Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
10-07-2013, 05:33 PM   #1
Senior Member
amateur6's Avatar

Join Date: Mar 2007
Location: Rhode Island
Photos: Gallery
Posts: 176
Sorry if it's been addressed somewhere... Sensor size?

I haven't read through ALL of the K-3 threads, so feel free to shut me down if this has been discussed:

I just (today!) finished telling my students in an Intro to Digital Photo class about the "Megapixel Myth" (see Ken Rockwell, David Pogue, et al.) -- and lo and behold, I also happen to spot the news about the K-3 (well, I'm not really in the market for a new body, sooooo). "Well," I think, "I wonder what size sensor it is?"

And it's an APS-C, 23.5 x 15.6mm -- JUST like all of the other K-series. But it boasts 24mp, compared to the K-5 and K-50's 16mp (and the K-r's 12!)...

Considering that I just finished telling my students that given an equal sensor size, the better quality image will always be the LOWER mp -- what gives? Now, granted there are lots of other things that make the K-3 a vast improvement over the others -- but why boost the mp count? Am I missing something?

Thanks, folks!

10-07-2013, 05:37 PM   #2
Pentaxian
twitch's Avatar

Join Date: Nov 2009
Posts: 4,571
QuoteOriginally posted by amateur6 Quote
Considering that I just finished telling my students that given an equal sensor size, the better quality image will always be the LOWER mp -- what gives? Now, granted there are lots of other things that make the K-3 a vast improvement over the others -- but why boost the mp count? Am I missing something?

Thanks, folks!
It's a common myth that increasing MP increases noise, it's not true, not when output size is equalised.
10-07-2013, 05:48 PM   #3
Site Supporter
Site Supporter
jatrax's Avatar

Join Date: May 2010
Location: Oregon
Photos: Gallery | Albums
Posts: 10,825
QuoteOriginally posted by amateur6 Quote
given an equal sensor size, the better quality image will always be the LOWER mp -- what gives?
I've no idea if that is or was ever true.

But in this case we may be dealing with the same sensor size, but most definitely not with the same sensor generation. It is far too early to make any IQ claims, but what has been posted so far seems to indicate a modest improvement over the 16MP Sony sensor used in previous cameras. So assuming your theory is correct and the smaller MP has better quality then this sensor must be dramatically better than the Sony 16MP in order to get the same IQ with 50% more MP.
10-07-2013, 05:58 PM   #4
Pentaxian




Join Date: Oct 2009
Location: Far North Qld
Posts: 3,294
Has the same 24mp Toshiba sensor as the D7100. So I suppose we could use the images produced by the D7100 as a bit of a guide until we see samples from the K-3?
Full-size sample photos from Nikon D7100

10-07-2013, 06:30 PM   #5
Administrator
Site Webmaster
Adam's Avatar

Join Date: Sep 2006
Location: Arizona
Photos: Gallery | Albums
Posts: 43,668
QuoteOriginally posted by amateur6 Quote
I haven't read through ALL of the K-3 threads, so feel free to shut me down if this has been discussed:

I just (today!) finished telling my students in an Intro to Digital Photo class about the "Megapixel Myth" (see Ken Rockwell, David Pogue, et al.) -- and lo and behold, I also happen to spot the news about the K-3 (well, I'm not really in the market for a new body, sooooo). "Well," I think, "I wonder what size sensor it is?"

And it's an APS-C, 23.5 x 15.6mm -- JUST like all of the other K-series. But it boasts 24mp, compared to the K-5 and K-50's 16mp (and the K-r's 12!)...

Considering that I just finished telling my students that given an equal sensor size, the better quality image will always be the LOWER mp -- what gives? Now, granted there are lots of other things that make the K-3 a vast improvement over the others -- but why boost the mp count? Am I missing something?

Thanks, folks!
I've written an article that gets very technical and talks about noise performance and all that.. It should help answer some of your questions

https://www.pentaxforums.com/forums/photography-articles/228535-how-why-senso...s-compact.html

Adam
PentaxForums.com Webmaster (Site Usage Guide | Site Help | My Photography)



PentaxForums.com's high server and development costs are user-supported. You can help cover those costs by donating. Or, buy your photo gear from our affiliates, Adorama, B&H Photo, or Topaz Labs, and get FREE Marketplace access - click here to see how! Trusted Pentax retailers:

10-07-2013, 06:38 PM   #6
Pentaxian




Join Date: Jul 2010
Location: Oregon
Photos: Albums
Posts: 1,418
QuoteOriginally posted by amateur6 Quote
given an equal sensor size, the better quality image will always be the LOWER mp
This is obviously incorrect. What if an APS-C sensor had five pixels? Would it have the highest image quality ever?
10-07-2013, 07:20 PM   #7
Veteran Member




Join Date: Sep 2013
Location: Pasadena, CA
Photos: Albums
Posts: 1,118
I think the per-pixel noise can increase with the pixel density, but it's not a simple relation. It will be a matter of a crosstalk between the pixels and amplifying circuitry, shot noise (ie number of photons on the photo site) and the quality of the amplifiers. If there are fixed size elements that need to be added around the pixels, they will take away the net photosensitive area, that will affect the shot noise. This is probably not an issue with the Sony sensor's since they have all the circuitry on another layer, below photosites. If the pixels are close to each other, there could be an electromagnetic field affecting the charge in neighbouring photosites. Then there are other issues like how good of a microlens can you make when it gets smaller and smaller - will it be as effective in concentrating the photons on a photosite?

When you normalize for the output, you will average the noise out, but whether it will be the same or different, will depend on all those details above, and surely some more as well. Technically it should be the same, but hardware binning seems to be better than software binning. I am not sure why at the moment, but I would bet it happens due to digitization of a signal - adding two analog signals and then digitizing is going to be better than adding two digital signals, especially with few bits of depth (ie in the dark areas of an image).
10-08-2013, 09:38 AM   #8
Forum Member




Join Date: Apr 2012
Posts: 84
You forgot that technology keeps improving everyday. Giving sensors in the same generation, your thought is correct only in poor light condition. If you are shooting in good light condition, more pixels mean more details.

10-09-2013, 12:24 AM   #9
Veteran Member




Join Date: Sep 2013
Location: Pasadena, CA
Photos: Albums
Posts: 1,118
QuoteOriginally posted by Reltih Quote
You forgot that technology keeps improving everyday. Giving sensors in the same generation, your thought is correct only in poor light condition. If you are shooting in good light condition, more pixels mean more details.
I don't know if you are replying to my post, but I am just talking about the noise, not image quality overall.
10-09-2013, 12:30 AM   #10
Administrator
Site Webmaster
Adam's Avatar

Join Date: Sep 2006
Location: Arizona
Photos: Gallery | Albums
Posts: 43,668
QuoteOriginally posted by rrstuff Quote
I don't know if you are replying to my post, but I am just talking about the noise, not image quality overall.
Noise is directly influenced by other aspects of the sensor, such as dynamic range and color depth. Generally speaking, noise performance gets better in newer-gen sensors, but there will always be a big gap between FF and APS-C assuming the generation is the same.

Now, as for the 24Mp sensor, all it needs is slightly newer tech and it should be able to have a better signal to noise ratio than the 16Mp chip when scaled to the same size. 100% crops will probably be very close, if not noisier than the 16mp's.

Have you seen the article I linked above?

Adam
PentaxForums.com Webmaster (Site Usage Guide | Site Help | My Photography)



PentaxForums.com's high server and development costs are user-supported. You can help cover those costs by donating. Or, buy your photo gear from our affiliates, Adorama, B&H Photo, or Topaz Labs, and get FREE Marketplace access - click here to see how! Trusted Pentax retailers:

10-09-2013, 01:03 AM   #11
Veteran Member




Join Date: Sep 2013
Location: Pasadena, CA
Photos: Albums
Posts: 1,118
QuoteOriginally posted by Adam Quote
Noise is directly influenced by other aspects of the sensor, such as dynamic range and color depth. Generally speaking, noise performance gets better in newer-gen sensors, but there will always be a big gap between FF and APS-C assuming the generation is the same.

Now, as for the 24Mp sensor, all it needs is slightly newer tech and it should be able to have a better signal to noise ratio than the 16Mp chip when scaled to the same size. 100% crops will probably be very close, if not noisier than the 16mp's.

Have you seen the article I linked above?
Yes, but I don't think we are talking about the same thing. I listed some of the variables and design constraints that affect sensor noise with increasing pixel density, purely from an engineering perspective. I am not really giving any opinion on whether it will be better or not with the next generation sensors.
10-09-2013, 01:10 AM   #12
Veteran Member




Join Date: Sep 2013
Location: Pasadena, CA
Photos: Albums
Posts: 1,118
If I was to give an opinion about whether 24mpix will look better on the print or not, purely in terms of noise, I would look into the comparison of D7100 and D7000, which use the sensors in K-3 and K-5 respectively. Just following DXO data, the ISOs on D7100 seem to be set a bit lower than on D7000 to maintain the same signal to noise ratio. So overall, the camera needs more light to get the same ratio of signal to noise, which indicates that D7100 is a bit more 'noisy' - as the amount of noise decreases with greater amount of light.

I don't find that surprising because the 24mp and 16mp sensors are designed using a very similar technology.
Tests and reviews for the camera Nikon D7000 - DxOMark
Tests and reviews for the camera Nikon D7100 - DxOMark

It won't matter for practical purposes though. The differences are very slight.
10-09-2013, 02:04 AM - 1 Like   #13
Forum Member




Join Date: Apr 2012
Posts: 84
QuoteOriginally posted by rrstuff Quote
I don't know if you are replying to my post, but I am just talking about the noise, not image quality overall.
I intended to reply amateur6. With new technology, noise will be better so we may have same noise on sensor with more MP (assuming the MP gap is not very big for example from 16 to 24).
10-09-2013, 02:22 AM   #14
Pentaxian
Digitalis's Avatar

Join Date: Mar 2009
Location: Adelaide.
Posts: 8,878
QuoteOriginally posted by twitch Quote
It's a common myth that increasing MP increases noise, it's not true, not when output size is equalised.
+1 this, though with a few caveats: this premise works under the assumption that sensor technologies are improving - eventually we will hit the ceiling. And at that point a physically bigger sensor will always be better than a smaller one (assuming the same or better tech has been used)
the funny thing is at high ISO settings the Canon 1Ds MK II* performs worse than the K5IIs, even though the MKII has a full frame sensor with basically the same pixel count as the K5IIs the newer tech used in the K5IIs wipes the floor with the older camera.

* The differences are most noticeable at 1600>3200 as the MKII tops out at ISO 3200, and the K5IIs is just getting started in that ISO range. The Canon 1Ds MK II camera was announced in 2004, released for review in 2005.

Last edited by Digitalis; 10-09-2013 at 02:35 AM.
10-09-2013, 02:44 AM   #15
Site Supporter
Site Supporter
Stone G.'s Avatar

Join Date: Jun 2010
Location: North Zealand, Denmark
Photos: Gallery | Albums
Posts: 1,516
QuoteOriginally posted by amateur6 Quote
Considering that I just finished telling my students that given an equal sensor size, the better quality image will always be the LOWER mp -- what gives?
Image Quality cannot really be measured quantitavely and there are many things that contribute towards "better quality images" as we percieve them.

Leaving the lens focal length and absolute aperture out of the equation (which in itself is an over-simplification) and only looking at the sensor properties, you will have to consider absolute size, pixel size and resolution (sampling), statistical and electrical noise, material properties/sensor sensitivity (which is not "ISO"). And you further have to consider how manufacturers sample and process their raw files (because no raw file will ever be truly "RAW").

But in one aspect you have a certain point: The fundamental laws of nature cannot be changed and the inevitable, statistical noise will always be there and will always depend upon the pixel size/pitch:

Clarkvision: Does Pixel Size Matter
Reply

Bookmarks
  • Submit Thread to Facebook Facebook
  • Submit Thread to Twitter Twitter
  • Submit Thread to Digg Digg
Tags - Make this thread easier to find by adding keywords to it!
dslr, k-3, k3, mp, pentax k-3, sensor, size, students
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Nature It's Been A While...... Rupert Post Your Photos! 10 11-20-2012 07:39 AM
Sensor Size and Pixel Size interested_observer Photographic Technique 5 02-04-2012 07:59 AM
Macro It's here somewhere! eaglem Post Your Photos! 4 10-30-2010 04:08 PM
16-45 or 17-70? (and sorry if this has been discussed before) Finn Pentax SLR Lens Discussion 15 01-02-2009 04:59 AM



All times are GMT -7. The time now is 05:22 AM. | See also: NikonForums.com, CanonForums.com part of our network of photo forums!
  • Red (Default)
  • Green
  • Gray
  • Dark
  • Dark Yellow
  • Dark Blue
  • Old Red
  • Old Green
  • Old Gray
  • Dial-Up Style
Hello! It's great to see you back on the forum! Have you considered joining the community?
register
Creating a FREE ACCOUNT takes under a minute, removes ads, and lets you post! [Dismiss]
Top