Reading mode: Light Dark

K5 dxo mark up, remarkable score, 1160 high ISO

Go to last post
It's true, actually internal flare of lens and body already limits DR, this I guess they measure directly from sensor readings.

Now, it seems Pentax is doing something at 3200 ISO and up, the graphs show it.
[U] sem said:
rhlpetrus said:
I guessed the high ISO right, exactly 1 stop below D700's, using D7000's RAW files. Now, 14+ EVs of DR, wow, Sony knows how to make a good DR sensor, and it's getting better all the time!
Click to expand...
Well, good news, and the DxO site hardly manages ;)

I think the bare highest DR number at low ISO is not so relevant because other things are limiting factors in practice. The SNR, TR, CS curves almost overlap with D90 in the upper medium range. The useful improvement is that it does get better at lower ISO (when one has enough light to afford it). It doesn't seem to get close to D3s in limited light if the in-camera NR is discounted.
Click to expand...
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)[/U]
 
You may have already answered this for me, but wouldn't +1EV overexposure result in -1EV of dynamic range, making one lower on the Y-axis?
falconeyes said:
binary_eye said:
Why isn't the extended ISO 50 of the D3X included?
Click to expand...
Because D3X has no extended ISO 50. It's just +1EV overexposure. Both D3X ISO 100 and ISO 50 are true ISO 78. So, both points are plotted but appear as a single point in the plot.
Click to expand...
--
Correct my incorrect photography answers at:
http://photo.stackexchange.com/users/389/eruditass

Please update the Camera Shutter Life Expectancy Database with your experiences:
http://www.olegkikin.com/shutterlife/
 
Apparently not, since all of its measurements improve at 80 ISO. It very well may have a base ISO between 80 and 100. No one knows why Pentax markets it as extended, but the K-x was like this as well.
binary_eye said:
falconeyes said:
Because D3X has no extended ISO 50. It's just +1EV overexposure.
Click to expand...
I understand that, but how is the K-5 different? Is its ISO 80 not simply ISO 100 overexposed? If it's not, why does Pentax consider it an expanded ISO setting?
Click to expand...
--
Correct my incorrect photography answers at:
http://photo.stackexchange.com/users/389/eruditass

Please update the Camera Shutter Life Expectancy Database with your experiences:
http://www.olegkikin.com/shutterlife/
 
They should drop the extended isos below, it doesn't make sense. If one wants to overexpose image for some reason, just use manual exposure and do it. The D7000 and D3100 have dropped it. Canon don't use it as well. Just calibrate camera so that best DR is attained at lowest available ISO, period.
johnnyz86 said:
Apparently not, since all of its measurements improve at 80 ISO. It very well may have a base ISO between 80 and 100. No one knows why Pentax markets it as extended, but the K-x was like this as well.
binary_eye said:
falconeyes said:
Because D3X has no extended ISO 50. It's just +1EV overexposure.
Click to expand...
I understand that, but how is the K-5 different? Is its ISO 80 not simply ISO 100 overexposed? If it's not, why does Pentax consider it an expanded ISO setting?
Click to expand...
--
Correct my incorrect photography answers at:
http://photo.stackexchange.com/users/389/eruditass

Please update the Camera Shutter Life Expectancy Database with your experiences:
http://www.olegkikin.com/shutterlife/
Click to expand...
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
Very impressive and extremely tempting. Until now I thought I would change from Canon to Nikon D7000 but I guess the D7000 won't achieve the marks of the K-5 because the Sony SLT A-55 wasn't even as good as the D90.

But there is one thing that worries me about the K-5: The choice of lenses is by far not as versatile and as good as the Nikon or Canon glasses.

Hmm, will be a tough decision!
 
quintana said:
Very impressive and extremely tempting. Until now I thought I would change from Canon to Nikon D7000 but I guess the D7000 won't achieve the marks of the K-5 because the Sony SLT A-55 wasn't even as good as the D90.
Click to expand...
The SLT loses 30% of the light, though. D7000 does not.

However, D7000 does have a lot more hot pixels, K-5 does not.
 
Yes, they are here to stay!

Which is the difference between A55 and K5? On one hand, the pellicle mirror, so determined percentadge of light gets lost, so we have a constant that applies to the full sensitivity range of the sensor. On the other hand, Sony has a 12bit process while Pentax has a 14bit process. So we may have a limiting factor, as long as capturing precission of the sensor exceeds what 12 bits are able to represent.

What I see on those charts is, on one hand, a constant, possibly the ammount of light that hits the sensor, that makes the K5 instantly better, constantly better, at any sensitivity. Lines on the charts are absolutely parallel.

On the other hand, once the dynamic range has the chance to exceed the 12 EVs it seems the K5 DESTROYS the A55. A55 tops at 12,39EV and K5 at 14.12EV. It can not be a coincidence.

Fortunately for us all, K5 and D7000 have MUCH MORE in common than K5 or D7000 with A55. Don't kill me if I'm wrong, but if I have to bet... I bet D7000 will have mostly identical charts than the K5. Probably with more noise starting at ISO 3200, but retaining more detail.
 
Daviii said:
Fortunately for us all, K5 and D7000 have MUCH MORE in common than K5 or D7000 with A55. Don't kill me if I'm wrong, but if I have to bet... I bet D7000 will have mostly identical charts than the K5. Probably with more noise starting at ISO 3200, but retaining more detail.
Click to expand...
By the way, Thorn claimed that Nikon has tweaked Sony's sensor for D7000. They might have "tweaked" it wrong way, given far more reported hotpixels :)

We have to wait for D7000 and A580 scores now.
 
Nikon cameras that share their sensor with Pentax score higher than their Pentax counter-part. Thus, if anything I expect the D7000 to score even higher than the K-5
quintana said:
Very impressive and extremely tempting. Until now I thought I would change from Canon to Nikon D7000 but I guess the D7000 won't achieve the marks of the K-5 because the Sony SLT A-55 wasn't even as good as the D90.

But there is one thing that worries me about the K-5: The choice of lenses is by far not as versatile and as good as the Nikon or Canon glasses.

Hmm, will be a tough decision!
Click to expand...
 
quintana said:
Very impressive and extremely tempting. Until now I thought I would change from Canon to Nikon D7000 but I guess the D7000 won't achieve the marks of the K-5 because the Sony SLT A-55 wasn't even as good as the D90.

But there is one thing that worries me about the K-5: The choice of lenses is by far not as versatile and as good as the Nikon or Canon glasses.

Hmm, will be a tough decision!
Click to expand...
while the K5's scores is impressive, always think about the system as a whole, unless of course you're only intent for your DSLR is taking casual snapshots.

if primes go pentax, if zooms, go nikon
--
'when 900 years you reach, look as good you will not'
-- master yoda

http://jordanpaw.zenfolio.com
 
They will all trade blows and even out in the future. Pick based on a system, and stick with it. For my purposes, I chose the Pentax system (Nikon was 2nd choice) and am happy. Others may choose other systems for other reasons.
quintana said:
Very impressive and extremely tempting. Until now I thought I would change from Canon to Nikon D7000 but I guess the D7000 won't achieve the marks of the K-5 because the Sony SLT A-55 wasn't even as good as the D90.

But there is one thing that worries me about the K-5: The choice of lenses is by far not as versatile and as good as the Nikon or Canon glasses.

Hmm, will be a tough decision!
Click to expand...
--
Correct my incorrect photography answers at:
http://photo.stackexchange.com/users/389/eruditass

Please update the Camera Shutter Life Expectancy Database with your experiences:
http://www.olegkikin.com/shutterlife/
 
Don't think bits is the issue, D300/s had 14 bits and D90 didn't and tehre was basically no difference.

This sensor may be different from the one in the A55. Or the ADc system may be different, not just re 12 x 14 bits.

And toppings also affect results.

The K5 graphs show some RAW manipulation at and above 3200 ISO and DxO mark seems to have detetcted it, they are using diffrent symbols for points in that range.
Daviii said:
Yes, they are here to stay!

Which is the difference between A55 and K5? On one hand, the pellicle mirror, so determined percentadge of light gets lost, so we have a constant that applies to the full sensitivity range of the sensor. On the other hand, Sony has a 12bit process while Pentax has a 14bit process. So we may have a limiting factor, as long as capturing precission of the sensor exceeds what 12 bits are able to represent.

What I see on those charts is, on one hand, a constant, possibly the ammount of light that hits the sensor, that makes the K5 instantly better, constantly better, at any sensitivity. Lines on the charts are absolutely parallel.

On the other hand, once the dynamic range has the chance to exceed the 12 EVs it seems the K5 DESTROYS the A55. A55 tops at 12,39EV and K5 at 14.12EV. It can not be a coincidence.

Fortunately for us all, K5 and D7000 have MUCH MORE in common than K5 or D7000 with A55. Don't kill me if I'm wrong, but if I have to bet... I bet D7000 will have mostly identical charts than the K5. Probably with more noise starting at ISO 3200, but retaining more detail.
Click to expand...
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
Daviii said:
Yes, they are here to stay!

Which is the difference between A55 and K5? On one hand, the pellicle mirror, so determined percentadge of light gets lost, so we have a constant that applies to the full sensitivity range of the sensor. On the other hand, Sony has a 12bit process while Pentax has a 14bit process. So we may have a limiting factor, as long as capturing precission of the sensor exceeds what 12 bits are able to represent.

What I see on those charts is, on one hand, a constant, possibly the ammount of light that hits the sensor, that makes the K5 instantly better, constantly better, at any sensitivity. Lines on the charts are absolutely parallel.

On the other hand, once the dynamic range has the chance to exceed the 12 EVs it seems the K5 DESTROYS the A55. A55 tops at 12,39EV and K5 at 14.12EV. It can not be a coincidence.

Fortunately for us all, K5 and D7000 have MUCH MORE in common than K5 or D7000 with A55. Don't kill me if I'm wrong, but if I have to bet... I bet D7000 will have mostly identical charts than the K5. Probably with more noise starting at ISO 3200, but retaining more detail.
Click to expand...
What people over and over keep forgetting :

1) The DXOMark data is not coherent with actual results in finished images/prints,

2) Other factors like the quality of the noise and retained level of detail , lens and aperture used, AF (speed and focus accuracy) exposure, scene (contrast, color temperature), raw conversion (raw converter and actual settings in that), additional NR and sharpening, printer and paper, viewing distance etc determine the apperance of the finished image/print,

3) Small differences (concerning resolution or less than 1/2 stop about noise etc) will never ever be visible in the finished images/prints.

4) Differences seen in 100% crops have no value determining what (differences) will be seen in the final image/print.
 
Yep, it will certainly be interesting to see the D7000. Nikon usually does the best job of holding onto DR, tonal range, and color sensitivity as the ISO increases.
Jon Rty said:
Nikon cameras that share their sensor with Pentax score higher than their Pentax counter-part. Thus, if anything I expect the D7000 to score even higher than the K-5
quintana said:
Very impressive and extremely tempting. Until now I thought I would change from Canon to Nikon D7000 but I guess the D7000 won't achieve the marks of the K-5 because the Sony SLT A-55 wasn't even as good as the D90.

But there is one thing that worries me about the K-5: The choice of lenses is by far not as versatile and as good as the Nikon or Canon glasses.

Hmm, will be a tough decision!
Click to expand...
Click to expand...
--
Correct my incorrect photography answers at:
http://photo.stackexchange.com/users/389/eruditass

Please update the Camera Shutter Life Expectancy Database with your experiences:
http://www.olegkikin.com/shutterlife/
 
Jon Rty said:
Nikon cameras that share their sensor with Pentax score higher than their Pentax counter-part. Thus, if anything I expect the D7000 to score even higher than the K-5
Click to expand...
Let's hope so. I browsed through the lens choice of Pentax and I am pretty sure that I won't be happy with it.

So I hope for really impressive marks with the D7000 because I guess I will buy it because of the far better lens choice and I don't always want to look at the K-5 with envious eyes. ;-)
 
Once you go above 12 stops of dynamic range (per pixel), you are limited by a 12-bit RAW file. In other words, you can't go above 12 stops per pixel with a 12-bit RAW file.
rhlpetrus said:
Don't think bits is the issue, D300/s had 14 bits and D90 didn't and tehre was basically no difference.

This sensor may be different from the one in the A55. Or the ADc system may be different, not just re 12 x 14 bits.

And toppings also affect results.

The K5 graphs show some RAW manipulation at and above 3200 ISO and DxO mark seems to have detetcted it, they are using diffrent symbols for points in that range.
Daviii said:
Yes, they are here to stay!

Which is the difference between A55 and K5? On one hand, the pellicle mirror, so determined percentadge of light gets lost, so we have a constant that applies to the full sensitivity range of the sensor. On the other hand, Sony has a 12bit process while Pentax has a 14bit process. So we may have a limiting factor, as long as capturing precission of the sensor exceeds what 12 bits are able to represent.

What I see on those charts is, on one hand, a constant, possibly the ammount of light that hits the sensor, that makes the K5 instantly better, constantly better, at any sensitivity. Lines on the charts are absolutely parallel.

On the other hand, once the dynamic range has the chance to exceed the 12 EVs it seems the K5 DESTROYS the A55. A55 tops at 12,39EV and K5 at 14.12EV. It can not be a coincidence.

Fortunately for us all, K5 and D7000 have MUCH MORE in common than K5 or D7000 with A55. Don't kill me if I'm wrong, but if I have to bet... I bet D7000 will have mostly identical charts than the K5. Probably with more noise starting at ISO 3200, but retaining more detail.
Click to expand...
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
Click to expand...
--
Correct my incorrect photography answers at:
http://photo.stackexchange.com/users/389/eruditass

Please update the Camera Shutter Life Expectancy Database with your experiences:
http://www.olegkikin.com/shutterlife/
 
for me it depends whatu want and what are uor needs...if u need long pro tele better stay nikon or canon...if u want a cropped system sealed oìr with a lot of mini fast and limited primes go pentax.

pentax DA* sealed lenses

16-50
50-135
60-250
thre pretty good zoom covering a good focal
DA* prime
55 200 2,8
300 4 and un upcoming 400 4,5 or 4 probably with tele converter

all these are sealed lenses , u dont have cropped sealed lenses in nikon lineup only ff camera

prime macro 50 100
aother prime 15 21 35 35 f2 31 55 40 77 70
other zoom sealed : 18-135 18-55
zoom not sealed: 17-70 10-17 12 -24

apart long tele what we are missing? tilt and shift? i use zork psa with pentax 645 much more flexible, and multifocus tilt and shift or other tilt and shift adapter.

i think next year will see some release of lenses.

personally i will buy nikon only for d3s or d700 and fast prime and long tele.
quintana said:
Very impressive and extremely tempting. Until now I thought I would change from Canon tofor me it depends Nikon D7000 but I guess the D7000 won't achieve the marks of the K-5 because the Sony SLT A-55 wasn't even as good as the D90.

But there is one thing that worries me about the K-5: The choice of lenses is by far not as versatile and as good as the Nikon or Canon glasses.

Hmm, will be a tough decision!
Click to expand...
 
12 bit files allow for a maximum of 13.79 stops of dynamic range.
johnnyz86 said:
Once you go above 12 stops of dynamic range (per pixel), you are limited by a 12-bit RAW file. In other words, you can't go above 12 stops per pixel with a 12-bit RAW file.
rhlpetrus said:
Don't think bits is the issue, D300/s had 14 bits and D90 didn't and tehre was basically no difference.

This sensor may be different from the one in the A55. Or the ADc system may be different, not just re 12 x 14 bits.

And toppings also affect results.

The K5 graphs show some RAW manipulation at and above 3200 ISO and DxO mark seems to have detetcted it, they are using diffrent symbols for points in that range.
Daviii said:
Yes, they are here to stay!

Which is the difference between A55 and K5? On one hand, the pellicle mirror, so determined percentadge of light gets lost, so we have a constant that applies to the full sensitivity range of the sensor. On the other hand, Sony has a 12bit process while Pentax has a 14bit process. So we may have a limiting factor, as long as capturing precission of the sensor exceeds what 12 bits are able to represent.

What I see on those charts is, on one hand, a constant, possibly the ammount of light that hits the sensor, that makes the K5 instantly better, constantly better, at any sensitivity. Lines on the charts are absolutely parallel.

On the other hand, once the dynamic range has the chance to exceed the 12 EVs it seems the K5 DESTROYS the A55. A55 tops at 12,39EV and K5 at 14.12EV. It can not be a coincidence.

Fortunately for us all, K5 and D7000 have MUCH MORE in common than K5 or D7000 with A55. Don't kill me if I'm wrong, but if I have to bet... I bet D7000 will have mostly identical charts than the K5. Probably with more noise starting at ISO 3200, but retaining more detail.
Click to expand...
--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
Click to expand...
--
Correct my incorrect photography answers at:
http://photo.stackexchange.com/users/389/eruditass

Please update the Camera Shutter Life Expectancy Database with your experiences:
http://www.olegkikin.com/shutterlife/
Click to expand...
 
Daviii said:
Yes, they are here to stay!

Which is the difference between A55 and K5? On one hand, the pellicle mirror, so determined percentadge of light gets lost, so we have a constant that applies to the full sensitivity range of the sensor. On the other hand, Sony has a 12bit process while Pentax has a 14bit process. So we may have a limiting factor, as long as capturing precission of the sensor exceeds what 12 bits are able to represent.

What I see on those charts is, on one hand, a constant, possibly the ammount of light that hits the sensor, that makes the K5 instantly better, constantly better, at any sensitivity. Lines on the charts are absolutely parallel.

On the other hand, once the dynamic range has the chance to exceed the 12 EVs it seems the K5 DESTROYS the A55. A55 tops at 12,39EV and K5 at 14.12EV. It can not be a coincidence.
Click to expand...
12 bit files can have a maximum of 13.79 stop of dynamic range, so it looks like something else is going on.
Daviii said:
Fortunately for us all, K5 and D7000 have MUCH MORE in common than K5 or D7000 with A55. Don't kill me if I'm wrong, but if I have to bet... I bet D7000 will have mostly identical charts than the K5. Probably with more noise starting at ISO 3200, but retaining more detail.
Click to expand...
I doubt it (the D7000 certainly won't have the high ISO smoothing), I expect the D7000 and A580 to be similar though.
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top