p.2 #1 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Scott Stoness wrote:
And thats not all
Interesting to look at the review of R6 - it shows that using electronic shutter (12bit) drops the ability to push shadows by about 1- 2 stops (my assessment) vs mechanical shutter (14 bit) non plus (8fps).
This suggests that R5 pushing at low iso's will be diminished significantly? when using electronic shutter or H plus.
[I suspect that this is also true of Sony and Nikon - DPReview has just not presented this previously ]
So shoot at slow fps (14bit) at 8fps if you can on R5. And don't bother with the A9 - it can't do it (14bit continous) ...Show more →
Bill Claff measured that already for the R5 (https://www.photonstophotos.net/Charts/PDR.htm). Compared to the A9, for example, DR drops significantly in ES for Canon. His A9 results showed hardly any difference.
p.2 #3 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Holger wrote:
Bill Claff measured that already for the R5 (https://www.photonstophotos.net/Charts/PDR.htm). Compared to the A9, for example, DR drops significantly in ES for Canon. His A9 results showed hardly any difference.
Thanks - Sony A9 does not drop but it was not as high as Canon was in dynamic range to drop from. I presume because its always on 12 bit whereas Canon is on 14 bit at 8fps and drops to 13 bit at 14fps and 12 bit at 20fps? In other words canon has better dynamic range than a9 if you can keep your fps to 8fps (14 bit) and <=iso800. Otherwise, at 20fps or =iso800 they are even on dynamic range. And at 14fps (13bit) <=iso800, canon is also better than a9.
p.2 #4 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
I've been shooting nearly exclusively in e-shutter, but haven't shot anything that is remotely challenging to get the "rolling banding" seen in the opening post.
On the other side of the testing, here is someone else's a LED wall test to confirm the actual read out speed based on a "normal" frequency of an LED light. I've always been curious, are these tests relatively accurate?
p.2 #5 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Holger wrote:
Bill Claff measured that already for the R5 (https://www.photonstophotos.net/Charts/PDR.htm). Compared to the A9, for example, DR drops significantly in ES for Canon. His A9 results showed hardly any difference.
Do you know if the A9 captures 14-bit at any time? I read somewhere the e-shutter is captured at 12-bit, but I can't find whether the slower modes use a 14-bit depth.
p.2 #6 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Mike Jacks0n wrote:
Do you know if the A9 captures 14-bit at any time? I read somewhere the e-shutter is captured at 12-bit, but I can't find whether the slower modes use a 14-bit depth.
The tests of a9 suggest that its always at 12 bit.eg no bump in dynamic range between electronic and mechanical.
hmmmm from the same review " * We confirmed that continuous modes were 12-bit, while single shot modes were 14-bit, by comparing histograms of respective Raw files. The 14-bit single drive files have 14-bits of data compared to the 12-bit files (the histogram shows Raw values 1, 2, and 3 are vacant while the 14-bit files do have pixels with these values)."
I guess Bill tested a9 at ms/ continuous not single or this is wrong.
p.2 #7 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Scott Stoness wrote:
Thanks - Sony A9 does not drop but it was not as high as Canon was in dynamic range to drop from. I presume because its always on 12 bit whereas Canon is on 14 bit at 8fps and drops to 13 bit at 14fps and 12 bit at 20fps? In other words canon has better dynamic range than a9 if you can keep your fps to 8fps (14 bit) and <=iso800. Otherwise, at 20fps or =iso800 they are even on dynamic range. And at 14fps (13bit) <=iso800, canon is also better than a9.
14-bit stills is not an advantage in most real-world shooting, especially with sports cameras like the a9/a9ii that are designed from scratch to perform better at higher iso levels.
"Raw bit depth is often discussed as if it improves image quality and that more is better, but that's not really the case. In fact, if your camera doesn't need greater bit depth then you'll just end up using hard drive space to record noise.
...A 14-bit Raw file won't generally give extra highlight capture, it'll mean having sufficient Raw numbers left to be able to capture detail in the shadows. And if your camera is swamped by noise before you get to 14EV (most are), all this extra data will effectively be used to record shadow noise.
see this quote from the link above: "In other words, 12-bits provides enough room to encode roughly 12 stops of dynamic range, while 14 bits gives the extra space to retain up to around 14EV."
as i understand it, i don't see where the r5 ever needs more than 12-bits in any stills situation, because it always has less than 12 stops of d.r... the hassleblad camera on the other hand, kills the r5 by a stop of better pdr, so it could benefit from 14-bit files.
p.2 #8 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
osv2 wrote:
14-bit stills is not an advantage in most real-world shooting, especially with sports cameras like the a9/a9ii that are designed from scratch to perform better at higher iso levels.
