-
UPDATE
The forum will be moving to a new domain in the near future (canonrumorsforum.com). I have turned off "read-only", but I will only leave the two forum nodes you see active for the time being. I don't know at this time how quickly the change will happen, but that will move at a good pace I am sure. ------------------------------------------------------------
Any thing Shot with a 5ds/r
- Thread starter gary samples
- Start date
Great shots Chuck!
Some are concerned with how fast the shutter speed need to be, to give sharp images. I don´t believe I have a particularly steady hand, so if I can do it, most of you can. I also read that this is not the camera for bird photography. I agree that high fps, action oriented shooting is not what it does best, but for most of what I do (as long as I don´t push beyond ISO6400, it works well.
This was shot handheld, with the 600mm f4L IS II at 1/160s, f6.3, ISO400. I´ll add a crop of its head below. It is not tripod sharp, but on the other hand, the bird was swimming unsteady, so there is some movement also.
Some are concerned with how fast the shutter speed need to be, to give sharp images. I don´t believe I have a particularly steady hand, so if I can do it, most of you can. I also read that this is not the camera for bird photography. I agree that high fps, action oriented shooting is not what it does best, but for most of what I do (as long as I don´t push beyond ISO6400, it works well.
This was shot handheld, with the 600mm f4L IS II at 1/160s, f6.3, ISO400. I´ll add a crop of its head below. It is not tripod sharp, but on the other hand, the bird was swimming unsteady, so there is some movement also.
Upvote
0
Dylan777 said:
TY!
the church was all natural light no flash --- the outdoor shots, first 2 natural light, the last one had one off camera light (camera right)
And the reception had 2 off cam lights - and the second one btw has zero post processing (857)...
Upvote
0
Great to see one of your images again Edward. Even those without teeth and claws look great 
Your first selfie looks great Chuck. Guess you had a long selfie stick
I have not tested the 5DSR with the Zeiss lenses yet. I´m missing an S-screen for focusing and I hate live view. But today I start on a two-week trip up the northwest coast of Norway and, weather permitting (all the water evaporating in the south seems to land in our heads ...), I will get some good opportunities there.
Your first selfie looks great Chuck. Guess you had a long selfie stick
I have not tested the 5DSR with the Zeiss lenses yet. I´m missing an S-screen for focusing and I hate live view. But today I start on a two-week trip up the northwest coast of Norway and, weather permitting (all the water evaporating in the south seems to land in our heads ...), I will get some good opportunities there.
Upvote
0
Here´s a pano of the breakfast table view from Crete. Below you have the dropbox link, which I believe should give you access to the full size file. If you want resolution, I guess this is it
70-200 f2.8L IS II @70mm (don´t remember the number of shots, but believe it is 5), f6.3, 1/200s, ISO100.
https://dl.dropboxusercontent.com/u/109462200/_23A0108-Pano-2-2.jpg
70-200 f2.8L IS II @70mm (don´t remember the number of shots, but believe it is 5), f6.3, 1/200s, ISO100.
https://dl.dropboxusercontent.com/u/109462200/_23A0108-Pano-2-2.jpg
Attachments
Upvote
0
First shot with the 24 TS-E f3.5L II
This is one of the more famous walls in the world, called Trollveggen (The Troll Wall). A number of base jumpers have ended their lives in that wall. It is always difficult to give a full impression of how big it is, but longest straight vertical is 1.000 meters, with a 50 overhang, quite scary ...
1/80s, f8.0, ISO100, full shift.
This is one of the more famous walls in the world, called Trollveggen (The Troll Wall). A number of base jumpers have ended their lives in that wall. It is always difficult to give a full impression of how big it is, but longest straight vertical is 1.000 meters, with a 50 overhang, quite scary ...
1/80s, f8.0, ISO100, full shift.
Upvote
0
eml58 said:
Some interesting physical geography there ! I think that is the sort of picture that would really come into its own on a 2 m wide canvas !
3 frames ? So landscape orientation I presume. You've shot it with the highest resolution FF camera, with the highest available resolving lens and made a format that is larger than DMF: so assuming your technique is on the money (which I know it is !) that picture must be the very highest possible IQ that we can currently achieve ! I'd love to see the full size image.
Upvote
0
Eldar, thank you for posting this panorama in full size. I always wondered if this 5DsR is really as good as some people claim. The level of details and sharpness is!
I shoot mainly landscapes and hate the noise of the 5D II, especially in the sky. No matter which ISO I choose, there is always some noise. Often I use grad ND filters and later one need to brighten up parts of the sky by a 1/3 or even 1 stop. Then noise can become a problem for a pixel peeper....
Now I saw that your pano was shot at ISO 100. So I took a crop to compare it with a shot from my 5D II. For this purpose I converted with Canon DPP one of my Raw files into a JPEG (just a conversion, nothing else done in DPP). I have no full size JPEGs out of came available and it's late at night. So a Raw to JPEG conversion must do. Attached is the comparison of the sky at 100% view.
BTW Eldar, a 242.5 MP pano is very impressiv. I guess you can easily print it several meters wide without noticing any noise at all. The details and sharpness of this pano are amazing. With Photoshop I reduced the size from 30548 to just 12000 pixel long (with auto rendering applied). With 37 MP it still has amazing details and sharpness and now the noise is comparable to my 5D II. So your picture was very useful for me to make some tests. Thank you.
I shoot mainly landscapes and hate the noise of the 5D II, especially in the sky. No matter which ISO I choose, there is always some noise. Often I use grad ND filters and later one need to brighten up parts of the sky by a 1/3 or even 1 stop. Then noise can become a problem for a pixel peeper....
Now I saw that your pano was shot at ISO 100. So I took a crop to compare it with a shot from my 5D II. For this purpose I converted with Canon DPP one of my Raw files into a JPEG (just a conversion, nothing else done in DPP). I have no full size JPEGs out of came available and it's late at night. So a Raw to JPEG conversion must do. Attached is the comparison of the sky at 100% view.
BTW Eldar, a 242.5 MP pano is very impressiv. I guess you can easily print it several meters wide without noticing any noise at all. The details and sharpness of this pano are amazing. With Photoshop I reduced the size from 30548 to just 12000 pixel long (with auto rendering applied). With 37 MP it still has amazing details and sharpness and now the noise is comparable to my 5D II. So your picture was very useful for me to make some tests. Thank you.
Attachments
Upvote
0
Robert, I´m happy you liked the pano. It is a very fast job in post, but quite impressive from a resolution perspective. But I´m not sure what you´ve done, even thogh I´m sure it has a purpose. I ´ll leave it to the experts to comment. All I can say is that the 5DSR has exceeded my expectations by a significant margin. And if your post was to compare the 5DSR performance to that of the 5DII and also to indicate that the 5DII was cleaner ... That Is Wrong. The 5DII is not even up to the socks of the 5DSR.RobertG. said:
Upvote
0
Eldar, thank you for your reply. I just want to see how the 5dsR performs. I'm looking for a new camera body and I'm torn between the 5DsR and the Sony A7r II. I shoot panos and so your pano in full size seemed to be a good way to see what is possible with the 5DsR. After your reply I was looking for 5DsR images at flickr. I found one with a bright blue sky at ISO 100 and downloaded the full size version. It looks much cleaner at 100% view than yours. It actually is very clean but I don't know how much post processing was done with it. I can only guess that the software used to stitch the pano introduced some noise. I know that Color Autopano does.
I make sometimes pretty large prints and lack of resolution as well as noise is a problem for me with the 5D II. I shoot a lot of panoramas but in some situations it is not possible. So a new cam with huge clean images is what I'm looking for. I'm impressed by the PENTAX 645Z but can't effort it. I have a lot of Canon lenses (which I like a lot, especially the TS-E lenses) and so I would like to get a cam which performs at least as good as the 5D II with the Canon lenses but offers much larger and better files. I hope that explains my point.
I make sometimes pretty large prints and lack of resolution as well as noise is a problem for me with the 5D II. I shoot a lot of panoramas but in some situations it is not possible. So a new cam with huge clean images is what I'm looking for. I'm impressed by the PENTAX 645Z but can't effort it. I have a lot of Canon lenses (which I like a lot, especially the TS-E lenses) and so I would like to get a cam which performs at least as good as the 5D II with the Canon lenses but offers much larger and better files. I hope that explains my point.
Upvote
0
RobertG. said:
Are you normalizing the images before comparing crops? At native size, the 5Ds has considerably smaller pixels. You need to first normalize the images (downsample the 5Ds image to the 5D II size), then crop and compare. While I am not sure that will actually result in the 5Ds being cleaner than the 5D II, it should certainly reduce the amount of noise your seeing in comparison, possibly make the two images look the same (as far as noise goes). We can actually calculate what the results might be.
The 5D II is gathering more light per pixel, so the signal is stronger in each pixel, and the noise is in relative terms less (noise = SQRT(signal)). If the 5Ds has 1000e- signal in a sky pixel, it's noise is (ignoring RN for the moment) 31.63e-. That is an SNR of 31.63:1 (again, ignoring read noise). Now, if the 5D II has 4000e- signal in a sky pixel, it's noise is 63.25e-. Seems like it's higher...the key is, that's an SNR of 63.25:1. The RATIO is larger with the 5D II, so the image looks cleaner. Downsampling the 5Ds image will average pixel data together. That will increase signal strength, thus increasing SNR, and the differences should drop.
Now, there is read noise to contend with in the real world. The 5D II has 30e- RN at ISO 100, and an FWC of 61072e-. Assuming the blue sky is somewhere around a mitdone gray level, the signal strength of a blue sky pixel in the 5D II would be 30536e-. The SNR would be 30536/SQRT(30536 + 30^2), or 172.23e-:1, which in terms of decibels is ~44dB (out of a maximum of 47.8dB). The 5Ds on the other hand probably has (using 7D II data for now, since we don't have Sensorgen.info for the 5Ds yet and that the 5Ds non-normalized SNR is very close to the 7D II non-normalized SNR), 13e- RN at ISO 100, and an FWC of 29544e-. An 18% midtone blue sky is going to have a signal strength of 14772e-. So SNR would be 14772/SQRT(14772e + 13^2), or 120.85:1, ~41.5dB.
The 5Ds is newer technology than the 5D II. We should hope that, at the very least, Canon has improved the sensor technology between the 5D II and now. I suspect at the very least there will be an increase in Q.E. and a reduction in dark current (which is generally negligible these days anyway, so I'm not accounting for it here.) The 5Ds should actually be gathering more light PER UNIT AREA than the 5D II, so it has the potential to produce better normalized images. I can't say that for sure, and Canon's noise characteristics are usually less than ideal, so only an actual test will say for sure. However, theoretically...
Once you downsample, you should see an improvement, a solid improvement. The ratio of pixel areas between these two cameras is 2.44x (6.4^2/4.1^2). So, if we do: (14772 * 2.44)/SQRT(2.44 * (14772 + 13^2)), we get an SNR closer to what a downsampled image would have. In this case, it's 188.77:1, or 45.5dB. That is actually better than the 5D II! Well, it would be, if this was actually using 5Ds statistics...I substituted the 7D II statistics, so, I can't say for sure. It is marginally better, we aren't even talking a third of a stop here. Plus, how much the data improves often depends on the scaling algorithm, and there are a lot of those out there (Photoshop will just do a basic BiCubic).
Anyway...the increase in noise with the 5Ds is probably because you are comparing crops of sky at native size. Normalize the images, then crop, and the results may actually tilt in favor of the 5Ds.
Upvote
0
I didn't un-box the 5DS until late night, so the very first shot is in the home office.
The lens is my old 24-105L. so not the best super sharp lens. I opening the shot in LR6 and set the sharpen / noise to zero. I used a tripod, live view & remote release. I was pleased with the overall quality and resolution.
5DS test 1 camera display © Keith Breazeal by Keith Breazeal, on Flickr
Screen shot of the crop
5DS testing Camera display zoomed © Keith Breazeal by Keith Breazeal, on Flickr
The lens is my old 24-105L. so not the best super sharp lens. I opening the shot in LR6 and set the sharpen / noise to zero. I used a tripod, live view & remote release. I was pleased with the overall quality and resolution.
5DS test 1 camera display © Keith Breazeal by Keith Breazeal, on FlickrScreen shot of the crop
5DS testing Camera display zoomed © Keith Breazeal by Keith Breazeal, on Flickr
Upvote
0
Jon, thank you for taking the time for this detailed explanation. I really appreciate it. I compared both images at 100% view. I understood from your explanation that there should be comparable or slightly less noise when I downsize the 5Ds' image to the size of the image of a 5d II. But why should I do this? I need larger files straight out of the cam because stitching is not always possible (and makes things more complicated, takes extra time, causes additional noise etc.). I need larger files for larger prints or larger crops from these files. So larger files with less noise is what I'm looking for. I don't need the best autofocus and 5 fps. I shoot landscapes, most of the time at ISO 100 till 400. I use manual focus lenses for the majority of my important shots. I just need large clean files, which can be lifted in post by at least 1 stop without increasing noise too much. I hope that the 5Ds offers them but I'm not conviced yet. The sharpness and details are amazing, I have to admit.
BTW, I thought the purpose of these "Any thing shot with ..." threads is to show how capable this particular camera or lens is. Correct me if I'm wrong.
BTW, I thought the purpose of these "Any thing shot with ..." threads is to show how capable this particular camera or lens is. Correct me if I'm wrong.
Upvote
0
