Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

Canon 6D Mark II Sensor Measurement Completed

astrophotography dslr
  • This topic is locked This topic is locked
61 replies to this topic

#1 Jim Waters

Jim Waters

    Viking 1

  • *****
  • topic starter
  • Posts: 547
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 15 July 2017 - 06:01 PM

FYI - Some results are being published.

 

https://www.dpreview...s/post/59833872

 

http://www.photonsto.../Charts/PDR.htm

 

 


 

#2 leveye

leveye

    Gemini

  • *****
  • Posts: 3403
  • Joined: 06 Feb 2013
  • Loc: Central Oregon Coast

Posted 15 July 2017 - 07:43 PM

Looks like ISO 800 is still the best setting.


 

#3 Jim Waters

Jim Waters

    Viking 1

  • *****
  • topic starter
  • Posts: 547
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 15 July 2017 - 10:53 PM

I was hoping for greater Dynamic Range.  I will be Ha modifying my existing 6D and using my next camera for landscape and nature photography.  I am reconsidering the 6D Mk II.  I may get the 5D Mk IV.


 

#4 MCovington

MCovington

    Gemini

  • *****
  • Posts: 3255
  • Joined: 13 May 2014
  • Loc: Michael Covington - Athens, Georgia

Posted 15 July 2017 - 11:59 PM

Yes, it looks like this is an old-style DSLR sensor (like the 60D), not new-style (like the 80D and many Nikon products).

A good one, but not "ISOless."


 

#5 Jerry Lodriguss

Jerry Lodriguss

    Vendor

  • *****
  • Vendors
  • Posts: 6034
  • Joined: 19 Jul 2008
  • Loc: Voorhees, NJ

Posted 16 July 2017 - 01:59 AM

Hi Jim,

 

Thanks for the post and heads up!

 

Wow, look a the read noise at Bill Claff's site

 

It's even a little worse than the first 6D, and not close to the 80D.

 

It is certainly not ISOless as Michael notes, and I would say a disappointment if it's correct, and Bill Claff knows what he is doing.

 

I'm going to say I'm shocked.

 

Jerry


 

#6 Jim Waters

Jim Waters

    Viking 1

  • *****
  • topic starter
  • Posts: 547
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 16 July 2017 - 02:16 AM

I am disappointed too...  I am calling B&H Sunday and cancel my order.  I will likely get a 5D4 for terrestrial use.


 

#7 Jon Rista

Jon Rista

    Hubble

  • *****
  • Posts: 18116
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 16 July 2017 - 03:07 AM

WOW...


 

#8 jforkner

jforkner

    Mariner 2

  • -----
  • Posts: 245
  • Joined: 04 Aug 2015
  • Loc: Copperopolis, CA

Posted 16 July 2017 - 07:39 AM

Since the camera has not been officially released, I’m dubious of all this analysis at fredmiranda.com & photonstophotos.net.  The analysis at Fred Miranda is clearly opinion based on pre-production models & documented specs.  And I find the analysis at Photon to Photos suspect because of statements like these:

 

▪ “As far as I know, production…”

 

▪ “I have one complete set of data from dpreview which I believe is from a "production" body and I have addition files from elsewhere.”

 

▪ “These raw files were prepared especially for me. They cannot be released publicly because of the embargo; but the results of the analysis can be shared publicly.”

 

▪ “All measurements at PhotonsToPhotos are performed using raw files.  Many of my files come from "ordinary" people such as yourself.”

 

It seems ironic that none of the traditional camera testing sites (e.g., DxOMark & DPReview, who supposedly provided images) have any posted results.  And all the data that’s getting people worked up comes from forum posts.  Yet, there are apparently a lot of “disappointed” folks out there willing to accept the alleged findings.

 

Not that anyone cares, but I remain skeptical of all this data until the camera is officially released (reportedly July 27th) & properly tested.