"Raw bit depth is often discussed as if it improves image quality and that more is better, but that's not really the case. In fact, if your camera doesn't need greater bit depth then you'll just end up using hard drive space to record noise.
...A 14-bit Raw file won't generally give extra highlight capture, it'll mean having sufficient Raw numbers left to be able to capture detail in the shadows. And if your camera is swamped by noise before you get to 14EV (most are), all this extra data will effectively be used to record shadow noise.
see this quote from the link above: "In other words, 12-bits provides enough room to encode roughly 12 stops of dynamic range, while 14 bits gives the extra space to retain up to around 14EV."
as i understand it, i don't see where the r5 ever needs more than 12-bits in any stills situation, because it always has less than 12 stops of d.r... the hassleblad camera on the other hand, kills the r5 by a stop of better pdr, so it could benefit from 14-bit files.
I need dynamic range most in bright conditions where iso200 or 400 is possible. Sometimes it is a dark bear with a bright sky or a white bear on dark background after dawn or significantly before dusk.
This is a real world situation for me.
The graphs showing R5 geting 12 stops at 14 bit, 13 stops at 13 bit and 11 stops at 12 bit suggests that at these higher bit depths are useful in real world situations. As compared to 11 stops at iso 100 for A9 at 12 bit (continuous).
The graphs are corroborated by the dpreview studio scene that compares e-shutter vs mechanical shutter for R6- the mechanical (14 bit) looks at least a stop better. And this suggests that for slower moving big animals, you might prefer the R5 over the a9 for events like polar bears, grizzly bears or lion safaries. Not to mention 45mpx vs 24 mpx.
Thus for me, the R5 would be more useful at 8fps for polar bears or grizzly bears or moose in higher contrast hours where low iso is possible.
I understand that the a7riv has 14 bit as long as you are not in compressed raw mode [which means like the R5 you will have reduced fps to get maximum dynamic range]. But not the a7r3 or a7r2.
p.2 #9 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Scott Stoness wrote:
Thanks - Sony A9 does not drop but it was not as high as Canon was in dynamic range to drop from. I presume because its always on 12 bit whereas Canon is on 14 bit at 8fps and drops to 13 bit at 14fps and 12 bit at 20fps? In other words canon has better dynamic range than a9 if you can keep your fps to 8fps (14 bit) and <=iso800. Otherwise, at 20fps or =iso800 they are even on dynamic range. And at 14fps (13bit) <=iso800, canon is also better than a9.
The Canon cheats by using noise reduction. This is indicated by the triangles up to iso 640. Bill Claff has mentioned that NR will give Canon 2/3 stops in PDR ( (https://www.dpreview.com/forums/thread/4508716?page=2)). Just imagine Sony doing the same.
I prefer the possibility to use e-shutter in my a9ii for almost everything. Fantastic for events. Canons implementation is not really convincing with all the limitations coming with it.
p.2 #10 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Scott Stoness wrote:
I need dynamic range most in bright conditions where iso200 or 400 is possible. Sometimes it is a dark bear with a bright sky or a white bear on dark background after dawn or significantly before dusk.
This is a real world situation for me.
The graphs showing R5 geting 12 stops at 14 bit, 13 stops at 13 bit and 11 stops at 12 bit suggests that at these higher bit depths are useful in real world situations. As compared to 11 stops at iso 100 for A9 at 12 bit (continuous).
The graphs are corroborated by the dpreview studio scene that compares e-shutter vs mechanical shutter for R6- the mechanical (14 bit) looks at least a stop better. And this suggests that for slower moving big animals, you might prefer the R5 over the a9 for events like polar bears, grizzly bears or lion safaries. Not to mention 45mpx vs 24 mpx.
Thus for me, the R5 would be more useful at 8fps for polar bears or grizzly bears or moose in higher contrast hours where low iso is possible.
I understand that the a7riv has 14 bit as long as you are not in compressed raw mode [which means like the R5 you will have reduced fps to get maximum dynamic range]. But not the a7r3 or a7r2....Show more →
Jim Kasson measured 12 bit vs. 14 bit for the Nikon D810 using the photon transfer curve. From iso 200 on you don't get ANY improvement in DR due to the noise floor (https://blog.kasson.com/the-last-word/nikon-12-bit-raw-mode/). Canon's use of noise reduction "cheats" here by boosting DR artificially.
If you only want 8fps, my A7riv is fantastic, too, giving me even 61MP. Fully sufficient for bears, lions, etc. Makes no sense buying into an additional system in my opinion if you are invested in Sony already for this.
p.2 #15 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
osv2 wrote:
...
just like the r5 tho, there is little to no advantage in real-world use, especially at the iso200-iso400 range that you are shooting at.
with canon using nr on the r5, i would also be concerned about loss of detail.
