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

New Canon 7DII has one-tenth the dark current

This topic has been archived. This means that you cannot reply to this topic.
32 replies to this topic

#1 Lorien

Lorien

    Sputnik

  • -----
  • topic starter
  • Posts: 41
  • Joined: 01 Dec 2013

Posted 01 December 2014 - 05:05 PM

The new Canon 7DII appears to have less than one tenth the dark current of previous Canon cameras including the 6D.  Assuming this is confirmed by other testers, this is game-changing for DSLR astrophotography.  The author also states that the 7DII unmodified shows better Ha sensitivity than other unmodified Canon DSLRs.  Full review at: http://www.clarkvisi...ion-canon-7dii/.

 

dark-current-compared_2-6c-v1.gif


Edited by Lorien, 01 December 2014 - 05:29 PM.


#2 austin.grant

austin.grant

    Vendor - Modified DSLR Cameras

  • *****
  • Vendors
  • Posts: 1444
  • Joined: 18 Oct 2010

Posted 01 December 2014 - 05:33 PM

I was under the impression it is the same sensor as in the 70D. I don't see where Roger Clark has tested that camera, but I wonder how they compare? At half the price, if the 70D has the same noise characteristics, it would be the obvious choice.



#3 jerry10137

jerry10137

    Viking 1

  • *****
  • Vendors
  • Posts: 780
  • Joined: 26 Jan 2013

Posted 01 December 2014 - 05:36 PM

Low noise dslr cameras will no doubt start giving the CCD guys a run for their money.  That 7DII is supposed to be very friendly in the noise department and it looks like results to back that up are starting to come out. A CCD camera does not have an adjustable setting for sensitivity like the dslrs do. I've been working a little with an SBIG osc stl11000 camera and I have to expose for 10 or 15 minutes on a single sub to gather the amount of signal equal to about 180 seconds of ISO 800 dslr.  I have a 6D and it is an oracle compared that SBIG osc.



#4 mmalik

mmalik

    DSLR camera modifications

  • *****
  • Vendors
  • Posts: 9890
  • Joined: 13 Jan 2012

Posted 01 December 2014 - 05:42 PM

Not sure what to make of this? Horsehead picture posted looks grossly over-saturated and not well resolved. Would like to see more peer reviews. Regards


Edited by mmalik, 03 December 2014 - 05:52 AM.


#5 kd4pbs

kd4pbs

    Apollo

  • *****
  • Posts: 1139
  • Joined: 13 Feb 2013

Posted 04 December 2014 - 11:49 AM

I was wondering the same thing about the Horsehead region in that review... like the camera has been modified.  I don't see a stock DSLR taking that much of a saturated image, no matter how superior the noise floor is.  I can put my unmodified 7d on that subject for the same amount of integration time and get nowhere near the amount of nebula that was done here.  Perhaps there's more to the 7d Mark II than increased quality of the sensor?  Perhaps they've changed the IR cut filter response so that it admits more into the IR band?

Certainly interesting for sure...  I've had a love-love relationship with my 7d, but that's after coming from a 20d.  It's still far less noisy than the old 20d, and I use it for more than just AP, but it looks like the 7D II might be a real performer!  Maybe in a few years when they can be had second-hand on the cheap...

EDIT:
Well, I guess I should have read the whole article before posting that.  The author explains this nicely; "Another factor for astrophotographers is the sensitivity to red hydrogen-alpha light. The image of the Horsehead nebula above illustrates quite respectable hydrogen alpha sensitivity." He then elaborates on this.  Indeed, this looks like, as he puts it, a "game changing" camera indeed!


Edited by kd4pbs, 04 December 2014 - 12:43 PM.


#6 DuncanM

DuncanM

    Mercury-Atlas

  • -----
  • Posts: 2657
  • Joined: 03 Nov 2009

Posted 04 December 2014 - 05:54 PM

We could restate this as:

 

"Older Canons have ten times the dark current..."

 

The 7DII is the first Canon to match Nikon/Sony in dark current, but I note that read noise is still higher at low ISO. 



#7 17.5Dob

17.5Dob

    Fly Me to the Moon

  • ****-
  • Posts: 5394
  • Joined: 21 Mar 2013

Posted 04 December 2014 - 09:01 PM

We could restate this as:

 

"Older Canons have ten times the dark current..."

 

The 7DII is the first Canon to match Nikon/Sony in dark current, but I note that read noise is still higher at low ISO. 

+1 !

Canon is just now catching up to Sony/Nikon as far as dark current. Canon read noise has always been 10 X's Nikons'  and is one of the biggest faults with Canon.



#8 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10332
  • Joined: 03 Jul 2004

Posted 04 December 2014 - 09:23 PM

Re: New Canon 7DII has one-tenth the dark current

 

Which is measured through RAW digital output from the DSLR. 

Nowadays, with more powerful DSP in use, sensor data got modified (Nikon and SONY have been doing that for a long time, Nikon is getting better on that.)

 

C, on the other hand, has abandoned sending out senor raw as the straight raw output but Camera (DSP assisted) RAW.

 

My only point is we are not comparing sensor to sensor but sensor plus light touches from the DSP.

I am happy that 7Dii is fairing great or excellence.

 

I suspect there will be trick mode is M.L. which can extract raw sensor out.  That would be a good basis for sensor to sensor comparisons.

 

Clear Skies!

 

ccs_hello



#9 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7217
  • Joined: 26 Aug 2012

Posted 04 December 2014 - 11:28 PM

It sounds like a big improvement in Canon APS-C sensors.  The good news is that this technology should move up to the full frame sensors as well.  My concern is that the 4 um pixel pitch is a bit on the small side for longer F/#s and that it will have a noticeably lower well depth than the 6.4 micron sensors in the 6D.  Frankly, when they come out with a full frame sensor with 5-6 micron pixels and sufficiently low noise to go to a 16 bit ADC, I'll be all over that product.  Get the thermal noise down into the 14th or 15th LSB with a 16 bit ADC and CMOS will finally be able to compete head to head with most CCDs.  Unfortunately without cooling, that goal may be hard to reach but this appears to be a step in the right direction.  What we need are CCD camera makers to start to pay attention to CMOS sensor.  I'd love a camera that is as easy to use as a DLSR (i.e. no computer) with integrated cooling and 15 bits (or more) of clean signal.

John



#10 fetoma

fetoma

    Gemini

  • *****
  • Posts: 3017
  • Joined: 26 Sep 2006

Posted 05 December 2014 - 09:32 AM

What we need are CCD camera makers to start to pay attention to CMOS sensor.  I'd love a camera that is as easy to use as a DLSR (i.e. no computer) with integrated cooling and 15 bits (or more) of clean signal.

John

 

This would be wonderful * if * we could get the CMOS sensors to go mono without the bayer layer.



#11 Samir Kharusi

Samir Kharusi

    Mercury-Atlas

  • *****
  • Posts: 2598
  • Joined: 14 Jun 2005

Posted 06 December 2014 - 03:29 AM

It sounds like a big improvement in Canon APS-C sensors.  The good news is that this technology should move up to the full frame sensors as well.  My concern is that the 4 um pixel pitch is a bit on the small side for longer F/#s and that it will have a noticeably lower well depth than the 6.4 micron sensors in the 6D.  Frankly, when they come out with a full frame sensor with 5-6 micron pixels and sufficiently low noise to go to a 16 bit ADC, I'll be all over that product.  Get the thermal noise down into the 14th or 15th LSB with a 16 bit ADC and CMOS will finally be able to compete head to head with most CCDs.  Unfortunately without cooling, that goal may be hard to reach but this appears to be a step in the right direction.  What we need are CCD camera makers to start to pay attention to CMOS sensor.  I'd love a camera that is as easy to use as a DLSR (i.e. no computer) with integrated cooling and 15 bits (or more) of clean signal.

John

Complaints, complaints  ;)  A pixel pitch of 4.09 microns implies 1.25arc-sec/pixel using a C14 Hyperstar (yours too?). Perfect for any average night with seeing around 2.5 arc-sec or worse... My complaint: I am still waiting for 1 arc-sec per pixel for those super still nights, then no need for slower f-ratios.



#12 whwang

whwang

    Mercury-Atlas

  • *****
  • Posts: 2911
  • Joined: 20 Mar 2013

Posted 06 December 2014 - 03:38 AM

 

What we need are CCD camera makers to start to pay attention to CMOS sensor.  I'd love a camera that is as easy to use as a DLSR (i.e. no computer) with integrated cooling and 15 bits (or more) of clean signal.

John

 

This would be wonderful * if * we could get the CMOS sensors to go mono without the bayer layer.

 

 

http://www.nikon.com...2_ds-qi2_01.htm

 

Such a thing already exists.  The question is just whether the prosumer market is large enough for

N/C/S to produce such a camera in a DSLR form factor at an affordable price.



#13 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7217
  • Joined: 26 Aug 2012

Posted 06 December 2014 - 01:04 PM

That camera looks very interesting indeed!  I couldn't find the ADC level, spectral response or the price but I've sent Nikon an inquiry so I'll see what I can find out.  It does look like the maximum exposure time is limited to 60 seconds.  It would sure be nice if Canon or Nikon would configure a camera just like this toward long exposure, low noise AP applications.

John



#14 whwang

whwang

    Mercury-Atlas

  • *****
  • Posts: 2911
  • Joined: 20 Mar 2013

Posted 06 December 2014 - 01:37 PM

You should be able to find those information if you google around.  Everything, including the QE curve, is very impressive.

What's not impressive is its price and control interface.  It is more suitable for research labs, but not amateur astronomy.

 

All we can do is to wait, or to write to Nikon telling them that we need a mono DSLR.



#15 mmalik

mmalik

    DSLR camera modifications

  • *****
  • Vendors
  • Posts: 9890
  • Joined: 13 Jan 2012

Posted 06 December 2014 - 03:09 PM

All we can do is to wait, or to write to Nikon telling them that we need a mono DSLR.

 

Nikon has yet to produce an 'a' version, let alone a mono. I wouldn't hold my breath on Nikon. Canon has shown some regard for astronomy and recognized 'our' existence at least. Regards



#16 whwang

whwang

    Mercury-Atlas

  • *****
  • Posts: 2911
  • Joined: 20 Mar 2013

Posted 06 December 2014 - 06:05 PM

I totally agree that Canon is much more astro-friendly.  But this probably has nothing to do with

whether or not Canon will produce a mono DSLR for astro fans.

 

20Da and 60Da are nice, but not the best choices for the most serious DSLR astrophotographers.

Canon can sell good numbers of them because many of the buyers are those who just start to get

serious about astrophotography.  Those are the main target customers of 20Da and 60Da.

 

On the other hand, those who really want monochrome DSLRs for astrophoto are much more

serious fans.  There are just not enough of them to drive the market.  No matter Canon or Nikon

or Sony, whoever produces a mono DSLR (Leica did it), it won't be targeting the astronomical

market.  It must be for normal photography. 



#17 piaras

piaras

    Apollo

  • -----
  • Posts: 1107
  • Joined: 26 Jan 2009

Posted 06 December 2014 - 08:39 PM

The Canon equivalent to the Nikon, is the M15P-CL. It has a Nikon mount due to Nikon basically owning the industrial market.

Pierre



#18 zsb04

zsb04

    Ranger 4

  • -----
  • Posts: 332
  • Joined: 05 Jun 2014

Posted 07 December 2014 - 08:15 PM

do you guys think this would be a nice upgrade from t3i???



#19 shan1987

shan1987

    Viking 1

  • -----
  • Posts: 554
  • Joined: 01 Apr 2013

Posted 08 December 2014 - 12:26 AM

do you guys think this would be a nice upgrade from t3i???


This would be a good upgrade to any other canon asp-c camera on the market if you have the money.... I an hoping to buy one soon to upgrade from my t3

#20 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23857
  • Joined: 10 Jan 2014

Posted 08 December 2014 - 12:33 AM

Has anyone measured the dark current of Sony and Toshiba sensors? I was under the impression they were already in this class as far as dark current levels go. Combined with the significantly lower read noise of an Exmor, it sounds like they are still the better bet overall for astrophotography, no? I think it is impressive that Canon has finally made some strides in the sensor department, though. It's been too long that they've just been....stagnating.



#21 whwang

whwang

    Mercury-Atlas

  • *****
  • Posts: 2911
  • Joined: 20 Mar 2013

Posted 08 December 2014 - 01:14 AM

Nikon D800 (which has a Sony sensor) has dark current roughly 3x lower than 6D,

and 3x higher than 7D2.  7D2 has certainly pushed the performance of CMOS to

a new level.



#22 DuncanM

DuncanM

    Mercury-Atlas

  • -----
  • Posts: 2657
  • Joined: 03 Nov 2009

Posted 08 December 2014 - 02:26 AM

Nikon D800 (which has a Sony sensor) has dark current roughly 3x lower than 6D,

and 3x higher than 7D2.  7D2 has certainly pushed the performance of CMOS to

a new level.

 

I think the jury is still out on exact figures for dark current. The temperatures on Clark's website are measured on sensor by the camera's firmware and as such measure the sensor as it is warmed by internal heating. A direct comparison is only possible if both cameras are measured side by side with the same external temperature. Additionally, the 7DII may be "cooking" it's RAW somewhat more than the Nikon with the hacked firmware.



#23 whwang

whwang

    Mercury-Atlas

  • *****
  • Posts: 2911
  • Joined: 20 Mar 2013

Posted 08 December 2014 - 03:22 AM

Interesting.  I didn't know that Roger quote the sensor temperature.  My D800's dark current 

measurements are referring to ambient temperature.

 

On my 5D2, the sensor temperature is typically 10 to 14 deg C higher than the ambient.

Let's give 7D2 some advantage and say its sensor temperature is only 8 deg C higher

than the ambient.  After this adjustment, according to Roger's graph, at an ambient

temperatures 6 deg C (14 deg C sensor), the dark rate is roughly 0.02 to 0.03 e- per

pixel per second.  At an ambient temperature of 25 deg C, the dark rate is roughly

0.4 e- per pixel per second.  Translating these to dark rates per surface area, it is

1350 e- per second per square-mm at 6 deg C, and 21600 e- per second per square-mm

at 25 deg C.  D800's numbers are 1400 at 6 deg C and 12803 at 25 deg C (ambient).

 

They are basically on par with each other.  (D800 and 7D2 have similar pixel sizes,

4.8 um vs. 4.3 um.  So even if we just look at dark rate per pixel rather than per

square-mm, the results won't be too different.)

 

On the other hand, Roger's measurement method is robust against what Canon did

to alter the raw image.  (He is not really measuring dark rates.  He measures dark

noise, which is what we really care about.)  You can trust his measurements.

 

Cheers,

Wei-Hao



#24 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23857
  • Joined: 10 Jan 2014

Posted 08 December 2014 - 04:42 AM

I do believe that the 7D II is cooking the RAW a fair bit, given that is exactly what DIGIC 6 was designed to do. I wonder how much that affects the readings of dark current noise...and if it does, does it even matter? I think in-camera RAW processing is the way of the future. Nikon has been doing it a while. Sony is doing it with their Bionz X chips (i.e. A7s), and now Canon is doing it with the DIGIC 6 (well, they started doing it with DIGIC 6 a while ago, just on compacts.) 

 

Is it better to have the camera process the RAW, or is it better to leave the signal be and let the imager process the RAW? Given how important linearity in a CCD is, I suspect the latter is still true...but, I wonder at what point the data coming from CMOS cameras is going to be so clean that it'll have lower noise across the board than CCD cameras. (As it stands now, is that already true with the 7D II and D800...if they have ~0.002e-/s/px DC at around -2C and 3e- or less RN...that already sounds better than even a Sony ICX, which has around 0.003e- DC at -10C and 5e- RN...!! :\)



#25 austin.grant

austin.grant

    Vendor - Modified DSLR Cameras

  • *****
  • Vendors
  • Posts: 1444
  • Joined: 18 Oct 2010

Posted 08 December 2014 - 09:56 AM

More in-camera RAW processing seems like a bad thing. Reminds me of the Nikon "star-eating" fiasco. If this is how the dark current has been reduced, I don't think it's such a big step forward. Wouldn't we rather see the changes happen at the sensor level?




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.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics