Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
04-14-2013, 07:53 PM   #31
Site Supporter
jpzk's Avatar

Join Date: Jan 2009
Location: Québec
Photos: Gallery | Albums
Posts: 6,975
The story repeats itself: the OP posts a message and does not bother with the replies.
I suppose he(she) was serious about: "I think I'm finally over you Pentax. It was fun while it lasted."


JP

04-14-2013, 10:46 PM   #32
Pentaxian




Join Date: Jun 2009
Location: GMT +10
Photos: Albums
Posts: 10,576
QuoteOriginally posted by Sol Invictus Quote
Suddenly, it's a "synthetic" measure. Essentially what Cannon shooters have been saying for years, while us Pentaxians chuckled smugly.
If I remember correctly, Pentaxians were once as grumpy and in denial as Canon shooters are today about DxOMark, in the days of the K200D/K20D, then the K-7 and K-m. Only when the K-x, the 645D, and especially the K-5 get measured by DxO and did well, did Pentax folks start to trust DxOMark a lot more.

I like DxO, so whatever they say about any new Pentaxes will be worth paying attention to. I think Pentax certainly know about DxOMark too, and the marketing advantages of scoring well on DxOMark nowadays. A case in point was Pentax letting the K-5 ISO range run down to ISO 80, which was I think a direct play at helping the K-5 beat by a mere nose the DxO DR score of the D7000.
04-15-2013, 02:03 AM   #33
Pentaxian




Join Date: Apr 2007
Location: Romania
Posts: 8,413
QuoteOriginally posted by Sol Invictus Quote
I find it funny that when the Sony 16 MP sensor was at the top of DxOMark, people wouldn't stop pointing out that fact. Now that the sensor on a mid-level Nikon is beating it, all of a sudden DxOMark isn't that relevant anymore. Suddenly, it's a "synthetic" measure. Essentially what Cannon shooters have been saying for years, while us Pentaxians chuckled smugly.
Not suddenly, it was like this from the beginning; and it doesn't mean their measurements are invalid, just that they're incomplete.
Indeed, the 16MP Sony sensor's DR is real; but this alone is not what makes the sensor what I consider to be "the best APS-C". It's pretty much faultless while obtaining very good benchmark scores, and this is important.
The 24MP sensors, due to a scoring method which factors in the number of pixels, have very slightly higher scores; but they're not faultless.

Last edited by Kunzite; 04-15-2013 at 02:45 AM.
04-15-2013, 02:52 AM   #34
Pentaxian




Join Date: Jun 2009
Location: GMT +10
Photos: Albums
Posts: 10,576
QuoteOriginally posted by Sol Invictus Quote
Now that the sensor on a mid-level Nikon is beating it,
You are exaggerating. The DxOMark results of the D7100 vs the K-5's and even the D7000 reveal minimal differences. Just look at the SNR chart for example for the D7100/D7000/K-5. It is a real case of spot the difference:



04-17-2013, 06:49 AM   #35
Veteran Member
Sol Invictus's Avatar

Join Date: Aug 2011
Location: Calgary, Alberta
Photos: Gallery
Posts: 392
QuoteOriginally posted by rawr Quote
You are exaggerating. The DxOMark results of the D7100 vs the K-5's and even the D7000 reveal minimal differences. Just look at the SNR chart for example for the D7100/D7000/K-5. It is a real case of spot the difference:
I've always said that as well. The difference between a D7000 sensor and the K5 sensor was so tiny as to be meaningless. But that didn't stop us from claiming the K5 had the best sensor in the world, full stop. If we're using the same margin to call a winner one way, it works the other way as well. That's all I'm saying.
04-17-2013, 08:20 AM   #36
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 23,731
QuoteOriginally posted by Sol Invictus Quote
I've always said that as well. The difference between a D7000 sensor and the K5 sensor was so tiny as to be meaningless. But that didn't stop us from claiming the K5 had the best sensor in the world, full stop. If we're using the same margin to call a winner one way, it works the other way as well. That's all I'm saying.
Or to look at it another way... if you're tied with the best, you're still the best.

When researching a couple of Canon models people spoke of the rich colours etc...I'm not sure if that's a comparative observation, but Canon's obviously have something to offer as well. And that's the issue, if the camera is so bad you can't get the job done maybe there's an issue. I've heard people say they prefer Pentax ergonomics etc. but no one has said they couldn't get the job done, no matter what the DxO scores say. Which begs the question, what are the DxO scores really about? How do you have a system that rates different systems higher than one another based on a 2% difference in dynamic range, and colour depth. DxO has never produced a single image to demonstrate what such close stats mean in the real world. And until we know that, "we know nuffink, nuffink I say."
04-17-2013, 06:12 PM   #37
Pentaxian




Join Date: Jun 2009
Location: GMT +10
Photos: Albums
Posts: 10,576
QuoteOriginally posted by normhead Quote
How do you have a system that rates different systems higher than one another based on a 2% difference in dynamic range, and colour depth.
It's just a metric. They provide the data and explain pretty clearly how they derive it.

I imagine displaying the [usually marginal] differences between sensors via comparison images of 'real world' scenes for each DxO sensor evaluation would be a problem when you factor in people's different monitor calibrations, display resolution, not to mention 'real world' subject lighting variations, lens quality, focus variations etc etc. I can understand why they steer clear of doing that.
04-17-2013, 06:52 PM   #38
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 23,731
QuoteOriginally posted by rawr Quote
It's just a metric. They provide the data and explain pretty clearly how they derive it.

I imagine displaying the [usually marginal] differences between sensors via comparison images of 'real world' scenes for each DxO sensor evaluation would be a problem when you factor in people's different monitor calibrations, display resolution, not to mention 'real world' subject lighting variations, lens quality, focus variations etc etc. I can understand why they steer clear of doing that.
Well I can see you thinking that might be it, I was thinking more along the lines of it wouldn't be possible to demonstrate a difference that would be meaningful in any way.

04-18-2013, 04:34 AM   #39
Pentaxian




Join Date: Jun 2009
Location: GMT +10
Photos: Albums
Posts: 10,576
QuoteOriginally posted by normhead Quote
a difference that would be meaningful in any way
I think they would probably only be able to easily demonstrate visual differences that anyone could see if the specs were miles apart - eg what the SNR at 3200 ISO looks like between a 12MP Nikon D700 and a 12 MP Canon Powershot S100 when shot at the same shutter speed and low-light level.

But even then, what would folks like to look at in order to judge what meaningfully 'looks best'? Images resized on a computer screen to a standard 1024px on the wide side? 100% image crops viewed at a minimum of 4 feet away from the monitor? Images printed on 8x10 glossy paper at 150 dpi and viewed 2 feet away?

You start to run into perceptual issues like those that David Pogue posted about in 2006 in his 'The Truth About Digital Cameras' blog post and its subsequent follow-ups.
04-18-2013, 06:50 AM   #40
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 23,731
QuoteOriginally posted by rawr Quote
I think they would probably only be able to easily demonstrate visual differences that anyone could see if the specs were miles apart - eg what the SNR at 3200 ISO looks like between a 12MP Nikon D700 and a 12 MP Canon Powershot S100 when shot at the same shutter speed and low-light level.

But even then, what would folks like to look at in order to judge what meaningfully 'looks best'? Images resized on a computer screen to a standard 1024px on the wide side? 100% image crops viewed at a minimum of 4 feet away from the monitor? Images printed on 8x10 glossy paper at 150 dpi and viewed 2 feet away?

You start to run into perceptual issues like those that David Pogue posted about in 2006 in his 'The Truth About Digital Cameras' blog post and its subsequent follow-ups.
It basically comes down to do your research, buy the camera you think you might like, if it doesn't work for you in some meaningful way, buy something else. With modern cameras, all are so good, it's other factors besides technical specs that usually make the difference.
04-18-2013, 08:27 AM   #41
Pentaxian




Join Date: Jun 2009
Location: GMT +10
Photos: Albums
Posts: 10,576
QuoteOriginally posted by normhead Quote
With modern cameras, all are so good, it's other factors besides technical specs that usually make the difference.
Very true. You also have to recognise what sort of shooter you prefer to be.
Not much point lusting after a Canon 1DX or Nikon D4 if you are at heart a family-kids-and-pets shooter, or a travelling-incognito-street-shooter, for example.

In my heart I am a point-and-shoot snapshooter at best. If I can shoot a decent shot of a bird or two from time to time with a decent DSLR and lens that's a bonus.
04-18-2013, 07:04 PM   #42
Site Supporter




Join Date: Feb 2011
Posts: 4,892
QuoteOriginally posted by rawr Quote
I think they would probably only be able to easily demonstrate visual differences that anyone could see if the specs were miles apart - eg what the SNR at 3200 ISO looks like between a 12MP Nikon D700 and a 12 MP Canon Powershot S100 when shot at the same shutter speed and low-light level.
They do show that.
04-19-2013, 07:10 AM   #43
Pentaxian
normhead's Avatar

Join Date: Jun 2007
Location: Near Algonquin Park
Photos: Gallery | Albums
Posts: 23,731
YOu can only go with what you know. My Pentax k20D is rated at 67. My K-5 is rated at 82. My wife and I shot for about 4 months side by side with both cameras. There were many situations where the K-20D produced identical images to the K-5. In sunset however, one of our favourite subject, the shadow detail in the K-5 made it clearly superior. As long as there was enough light, the K20D was as good.

There's a 14 point difference between a K-5 and a D800 and a 12 point difference between a K-5 and D600. But I have no idea why. Taking on past experience I cab guess, something is going to be better.

QuoteQuote:
5 points is roughly the smallest visible difference in actual photos (DxO: "equivalent to 1/3 stop"). The measurements themselves appear to be repeatable to within one or two points[1].
From Luminous Landscape.

"Repeatable within one or two points. SO in other words at it's low end two cameras ranked 4 points apart could score identically on the same test if the lower rated camera scored at it's high end and a lower rated camera scored at it's low end.

I look at a D800 and my K-5 and the largest improvement between the two is in high ISO performance. But high ISO really only affects lower quality images. As ISO increases, DR decreases. Maybe D800 is better at handling the decrease in IQ, but if I'm shooting ISO 100-800 and a D800 user is shooting between 100-1600 ISO we are essentially on a level playing field. I'm essentially losing one stop. But if I don't use high ISO because of image degradation, then it's really a moot point. I wouldn't be happy using my K-5 at 800, and I wouldn't be happy using my D800 at 800 or 1600, what's the advantage of having another rating on a scale with which I'm not happy? You only get the full dynamic range of the sensor shooting at 100 ISO or lower. Shooting at higher ISO reduces your dynamic range, even if the noise is controlled and the image is judged "acceptable" by some technical criteria.

So at that point the question becomes, how much will you pay for one stop of useful ISO performance, because that is the only meaningful difference between a K-5 and D800. You'll notice, one stop is also the difference between my K20D and my K-5. The difference between the K20D and the K-5 was true for a year and a half before I bought the first K-5. The upgrade cost me $1000. The upgrade, to a Nikon D800 would cost me $2700 not including lenses. SO I can safely say, FF has not approached my price point yet.

Seeing as the K-5 gives my practically identical dynamic range and the colour depth of the K-5 is close enough to the D800 it's not an issue. We know cameras with a colour depth of 8 do very poorly compared to cameras with a colour depth of 16 bits. But do we know pictures taken with a colour depth of 23.8 look different than those with a colour depth of 25.6, I don't know. More to the point, with PP can images taken with 23.8 be made comparable to images taken at 25.6 bits of depth when most printers are 8 bit and the best are 16. The answer there is they certainly can. Both formats have more information than they can functionally use.

Any way, thanks for this little foray into the world of DxO sensor scores. All you have to remember is the margin of error is 5 points. Cameras within 5 points are capable of producing the same results. And if you do most of your shooting between the ISO 100 and 800, the D800- D600 advantage is very small, if in fact it exists at all. But if your shooting style depends on a lot ISO 1600 shots or higher, you can do better with other brands beside Pentax.

04-19-2013, 11:21 AM   #44
Site Supporter
cali92rs's Avatar

Join Date: May 2011
Location: Long Beach, CA
Posts: 3,195
QuoteOriginally posted by normhead Quote
I look at a D800 and my K-5 and the largest improvement between the two is in high ISO performance. But high ISO really only affects lower quality images. As ISO increases, DR decreases. Maybe D800 is better at handling the decrease in IQ, but if I'm shooting ISO 100-800 and a D800 user is shooting between 100-1600 ISO we are essentially on a level playing field. I'm essentially losing one stop. But if I don't use high ISO because of image degradation, then it's really a moot point. I wouldn't be happy using my K-5 at 800, and I wouldn't be happy using my D800 at 800 or 1600, what's the advantage of having another rating on a scale with which I'm not happy? You only get the full dynamic range of the sensor shooting at 100 ISO or lower. Shooting at higher ISO reduces your dynamic range, even if the noise is controlled and the image is judged "acceptable" by some technical criteria.

So at that point the question becomes, how much will you pay for one stop of useful ISO performance, because that is the only meaningful difference between a K-5 and D800. You'll notice, one stop is also the difference between my K20D and my K-5. The difference between the K20D and the K-5 was true for a year and a half before I bought the first K-5. The upgrade cost me $1000. The upgrade, to a Nikon D800 would cost me $2700 not including lenses. SO I can safely say, FF has not approached my price point yet.

Seeing as the K-5 gives my practically identical dynamic range and the colour depth of the K-5 is close enough to the D800 it's not an issue. We know cameras with a colour depth of 8 do very poorly compared to cameras with a colour depth of 16 bits. But do we know pictures taken with a colour depth of 23.8 look different than those with a colour depth of 25.6, I don't know. More to the point, with PP can images taken with 23.8 be made comparable to images taken at 25.6 bits of depth when most printers are 8 bit and the best are 16. The answer there is they certainly can. Both formats have more information than they can functionally use.

Any way, thanks for this little foray into the world of DxO sensor scores. All you have to remember is the margin of error is 5 points. Cameras within 5 points are capable of producing the same results. And if you do most of your shooting between the ISO 100 and 800, the D800- D600 advantage is very small, if in fact it exists at all. But if your shooting style depends on a lot ISO 1600 shots or higher, you can do better with other brands beside Pentax.
What that illustrates to me is how awesome the D800 really is, because you get over double the resolution WITHOUT the degradation of high ISO performance (actually even better high ISO performance).
You get your cake and are able to eat it too (if your cake isn't dependent on size or price).
04-19-2013, 01:13 PM   #45
Site Supporter




Join Date: Feb 2011
Posts: 4,892
QuoteOriginally posted by normhead Quote
I look at a D800 and my K-5 and the largest improvement between the two is in high ISO performance. But high ISO really only affects lower quality images. As ISO increases, DR decreases. Maybe D800 is better at handling the decrease in IQ, but if I'm shooting ISO 100-800 and a D800 user is shooting between 100-1600 ISO we are essentially on a level playing field.
The D800 at 100-1600 would be comparable to the K-5 at 50-800, so 'essentially' you're comparing a stop apart, as you mention.

QuoteOriginally posted by normhead Quote
But if I don't use high ISO because of image degradation, then it's really a moot point. I wouldn't be happy using my K-5 at 800, and I wouldn't be happy using my D800 at 800 or 1600, what's the advantage of having another rating on a scale with which I'm not happy?
Sometimes I wish I could shoot the K-5 at ISO 50 (or 25...).



QuoteOriginally posted by normhead Quote
So at that point the question becomes, how much will you pay for one stop of useful ISO performance, because that is the only meaningful difference between a K-5 and D800.
And a stop of color depth, and a stop of dynamic range (you were wrong when you claimed otherwise) and about 1.5x better resolution. Those might not be required for a person's photos - or you could pay $1000's of dollars for premium lenses with APS-C - but please don't mislead everyone else on this forum. They should be able to determine the best camera for themselves without any wrong information.
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!
full-frame, pentax
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
No April 1 FF rumours this year? normhead Photographic Industry and Professionals 8 04-03-2013 12:43 PM
No FF for at least two years. normhead Pentax DSLR Discussion 22 10-02-2012 08:27 AM
Not trolling, but who buys Pentax FF if the lens selection is inadequate? weave2d2 Pentax SLR Lens Discussion 136 07-08-2012 04:34 AM
Kit lens, but no AF helmut Troubleshooting and Beginner Help 7 08-15-2011 09:06 AM
No FF from Pentax but it is 3000 Euro RaduA Pentax News and Rumors 10 11-14-2008 04:54 PM



All times are GMT -7. The time now is 01:23 PM. | 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