Are you saying you don't think the R5 benefits from 14 bit capture in the 200-400 range or are you saying the A9 II doesn't benefit from it and that range?
Also, any NR issues you are worried about on the R5 I don't really notice. That's not to say it doesn't exist, just to say its very well controlled and a lack of detail is not an issue that I'm finding.
just like the r5 tho, there is little to no advantage in real-world use, especially at the iso200-iso400 range that you are shooting at.
with canon using nr on the r5, i would also be concerned about loss of detail.
Interesting about the A9II vs A9. I can't recall reading about that change when the A9II came out.
It also lists it doing 14-bit in E-shutter in Uncompressed RAW.
p.2 #17 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Mike Jacks0n wrote:
Are you saying you don't think the R5 doesn't benefits from 14 bit capture in the 200-400 range or are you saying the A9 II doesn't benefit from it and that range?
Also, any NR issues you are worried about on the R5 I don't really notice. That's not to say it doesn't exist, just to say its very well controlled and a lack of detail is not an issue that I'm finding.
I don't understand the statement either - ISO 100/200/400 is used in later morning or earlier evening in the day sometimes and 14 bit is better than 12 bit (as shown in my link above) for dynamic range. That is real world usage. Dynamic range challenges are greater in day than the night when low iso's are possible.
[Now based on the link its clear to me that a9ii can do 14bit in continuous but the a9 cannot. So absent the very low mpx of a9 (and thats a big absent) its as good as the R5. But the a9 is 12 bit for continuous. So I personally would not buy either of them because I am not a 20fps shooter. Per the chart though a7riii and a7riv would be useful to me, if I did not have a bunch of big white canon's and 5DSR and a dislike for sony menu/battery, dust control, cold weather performance....]
Sep 17, 2020 at 11:38 AM
osv2 Offline • • • • [X]
p.2 #18 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Mike Jacks0n wrote:
Are you saying you don't think the R5 doesn't benefits from 14 bit capture in the 200-400 range or are you saying the A9 II doesn't benefit from it and that range?
afaik, the data that has been presented so far indicates that no ff camera has good enough d.r. vs. noise at iso200-iso400, to benefit from 14 bit files.
because dr always drops when iso goes up, see the bill claff pdr data.
Sep 17, 2020 at 12:21 PM
osv2 Offline • • • • [X]
p.2 #19 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Scott Stoness wrote:
I don't understand the statement either - ISO 100/200/400 is used in later morning or earlier evening in the day sometimes and 14 bit is better than 12 bit (as shown in my link above) for dynamic range. That is real world usage.
i haven't seen any links proving that 14 bit is better at iso200-iso400, can you re-post whatever it is that you are referring to.
p.2 #20 · Electronic Shutter Artifacts in Sony A9 and Canon 1DX3 Images
Scott Stoness wrote:
I don't understand the statement either - ISO 100/200/400 is used in later morning or earlier evening in the day sometimes and 14 bit is better than 12 bit (as shown in my link above) for dynamic range. That is real world usage. Dynamic range challenges are greater in day than the night when low iso's are possible.
[Now based on the link its clear to me that a9ii can do 14bit in continuous but the a9 cannot. So absent the very low mpx of a9 (and thats a big absent) its as good as the R5. But the a9 is 12 bit for continuous. So I personally would not buy either of them because I am not a 20fps shooter. Per the chart though a7riii and a7riv would be useful to me, if I did not have a bunch of big white canon's and 5DSR and a dislike for sony menu/battery, dust control, cold weather performance....]...Show more →
If you actually measure the DR, you compare against the noise floor. Measurements show that from iso200 on 14 bits only helps digitizing the noise floor better, with no real benefit (read noise dithering overwhelms the bit difference). This can be clearly seen from looking at the photon transfer curve.
If you determine engineering DR, for example, the low end point is determined by where the SNR curve crosses the value of 1. Now if you do subtle NR at low isos, you can influence this point and the measurements show higher DR than you would normally get. This is what Canon does and Bill Claff estimates it to give them 2/3 of a stop.
When Jim Kasson did the DR test using the D810 which has a real iso 64, you could indeed measure(!) an advantage using iso 64 in 14 vs. 12 it mode. You couldn't for iso >200.
It is super hard to get real world images showing a difference of 12 vs. 14 bit. I would love to see technically perfect shots of a situation demonstrating this for iso 400, for example unless(!) the noise reduction is changed according to which mode you are in, i.w. 12 or 14 bit. Without knowing that it will be difficult to explain differences.
