Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Canon Forum | Join Upload & Sell

1       2      
3
       4              7       8       end
  

Archive 2016 · Canon 80D dynamic range: good news! (quick test with RAW images)

  
 
GC5
Offline
• • • •
Upload & Sell: Off
p.3 #1 · p.3 #1 · Canon 80D dynamic range: good news! (quick test with RAW images)


Trolling about new Nikons aside, has anyone evaluated the files next to 7D2 or 70d files? I'd really like to know how much, if any, improvement can be expected in high iso quality and DR...


Mar 20, 2016 at 04:39 PM
arbitrage
Offline
• • • • • • •
Upload & Sell: On
p.3 #2 · p.3 #2 · Canon 80D dynamic range: good news! (quick test with RAW images)


GC5 wrote:
Trolling about new Nikons aside, has anyone evaluated the files next to 7D2 or 70d files? I'd really like to know how much, if any, improvement can be expected in high iso quality and DR...


As I said back in post #4, I looked at ISO 1600-6400 against the 7D2 and noise was very similar. When lifting shadows on the high ISO files the 80D had no weird purple and green colours in the blacks where as the 7D2 did.

In post #5, the DR is shown for all ISOs vs the 7D2.



Mar 20, 2016 at 06:48 PM
Pixel Perfect
Offline
• • • • • • •
Upload & Sell: On
p.3 #3 · p.3 #3 · Canon 80D dynamic range: good news! (quick test with RAW images)


I downloaded 70D and 80D RAW's, processed them identically, then remapped the 80D to 20MP and at ISO 6400 I could not tell the difference between the two. The per pixel noise on the 80D is a tad worse, but the detail is not lost and equalised it's a deadheat. I haven't tried using ISO 100 RAW's to do shadow pushing yet, but I'll try that today.


Mar 20, 2016 at 07:16 PM
thw2
Offline
• • • •
Upload & Sell: Off
p.3 #4 · p.3 #4 · Canon 80D dynamic range: good news! (quick test with RAW images)


Pixel Perfect wrote:
I downloaded 70D and 80D RAW's, processed them identically, then remapped the 80D to 20MP and at ISO 6400 I could not tell the difference between the two. The per pixel noise on the 80D is a tad worse, but the detail is not lost and equalised it's a deadheat. I haven't tried using ISO 100 RAW's to do shadow pushing yet, but I'll try that today.


Same here. Using the Imaging Resource files, I couldn't tell too much difference at ISO 100 too. Weird. Perhaps someone could post his results.



Mar 20, 2016 at 10:16 PM
EB-1
Offline
• • • • • • •
Upload & Sell: Off
p.3 #5 · p.3 #5 · Canon 80D dynamic range: good news! (quick test with RAW images)


If it's not better than the 5DsR at ISO 1600, then maybe I should cancel the pre-order. How do I compare that?

EBH



Mar 20, 2016 at 10:46 PM
Pixel Perfect
Offline
• • • • • • •
Upload & Sell: On
p.3 #6 · p.3 #6 · Canon 80D dynamic range: good news! (quick test with RAW images)


EB-1 wrote:
If it's not better than the 5DsR at ISO 1600, then maybe I should cancel the pre-order. How do I compare that?

EBH


Go to imaging-resource, download studio RAWs from each camera at ISO 1600, process them the same as much as possible, resample 5DsR to 24.2MP

I did this for ISO 1600 and 5DsR is a bit better with finer noise. I made 100% crops from test scene for perusal. First two are 80D with defaults and then boosted contrast and sharpening, third is 5DsR. Note that the ACR profiles for 5DsR are very contrasty so I knocked contrast down 30 and added +10 to blacks,

















Mar 20, 2016 at 11:34 PM
EB-1
Offline
• • • • • • •
Upload & Sell: Off
p.3 #7 · p.3 #7 · Canon 80D dynamic range: good news! (quick test with RAW images)


Thanks. I would not resample anything, because the whole point is to compare options for the 5DsR at 1.6x crop and 80D. The 70/7D II have almost the same pixel pitch as the 5Ds/R and the 80D has about 10% better linear resolution. Obviously the extra resolution is not worth much, especially with the AA filter.

