that pixel peep satisfaction), and see no reason why one would want to add artifacts for a 1-2% resolution increase. Then I think it's better to produce highest quality files possible now and wait with the resolution increase until an even higher resolution sensor is released, and they usually don't add only 2% extra, I would guess the next step is 70-80 megapixels.

50 megapixel is indeed a lot of pixels, 8688x5792, which means that a 100 ppi print will be 57". I think it's only below 100 ppi one needs to start to worry about pixelation aliasing, so in light of that it's not a big issue for most. On the other hand with that huge amount of resolution, why not secure the quality by adding the AA. Moire can happen in landscape, moving water is a classic. I've seen prints in books by well-known landscape photographers with water moire...
 
Last edited:

jpaulmoore

Active member
Do you think that the R version is going to present real world issues with Architecture (exteriors) and interior photography where fabrics are the norm?
Thanks,
J. Paul
 

gazwas

Active member
Do you think that the R version is going to present real world issues with Architecture (exteriors) and interior photography where fabrics are the norm?
Thanks,
J. Paul
+1

I would like to know the exact same thing however its only a guess until the cameras are in the hands of users. I was going to pre order a 5DsR but in light of torgers excellent comments I'm now going to wait until the cameras are released and evaluate then. If the Sony is introduced during the wait then all the more choice.
 

torger

Active member
I would just love to rent a 5DS and a 5DS R and really make a thorough A/B testing. Before that I don't really know how it will behave exactly. As said earlier I'm convinced the 5DS R will alias less than my H4D-50 because 1) even if Canon lenses are great and sharp, they are not as sharp as my tech lenses, 2) the 5DS R has modern microlenses, my H4D-50 has no microlenses at all which means less of the pixel sampled and more aliasing.

But if it's so little that it's negligible I don't know, and if it's negligible for person A doesn't mean it's negligible for person B, so a thorough test with examples so one can make an own decision from what is seen would be great. Although I think resolution-wise it's next to meaningless with the AA-filter less version and I think that is quite easy to demonstrate, there might be other advantages, such as a sharper live view which makes live view focusing easier, and it will be easier to really get that forensic analysis of lens sharpness, and I can't deny pixel peeping is more fun when it's sharper.
 

pedro39photo

New member
Last 3 years spend to much on High MP war, and just 20% of my clients saw or care about the difference against my 5DM3. Some H3D 39MP beautiful files got wasted in poor large prints made in a run for 2 or 3 days of booth fairs...

I make my run in DMF - mamiya zd - H3D 22MP - H3DII 39MP but i feel that in the last 5-6 years just pentax was bold to bring something new to this medium, and with the "right" prices and best body technology.

Now the problem its not about resolution its about motion ! i saw in the last 2 years a swift in my clients to the motion-movies.
My industrial clients now don't have large prints on there fair booths they have LCDs with movies !!!

I go to the shopping mall today and the fashion shops don't have beautiful large prints ! they have large vertical LCDs with fashion videos!!!

My next buy sad to say, will be a Canon C100, and today i spend more time on cinematographer discussion forums and less on medium format or pursuit of a High MP camera.

There will hallways be room for big resolution photography, but in the corporate business-some fashion i see a swift for more video/motion

PN
 

gazwas

Active member
Was interesting to read the below posted on Northlight Images Canon rumour site.

.......testing a pre-release 5Dsr with a test version of DPP4. Low ISO DR is put at 1.5-2 stops better than the 1D X, but high ISO performance (6400) falls marginally behind the 7Dmk2.
The sensor is similar to the 7D2 technology, but has apparently been tuned to maximise low ISO performance.


I know this is only a rumour but there still is hope Canon may have pulled at least 13 stops of DR out of the 50Mpix sensor which I'd happily live with.
 

jonoslack

Active member
I don't believe that there is any global advantage of not having an AA filter, that is if you zoom out or look at a distance so you don't see the pixel peep level then you won't see a difference. This is one of those religion type of things though. Before the D800 when there was a bigger gap between MFD and 135 the "no AA filter = feature" view grew strong, but I think it has more to do with that it was a differentiator than it actually provides a real advantage (except for pixel crisp).
Your arguments are refreshing and interesting BUT - don't you think that it's easy to extrapolate and just decide the 20mp is enough?
 

Guy Mancuso

Administrator, Instructor
Be it this is the pot of gold or not it pushes the industry for better. That's the real take away. I'll sit it out until it hits the streets and see if it makes sense but going back to a DSLR is not something I want to do. BTW I'm not buying it has not improved in the DR area. Lot of info going on without any REAL hands on testing with a production model. I'm also waiting to see what Sony does.
 

torger

Active member
Your arguments are refreshing and interesting BUT - don't you think that it's easy to extrapolate and just decide the 20mp is enough?
From 20 to 50 there's more than a doubling of pixels. From 50 with AA to 50 without AA there's corresponds to an increase of say 50 to 55 or so.

Trading that small resolution increase for risks of false color and moire artifacts is considerably harder to motivate.

But sure, also 20 megapixels reaches very far. I'm sure there are many users that would have preferred that they had stayed at 20 megapixels and instead achieved a record breaking dynamic range.
 

anGy

Member
:worthless:

I'm seduced by Torger's way of seeing things, until I see this kind of things (hope it survives the copy, paste, send to forum operations):



I enjoy sensor crispness better than softer results needing a software sharpness push. Even if this is not the better way to go, it's just more exiting :shocked:...
 
Last edited:

torger

Active member
Yes, I can't deny the pixel peep satisfaction :)

However both of those images are quite soft (with that kind of softness aliasing is unlikely), and you would in a real situation apply more sharpening on the AA-filtered version.

Technique in both shooting and raw conversion will be important to make the cameras perform as good as they can. Standard jpegs and indeed standard raw conversion is generally not tuned to make it as pixel peep sharp as possible.

The jpeg compression artifacts around the text suggests a quite heavily compressed jpeg which too will affect sharpness negatively.
 

MGrayson

Subscriber and Workshop Member
Once upon a time, there was a sound recording medium called the LP. But the LP had a problem - hiss. It had a lot of high frequency noise. So the engineers came up with a partial solution. They would BOOST the high frequencies on recording, and REDUCE them on playback. That way, the hiss would be reduced and you would hear more of the good high frequency sounds.

In particular, this meant that the RAW (see what I did there?) recorded sound was deliberately NOT what anyone would want to hear, nor should they EVER listen to it unless for some particular diagnostic purpose. No one would ever, say compare the RIAA encoded signal to the unencoded signal of a different method and say that one was better than the other. It would be crazy.

Best,

Matt (Who usually prefers non-AA sensors, but that's because it's the system that counts and other factors often outweigh this one. I have had some hellish aliasing on the M9, but I do this for fun. And I fall for that pixel-sharp look every time. Sigh.)
 

gazwas

Active member
Those pictures are a very compelling argument for the R version as the S just looks like a standard Canon file - a little soft and mushy just like my 1Ds3 (but do sharpen well).

However, if every bitingly sharp R image is accompanied by these funny coloured highlights (false colour as explained by Torger) as illustrated in the Hippo Canon test shot then who knows which is the best solution. This probably boils down to subject matter etc. I personally think its shaping up to be a great camera especially if the base ISO 1.5 - 2 stop increased DR rumours are true.

I can't wait for release so I can test for myself.

 

torger

Active member
I would guess that the Hippo processed in Capture One would render better highlights, the problem with pixel neighbors to clipped specular highlights is quite easy to solve. Phase One has long experience with AA-filter-free cameras and makes substantially less artifacts than for example Lightroom, and since Canon is doing their first(?) AA-filter-free camera I would not be surprised if their demosaicer in-camera is worse at hiding aliasing artifacts than Capture One.

So we can get to very different results depending on subjects, lenses, software, sharpening workflow and personal taste.
 

torger

Active member
Worth noting is that the 5Ds R vs 5Ds image with the flowers and blue apple has the following data:
Canon EOS 5Ds R w/ EF 24-70mm F2.8L II USM lens, ISO 400, f/11, 1/15 sec

It's an f/11 shot which means that there will be quite some diffraction onset. The crop is also far out to the left in the frame, and it's a zoom lens so it's not surprising that it's a little soft, could be hand-held too. At f/11 most aliasing will be gone due to diffraction, even with a perfectly sharp lens.

f/11 on the 24x36mm sensor corresponds to about f/16 on a 49x37mm sensor.

The cityscape from the air is also a f/11 shot (won't see much aliasing), while the hippo is a f/4 shot and there it will become more apparent. I think you will need to shoot at f/8 or larger aperture to see aliasing.

Here are some more sample shots using the 5Ds R, but unfortunately they have been scaled down:
http://www.ephotozine.com/article/canon-eos-5ds-5ds-r-hands-on-sample-photos-26928
They look promising though in terms of that real-world results in terms of aliasing will be considerably better than we're used to from medium format, because if it had been really bad we would have seen it also in the scaled down samples.
 

gazwas

Active member
Here are some more sample shots using the 5Ds R, but unfortunately they have been scaled down:
Canon EOS 5DS 5DS R Hands-On Sample Photos
They look promising though in terms of that real-world results in terms of aliasing will be considerably better than we're used to from medium format, because if it had been really bad we would have seen it also in the scaled down samples.
They look excellent and the construction images on the 5DsR have no signs of moire so a good indication for us Architectural shooters.
 
Top