Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
12-28-2015, 12:53 PM   #76
Pentaxian
RonHendriks1966's Avatar

Join Date: Jan 2009
Photos: Gallery
Posts: 8,556
QuoteOriginally posted by falconeye Quote
I checked one of your results (K-3, 1640 e-/Ám^2). So, looks correct. Anyway, these numbers are derived from DxO measurements by fitting a physics model. I'd take them with a grain of salt, systematical errors are at least 10%.

And no, it isn't compensated by more pixels as you already normalized per unit area. The K-5 seems to have more well capacity and it shows in the DxO landscape score. BTW, smaller pixels do not, by themselves, lead to smaller or larger well capacity per unit area. Rather, it is a result of pixel design, silicon design rules and the manufacturing process. More well capacity increases cost of manufacture.
Realized that it was per unit area. Is this where the magic of back iluminated would kick in. More pixels, is more wires and thus old fashion sensors like K-3 have less area to catch light? But still, 25 % better performance for the K-5 is big. The K-01 is great, just missing the iso80 setting of the K-5, whiel not fullfilling the top-tier camera spot.

Did they cheap out on the sensor specifications for the K-3 to make the camera cheaper? And thus being behind a camera like the Nikon D7200?

12-28-2015, 01:09 PM   #77
Pentaxian




Join Date: Nov 2013
Posts: 4,555
QuoteOriginally posted by blende8 Quote
No.
According to my logic the K5 should produce the better result, because every single pixel gets more light.

I am not sure if I am right, it does not sound right, but I do not find the error.
At same iso, the light intensity is constant per surface area so if you have more surface area, you get more light and so a stronger signal.

Regardless of photosite size, the whole picture will gather 2.3 time more light on an FF than on an APSC because the FF surface is 2.3 time greater. That's why basically a D800 has 2.3 time the high iso score as a K5 but this is also the case of D600 that has bigger pixels. Pixel size has a smaller influence.
12-28-2015, 01:11 PM   #78
Pentaxian
falconeye's Avatar

Join Date: Jan 2008
Location: Munich, Alps, Germany
Photos: Gallery
Posts: 6,862
QuoteOriginally posted by RonHendriks1966 Quote
Is this where the magic of back iluminated would kick in. ...
Don't know the answers.
BSI does primarily help with the fill factor for smaller-sized pixels, not with well capacity.
12-28-2015, 01:27 PM   #79
Pentaxian
RonHendriks1966's Avatar

Join Date: Jan 2009
Photos: Gallery
Posts: 8,556
QuoteOriginally posted by falconeye Quote
Don't know the answers.
BSI does primarily help with the fill factor for smaller-sized pixels, not with well capacity.
We will see what the K-1 will bring.

12-28-2015, 01:50 PM   #80
Pentaxian
thibs's Avatar

Join Date: Jun 2007
Location: Belgium
Photos: Albums
Posts: 5,161
The difference between K-5 and K-01 is puzzling IMO
12-28-2015, 02:25 PM   #81
Pentaxian
RonHendriks1966's Avatar

Join Date: Jan 2009
Photos: Gallery
Posts: 8,556
QuoteOriginally posted by thibs Quote
The difference between K-5 and K-01 is puzzling IMO
Basicly that is due to the iso80 on the K-5.
12-28-2015, 03:36 PM - 1 Like   #82
Senior Member
Matthew Saville's Avatar

Join Date: Jan 2014
Location: Southern California
Posts: 250
Now, if only the full-frame Pentax has similarly impressive high ISO and dynamic range performance, I'll be a very happy man.
12-29-2015, 01:38 AM   #83
Pentaxian




Join Date: Nov 2013
Posts: 4,555
QuoteOriginally posted by Matthew Saville Quote
Now, if only the full-frame Pentax has similarly impressive high ISO and dynamic range performance, I'll be a very happy man.
There no reason for it to perform any better than latest Sony FFs sensors so basically in the 2400-3500 range.

12-29-2015, 03:35 AM - 1 Like   #84
Loyal Site Supporter
clackers's Avatar

Join Date: Jul 2013
Location: Melbourne
Photos: Albums
Posts: 7,723
QuoteOriginally posted by Nicolas06 Quote
.

Regardless of photosite size, the whole picture will gather 2.3 time more light on an FF than on an APSC because the FF surface is 2.3 time greater. That's why basically a D800 has 2.3 time the high iso score as a K5 but this is also the case of D600 that has bigger pixels. Pixel size has a smaller influence.
Oh, Nicolas, you're too smart to fall for this Total Light BS.

Your picture is made up of pixels - individual wells.

Their capacity, the number of them, the electronics behind them, the dissipation of heat in the camera body and suppression of stray electric currents in the circuitry determine the noise in the image.

Sensor size is *not* the reason, because light intensity is independent of area.

If you take a crop of a FF image, the noise is exactly the same.
12-29-2015, 04:09 AM   #85
Loyal Site Supporter




Join Date: Mar 2009
Location: Southern Indiana
Photos: Gallery
Posts: 14,957
QuoteOriginally posted by clackers Quote
Oh, Nicolas, you're too smart to fall for this Total Light BS.

Your picture is made up of pixels - individual wells.

Their capacity, the number of them, the electronics behind them, the dissipation of heat in the camera body and suppression of stray electric currents in the circuitry determine the noise in the image.

Sensor size is *not* the reason, because light intensity is independent of area.

If you take a crop of a FF image, the noise is exactly the same.
Are you talking about dynamic range or SNR? Because looking at the graphs, SNR is pretty similar between full frame sensors of a given generation, but dynamic range at various isos can vary quite a bit, based (I guess) on the photon well size, etc. I really have no idea what all this means, except that with sensors with more dynamic range, I have found that I can pull the shadows up a lot harder than with cameras with less dynamic range.
12-29-2015, 04:14 AM   #86
Veteran Member
blende8's Avatar

Join Date: Feb 2007
Location: Bremen, Germany
Photos: Albums
Posts: 1,484
QuoteOriginally posted by Nicolas06 Quote
At same iso, the light intensity is constant per surface area so if you have more surface area, you get more light and so a stronger signal.
No.
Photographing the same scene, photon/pixel ratio gets higher with smaller sensor.
12-29-2015, 04:38 AM   #87
Pentaxian




Join Date: Dec 2011
Posts: 1,157
QuoteOriginally posted by clackers Quote
Sensor size is *not* the reason, because light intensity is independent of area.
If you take a crop of a FF image, the noise is exactly the same.
You are certainly right. The whole pseudo-scientific BS we read here again and again obviously doesn't help people understand that the actual logics (not theories) you need to know are extremely simple.

1. "Noise" results in "ugly spots" on a photo. If you enlarge an ugly spot it gets uglier because: You perceive it more easily.

As a 36x24 (small format or "FF") sensor is smaller than a 44x33mm sensor you need to magnify the image more if you want to hang it on your wall in the same size. Thus you perceive more noise (the spots being larger) while the actual noise level per area is the exact same when using the same exposure. It's all about magnification of the final image if you want an easy to understand explanation.

If you would simply print your 36x24mm picture smaller by the crop factor 1,3 and hang it right beside the other one but still watch it from the same chair in the room, even the perceived noise stays the same.

2. The actual noise level per area only depends on technology and this one changes with each and every sensor.

That is why the max dynamic range of an old APSC sensor in the K-3 matches the max dynamic range of the brandnew A7RII. This has nothing to do with logic just with (IMHO) poor Sony chip design. In all "theories" the A7RII would require to be much better. But it isn't. That's why you need to look at actual measurements in every case and why formulae don't really help beyond old fashioned rule of thumb quality level.

3. Then we have pixel size. Actually this directly links to item #1 above and to the pixel count usefulness. Obviously a large pixel per se is better in measurements than a small one as it can capture more photons. But that is counterered by getting fewer of them on the same total space.
The one bit few people understand is that your eyes can only resolve so much (about 6 MPx for good sighted person viewing with a 43mm FF viewing angle = distance is the same as image diagonal).
So if you want to benefit anything from your marketing gaga 36/42/50 Mpx pixel count beyond what you get from a straight 6 MPx picture, you need to get closer to the image. And that is? Right, magnifying it.
And what is the result of this magnification? Right, more perceived noise.
So: Once you benefit (this absolutely requires to get closer) from any resolution beyond 6 x 2,25 = 13,5 MPx you actually get to perceive (remember the per area noise on same tech level is completely ignorant of sensor size) more noise on FF than on APSC if printed to the same size.

Yes, the ability to crop a picture in post (=digital zoom) deliberately can be a useful tool by itself. But it has the same consequences as moving closer: magnification --> more noise, less dynamic range.

And as cropping equals digital zoom obviously any 100% viewing means a lot of magnification. The more pixel count the more magnification.

If you followed this you will understand that instead of any funny theories all you need to understand is:
If you magnify a picture from sensor capture to result, the perceived (!) noise goes up, the perceived dynamic range goes down, the perceived background blur goes up. It's all about perception.
You then just need to check if your scenario involves magnification. And if you do comparisons you check which one magnifies more. With more magnification you can predict (that is: rule of thumb as the tech will be different in every camera) more perceived noise, less dynamic range.
12-29-2015, 06:47 AM   #88
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 23,772
QuoteQuote:
The argument is that if both sensor share similar technology you would expect the results to match by the rougly the scale factor of the surface area of the sensor. Either same isos for both, the 645Z is better by the factor applied to the different metrics either you apply the factor on isos and you get the same results. It is exactly the same concepts that give the equivalent apperture, focal length and high iso performance.
The forum consensus was that 5 points on DxO was observable, less than that, not so much. The K-5 was noticeably better than the D7000, the D7200 is noticeably better than the K-3. A camera system is a lot more than a sensor. IN fact if you look at systems like the D4x 5d mkII, they have relatively low MP but relatively high resolution and performance. There's a lot that can be done to an image in between the raw data being captured and a DNG file being output. The more expensive cameras tend to have more expensive image processors.

So, within certain limits yes, but those limits can be observable without micro-analysis. The K-5 image processing was the on APS-c best at the time. The K-3 has fallen behind. No one knows what Pentax will come up with for the FF.

But just as most Canon users could care less if their sensors are DxO poorly rated, this is not something you have to care about, in fact it's something you shouldn't care about. If folks can take great pictures with those 12 EV DR Canons, and a lot of pros do, obviously there's a lot more to taking pictures than a high DxO score.

That would be my first question to ask DxO.

"How is it that so many people shooting with cameras that don't rate highly on your scales, are making their livings from the images they sell, even when competing with people who shoot with cameras rated highly on you scales. Surely the bottom line is not a DxO rating, but "are you taking picture you and other people like." People will claim DxO ratings have something to do with that, but they can't prove it.

It's just something geeks like to talk about. It isn't real world knowledge.

Last edited by normhead; 12-29-2015 at 06:53 AM.
12-29-2015, 07:00 AM   #89
Senior Member
romeck's Avatar

Join Date: Oct 2012
Location: Piekary Śląskie
Photos: Albums
Posts: 156
QuoteOriginally posted by falconeye Quote
I checked one of your results (K-3, 1640 e-/Ám^2).....
Where you see these data?
Somewhere on sensorgen.info?
12-29-2015, 07:11 AM   #90
Pentaxian




Join Date: Nov 2013
Posts: 4,555
QuoteOriginally posted by blende8 Quote
No.
Photographing the same scene, photon/pixel ratio gets higher with smaller sensor.
The photon per surface area is fixed by exposure only. So as long as you keep the same apperture and shutter speed, regarless of framing, sensor size or focal length, the photon/pixel is strictly dependant of the physical size of pixels.

Because on a D800 photosites are rougly same size as on K5, the photons/photosite are also the same as long the apperture/shutter speed is kept identical. And on a D600 photosites are bigger so there more photons/pixel. This doesn't prevent the D810 to be slightly better.

Even if you change the apperture to keep the same deph of field, you'll need to change shutter speed to keep the same exposure anyway so the photon per surface area will remain constant for the same exposure and the photos per pixel will be strictly dependant of pixel size.
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!
645z, apples, base, capacity, claims, d810, design, dr, dxo, dxomark, files, ii, iso, k-3, k-5, k-7, lossless, noise, pentax, pentax 645z scores, pentax news, pentax rumors, pixels, score, scores, sensor, technology, unit
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Pentax 645z in DXOMark 101 points..., is it true? Ventzy Pentax Medium Format 32 01-01-2016 12:08 PM
K-S1 scores on DxOMark 6BQ5 Pentax K-S1 & K-S2 34 11-14-2014 07:52 AM
Red Epic Dragon jumps to top of DxOMark Sensor charts with score of 101 ! jogiba Non-Pentax Cameras: Canon, Nikon, etc. 14 03-10-2014 01:50 PM
Arrggh - Nikon V2 scores lower than V1 or RX100 on dxomark Christine Tham Non-Pentax Cameras: Canon, Nikon, etc. 36 12-24-2012 04:41 PM
DXOMark scores for the Q posted. . . snostorm Pentax Q 8 12-03-2011 01:47 PM



All times are GMT -7. The time now is 04:57 AM. | See also: NikonForums.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