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       end
  

Archive 2018 · EOS R dynamic range (and my disillusion with EOS R)

  
 
cgarcia
Offline
• •
Upload & Sell: Off
p.1 #1 · p.1 #1 · EOS R dynamic range (and my disillusion with EOS R)


Continuing my old "tradition" of being the first publishing this data I just have evaluated some RAWs posted in CR... the news is that there aren't news... the DR is the same (or in the worse case, maybe slighly lower):

https://www.canonrumors.com/forum/index.php?threads/come-play-with-the-canon-eos-r-raw-files.35809/#post-742829

Well, that is okayish and expected (no 6D2 frights here).

And here is my take on EOS R... of course, I'm not trolling!. I like Canon a lot. I use Canon lenses. There is a lot of talent in their engineers. But one can't understand how their managers drive the company. Canon is a big company (e.g. compared with Nikon) and can do much better. They perfectly know that the future is mirrorless. And rather sooner than later.

Here in FM, the Canon forum last post in the first page is 7 days old (despite the EOS R launching). The same post is only 2 days old in Sony. There are 3 topics with more than 500000 views in the Sony forum. None to that level in the Canon. This is the same everywhere (dpreview). Canon is losing their "core" base. Many maybe only amateurs today. Canon still sells, but each time less and less people is interested on them.

After the EOS R launching I got disillusioned. I like a lot Canon lenses and engineering. And the ergonomics (I find too "acute" the A7R3 corners, while the new EOS R is more rounded). But Mirrorless seems to be a second class citizen for Canon. There is no IBIS promise for the future. The EOS R still lacks spot metering linked to AF point, and has the classic metering modes, despite the sensor is now always switched on. While Sony has RAW zebras for perfect ETTR. They could have used the EF-M mount, in fact 1mm wider than Sony FE (and same flange)... but opted instead for a new an incompatible mount suitable for high quality huge aperture lenses. Just as Nikon, perhaps because of marketing. The fact that the "huge" EF-M mount (too big for normal aperture APS-C lenses) was designed to support full frame lenses on a full frame sensor, seems to indicate that the original Canon plans for mirrorless may have changed. Now they seek for the excellence. But maybe a medium format 40-100/2.8 would be more compact than the new 28-70/2, as in fact it is more compact the Canon 24-70/4 compared to the four thirds Olympus Zuiko 14-35/2, optically equivalent (huge apertures to compensate small sensors seem to not optimize the weight).

I bought the A7R3 in January. During the past months I felt as a Canon user with a Sony body, enjoying IBIS and viewfinder zoomed manual focus with my TSE. Since a week ago, however, I now feel as a Sony user with Canon lenses. No longer waiting for what Canon is about to come with. Definitely now I'll buy some of those incredible Loxia, Batis and Voigtlander I have discovered the past months. And hope that Canon continues releasing its marvelous TSE lenses for EF for a while (I think that movements could interfere with the grip if designed for R distance).


Edited on Sep 09, 2018 at 08:24 PM · View previous versions



Sep 09, 2018 at 02:27 PM
RobDickinson
Offline
• • • •
Upload & Sell: Off
p.1 #2 · p.1 #2 · EOS R dynamic range (and my disillusion with EOS R)


I expected it to be the 5d4 sensor, I'm OK with that to replace my 5d4, hoping for an a7r3 competitor next 6.

As for the mount I'm confused too, the efm would have done just fine.

Why do we need a huge mount for what should be a smaller lighter system.



Sep 09, 2018 at 03:33 PM
snapsy
Offline
• • • • • •
Upload & Sell: On
p.1 #3 · p.1 #3 · EOS R dynamic range (and my disillusion with EOS R)


Here's a visual depiction of the R's DR:

Canon R, Full-Sized Image, Orig
Canon R, Full-Sized Image, +7EV Push



Sep 09, 2018 at 04:08 PM
gfiksel
Offline
• • • •
Upload & Sell: Off
p.1 #4 · p.1 #4 · EOS R dynamic range (and my disillusion with EOS R)


RobDickinson wrote:
I expected it to be the 5d4 sensor, I'm OK with that to replace my 5d4, hoping for an a7r3 competitor next 6.

As for the mount I'm confused too, the efm would have done just fine.

Why do we need a huge mount for what should be a smaller lighter system.


Doesn’t it provide a more normal incidence, especially at the edges?



Sep 09, 2018 at 04:45 PM
gfiksel
Offline
• • • •
Upload & Sell: Off
p.1 #5 · p.1 #5 · EOS R dynamic range (and my disillusion with EOS R)


If it’s like 5d4, it’s great! Combined with -6EV focusing and I couldn’t care less about FPS.


Sep 09, 2018 at 04:47 PM
philip_pj
Offline
• • • • •
Upload & Sell: On
p.1 #6 · p.1 #6 · EOS R dynamic range (and my disillusion with EOS R)


'There is no IBIS promise for the future.' and 'enjoying IBIS and viewfinder zoomed manual focus..'

There will (very soon) be a vast outpouring of gushing material extolling the tremendous image quality of Canon's new lenses, yet the stability elephant-in-the-room will go almost totally unnoticed - just watch.

We must presume the upside-down world of providing IS on low-moderate quality lenses while omitting it on their high end regular focal length lenses made good sense in the board rooms of Canon management. But if users want handholding performance in moderate-low light levels, the potential IQ gain promised by 'faster=better' lenses than the opposition simply will not eventuate.

It's not new. Look at their current EF lenses to see that all the fine new II series lenses all lack IS - the exception being the new 85/1.4. Comparisons with other FF MILCs will not flatter Canon, those makers are also adept at good fast lenses.

Handholding with good technique approaches tripod levels of quality at s/s of around 8x to 10x of focal length .. so a 100mm lens gives you that at 1/1000s, for example. IBIS takes that IQ level to a far more light-manageable 1/250s if you agree with the rather moderate claim of two stops. You can also make gains in lower ISO (better everything) and/or stronger apertures (better IQ moving from f2 > f2.8) if you choose that 'good light' path instead of faster shutter speeds. You always win.

Had Sony not debuted IBIS on FF early (in 2015) Canon would have got this glaring omission past many more people. With the surge of ex-Canon users now with Sony, the cat is out of the bag and running away - fast. This is a far bigger final quality issue than shot time features (fps, EV AF) or sensor performance, which is actually quite close (5DIV vs modern Sonys).

[As you say, manual focus is another 'no contest' comparison. Even the VF is stabilized (sensor feed), so that 100mm does not shake in the EVF, like a ~1200mm lens at 12x.]



Sep 09, 2018 at 05:22 PM
fish_shooter
Offline
• •
Upload & Sell: Off
p.1 #7 · p.1 #7 · EOS R dynamic range (and my disillusion with EOS R)


I imported these DNGs into Lightroom and noted that the file size is a bit low for the megapixels. The one with the underexposed foreground is especially small at 17.67 MB. So I wonder if an uncompressed raw file would be better?


Sep 09, 2018 at 05:47 PM
skibum5
Offline
• • • • • •
Upload & Sell: Off
p.1 #8 · p.1 #8 · EOS R dynamic range (and my disillusion with EOS R)


Wow, so even with this Canon still have not managed to catch up to Sony sensors for DR at all? And what is a mirrorless for but often landscape type stuff where DR can be tricky and occur in complex ways. I probably shouldn't be surprised anymore (I was among the first to point out that this is where they were aimed, already a full decade ago), but I guess I am. Anyway I guess just as well I stopped waiting a while ago.

A shame, since I do like a true DSLR and the UI of Canon best of all by far (and they do have nice lenses, although they can be used with Sony stuff at least even if it is not quite ideal) and all, but yeah, sadly, I can't remember the date that I last purchased something from Canon (and I had been a Canon only guy since the late 80s).
I've even sold some Canon stuff.




