CCD VS CMOS

The place for questions about shooting with Blackmagic Cameras.
  • Author
  • Message
Offline

paulkosmala

  • Posts: 148
  • Joined: Mon Sep 17, 2012 10:50 am

CCD VS CMOS

PostTue Dec 04, 2012 6:16 am

Age old discussion. I bring it up, because a new camera is in the works using a 4k CCD sensor.


Pros for CCD:
3 chip gives true color information vs debayering.
no moire due to lack of pixel space
technically higher image quality.

Pros for CMOS:
cheap.
'better' low light.
easier to make larger.

At 4/3s a 3chip CCD sensor might be an interesting take on 2.5k raw (actually 2.5k too)... just saying. I'm not an engineer, but Is there a horrible flaw in this idea?
Offline
User avatar

Lorenzo Straight

  • Posts: 267
  • Joined: Wed Aug 22, 2012 8:52 am
  • Location: Atlanta, Ga

Re: CCD VS CMOS

PostTue Dec 04, 2012 7:46 am

What camera?
MSI GS60 Ghost Pro-4K-605-2.7 GHz Intel Core i7-5700HQ (Broadwell)-16GB of 1600 MHz DDR3L RAM-1TB 7200 rpm HDD + 128GB M.2 SSD-NVIDIA GeForce GTX 970M GPU (6GB GDDR5)-15.6" UHD + 4K IPS Display-3840 x 2160
-Windows 8.1
Offline

Nick Smith

  • Posts: 76
  • Joined: Sun Aug 26, 2012 5:37 pm

Re: CCD VS CMOS

PostTue Dec 04, 2012 2:31 pm

I think 'cheap' is the main pro as for why CMOS has been used. It really is an awfully cheap camera for what it is.
Offline

Bocian

  • Posts: 2
  • Joined: Tue Nov 13, 2012 12:29 pm

Re: CCD VS CMOS

PostTue Dec 04, 2012 2:37 pm

I could be wrong, but I think that all CCD sensors have a global shutter.
Offline
User avatar

sean mclennan

  • Posts: 1435
  • Joined: Wed Aug 22, 2012 8:28 pm
  • Location: Toronto, ON

Re: CCD VS CMOS

PostTue Dec 04, 2012 4:15 pm

CCDs use WAY more power than CMOS sensors. That's about all I know.

Zach Brinkerhoff

Re: CCD VS CMOS

PostWed Dec 05, 2012 11:08 am

Here is a comparison video between CCD vs. CMOS

For creating Visual Effects on live action footage, I hate Rolling Shutter.

Offline

paulkosmala

  • Posts: 148
  • Joined: Mon Sep 17, 2012 10:50 am

Re: CCD VS CMOS

PostTue Jan 08, 2013 3:58 am

Lorenzo Straight wrote:What camera?



the aaton penelope delta

It's 16-bit uncompressed raw with a s35 sensor.

took me a long time to find it again. there's a vimeo video for it.
Offline

paulkosmala

  • Posts: 148
  • Joined: Mon Sep 17, 2012 10:50 am

Re: CCD VS CMOS

PostTue Jan 08, 2013 4:01 am

paulkosmala wrote:
Lorenzo Straight wrote:What camera?



the aaton penelope delta

It's 16-bit uncompressed raw with a s35 sensor.

took me a long time to find it again. there's a vimeo video for it.


also 3.5k and 14stops dynamic range. mechanical shutter, and optical viewfinder.
Offline

paulkosmala

  • Posts: 148
  • Joined: Mon Sep 17, 2012 10:50 am

Re: CCD VS CMOS

PostTue Jan 08, 2013 4:04 am

paulkosmala wrote:
paulkosmala wrote:
Lorenzo Straight wrote:What camera?



the aaton penelope delta

It's 16-bit uncompressed raw with a s35 sensor.

took me a long time to find it again. there's a vimeo video for it.


also 3.5k and 14stops dynamic range. mechanical shutter, and optical viewfinder.


also - it apparently shakes the sensor to randomize the noise structure and give more spacial information.
http://nofilmschool.com/2012/10/aatons- ... ta-camera/


and it'll cost about $120,000
Offline

John Brawley

  • Posts: 4289
  • Joined: Tue Aug 21, 2012 7:57 am
  • Location: Los Angeles California

Re: CCD VS CMOS

PostTue Jan 08, 2013 11:45 am

paulkosmala wrote:
also - it apparently shakes the sensor to randomize the noise structure and give more spacial information.
http://nofilmschool.com/2012/10/aatons- ... ta-camera/


and it'll cost about $120,000


It doesn't shake it. It shifts it frame to frame to offset by a pixel and a 7K file can be interpolated in post from a 3.5K file.

jb
John Brawley ACS
Cinematographer
Currently - Los Angeles
Offline

paulkosmala

  • Posts: 148
  • Joined: Mon Sep 17, 2012 10:50 am

Re: CCD VS CMOS

PostTue Jan 08, 2013 12:31 pm

John Brawley wrote:
paulkosmala wrote:
also - it apparently shakes the sensor to randomize the noise structure and give more spacial information.
http://nofilmschool.com/2012/10/aatons- ... ta-camera/


and it'll cost about $120,000


It doesn't shake it. It shifts it frame to frame to offset by a pixel and a 7K file can be interpolated in post from a 3.5K file.

jb


does it physically move? or if not, then wouldn't it be possible to do the same process in post?
Offline

John Brawley

  • Posts: 4289
  • Joined: Tue Aug 21, 2012 7:57 am
  • Location: Los Angeles California

Re: CCD VS CMOS

PostTue Jan 08, 2013 12:33 pm

paulkosmala wrote:
John Brawley wrote:
paulkosmala wrote:
also - it apparently shakes the sensor to randomize the noise structure and give more spacial information.
http://nofilmschool.com/2012/10/aatons- ... ta-camera/


and it'll cost about $120,000


It doesn't shake it. It shifts it frame to frame to offset by a pixel and a 7K file can be interpolated in post from a 3.5K file.

jb


does it physically move? or if not, then wouldn't it be possible to do the same process in post?


Some companies like panasonic do it without moving the sensor. (pixel shifting)

Aaton physically move the sensor.

There are other more interesting features...i like the multi finger shutter design...

http://johnbrawley.wordpress.com/2012/0 ... -penelope/

jb
John Brawley ACS
Cinematographer
Currently - Los Angeles
Offline

paulkosmala

  • Posts: 148
  • Joined: Mon Sep 17, 2012 10:50 am

Re: CCD VS CMOS

PostTue Jan 08, 2013 12:36 pm

paulkosmala wrote:
John Brawley wrote:
paulkosmala wrote:
also - it apparently shakes the sensor to randomize the noise structure and give more spacial information.
http://nofilmschool.com/2012/10/aatons- ... ta-camera/


and it'll cost about $120,000


It doesn't shake it. It shifts it frame to frame to offset by a pixel and a 7K file can be interpolated in post from a 3.5K file.

jb


does it physically move? or if not, then wouldn't it be possible to do the same process in post?


shake was a bit of hyperbole. if it is offsetting the sensor after each frame than I guess 'vibrate' would be a more accurate word. Right now there's not a lot of literature of that feature.
Offline

paulkosmala

  • Posts: 148
  • Joined: Mon Sep 17, 2012 10:50 am

Re: CCD VS CMOS

PostTue Jan 08, 2013 12:40 pm

jb[/quote]

does it physically move? or if not, then wouldn't it be possible to do the same process in post?[/quote]

Some companies like panasonic do it without moving the sensor. (pixel shifting)

Aaton physically move the sensor.

There are other more interesting features...i like the multi finger shutter design...

http://johnbrawley.wordpress.com/2012/0 ... -penelope/

jb[/quote]

the new shutter to reduce iso... Its good enough to wonder why no one else thought of that...
it's 2013, really... why has no one else thought of that in 10+ years of digital cinema?
Offline

John Brawley

  • Posts: 4289
  • Joined: Tue Aug 21, 2012 7:57 am
  • Location: Los Angeles California

Re: CCD VS CMOS

PostTue Jan 08, 2013 12:42 pm

Well it's not really the correct language. It doesn't move it during the exposure. It moves it INBETWEEN frames a tiny amount. That's not vibrating it or shaking it which infers its more random and done during exposures.

Many early 35mm motion picture lens relays and adapters featured a vibrating ground glass to try and hide the texture of the glass.

This is nothing like that.

JB.
John Brawley ACS
Cinematographer
Currently - Los Angeles
Offline

John Brawley

  • Posts: 4289
  • Joined: Tue Aug 21, 2012 7:57 am
  • Location: Los Angeles California

Re: CCD VS CMOS

PostTue Jan 08, 2013 12:45 pm

paulkosmala wrote:
the new shutter to reduce iso... Its good enough to wonder why no one else thought of that...
it's 2013, really... why has no one else thought of that in 10+ years of digital cinema?


Well...

Aaton have always had been the better designed cameras in my opinion, but most of the world disagree with me and choose an Arri instead.

Aaton did timecode on film first. They did Super 16 first. They did magnetic magazine drives, electrostatic heated viewfinders, electronic magazine identification, counter rotating viewfinder shutter, lateral film support, film movements that DON't use a registration pin...the list goes on....

jb
John Brawley ACS
Cinematographer
Currently - Los Angeles
Offline
User avatar

Noel Sterrett

  • Posts: 521
  • Joined: Wed Aug 22, 2012 9:12 pm
  • Location: Atlanta

Re: CCD VS CMOS

PostTue Jan 08, 2013 1:08 pm

John Brawley wrote:It doesn't shake it. It shifts it frame to frame to offset by a pixel and a 7K file can be interpolated in post from a 3.5K file.

jb

It's a lovely camera, but if it shifts frame to frame then unless there is no movement whatsoever between frames, the resolution is not changed.

As an analogy, 60i = 30P if and only if there is no movement between fields.

Cheers.
Admit One Pictures
Resolve Studio 18 | Linux Lint 21 | Nvidia 515 | Xeon | iCore | Ryzen
Offline
User avatar

Michael Sandiford

  • Posts: 308
  • Joined: Wed Aug 22, 2012 8:48 am
  • Location: Hereford

Re: CCD VS CMOS

PostTue Jan 08, 2013 1:14 pm

They also make the best looking cameras for me as well. I mean look at it
Image
It's like Frank Gehry tries Bauhaus
Offline

John Brawley

  • Posts: 4289
  • Joined: Tue Aug 21, 2012 7:57 am
  • Location: Los Angeles California

Re: CCD VS CMOS

PostTue Jan 08, 2013 1:18 pm

Noel Sterrett wrote:
John Brawley wrote:It doesn't shake it. It shifts it frame to frame to offset by a pixel and a 7K file can be interpolated in post from a 3.5K file.

jb

It's a lovely camera, but if it shifts frame to frame then unless there is no movement whatsoever between frames, the resolution is not changed.

Cheers.


My understanding is that it's an INTERPOLATED 7K image from a 3.5k file by ERGON, Aaton's post production workstation.

jb
John Brawley ACS
Cinematographer
Currently - Los Angeles
Offline
User avatar

Noel Sterrett

  • Posts: 521
  • Joined: Wed Aug 22, 2012 9:12 pm
  • Location: Atlanta

Re: CCD VS CMOS

PostTue Jan 08, 2013 1:31 pm

John Brawley wrote:My understanding is that it's an INTERPOLATED 7K image from a 3.5k file by ERGON, Aaton's post production workstation.

jb

I understand. But the fact is it has the same problem as HDR - the timing is off.

"Skyfall" is on its way to $1B with 2.8K. I'd rather better pixels than just more, and Aaton has that without parlor tricks.

Cheers.
Admit One Pictures
Resolve Studio 18 | Linux Lint 21 | Nvidia 515 | Xeon | iCore | Ryzen
Offline

John Brawley

  • Posts: 4289
  • Joined: Tue Aug 21, 2012 7:57 am
  • Location: Los Angeles California

Re: CCD VS CMOS

PostTue Jan 08, 2013 1:36 pm

Noel Sterrett wrote:
John Brawley wrote:My understanding is that it's an INTERPOLATED 7K image from a 3.5k file by ERGON, Aaton's post production workstation.

jb

I understand. But the fact is it has the same problem as HDR - the timing is off.


Perhaps it happens with the frame cycle.

I've not actually seen a 7K file from it yet as that's still "secret" and I may well be mis-representing how it does it.

Noel Sterrett wrote:"Skyfall" is on its way to $1B with 2.8K. I'd rather better pixels than just more, and Aaton has that without parlor tricks.


It's a moot point isn't it. We haven't really seen what motion footage looks like yet. It seems like it might be better on paper...

jb
John Brawley ACS
Cinematographer
Currently - Los Angeles
Offline
User avatar

Noel Sterrett

  • Posts: 521
  • Joined: Wed Aug 22, 2012 9:12 pm
  • Location: Atlanta

Re: CCD VS CMOS

PostTue Jan 08, 2013 1:56 pm

John Brawley wrote:It's a moot point isn't it. We haven't really seen what motion footage looks like yet. It seems like it might be better on paper...

jb

They did post a few DNG clips, one with motion, but it will be interesting to see how their shutter affects various kinds of motion.

Timing is everything.

Cheers.
Admit One Pictures
Resolve Studio 18 | Linux Lint 21 | Nvidia 515 | Xeon | iCore | Ryzen

Return to Cinematography

Who is online

Users browsing this forum: No registered users and 119 guests