 

Jack


 

#9 andysea

andysea

    Mercury-Atlas

  • *****
  • Posts: 2686
  • Joined: 03 Sep 2010
  • Loc: Seattle, WA

Posted 16 July 2017 - 07:39 AM

I am disappointed too... I am calling B&H Sunday and cancel my order. I will likely get a 5D4 for terrestrial use.

i did the same thing
 

#10 Ron359

Ron359

    Viking 1

  • *****
  • Posts: 757
  • Joined: 21 Apr 2008
  • Loc: -105 +39

Posted 16 July 2017 - 09:47 AM

Since the camera has not been officially released, I’m dubious of all this analysis at fredmiranda.com & photonstophotos.net.  The analysis at Fred Miranda is clearly opinion based on pre-production models & documented specs.  And I find the analysis at Photon to Photos suspect because of statements like these:

 

▪ “As far as I know, production…”

 

▪ “I have one complete set of data from dpreview which I believe is from a "production" body and I have addition files from elsewhere.”

 

▪ “These raw files were prepared especially for me. They cannot be released publicly because of the embargo; but the results of the analysis can be shared publicly.”

 

▪ “All measurements at PhotonsToPhotos are performed using raw files.  Many of my files come from "ordinary" people such as yourself.”

 

It seems ironic that none of the traditional camera testing sites (e.g., DxOMark & DPReview, who supposedly provided images) have any posted results.  And all the data that’s getting people worked up comes from forum posts.  Yet, there are apparently a lot of “disappointed” folks out there willing to accept the alleged findings.

 

Not that anyone cares, but I remain skeptical of all this data until the camera is officially released (reportedly July 27th) & properly tested.

 

Jack

I agree, the "data" plots make little sense.  If you plot their comparison data to other models it differs little from other earlier models. Even the current 6D are proven performers.  


 

#11 Ron359

Ron359

    Viking 1

  • *****
  • Posts: 757
  • Joined: 21 Apr 2008
  • Loc: -105 +39

Posted 16 July 2017 - 09:58 AM

Hi Jim,

 

Thanks for the post and heads up!

 

Wow, look a the read noise at Bill Claff's site

 

It's even a little worse than the first 6D, and not close to the 80D.

 

It is certainly not ISOless as Michael notes, and I would say a disappointment if it's correct, and Bill Claff knows what he is doing.

 

I'm going to say I'm shocked.

 

Jerry

To me the notes (caveats) at the bottom of the graph negate any possibility this is any real test:  

 

"The conversion is for analog gain only. Digital gain and other signal processing are supressed."    

 

So, its not an actual test of what in camera or actual output processing or noise supression there might be.  

 

"These raw values are not appropriate for comparing camera models because they are not adjusted for area."  

 

 Then why are they shown comparing models?!  How can you conclude anything relative to other models with a statement like that!  

 

"Flat spots in groups of three in the analog region usually indicate where digital intermediate ISO gain is present."   

 

"Hi" ISO values can behave strangely once signal processing such as noise reduction kicks in."   

 

Yeah.  So, the entire "data" set are invalid.  Give me a break.  


 

#12 Jim Waters

Jim Waters

    Viking 1

  • *****
  • topic starter
  • Posts: 547
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 16 July 2017 - 10:53 AM

My order is canceled for now.  I will be following the discussions on this for the next 2 to 3 weeks and then decide what to do. 


 

#13 MCovington

MCovington

    Gemini

  • *****
  • Posts: 3255
  • Joined: 13 May 2014
  • Loc: Michael Covington - Athens, Georgia

Posted 16 July 2017 - 11:10 AM

I think the tests on PhotonsToPhotos and DxOmark are quite appropriate.  This is a good camera, but it uses Canon's traditional type of sensor, like the 60D, rather than the new nearly-ISOless sensor technology of the 80D.  It is not a bad camera.  The 60D is not a bad camera.  But it is not the latest wave of sensor performance.


 

#14 StuartJPP

StuartJPP

    Mariner 2

  • -----
  • Posts: 253
  • Joined: 16 Oct 2015
  • Loc: Leeds, United Kingdom

Posted 16 July 2017 - 12:28 PM

According to the site above there is negligible difference between the Canon 600D and the Canon 6D at ISO 800.

http://www.photonsto...Canon EOS 6D_14

 

Yet in practice the actual captured results (i.e. what actually counts) are MILES apart, chalk and cheese, black and white. I own both and there is no way I would use the 600D except if I wanted to put more pixels on a target using the same OTA, but I'd have to spend a load more time on the target to get the noise down to the same level as the 6D.

 

If we were to just focus on this single aspect we'd all be imaging with iPhone 7's:

http://www.photonsto...Canon EOS 6D_14


 

#15 jforkner

jforkner

    Mariner 2

  • -----
  • Posts: 245
  • Joined: 04 Aug 2015
  • Loc: Copperopolis, CA

Posted 16 July 2017 - 01:34 PM

My order is canceled for now.  I will be following the discussions on this for the next 2 to 3 weeks and then decide what to do. 

Probably a smart move under the circumstances.  But until the production camera is released, any discussion on the subject is pure speculation.

 

Jack


 

#16 Jim Waters

Jim Waters

    Viking 1

  • *****
  • topic starter
  • Posts: 547
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 16 July 2017 - 03:26 PM

 

My order is canceled for now.  I will be following the discussions on this for the next 2 to 3 weeks and then decide what to do. 

Probably a smart move under the circumstances.  But until the production camera is released, any discussion on the subject is pure speculation.

 

Jack

 

You are probably right.  The question is will the data set be updated on these sites?


 

#17 Spica010

Spica010

    Sputnik

  • -----
  • Posts: 26
  • Joined: 03 Dec 2014

Posted 16 July 2017 - 03:53 PM

Sorry for putting my two cents into this. I am sure quite a number of forum members are longer in the game...

But, Canon explicitly advertises the new camera sensor for its dynamic range and the achievable fine tonal gradation.

(For those who speak Japanese: http://cweb.canon.jp...2/fullsize.html).

Now, does Canon lag behind Nikon and Sony as far as sensor performance (read noise, DR etc) is concerned? Surely.

But are they so stupid to hype up a new camera and advertise it for some key parameters only to find that some real world measurements show it to be worse than the very camera it is supposed to substitute? Difficult to believe.

I think patience is the key: The officially released version of the camera needs to be properly tested in order to judge it.

BTW, playing early adopter is never a good idea, not even with Canon and not only because of the higher price tag. All the companies have QA departments and little problems need to be ironed out in the months following a product release many of which can stay hidden from the public view for quite some time. (HW) Changes to the product can be introduced without changing the name of the product.


Edited by Spica010, 16 July 2017 - 03:55 PM.

 

#18 Jim Waters

Jim Waters

    Viking 1

  • *****
  • topic starter
  • Posts: 547
  • Joined: 21 Oct 2007
  • Loc: Phoenix, AZ USA

Posted 16 July 2017 - 04:49 PM

The DP Review link I posted above looks like it was deleted...

 

 

"Forum thread deleted or withdrawn

 

The thread that you are looking for has been deleted by the moderators or withdrawn by the original poster. If you have a second, we would appreciate if you let us know in case you have reached this page via some existing and genuine link. We will investigate the problem as soon as possible."


 

#19 andysea

andysea

    Mercury-Atlas

  • *****
  • Posts: 2686
  • Joined: 03 Sep 2010
  • Loc: Seattle, WA

Posted 16 July 2017 - 04:51 PM

I agree, I decided not to be an early adopter and cancelled my order. I'll be watching the forums in the months ahead to see how the camera performs.
 

#20 sharkmelley

sharkmelley

    Apollo

  • -----
  • Posts: 1349
  • Joined: 19 Feb 2013

Posted 16 July 2017 - 05:28 PM

According to the site above there is negligible difference between the Canon 600D and the Canon 6D at ISO 800.

http://www.photonsto...Canon EOS 6D_14

 

Yet in practice the actual captured results (i.e. what actually counts) are MILES apart, chalk and cheese, black and white. I own both and there is no way I would use the 600D except if I wanted to put more pixels on a target using the same OTA, but I'd have to spend a load more time on the target to get the noise down to the same level as the 6D.

 

If we were to just focus on this single aspect we'd all be imaging with iPhone 7's:

http://www.photonsto...Canon EOS 6D_14

I wholeheartedly agree. 

 

For long exposure deep sky imaging, "input referred read noise" is just one factor of many. After all, sufficiently long exposures can deal with read noise differences.  More important in my view are quantum efficiency, dark current, various types of fixed pattern noise and raw file "cooking".  Give me 80-90% QE, low dark current, very low FPN and no star eating spatial filtering and I'll buy the 6D MkII tomorrow and put my Sony A7S in retirement.  I'm being absolutely serious.  Quite honestly, a slightly raised read noise and lack of improvement in dynamic range is of no consequence to me for deep sky imaging.

 

Let's see how the 6D MkII behaves on the criteria that really matter to us astroimagers.

 

Mark


Edited by sharkmelley, 16 July 2017 - 05:37 PM.

 

#21 sharkmelley

sharkmelley

    Apollo

  • -----
  • Posts: 1349
  • Joined: 19 Feb 2013

Posted 16 July 2017 - 05:31 PM

The DP Review link I posted above looks like it was deleted...

 

 

"Forum thread deleted or withdrawn

 

The thread that you are looking for has been deleted by the moderators or withdrawn by the original poster. If you have a second, we would appreciate if you let us know in case you have reached this page via some existing and genuine link. We will investigate the problem as soon as possible."

 

The PhotonsToPhotos data is still available.


Edited by sharkmelley, 17 July 2017 - 01:59 PM.

 

#22 jforkner

jforkner

    Mariner 2

  • -----
  • Posts: 245
  • Joined: 04 Aug 2015
  • Loc: Copperopolis, CA

Posted 16 July 2017 - 06:51 PM

The DP Review link I posted above looks like it was deleted...

 

 

"Forum thread deleted or withdrawn

 

The thread that you are looking for has been deleted by the moderators or withdrawn by the original poster. If you have a second, we would appreciate if you let us know in case you have reached this page via some existing and genuine link. We will investigate the problem as soon as possible."

Ought to tell you somethin'...


 

#23 Herra Kuulapaa

Herra Kuulapaa

    Messenger

  • -----
  • Posts: 475
  • Joined: 10 Dec 2014
  • Loc: Vantaa, Finland

Posted 17 July 2017 - 01:09 AM

 

The DP Review link I posted above looks like it was deleted...

 

 

"Forum thread deleted or withdrawn

 

The thread that you are looking for has been deleted by the moderators or withdrawn by the original poster. If you have a second, we would appreciate if you let us know in case you have reached this page via some existing and genuine link. We will investigate the problem as soon as possible."

Ought to tell you somethin'...

 

If one doesn't wish to speculate with data available so far, then making any assumptions based on this isn't very safe either..


 

#24 Jon Rista

Jon Rista

    Hubble

  • *****
  • Posts: 18116
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 17 July 2017 - 11:19 AM

People have been in this exact same position before. When the 5D III was about to be released, no one, including myself, believed the preliminary reports about it's noise and dynamic range, the amount of banding it had, etc. Thing is, there really isn't much difference in these kind of sensor characteristics between the models that are demoed before release, and the actual release model. Canon doesn't usually put out test models until the sensor work is thoroughly completed. The 5D III was just as disappointing when it actually hit the streets as the preliminary analyses indicated.

 

I don't see any reason there will be some magical improvement in the 6D II performance here, either. There is no reason for there to be an improvement. It's clear Canon is using their old architecture, which uses off-die ADC units, probably 4 or 8 of them, operating at a high clock frequency, rather than the newer architectures which move more of those components onto the sensor die, and increase their parallelism. Canon seems to take an entirely different approach to the market, and they haven't seemed as willing to invest in sensor technology and pushing the envelope with ultra low noise sensors like the rest of the industry. It just isn't in their M.O. Never has been. So don't get your hopes up too hight that the 6D II upon official release will suddenly hit 12 stops of DR on bclaff's charts, or hit 13.5 stops on DXO's. It just ain't gonna happen. 


 

#25 Jerry Lodriguss

Jerry Lodriguss

    Vendor

  • *****
  • Vendors
  • Posts: 6034
  • Joined: 19 Jul 2008
  • Loc: Voorhees, NJ

Posted 17 July 2017 - 01:44 PM

I'm surprised that people are doubting Bill Claff's analysis.

 

That guy knows what he is doing.

 

What is surprising is that Canon has the off-die ADC technology in the 80D and 1D MII and apparently the 5D MIV.

 

I guess they consider the 6D MII to be entry-level full frame and don't want a $2,000 camera to compete with the big boy 1D MII at $6,000 and the 5D MIV at $3,500.

 

Jerry


 


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.






Also tagged with one or more of these keywords: astrophotography, dslr



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics