p.5 #1 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
Using the same pixels/sensor from 7D2 to estimate 5DS performance at 12.44 ... Ok. Forgetting one thing. Tuned for Low ISO which according to northlight has produced a sensor "operating at significantly lower temperatures"
Low ISO tuning means Less heat in the signal path which means more operational efficiency and thereby less noise. Less noise means more usable DR making it to the ADCs. Something very different from both the 5D3, the 1DX, the 6D, and the 7D2. Never been done like this before. This is where the 1.5 to 2 stops are coming from I'm betting.
But yes, you're math is good. I'm just suggesting the variables are perhaps other than what you pluged in
p.5 #2 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
pureclassa wrote:
Using the same pixels/sensor from 7D2 to estimate 5DS performance at 12.44 ... Ok. Forgetting one thing. Tuned for Low ISO which according to northlight has produced a sensor "operating at significantly lower temperatures"
Low ISO tuning means Less heat in the signal path which means more operational efficiency and thereby less noise. Less noise means more usable DR making it to the ADCs. This is where the 1.5 to 2 stops are coming from I'm betting.
I would trust Chuck Westfall's statements about the 5Ds's DR over that report from NL.
p.5 #5 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
alundeb wrote:
ISO 12232 covers the so-called ISO-ratings and
ISO 15739 covers noise measurements including dynamic range
But I am ALL in for the caveats and to eat the pudding
Thanks for the heads up on ISO standards. Do DxO test to ISO 15739:2013 or any of the linked/named sensor scores?
I note DxO state
'Pre-cooked" RAW?
Certain manufacturers embed a small part of the processing directly in the sensor, which means that some degree of processing occurs before the RAW image is sent to the RAW converter. In this case, measurements for these "pre-cooked" RAW images can be biased by this processing.
To avoid any potential impact on our measurements, DxOMark always tests all cameras to detect any pre-processing of RAW images. A processed or pre-cooked RAW image has different characteristics from a genuinely unprocessed image. To some extent, these characteristics enable us to walk back the processing and reconstruct the original image to perform unbiased measurements, and we always inform the user about models with embedded pre-processing.
So, it is just as well DxO can detect any baking of RAWs and compensate to some extent for the manipulation of curves and noise reduction in RAW capture, otherwise some sensors would score higher marks
p.5 #7 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
alundeb wrote:
I am not sure about what data you refer to here. sensorgen.info?
My mistake! I was referring to Senscore data here. Specifically the DR scores for FF cameras. Reading their FAQ they use some computation that covers the entire ISO range so a 1DX benefits from its high ISO performance relative to say D800's performance across all ISOs. As I said I don't know how to interpret this data yet.
And thanks Snapsy for the reference in the other thread. Too many pages in too many threads for me to keep up with them all! So I think I'm correct that read noise and shadow noise are related. About as far as I'll theorize at this point.
p.5 #8 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
Jon Tainton wrote:
Do DxO test to ISO 15739:2013 or any of the linked/named sensor scores?
I havent seen DxO referring directly to that ISO standard, but they do indeed test dynamic range according to the definitions in ISO 15739:2013, unlike many other online sources.
The key elements in that definition is that they test on raw data, and they define dynamic range as the ratio of the highest luminance to the luminance where the signal to noise ratio is 1:1. That is a little different from clarkvision.com and others who use the ratio of the highest signal to the read noise. The latter method uses read noise with zero luminance. That gives a little bit lower floor than with SNR 1:1, because when you still have some luminance you also have som photon shot noise in addition to the read noise.
p.5 #10 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
alundeb wrote:
The story that the DxO data doesn't tell, is the strength of the AA filter or other sources of blur in the sensor filter stack. Now that the 7DII is out, everybody agrees that the 7D was not as "sharp" at the pixel level as the 7DII. When normalizing images for accurate noise evaluation, this is important to keep in mind. It is also important that the images are normalized to the same level of visual detail, otherwise you are comparing two parameters at a time.
Thanks Anders. When evaluating my cameras I'm normalizing to the way a full image looks on my 27" iMac. The 7D didn't fare as well as its 40D predecessor since pixel level sharpness differences weren't normalized out. Pixel level resolution is important in a wildlife camera. I've found pixel level sharpness to be pretty consistent with the 40D and all three generations of my 5Ds. I'm hoping a 7D2 would be as well.
p.5 #11 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
snapsy wrote:
I would trust Chuck Westfall's statements about the 5Ds's DR over that report from NL.
I also trust Chuck Westfall, but all he says is that Canon tells us that the dynamic range is "Equivalent to the 5DIII". There was no mention of low ISO or anything like that in the question he got, nor in his answer. So I trust him but I don't know what he told me
p.5 #12 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
snapsy wrote:
Hopefully an ISO 100 raw image will leak out before then. It'll only take one to measure the read noise/DR.
More speculation from me.
I'm guessing the Canon NDAs in place for the prototype photographers are quite explicit about file types and sizes, as well as only using Canon glass for now. The 8 "Official" jpegs on Flickr are the only full resolution files I have found so far. I would hope for a planned / scheduled gradual release of RAW image files as well as a relaxation of the restrictions on lens brands, based on firmware / hardware finalization in the coming months. Perhaps not. Sony and Nikon have yet to make their 50mp moves, and those rumors / releases would probably figure into earlier release of more data points. I think Canon is wise to take their time and not rush to production, and a consequence of that reluctance is a limited information release.
I am also in general agreement with those who say Canon are downplaying performance so as not to increase expectations amid an already highly skeptical and even hostile installed user base.
I for one, am already sold, whatever the DR turns out to be.
50mp for 4k! The downward pricing pressure on other 50mp makers could increase over the next year by a fair amount. The 5Ds R - 11-24 F/4 combination is just more than 1/2 of the least expensive 50mp CMOS MFDB.
And now a Sigma ART F/1.4 24mm.
An exciting year for everyone. Some may be not so happy, but still quite exited.
p.5 #13 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
So multiple professionals who Canon has entrusted pre-production models of the 5DS with to field test are automatically all dead wrong in saying the sensor operates at significantly lower temps, has a CFA much more akin to the 1Ds3, and appears to have 1.5 to 2 stops more usable DR because the noise has been greatly reduced .... because of Chuck Westfall!! Why is Westfall so infallible in your book? The 2 sensors DO have similar DR, but apparently the 5DS is making far more efficient use of it.
p.5 #14 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
pureclassa wrote:
So multiple professionals who Canon has entrusted pre-production models of the 5DS with to field test are automatically all dead wrong in saying the sensor operates at significantly lower temps, has a CFA much more akin to the 1Ds3, and appears to have 1.5 to 2 stops more usable DR because the noise has been greatly reduced .... because of Chuck Westfall!! Why is Westfall so infallible in your book? The 2 sensors DO have similar DR, but apparently the 5DS is making far more efficient use of it.
Because Check Westfall is Canon's technical adviser for the USA and has access to Canon engineers and technical documentation. Are you really saying you trust anonymous internet reports over what Canon's USA technical lead is saying?
p.5 #15 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
alundeb wrote:
I also trust Chuck Westfall, but all he says is that Canon tells us that the dynamic range is "Equivalent to the 5DIII". There was no mention of low ISO or anything like that in the question he got, nor in his answer. So I trust him but I don't know what he told me
+1
After looking at the tables @ sensorgen.info ... it becomes rather clear to me that the variations on how to approach "optimized" and "sweet spot" for noise vs. ISO vs. DR is not exactly consistent (even within a given mfr) across diff cameras. Some seem to try and hold steady in a tighter band, while others seem willing to incur sacrifice in order to tradeoff / shift some low for high or vice verse.
Trying to reduce it to a single answer ... starting to seem a bit folly.
p.5 #16 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
alundeb wrote:
I also trust Chuck Westfall, but all he says is that Canon tells us that the dynamic range is "Equivalent to the 5DIII". There was no mention of low ISO or anything like that in the question he got, nor in his answer. So I trust him but I don't know what he told me
It's a low ISO camera studio/landscape camera. Why would Chuck reference High ISO DR when asked about DR? Esp when the camera max ISO rating was dropped to 6400, which most likely occurred due to a reduction in High ISO DR from the extra pixels, same as it does for the D800/D810. I think everyone is overcomplicating this.
p.5 #17 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
pureclassa wrote:
So multiple professionals who Canon has entrusted pre-production models of the 5DS with to field test are automatically all dead wrong in saying the sensor operates at significantly lower temps, has a CFA much more akin to the 1Ds3, and appears to have 1.5 to 2 stops more usable DR because the noise has been greatly reduced .... because of Chuck Westfall!! Why is Westfall so infallible in your book? The 2 sensors DO have similar DR, but apparently the 5DS is making far more efficient use of it.
Avoid binary thinking. It leads you into blind alleys.
It shows up in exaggerated terms like "dead wrong" and "infallible".
The truth is in between.
The experts are seeing things that are hard to measure accurately outside of a laboratory, but have subjective feel to them that can be real and desirable.
Westfall must choose his words carefully because of his official position and may be speaking conservatively because of that.
p.5 #20 · 5DSR Low ISO DR 1.5-2 stops better than 1DX ...
AndereObjektiv wrote:
More speculation from me.
I'm guessing the Canon NDAs in place for the prototype photographers are quite explicit about file types and sizes, as well as only using Canon glass for now. The 8 "Official" jpegs on Flickr are the only full resolution files I have found so far. I would hope for a planned / scheduled gradual release of RAW image files as well as a relaxation of the restrictions on lens brands, based on firmware / hardware finalization in the coming months. Perhaps not. Sony and Nikon have yet to make their 50mp moves, and those rumors / releases would probably figure into earlier release of more data points. I think Canon is wise to take their time and not rush to production, and a consequence of that reluctance is a limited information release.
I am also in general agreement with those who say Canon are downplaying performance so as not to increase expectations amid an already highly skeptical and even hostile installed user base.
I for one, am already sold, whatever the DR turns out to be.
50mp for 4k! The downward pricing pressure on other 50mp makers could increase over the next year by a fair amount. The 5Ds R - 11-24 F/4 combination is just more than 1/2 of the least expensive 50mp CMOS MFDB.
And now a Sigma ART F/1.4 24mm.
An exciting year for everyone. Some may be not so happy, but still quite exited. ...Show more →
Disregard my previous questions. I just now saw this post. Thanks Andere
!! Why is Westfall so infallible in your book? The 2 sensors DO have similar DR, but apparently the 5DS is making far more efficient use of it.