Sep 09, 2018 at 05:59 PM
skibum5
Offline
• • • • • •
Upload & Sell: Off
p.1 #9 · p.1 #9 · EOS R dynamic range (and my disillusion with EOS R)


gfiksel wrote:
Doesn’t it provide a more normal incidence, especially at the edges?


yeah, and if the MTF are to believe it does let them get noticeably better far edge performance for landscape work much more easily and even better wide open uber ultra fast lens performance, so there are those plusses




Sep 09, 2018 at 06:01 PM
Pixel Perfect
Offline
• • • • • • •
Upload & Sell: On
p.1 #10 · p.1 #10 · EOS R dynamic range (and my disillusion with EOS R)


snapsy wrote:
Here's a visual depiction of the R's DR:

Canon R, Full-Sized Image, Orig
Canon R, Full-Sized Image, +7EV Push


5D4 sensor is limited to about +4EV push, Sony about +5EV, Nikon about +6EV, so +7EV way beyond the bounds of usability. Still +4EV is more than adequate I’d say for 99% of real world high contrast shots. Higher than this blended exposure would be better.

5D4 sensor readout is too slow and read noise still much higher than Sony/Nikon, even though its about their best sensor ever and pretty darn good in general.



Sep 09, 2018 at 06:04 PM
RobDickinson
Offline
• • • •
Upload & Sell: Off
p.1 #11 · p.1 #11 · EOS R dynamic range (and my disillusion with EOS R)


fish_shooter wrote:
I imported these DNGs into Lightroom and noted that the file size is a bit low for the megapixels. The one with the underexposed foreground is especially small at 17.67 MB. So I wonder if an uncompressed raw file would be better?


From what I've seen yes the cr3 compressed raws do loose some extremes for things like shadow detail. Are there any cr2 files around?



Sep 09, 2018 at 06:09 PM
Rajan Parrikar
Offline
• • • • •
Upload & Sell: On
p.1 #12 · p.1 #12 · EOS R dynamic range (and my disillusion with EOS R)


For a company that boldly introduced the groundbreaking EOS 5D in 2005, Canon's recent record of bringing meh bodies to market is disappointing. If Canon's next version of the mirrorless R line - with a high MP high DR sensor on par with or better than Sony's sensor at the time, suitable for landscape work - is as uninspiring as this introductory lemon, it will be hard to remain with Canon. For me the great attraction of Canon right now is the superb line of lenses, but Sony and others are fast catching up and Canon isn't helping out.




Sep 09, 2018 at 06:10 PM
snapsy
Offline
• • • • • •
Upload & Sell: On
p.1 #13 · p.1 #13 · EOS R dynamic range (and my disillusion with EOS R)


Pixel Perfect wrote:
5D4 sensor is limited to about +4EV push, Sony about +5EV, Nikon about +6EV, so +7EV way beyond the bounds of usability. Still +4EV is more than adequate I’d say for 99% of real world high contrast shots. Higher than this blended exposure would be better.

5D4 sensor readout is too slow and read noise still much higher than Sony/Nikon, even though its about their best sensor ever and pretty darn good in general.


These images weren't meant to demonstrate how many EVs the camera could be pushed but instead how the R's performance compares to the 5DM4, which can found using Dpreview's raw DR widget.



Sep 09, 2018 at 06:16 PM
fish_shooter
Offline
• •
Upload & Sell: Off
p.1 #14 · p.1 #14 · EOS R dynamic range (and my disillusion with EOS R)


RobDickinson wrote:
From what I've seen yes the cr3 compressed raws do loose some extremes for things like shadow detail. Are there any cr2 files around?


My understanding is that this camera will be generating CR3 files, CR2 is now or soon will be in the past.
However we are NOT looking at CR3 files here but DNG conversions (with a beta converter?) so I wonder if one should be too conclusive about the DR at this point.