EBH



Mar 21, 2016 at 12:04 AM
Pixel Perfect
Offline
• • • • • • •
Upload & Sell: On
p.3 #8 · p.3 #8 · Canon 80D dynamic range: good news! (quick test with RAW images)


EB-1 wrote:
Thanks. I would not resample anything, because the whole point is to compare options for the 5DsR at 1.6x crop and 80D. The 70/7D II have almost the same pixel pitch as the 5Ds/R and the 80D has about 10% better linear resolution. Obviously the extra resolution is not worth much, especially with the AA filter.

EBH


Well I do not have access to 80D for FL limited shooting so that test will have to be performed by others. If you know how 70D compares at high ISO to cropped 5DsR then 80D won't be any better, but won't be any worse.



Mar 21, 2016 at 12:09 AM
Schlotkins
Offline
• • • •
Upload & Sell: Off
p.3 #9 · p.3 #9 · Canon 80D dynamic range: good news! (quick test with RAW images)


Not that I would recommend the 80D for landscapes over the 5Ds, but does it have more DR at 100ISO? seems to be about 3/4 more stops at ISO100 if these are dxomark compatible...


Mar 21, 2016 at 08:38 AM
eyal.ma
Offline
• •
Upload & Sell: Off
p.3 #10 · p.3 #10 · Canon 80D dynamic range: good news! (quick test with RAW images)


I'm impressed with the 80D sensor.
Hopefully the 5D3 replacement can out-perform that as well



Mar 21, 2016 at 09:39 AM
skibum5
Offline
• • • • • •
Upload & Sell: Off
p.3 #11 · p.3 #11 · Canon 80D dynamic range: good news! (quick test with RAW images)


cgarcia wrote:
IR has just posted some RAW samples.

After a very quick analysis, it seems that fortunately our hopes have finally become true. It seems that ISO 100 achieves about 12.4 stops (13.2 in "DXO notation"), ISO 200 about 11.9 (12.7 in "DXO notation").

I'm not sure about the data because both histograms clip almost at the max value (16383 DN) and seem to start in 512 DN (usually Canon used 2048) . The read noise is 2.87 DN at ISO 100 and 3.995 at ISO 200: indeed a new milestone for Canon.

It seems that Sony will still keep a leadership, but the
...Show more

Yeah those numbers seem correct if they really did jump the clipping way up to 16383.
I got 2.88 read and 512 base value and didn't try to hunt for clipping value.

normalized DR=Log2((16383-512)/(sqrt(8/24.2)*2.88)=13.2

The banding seems to be essentially 100% zero.

Much better than say 60D results (and vastly compared to 50D which also had tons of banding). Nicely better than a 7D. Better than a 5D2 or 5D3. The best ever sensor tech from Canon I'd suggest.

On the downside, if this is the best they can do after all this time, it's rather disappointing since it's still 1.3-1.6 stops behind what Exmor has already done a couple years ago and around what Nikon had managed in their top like pro sports stuff non-Exmor cameras rather some time ago (but it is cool that they at least match or beat that).

One can hope they have an even better process for 1DX2 or 5D4 or whatnot, but I don't know.

It's a nice improvement and could help real world for sure, but considering that Exmor just barely squeezes in lots of natural lighting high DR scenes, especially like say sunlight streaming into a forest or post storm sunbeams into swirling mists), this new sensor would still be struggling a bit for all those scenes where you wanted a better sensor, although maybe you can get away with it now, just, if in a bit ugly fashion. It sure would have been nice to just to turn to Canon for everything, but unless the 5D4 does better, I'm not quite so sure. For those who do lots of wide DR scene shooting, might still be stuck with a messy Sony+Canon mix or going all Nikon and giving up the awesome Canon lenses and UI, others won't care. But too early to say the FF stuff is not tested or even out or even announced in the 5D4 case.

Anyway, say whatever you want, hate on me for having dared say it's not Exmor level, whatever.. I'm outta here and won't respond to any posts regardless of content.



Mar 21, 2016 at 02:17 PM
ggreene
Offline
• • • • •
Upload & Sell: Off
p.3 #12 · p.3 #12 · Canon 80D dynamic range: good news! (quick test with RAW images)


skibum5 wrote:
Anyway, say whatever you want, hate on me for having dared say it's not Exmor level, whatever.. I'm outta here and won't respond to any posts regardless of content.




If only you would stop whining over in the Canonrumors forum with your sock puppet.



Mar 21, 2016 at 03:03 PM
form
Offline
• • • •
Upload & Sell: Off
p.3 #13 · p.3 #13 · Canon 80D dynamic range: good news! (quick test with RAW images)


I don't mind if it isn't quite as good as the best exmor sensors, it just has to be significantly better than the last generation of sensors so I don't have to deal with Crayola chroma every time I lift shadows a little on black wedding jackets.


Mar 21, 2016 at 03:13 PM
Pixel Perfect
Offline
• • • • • • •
Upload & Sell: On
p.3 #14 · p.3 #14 · Canon 80D dynamic range: good news! (quick test with RAW images)


form wrote:
I don't mind if it isn't quite as good as the best exmor sensors, it just has to be significantly better than the last generation of sensors so I don't have to deal with Crayola chroma every time I lift shadows a little on black wedding jackets.


Not sure about 80D but it appears 1DXII has greatly improved chroma noise compared to 1DX, as well as offering banding free, greatly improved shadow DR.



Mar 21, 2016 at 04:30 PM
cgarcia
Offline
• •
Upload & Sell: Off
p.3 #15 · p.3 #15 · Canon 80D dynamic range: good news! (quick test with RAW images)


skibum5 wrote:
One can hope they have an even better process for 1DX2 or 5D4 or whatnot, but I don't know.


I highly suspect that these will have better results. In fact the current 5DS from ISO 1600 and up has about 1 full stop of DR advantage compared to the D80, and despite the old tech manages to be a mere 0.7 stops under it at ISO 100. Maybe Canon has room enough to use cheap electronics in D80 and still manage to be competitive.

Unless there is a serious drop at high ISO, simply by keeping a similar DR profile in the graph (that is, not the old sharp slope drop at low ISO) the new high end FF should achieve a very good level. And we can certainly expect at the least the latest excelent noise characteristics. This is a thrilling moment for all Canon glass owners.



Mar 21, 2016 at 04:46 PM
halse
Offline
• •
Upload & Sell: Off
p.3 #16 · p.3 #16 · Canon 80D dynamic range: good news! (quick test with RAW images)


PSA--> dprreview has postred a 70D/80D comparison; looks like 80D has a bit more than 1 EV advantage

http://www.dpreview.com/news/7168986570/canon-shows-dynamism-eos-80d-breaks-new-ground-for-canon-low-iso-dr



Mar 21, 2016 at 04:58 PM
Pixel Perfect
Offline
• • • • • • •
Upload & Sell: On
p.3 #17 · p.3 #17 · Canon 80D dynamic range: good news! (quick test with RAW images)


Canon has opened up the potential of the sensor it appears by finally moving away from the antiquated 500nm lithography to I presume either 180nm or even possibly 130nm which I know they were talking about at one stage. This allows for more complex circuitry on the back of the sensor such as column-parallel ADC especially as pixel size reduces.

Hey I just dug up something from a Canon interview from only last September. Here's the excerpt of interest


DE: This is actually a very technical question. I’m not sure if it’s one that you would be free to answer or not, but with sensor technology some have pointed to the analog-to-digital conversion implementations being very critical for image quality and dynamic range. Can you tell us whether Canon currently uses on-chip or off-chip A/D converters?

MM: Right now, we use both on-chip and off-chip, but recently I made the decision going forward to concentrate on the on-chip.

DE: Mm-hmm. Yes.

MM: The intent is to increase the performance. In terms of cost, this may be a little negative, but in terms of the direction to take, this will make us more competitive.

DE: So the net cost, the combined cost of the sensor and separate A/D is less than a sensor with A/D on it. Ah, I didn’t realize that!

[This is quite significant, and it's interesting that it was Mr. Maeda's decision to move to on-sensor A/D across the board. Some have pointed to the use of off-chip A/D converters as contributing to lower dynamic range numbers for Canon sensors, when compared to those of Nikon and others in tests such as those published on DxOMark.com. Because separate A/D converters require transmitting analog signals from the image sensor to the converter, more electrical noise can enter the signal chain, resuting in higher noise floors. On-sensor A/D also enables massive parallelism in the A/D conversion, and thereby higher frame rates and possibly lower rolling shutter artifacts. There was some speculation that Canon was locked out of this technology by patents held by Sony and others. Whether through unrelated technology or licensing agreements, it appears that this in fact isn't a limitation, or at least isn't one any longer. It's hard to say whether this will mean improved signal to noise ratios in Canon's cameras, but it certainly can't hurt, and this is the first time I've gotten a direct answer to that question from any Canon executive.]



I think then we are in no doubt now as to seeing the fruits of this development and almost certainly guarantees the 1DXII and 80D are the beginning of a new sensor era for Canon.



Mar 21, 2016 at 05:47 PM
technic
Offline
• • • •
Upload & Sell: Off
p.3 #18 · p.3 #18 · Canon 80D dynamic range: good news! (quick test with RAW images)


Dutchphotog wrote:
Great info, do you have similar Numbers for the d7200 that you coups add to the graph?
I Looked at the dxo numbers for the d7200 and 7d mark II and they are slichgly different then your numbers. D7200 scores 14.6 I think in print DR which is still 1.5 stops above the canon 80d


In the comparison scene from DPReview mentioned just above you can chose D7200 (or other camera) instead of 70D. Compared to 70D the 80D looks good, but compared to D7200 it looks very disappointing IMHO and roughly matches that 1.5 stop advantage. Even the much cheaper D5500 is still far ahead of 80D :-(



Mar 21, 2016 at 06:36 PM
thw2
Offline
• • • •
Upload & Sell: Off
p.3 #19 · p.3 #19 · Canon 80D dynamic range: good news! (quick test with RAW images)


technic wrote:
In the comparison scene from DPReview mentioned just above you can chose D7200 (or other camera) instead of 70D. Compared to 70D the 80D looks good, but compared to D7200 it looks very disappointing IMHO and roughly matches that 1.5 stop advantage. Even the much cheaper D5500 is still far ahead of 80D :-(


Look carefully at the black areas in D7200. See how the color shifts from black to brown as exposure compensation is increased from +1 to +5 eV. What's happening here?

I also did not realize how atrocious the D7100 (2013 release) banding is until I looked at the DPReview comparison. The Sony A77 MkII and A6000 (both 2014 releases) are inferior to Canon 80D.



Edited on Mar 22, 2016 at 03:42 AM · View previous versions



Mar 21, 2016 at 09:46 PM
phibes
Offline
• •
Upload & Sell: Off
p.3 #20 · p.3 #20 · Canon 80D dynamic range: good news! (quick test with RAW images)


maybe the dual-pixel tech takes its toll ... its a 48mp sensor, dont forget.


Mar 22, 2016 at 02:29 AM
1       2      
3
       4              7       8       end




FM Forums | Canon Forum | Join Upload & Sell

1       2      
3
       4              7       8       end
    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.