What’s Up?
Yesterday I answered more than 140 e-mails, many dealing with Used Photo Gear stuff; the Used Gear page here has really been hopping. I did, however, have time for a chilly swim–the pool was down to 71 degrees yesterday, some core exercises, and an ice bath.
If…
If what you read below inspires you to purchase a Canon EOS 5DS R I would of course greatly appreciate your using my B&H affiliate link for your purchase; it won’t cost you a penny more and is the best way to thank me for the work that I put into the blog every day. This blog post took well more than four hours to assemble.
The Streak
In spite of having been buried by travel, teaching, the exhibit, and several major writing projects for the last two months, today’s blog post marks 79 days in a row with a new educational blog post. Again, please remember to use our B&H links for your major gear purchases. Your doing so is always greatly appreciated. 🙂
|
|
Canon 7D Mark II sharpness test image courtesy of and copyright Patrick Sparkman |
Bill Lloyd Was Right: Canon EOS 5DS R versus EOS 7D Mark II
With regard to the side by side comparison image in yesterday morning’s blog post here, the left hand image–see the JPEG above–was made under stringent testing condition by good friend Patrick Sparkman with the Canon EOS 7D Mark II. The right hand image–see the JPEG below–was of course also made under stringent testing condition by good friend Patrick Sparkman with the Canon EOS 5DS R.
Here is Bill Lloyd’s comment on the original post: 5DS R on the right, 7DII on the left, betcha.
Kudos to Bill. 🙂
|
|
Canon EOS 5DS R sharpness test image courtesy of and copyright Patrick Sparkman |
The Results
Nearly all who commented felt that the right hand image was dramatically better on all counts.
These two comments summed up the consensus:
Lance Krueger: In answer to your five questions, the image on the right is dramatically better.
Jim Amato: Right photograph: Sharper, cleaner, higher degree of pleasing contrast, best image quality. Dramatic difference: Much sharper, tonal differences, more contrast with all tones. Letter E shadow and “lace” work more defined and pronounced. That is my 100 cents worth.
Patrick and I agree, but only 100%.
Patrick’s Testing Protocol
The camera body test was set up as follows: Tripod-mounted Canon 600 f/4L IS II, Mongoose Head locked down.
I clamped the dollar bill target to the Michael Tapes Lens Align calibration set-up mounted on a sturdy on a tripod to ensure True Parallel alignment.
The target was illuminated by ambient indoor light and was approximately 30 feet from the lens.
The lens had Image Stabilization and autofocus turned off. WB was set to 4200K. The lens was focused manually using Live View at the highest magnification, and was refocused for each camera body but was not moved otherwise. The images were created using manual exposure. Large RAW, 10-second with mirror lockup.
The images were processed in DPP 4, and all the of the settings were left at zero except for setting Sharpness at 3, and checking the boxes for Peripheral Illumination and Chromatic Aberration correction.
Each image was then similarly cropped to a 1200 pixel square; this approximated a 100% crop that was easily duplicated for each image. After the identical RAW conversions in DDP 4 the TIFFs were imported into Lightroom where I used the export utility to create a JPEG. No sharpening was applied; file sizes were limited to < 300kb.
Patrick’s Initial Reaction
After completing the test Patrick called my cell and said, “The results are amazing. The 7D II images look as if they were photographed through a fog filter when compared to the 5DS R images. The big thing is the improved contrast but the sharpness differences are astounding as well.
Simplifying Things
Let’s say that I am standing right next to you with the same lens photographing the same bird at the same distance and that our sharpness techniques are equal. I am using a 5DS R and you are using a 7D II. Your bird will of course be larger in the frame. But when I crop my image so that the bird is the same size in the frame as your bird, my image will be sharper with more pleasing contrast and be of overall higher quality. And those differences will be just as dramatic as they were in Patrick’s tests.
I am pretty sure that this statement is correct: the two images–your original and my cropped to the same subject size–will have virtually the same pixel dimensions. Please feel free to correct me if my understanding here is not correct.
Patrick pointed out the following: in the situation above if the bird flaps the guy (or gal) with the 7D II will clip the wings while the person with the 5DS R will get the shot.
More Thoughts From Patrick via e-mail
In my opinion, while the image quality of the 5DS R is certainly a bonus over the 7DII, the great advantage for my photography comes from the ability to have the equivalent of a 1.6 ratio zoom on the l600 II (or any lens) as compared to the 7DII. When I use the 600 with a 1.4 teleconverter, I will have an 840-1344 zoom versus the 7DII where I have only 1344. This will in fact greatly aid flight shooting; I can work with the 600 alone, be able to acquire AF on the subject easier due to the full frame view at 600mm, and then crop down to the equivalent 960mm of the 7DII (or to any effective focal length in between 960 and 600. All that while getting better image quality than we have ever had before. And when I can get closer to the birds image quality will be off the charts.
Thanks for loaning me the camera for the test.
My Original Gut Feelings
Right from the get-go I was floored by the quality of the 5DS R images. I have long claimed not to have a great eye for fine detail and that remains the case. But when I looked at the 5DS R images I knew that they were sharper, that they offered far more fine detail, and that the image quality was far superior to anything I had ever seen before. I did not, however, realize that the increase in pleasing contrast was a huge contributing factor (even though in retrospect that seems to be obvious).
In short, I was extremely gratified that the results of Patrick’s test confirmed my gut feelings.
The Big Question is Why?
So why are the 5DS R images so much better than the 7D II images? As far as I can figure, there several possibilities:
1-the softness and low contrast look of the 7D II images might be due to the low pass anti-aliasing filter that is immediately in front of the sensor. The converse with the 5DS R might be that the cancellation of the low-pass filter effect allows for sharper, higher quality image files with greater fine detail.
2-it is possible that the sensor (and thus the quality of the pixels) on the 5DS and the 5DS R cameras is much better than on the 7D II.
3-some combination of 1 and 2.
The bummer is that a friend in San Diego had a 5DS that could have been loaned to Patrick for the tests…
From Rudy Winston
I asked Rudy Winston, one of Canon USA’s most knowledgeable technical folks, why the 5DS R images were so superior to the 7D II images. He had lots of interest to say:
Assuming when you say “images were cropped to the same size on the dollar bill,” you mean that the same (or virtually the same) number of actual pixels were used to record both files, I’m not stunned by what you see here.
The absence of a low-pass filter will indeed give a somewhat sharper initial result, out of the camera. There are other things I don’t know about these files (in-camera JPEGs? If RAW files, how processed? etc.), but again, assuming equivalent handling, right off the bat, that’ll give a 5DS R file an edge in terms of how crisply it renders fine detail.
To some degree, that difference can be minimized for an EOS 5DS, EOS 7D Mark II, or other camera body shooters by using sharpening like Unsharp Mask to counter the impact of the softening from the low-pass filter immediately in front of the sensor on those cameras (and nearly all others in our line).
Here is another equalizer that–to some degree–many folks by-pass: the Digital Lens Optimizer feature that can be accessed when RAW images from a recent EOS camera and a compatible (i.e., recently-introduced) lens are processed in Canon’s DPP software. Digital Lens Optimizer is an optional form of “smart sharpening” (my choice of words, not Canon’s), that applies specific sharpening based on lens parameters for each individual image. When truly minute detail is needed/desired, it’s an interesting tool to try on a Canon RAW file.
Finally, in at least some situations, the mechanics of the 5DS series are designed to minimize the impact of mirror shock, and so on — which can contribute to slight changes in image quality when viewed extremely critically. I know you said that these were shot with mirrors locked-up and so on, which probably neutralized most if not all differences between the architecture of these two cameras, but it’s something to keep in the back of one’s mind when making other real-world comparative evaluations.
To your question about the sensors of 7D Mark II vs. the 5DS series: they’re definitely not the same, in terms of being identical, although the pixel size on each is VERY similar. Canon hasn’t revealed precisely what differences may exist at the sensor level between the 5DS and 7D II sensors, but it’s clear that they’re not exact twins at the pixel level. Any differences that might be contained within the full-frame 5DS-series sensor would, of course, be on top of the impact of no low-pass filter which we get on files from the 5DS R. But we’re not claiming any specific, quality-based differences at this granular level.
And, just to reiterate something I’m sure you have already been made aware of: the CMOS image sensor on the EOS 5DS and 5DS R *are* identical. The only difference is the cancellation of the low-pass filter effect, immediately in front of the sensor itself on the 5DS R.
Hope this is helpful to you! — Rudy Winston
What About Frame Rate (and other 5DS R negatives)?
Yes, the 5DS R is much slower than the 1D X and the 7D II. At times I will miss those higher frame rates but not nearly as much as you might think. Why? regular readers know that I have never been a hold-the-hammer-down type of shooter. For flight photography I rarely take more than two or rarely three at most images when the bird passes through the ideal zone. I will most miss the high frame rates of the 1D X and the 7D II during intense high interactions such as bird altercations or copulations. But for me the trade-off in image quality is well worth it. Remember also that my great skill is in designing gorgeous images of static subjects. With birds in flight and in action ? Not so good.
The only other 5DS R negative is the fact that if you are doing any video or Live View work you need to have an extra battery or two in your pocket or vest. Heck, that is recommended even if you are just having a great morning. In an effort to save weight, I have stayed away from the vertical battery grip with the 5DS R.
And yes, the 5DS R is a bit on the pricey side for hobbyists.
Am I Serious?
Very. Yesterday I ordered a second 5DS R. When I head to Japan it will be with two 5DS R bodies and a single 1D X as a back-up.
What about the cameras that I already own?
Please remember that images from all previous Canon cameras are fully capable of producing outstanding, high quality images that have been honored many hundreds of times in prestigious international photography contests. Many of my images have done so over the past 13 years. But the fact is that images from the 5DS R are simply better.
That said, please understand that your skills, your vision, your heart and determination, are of far greater importance to the quality of your images than either your lens or camera body choices. I have long railed against the fact that many folks believe that if they go out and buy a $12,000 lens and a $6000 camera body that they will instantly become a better photographer. Nothing could be further from the truth. To become a better shooter you need to study, practice, look at as many great images as possible every week (can you say BPN?) and subscribe to the BAA Blog :).
|
|
birds as art: The Avian Photography of Arthur Morris/The Top 100
|
birds as art: The Avian Photography of Arthur Morris/The Top 100
If you missed the announcement of the new e-book click here for complete info that includes details on getting a signed copy.
Feedback Requested
If you ordered via download or have already received your CD, I would love to hear what you think. Please do leave a comment.
Please Remember to use our Affiliate Links 🙂
To show your appreciation for my continuing efforts here, we ask, as always, that you get in the habit of using my B&H affiliate links on the right side of the blog for all of your photo and electronics purchases. Please check the availability of all photographic accessories in the BIRDS AS ART Online Store, especially the Mongoose M3.6 tripod heads, Gitzo tripods, Wimberley heads and plates, LensCoats and accessories, and the like. We sell only what I have used, have tested, and can depend on. We will not sell you junk. We know what you need to make creating great images easy and fun. And we are always glad to answer your gear questions via e-mail. I just learned that my account was suspended during my absence; it should be up and running by Monday at the latest.
I would of course appreciate your using our B&H affiliate links for all of your major gear, video, and electronic purchases. For the photographic stuff mentioned in the paragraph above we, meaning BAA, would of course greatly appreciate your business. Here is a huge thank you to the many who have been using our links on a regular basis and visiting the BAA Online store as well.
Be sure to like and follow BAA on Facebook by clicking on the logo link upper right. Tanks a stack!
Typos
In all blog posts and Bulletins, feel free to e-mail or to leave a comment regarding any typos or errors. Just be right 🙂
Holy Cow!
What happened to artistic vision and understanding light?
The tech stuff is important; however it is a support system for the image that the
Photographer produces….
I hope all this energy goes into everyone’s artistic talents and light understanding….
Jim, Please do not forget that that is what I and the blog are all about. Once in a while it is fun to delve into the technical/pixel side of things. And heck, even George Lepp liked this post. In addition, it has sure generated lots of civil, back and forth dialogue. artie
One more thing: Since this is the Birds As Art blog it was most fitting for Patrick to choose an image with an eagle on it. But the heretofore unasked question is: What is that blueish blob on the eagles tail feathers. None of the dollar bills in my wallet have this. Old money/new money?
No clue Glen. I will try to remember to ask Patrick. a
Artie, I thought we had moved on past the pixel question a few years back. Yes, it is fun to play these games, but they are not really useful in picking the right camera today. I would rather pay more for wider latitude and less noise than ultra high resolution (which can be dealt with in post processing).
Jack Goodman
Have not. Your statement above is unclear. It seems that you wrote that ultra high resolution can be dealt with in post processing though I am pretty sure that that is not what you meant.
What do you mean by wider latitude? And what do you know about noise with a 5DS R?
Are you suggesting that the 5DS R is comparable to the old 7D?
Have you ever photographed with a 5DS R?
artie
Greetings Art,
I watched and waited for the pandemonium of PS’s “Dollar Bill Test” to die down. I’ve seen these types of tests since they were used to compare 35mm to 2-1/4 to 4X5 film cameras and the end result is always the same. Whether it is binary bits or silver bits, every thing else being equal, bigger is better.
When someone does a test like it is no surprise that the critics show up in numbers. So here is mine! I do have to question PS’s methodology: comparing the two cameras at ISO 800 skewed the test in favor of the FX “winner”. Native ISO would have been in order for this test. If I read correctly, this was a 10 second exposure? A 600mm lens? Hmm, I think that for sharpness and image quality test that throws any real conclusions out the window no matter the tripod or head.
For what it’s worth when I looked at the two images I thought DX left, FX right regardless of camera make or model. I hope that PS does not take any of this personally, tests like this are hard to conduct outside of laboratory conditions, and I do applaud his effort. It was of value that it does confirm my earlier comment that bigger is better. Now how about the Art Morris 1dx v 5dsr comparison?
Hi Glen,
Not sure what you mean by bigger is better as each test image roughly contained the same number of pixels…
What is FX?
You may have a point on the ISO used. I actually have the lights and stands that would improve the testing protocol. If I get back to San Diego in March I just might ship it all to Patrick….
The exposures were both 1/20 sec. Not sure where you got the ten seconds from….
What is DX?
Taking comments personally is a choice, one that I try to avoid 🙂
I do not do tests as I am not smart enough. That’s why I was glad that Patrick volunteered. I go by my gut feeling.
artie
Hi Art, my bigger is better comment referred to the amount or percentage of enlargement required of a given image, digital or film. FX refers to full frame sensor and DX about half-frame depending on sensor manufacture. DX cameras really take a beating to their FX counterparts as the ISO increases so hence the native ISO stipulation. Turns out my 10 second comment was the delay for mirror up not the shutter speed which you indicated in your reply is 1/20th second. Better, but 1/20th of a second is still one of the more violent (vibration) shutter speeds on the dial. I’m with you on not doing these tests, eliminating the variables requires laboratory grade conditions and time that neither one of us seem to have.
P.S. This topic itself is a can of worms of large proportions but is fun none the less.
Thanks Glenn but that leaves me more confused. Only the 5DS R image was “enlarged” at least as far as I can figure… By what you are saying it seems that that should have given the advantage to the 7D II image….
We are working at getting some more sophisticated tests done 🙂
a
David. I never said that I would use the in camera crop. I would never use that and only crop in post processing. Obviously, if I could get closer to the subject, and fill the frame on the 5DS R I would. But that is not always possible, so the 5DS R allows for significant cropping in post and still maintain good image quality. I hope that answers your question.
As far as I know, there is no such thing as in-camera cropping as there is with some Nikon bodies… a
ps: thanks Patrick!
Artie: This from Imaging Resource: “Also, in what we believe is another EOS camera first, Canon has added crop shooting modes for both 1.3x and 1.6x crop factors.”
This is what I was talking about. Check your manual! 🙂
David
What and where is “Imaging Resource”? Which manual? a
Imaging Resource is a photography site that reviews cameras among other things. http://www.imaging-resource.com. I was referring to your camera’s manual. It does have the two crop modes, you can trust me on that.
David
Artie: This is from the Canon web site on the 5DS:
File Size
Full Size
Large: Approx. 50.3 Megapixels (8688 x 5792)
Medium 1: Approx. 39.3 Megapixels (7680 x 5120)
Medium 2: Approx. 22.1 Megapixels (5760 x 3840)
S1 (Small 1): Approx. 12.40 Megapixels (4320 x 2880)
S2 (Small 2): Approx. 2.50 Megapixels (1920 x 1280)
S3 (Small 3): Approx. 350,000 Pixels (720 x 480)
RAW: Approx. 50.3 Megapixels (8688 x 5792)
M-RAW: Approx. 28.50 Megapixels (6480 x 4320)
S-RAW: Approx. 12.40 Megapixels (4320 x 2880)
Crop – 1.3x
Large: Approx. 30.5 Megapixels (6768 x 4512)
Medium 1: Approx. 24.1 Megapixels (6016 x 4000)
Medium 2: Approx. 13.6 Megapixels (4512 x 3008)
S1 (Small 1): Approx. 7.6 Megapixels (3376 x 2256)
S2 (Small 2): Approx. 2.5 Megapixels (1920 x 1280)
S3 (Small 3): Approx. 0.35 Megapixels (720 x 480)
Crop – 1.6x
Large: Approx. 19.6 Megapixels (5424 x 3616)
Medium 1: Approx. 15.4 Megapixels (4800 x 3200)
Medium 2: Approx. 8.7 Megapixels (3616 x 2408)
S1 (Small 1): Approx. 4.9 Megapixels (2704 x 1808)
S2 (Small 2): Approx. 2.50 Megapixels (1920 x 1280)
S3 (Small 3): Approx. 0.35 Megapixels (720 x 480)
Patrick: Thanks for your reply. Here is what you wrote, which I interpreted as meaning you’d have the ability to use the 5DS-R in crop mode: “In my opinion, while the image quality of the 5DS R is certainly a bonus over the 7DII, the great advantage for my photography comes from the ability to have the equivalent of a 1.6 ratio zoom on the 600 II (or any lens) as compared to the 7DII. When I use the 600 with a 1.4 teleconverter, I will have an 840-1344 zoom versus the 7DII where I have only 1344. This will in fact greatly aid flight shooting; I can work with the 600 alone, be able to acquire AF on the subject easier due to the full frame view at 600mm, and then crop down to the equivalent 960mm of the 7DII (or to any effective focal length in between 960 and 600. All that while getting better image quality than we have ever had before. And when I can get closer to the birds image quality will be off the charts.” I’m sorry if I misinterpreted it, but perhaps you can see why I did. David
I said it couldn’t be a comparison between these bodies, because the pixel site and density is the same. but damn, i totally forgot about the low pass anti-aliasing filter!
Artie i must say that having just bought a 7D mkII this article filled me with a little panic and mild depression – until I remembered that 5Ds is coming in at $3800 v the 7D mkII at $1500. Much more bang for buck! And for us impoverished South Africans t hats important.
Artie: I want to get to this from you: “But when I crop my image so that the bird is the same size in the frame as your bird, my image will be sharper with more pleasing contrast and be of overall higher quality. And those differences will be just as dramatic as they were in Patrick’s tests.” And to Patrick’s “additional thoughts.” First you. When you make that crop, you’ll in effect be using a sensor that’s exactly the same size and with exactly the same pixel density as the 7D2’s sensor. You will have lost the huge image-quality (IQ) advantage that the 5Ds-R’s bigger sensor provides. So where could the cropped 5DS-R’s IQ advantage come from? One obvious place is the absence of the low-pass filter. (If you did this with a 5DS, there’s be no such advantage.) A second place is advantages in sensor technology. And finally, number 3, there might be an advantage in processor technology. I am skeptical that advantage #2 would be large. And advantage # 3 probably doesn’t exist because both cameras use dual DIGIC 6 processors. This is why I asked for you or Patrick to actually do this test.
I don’t understand Patrick’s “additional thoughts.” As far as I know–please correct me if I’m wrong–you don’t get any advantage by going to crop mode in the 5DS-R; no increase in frame rate, no increase in buffer, no improvement in AF speed or accuracy, no nothing. Well, maybe smaller file sizes, but c’mon, you bought a 5DS-R and so you’ve made the commitment to those file sizes. And when you crop on the computer, your file gets smaller anyway. But you are throwing away 30 MP and all that extra light. The “reach” advantage, if the above is true, doesn’t exist; you’d be at least as well off doing the cropping on your computer as in the camera, and probably better off. As far as I can tell, this is equivalent to what compact cameras used to call “digital zoom.” I just don’t get why Canon did this, unless it was to copy Nikon, which also did this and also gave no advantage from the crop mode.
David, I am confused by many things that you wrote.
Let’s try to pick them off one at a time. It seems that you are saying that I wrote this:
“When you make that crop, you’ll in effect be using a sensor that’s exactly the same size and with exactly the same pixel density as the 7D2’s sensor. You will have lost the huge image-quality (IQ) advantage that the 5Ds-R’s bigger sensor provides. So where could the cropped 5DS-R’s IQ advantage come from?”
If I am correct, please let me know exactly where I (or even patrick) said that.
thanks, a
Furthermore neither I nor Patrick said anything about using the 5DS R in “crop mode.” As far as I know, there is no such thing….
a
The crop mode people are talking about might only be on Nikon. It began with Nikon’s FX cameras having a “DX mode”, and now all of Nikon’s newer cameras, DX included, have crop modes. The crop mode decreases the amount of pixels captured and in some cases allows the camera to shoot at a higher frame rate.
Lyle,
Understood and correct as far as I know 🙂 So thanks!
a
Hi, Artie. I put what you wrote in quotes. It’s under “Simplifying Things.” Here it is again. “I am using a 5DS R and you are using a 7D II. Your bird will of course be larger in the frame. But when I crop my image so that the bird is the same size in the frame as your bird, my image will be sharper with more pleasing contrast and be of overall higher quality. And those differences will be just as dramatic as they were in Patrick’s tests.” All the rest is from me, including the material you put in quotes in your recent reply to mine. I am arguing that I don’t think the difference in this case can be as dramatic as the difference that Patrick found in his tests, and I’m asking for this additional test to be done.
David
Artie:
Very interesting and helpful info seeing as I shoot with both the 7D MK II and 5DS R and at times have wondered which made the most sense in the sharpness dept. This pretty much gives me the answer.
Can you provide a link so we can download the CR2 files and check for ourselves? Thanks!
Are you talking about the dollar bill images?
Yes.
We will be offering the two Raw files for $5 each for those who would like to partake. Formal announcement on Monday. a
Some clarifications requested here –
1. Were the exposures identical? Iso, Aperture and shutter speed?
2. You said that “the images were processed in DPP 4, and all of the settings were left at zero except for setting Sharpness at 3, and checking the boxes for Peripheral Illumination and Chromatic Aberration correction”. DPP 3 had a nasty habit of applying some processing to the raw files regardless of setting everything to zero unless an additional step of setting “viewing and saving raw images” in “general settings” was also set to high speed. What I am saying is that could it be that DPP 4 applies different processing to the images depending on which camera was used to create that image and that leads to some differences in the IQ of both images?
3. “After the identical RAW conversions in DDP 4 the TIFFs were imported into Lightroom where I used the export utility to create a JPEG” – there is another variable here. Could Lightroom too process the two images differently while generating the jpeg?
A very interesting experiment would be as follows –
1. Crop the 5DSR raw file to 5472 x 3648 pixels (which is also the size of the 7D II files).
2. Replace the exif tags within the 5DSR file with those of a 7D II image file. Exiftool can do this. Now this cropped and modified 5DSR file will look exactly like a 7D II file to any program that tries to process it.
3. Process this modified 5DSR file and the original 7D II file in DPP 4 and generate the two jpeg images directly from within DPP 4.
This will give you the best comparison of the two sensors.
The same thing can also be done by upscaling very slightly the 7D II images to 5,760 x 3840 pixels (which is the size of the JPG MEDIUM 2 (22 MP) files on the 5DSR) and then replacing the exif info within the 7D file with that of the 5DSR file. Now you can process both the modified 7D file and a cropped (5,760 x 3840 pixels) 5DSR file, both appearing as files coming from the 5DSR to DPP.
#1: Yes.
#2: No.
#3: Surely not but I know nothing about LR.
re: the interesting experiment: DPP does not make adjustments based on the camera unless you check the box for Digital Lens Optimizer.
artie
Thanks for the useful comparison, and also thanks to Rudy for mentioning the Digital Lens Optimizer. This is something I had somehow overlooked. I have started to read up on the DLO, and noticed that (in the Canon article I read… which was not dated) many cameras can use this feature, but the 6D is not on the list. Does anyone know if this is an oversight or if the 6D simply can’t work with it?
Just when I thought I had decided what camera to purchase, you give us this wonderful curve ball. I have a 1D mk II. I have realized, I need a new camera. I would love to get a 1Dx or its replacement. However, this is out of my budget. At this point my brain is over saturated. I think I will get a used 1D mk IV. This would let me buy great new lenses. The 100-400 v2 sounds great.
Thank you for getting Rudy Winston’s explanation.
He gives the effective absence of the low pass filter on the 5DS R credit for the advantage in sharpness and tends to confirm what I was saying in an earlier blog about the innate quality of the 7DII and 5DS R being different.
I think the only other test we would all love to see is the 5DS vs the 5DS R, of course. My guess is that the sharpness of the 5DS would be similar to the 7DII, but that the contrast and noise control would be more like the 5DS R, although my ‘gut’ tells me that sharpness and contrast are related to some degree.
I wonder if you have tried any of the techniques Rudy mentioned for DPP to mitigate the sharpness difference. I would love to see some results of that here or maybe these techniques are already in your DPP4 guide or if not, maybe you could put out an addendum.
Also, have you ever noticed any Moire or other aliasing effects in any of your 5DS R shots? If not, I have to wonder about why Canon (and everyone else) has been so insistent on having the filters to date.
Hi Steve, Thanks for your comment. And YAW on Rudy. Yeah, me too on comparing the 5DS R and the 5DS… I have tried with Lens Optimizer and my recollection is that there is always some sort of snag. I will, however, try again soon.
I have not seen anything resembling moire. I read one post where it took a guy days to come up with one 1/2 decent example of moire. He had to shoot a building with a wide angle lens from far away… So I am not too worried about it ;0
later and love, artie
Based on your experience with the 5DS R, I can’t wait to see what the come up with for the 1DX replacemet!!
The top 100 collection is almost too overwhelming to absorb at once. I will be looking at it many, many times to be able to drink it all in. I will say I never really appreciated blurs, but Cottonwood Morning Wave is one of my favorite images. The colors and the composition are superb. There is a real feeling of being present at a live event that a sharp photo would not have conveyed, so maybe you won me over. You give a 100 great examples of something for us to strive for. I especially appreciated that you annotated every image to get a sense of what it took to capture it. Brilliant. Congratulations and thanks for sharing and making it so affordable to own this collection of some of the world’s greatest bird images.
Mike. Many thanks for your more than kind words. artie
Thank you for this comparison. I just bought a 7D II and have been very happy with it so far. I’ve always been a full frame shooter for my portrait work, but went to the 7D for the crop factor thinking it would be more advantageous for wildlife work, which I’ve been doing more and more off lately. Seeing this, I’m thinking I could have the best of both worlds all in one body! Thanks for the post!
You are welcome. For portrait work I would not hesitate for one second to get a 5DS R. If you do, please use my affiliate link 🙂
many thanks, a
The problem I see in this test is that it compares one camera with a low pass filter against one without such filter using the same processing. Of course the camera without the filter looks better with the same processing. The low pass filter blurs the image so of course it won’t look as sharp or have the contrast of the non filtered (non blurred) image. The filter requires different processing so applying the same processing results in misleading results. The big question is whether there is a significant difference in results when processing is applied that is optimal to each camera. Personally, I think both images will be fine at low ISOs. It is the high ISO comparison I would like to see. David explained the situation very well previously but I would like to see a test before I lay out the cash for a 5DSR.
We said that above :). I would prefer to start with the sharpest possible, highest quality image that shows good contrast and maximum fine detail, both of which are difficult to maximize in Photoshop.
As I said additionally, zillions of great images have been made and will continue to be made with previous camera bodies, Canon and otherwise.
a
ps: I will surely be doing some high ISO shooting in Japan.
Hey, Bill Richardson, thank you for your kind words. If I might be so bold, may I suggest that YOU test the 5DS-R yourself? We all know that Artie can extract amazing images from almost any camera/lens combination. The question for the rest of us is whether WE can. For my own part, I have no doubt that the 5DS-R can produce better images under any circumstances than the 7D2. Artie’s tests and comments convince me completely. The question for me is whether the sacrifice in speed, “reach,” file size, and cost is worth the extra image quality. Only I can answer that for myself; only you can answer that for you; and I think a hands-on test would be very helpful. Thanks again to Artie for all you do here, and good luck to you, Bill Richardson.
Artie: to quote my favorite bird photographer, I agree with everything in today’s blog post, but only 100%. 🙂 I became convinced in researching and preparing my recent responses to Bill Richardson and others that sensor size is a huge factor in image quality, and I wonder what you’d see if you compared the 5DS-R against a 1DX or 5D3 or 6D (all “full frame” sensors around 20 mp). Am I right that you feel the 5DS-R blows the 1DX away in terms of image quality as well? As I’ve said before, my 20 MP 6D has much better image quality than my 20 MP 7D2, but both are capable of much better than my skill allows.
Though we have not done the dollar bill testing of the 1D X against the 5DS R we would expect similar results. a
David: I hope you are not interpreting Rudy’s explanation as supporting the position that sensor size and not pixel size is what affects noise control. His explanation actually hasn’t addressed the issue of noise at all except indirectly when he says that while the sensors have very similar pixel sizes, they are different “at the pixel level”. There is no mention of difference in the size of the overall sensor.
Considering two sensors with the same pixels (size and light to electrical signal conversion consistency), the larger one with more pixels obviously collects more light, but this is ONLY because of the extra pixels. Each individual pixel does not collect more light and the extra pixels outside the core pixels (pixel set equivalent to the smaller sensor) do not share their light with these core pixels. Sharing this light between pixels would be like adding a really drastic low pass filter to the sensor or implementing an averaging process in the software. The effect of either of these would be to severely reduce sharpness and contrast and essentially blur the image.
You really need to understand that the ONLY reason a larger sensor with the same MP’s as a smaller one is less noisy (again assuming the same pixel to pixel size and conversion consistency) is because the larger sensor has larger individual pixels and each individual pixel collects more light. It has nothing to do with the total amount of light collected by the overall sensor.
One way to think of this is that each pixel’s output signal is effectively an average of the light collected over its area for the exposure time. In this way each individual pixel acts like a tiny low pass filter and a larger pixel with more area averages more light which results in a larger absolute signal and more of a filtering effect. Both of these effects tend to reduce the noise.
If there are two sensors with the same pixel size and one is larger because it has more pixels, then the noise control will be the same unless there is some inherent difference in the quality of the signal conversion (light to electrical) and electronic amplification processes. I believe that this is the case with the 5DS R to 7DII comparison. The 5DS R sensor is one generation improved in basic pixel technology from the 7DII and this is why it does better with higher ISO settings.
I also think that Artie is correct when he suggests that a 5DS R – 1DX comparison would show that the 5DS R is sharper. I am not sure about high ISO performance as this will depend on how much the 5DS sensor technology has improved over the 1DX sensor.
Steve Soderling: We have had this discussion previously, and I am sure that you are mistaken; sensor size is more important than pixel size for noise control. I don’t want to clutter another one of Artie’s wonderful educational blog posts with our discussion, but I’d be happy to continue with you privately. davidpolicansky@gmail.com
Thank you for this! Very helpful.
Many thanks Brendan. It’s my job. I just wish that I had not had my head buried in the sand for so long 🙂 a
Interesting! I always assumed that the image quality should be identical because both sensors have roughly the same pixel size, the 7DII is a “pro” grade APS-C and they were released only a few months from each other. But apparently there is more to it… Unfortunately the 5DSR is not for me (indoor sports) because I need high ISO capability.
Things are not always as we assume. As I stated, it might be the cancellation of the low pass filter. In fact, I should have mentioned that that is the most likely cause…
I do not understand why you say, “the 5DSR is not for me (indoor sports) because I need high ISO capability.” What camera are you using?
I have been quite happy with the high ISO noise control with the 5DS R.
a