Sep 09, 2018 at 06:27 PM
EB-1
Offline
• • • • • • •
Upload & Sell: Off
p.1 #15 · p.1 #15 · EOS R dynamic range (and my disillusion with EOS R)


cgarcia wrote:
Continuing my old "tradition" of being the first publishing this data I just have evaluated some RAWs posted in CR... the news is that there aren't news... the DR is the same (or in the worse case, maybe slighly lower):

https://www.canonrumors.com/forum/index.php?threads/come-play-with-the-canon-eos-r-raw-files.35809/#post-742829

Well, that is okayish and expected (no 6D2 frights here).

And here is my take on EOS R... of course, I'm not trolling!. I like Canon a lot. I use Canon lenses. There is a lot of talent in their engineers. But one can't understand how their managers drive the company. Canon is a big company (e.g. compared with Nikon) and
...Show more

I think you mean disillusioned, not delusioned.

I'm disillusioned with the Sony a7R III and cannot imagine that a 5D series R would be worse to use. Canon should learn a bit from whatever Canon users don't like about the cheap, 6D grade body.
I was also thinking that the TSE lenses might have clearance issues with the R bodies. (They are nearly useless on the EF-S unless you have really tough hands or use tools.)

EBH

EBH

Edited on Sep 09, 2018 at 07:00 PM · View previous versions



Sep 09, 2018 at 06:52 PM
RobDickinson
Offline
• • • •
Upload & Sell: Off
p.1 #16 · p.1 #16 · EOS R dynamic range (and my disillusion with EOS R)


fish_shooter wrote:
I imported these DNGs into Lightroom and noted that the file size is a bit low for the megapixels. The one with the underexposed foreground is especially small at 17.67 MB. So I wonder if an uncompressed raw file would be better?

RobDickinson wrote:
From what I've seen yes the cr3 compressed raws do loose some extremes for things like shadow detail. Are there any cr2 files around?

fish_shooter wrote:
My understanding is that this camera will be generating CR3 files, CR2 is now or soon will be in the past.
However we are NOT looking at CR3 files here but DNG conversions (with a beta converter?) so I wonder if one should be too conclusive about the DR at this point.


I was assuming the normal Canon lossless raws are still cr2 and the c-raw are cr3, though could be wrong.

And whichever the dng convertor will have to work on one or other other of the raw files I assume this was on the c-raw version as from dp review




Sep 09, 2018 at 06:53 PM
cgarcia
Offline
• •
Upload & Sell: Off
p.1 #17 · p.1 #17 · EOS R dynamic range (and my disillusion with EOS R)


philip_pj wrote:
'There is no IBIS promise for the future.' and 'enjoying IBIS and viewfinder zoomed manual focus..'

There will (very soon) be a vast outpouring of gushing material extolling the tremendous image quality of Canon's new lenses, yet the stability elephant-in-the-room will go almost totally unnoticed - just watch.

We must presume the upside-down world of providing IS on low-moderate quality lenses while omitting it on their high end regular focal length lenses made good sense in the board rooms of Canon management. But if users want handholding performance in moderate-low light levels, the potential IQ gain promised by 'faster=better' lenses than the
...Show more

I don't like tripods, nor are allowed in some places. But there is a further reason for IBIS/IS shooting stills, many times forgotten: when the floor moves. As when taking a picture from a overpass (vibrations from people walking), in a boat or a plane, etc. No tripod will do the job there...



Sep 09, 2018 at 08:35 PM
cgarcia
Offline
• •
Upload & Sell: Off
p.1 #18 · p.1 #18 · EOS R dynamic range (and my disillusion with EOS R)


fish_shooter wrote:
My understanding is that this camera will be generating CR3 files, CR2 is now or soon will be in the past.
However we are NOT looking at CR3 files here but DNG conversions (with a beta converter?) so I wonder if one should be too conclusive about the DR at this point.



I don't know if these files have been lossy compressed (either the CR3 or the DNG). But we can't infere from their size that the original CR3 was lossy compressed (I think that CR3 can also use lossless compression, as CR2 always did).

I tested in the past a CR2 file converted to DNG, and dcraw obtained exactly the same PGM image from both (100% coincidence in all the bytes, including the masked pixels). I have not yet tested a lossy compressed CR3 file converted to DNG, though. I don't know how they are handled. For example if the masked areas (used to evaluate the DR) are excluded from lossy compression, as they should because can not be actually compressed that way (100% noise). Nor I don't know how compression would affect the measurements.

But since the results are so close to the 5D4, I assume that they are reliable and not just a coincidence.

I think that Canon had (and has) an excellent lossless compression. I think that the new lossy alternative is only intended to increase the performance (the camera processor is fast, its storage access is slow) but will never improve the IQ, nor will save a lot of money in today's deadly cheap storagement, nor will cause a noticeable improvement in our PCs fast SSDs. In fact we can only hope that image quality will not be damaged in a measurable way. While we are aware that this wasn't previously true both for Nikon and Sony. If I remember well, compression is not recommended in NEF files suitable for pushing back (after ETTR). Sony has even artifacts, and ended allowing to optionally disable it via a firmware update.


Edited on Sep 09, 2018 at 09:12 PM · View previous versions



Sep 09, 2018 at 08:59 PM
PicGuy
Offline
• • •
Upload & Sell: Off
p.1 #19 · p.1 #19 · EOS R dynamic range (and my disillusion with EOS R)


skibum5 wrote:
Wow, so even with this Canon still have not managed to catch up to Sony sensors for DR at all? And what is a mirrorless for but often landscape type stuff where DR can be tricky and occur in complex ways. I probably shouldn't be surprised anymore (I was among the first to point out that this is where they were aimed, already a full decade ago), but I guess I am. Anyway I guess just as well I stopped waiting a while ago.


With the introduction of the A7/3 and especially the A9, this isn't the case anymore. Sony is one generation away, IMO, of surpassing the AF abilities of DSLRs. They already offer
AF features like "Eye AF" that not DSLR can touch with OVF focus.And what is a mirrorless for but often landscape type stuff



Sep 09, 2018 at 09:06 PM
EB-1
Offline
• • • • • • •
Upload & Sell: Off
p.1 #20 · p.1 #20 · EOS R dynamic range (and my disillusion with EOS R)


fish_shooter wrote:
I imported these DNGs into Lightroom and noted that the file size is a bit low for the megapixels. The one with the underexposed foreground is especially small at 17.67 MB. So I wonder if an uncompressed raw file would be better?

RobDickinson wrote:
From what I've seen yes the cr3 compressed raws do loose some extremes for things like shadow detail. Are there any cr2 files around?

fish_shooter wrote:
My understanding is that this camera will be generating CR3 files, CR2 is now or soon will be in the past.
However we are NOT looking at CR3 files here but DNG conversions (with a beta converter?) so I wonder if one should be too conclusive about the DR at this point.

cgarcia wrote:
I don't know if these files have been lossy compressed (either the CR3 or the DNG). But we can't infere from their size that the original CR3 was lossy compressed (I think that CR3 can also use lossless compression, as CR2 always did).

I tested in the past a CR2 file converted to DNG, and dcraw obtained exactly the same PGM image from both (100% coincidence in all the bytes, including the masked pixels). I have not yet tested a lossy compressed CR3 file converted to DNG, though. I don't know how they are handled. For example if the masked areas
...Show more

Is the crappy CR3 format required or optional?

EBH



Sep 09, 2018 at 09:10 PM
1
       2       3       4       end




FM Forums | Canon Forum | Join Upload & Sell

1
       2       3       4       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.