Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
11-01-2013, 12:45 PM   #46
Veteran Member
audiobomber's Avatar

Join Date: May 2008
Location: Sudbury, Ontario
Photos: Albums
Posts: 6,806
QuoteOriginally posted by panoguy Quote
Here is a 100% pixel crop with no noise reduction from an image I shot today at the Royal Ontario Museum.
The focusing is leagues better than the K-5, but if I was doing a lot of nighttime work or high ISO work and didn't need 24MP, I'd stick with the K-5IIs for now. Just my .02 CAD...
The flaw in your analysis is the "100% pixel crop". If you compare both at 16mp, you shouldn't see a difference in noise, as shown by DXOMark results.

11-01-2013, 01:24 PM   #47
Pentaxian
panoguy's Avatar

Join Date: Sep 2007
Location: Washington, D.C.
Photos: Gallery | Albums
Posts: 3,327
Original Poster
QuoteOriginally posted by Andi Lo Quote
How fast is the review? on D600 it takes 1.5 seconds to get to this view. I'm guessing the K-3 is faster
Try "instant," and this to display part of 300 images.
Really, 1.5 sec. on the D600? That's nuts...

QuoteOriginally posted by audiobomber Quote
The flaw in your analysis is the "100% pixel crop". If you compare both at 16mp, you shouldn't see a difference in noise, as shown by DXOMark results.
There is no flaw in my "analysis," but rather in your assumptions.
I use all the pixels a camera gives me, and when pushing the shadows on a K-5 raw file, there is much less noise at comparable ISOs. If I stitch together multiple K-5 files to get the same MP as the K-3, the shadows will still have less noise. Downsampling a K-3 file to 16MP isn't something I will do. The noise is definitely related to the MP count on the sensor (photosite density), and perhaps DxO tries to equalize this in their equations. That's their analysis...

Last edited by panoguy; 11-01-2013 at 05:04 PM.
11-01-2013, 01:55 PM   #48
Veteran Member
audiobomber's Avatar

Join Date: May 2008
Location: Sudbury, Ontario
Photos: Albums
Posts: 6,806
QuoteOriginally posted by panoguy Quote
The noise is definitely related to the MP count on the sensor (photosite density), and perhaps DxO tries to equalize this in their equations. That's their analysis...
DXO equalize all results to 8mp in their "Print" results. This is for comparing noise at the same resolution on a computer screen, and at the same print size (8"x10"). That makes sense to me. At 8mp, the K-3 and K-5 II will show the same amount of noise. At 16mp, they will show the same amount of noise.

I understand what you're saying, you will sometimes crop to 100%. So will I, but it doesn't make sense to criticize the K-3 for something the K-5 II cannot do, i.e. resolve from 16-24 megapixels.
11-01-2013, 02:04 PM   #49
Veteran Member




Join Date: Jun 2009
Posts: 11,913
How does that same dark scene render in the software supplied with the camera?

Using PDCU 5 is probably the best way to examine the way the RAW's look, since that's the 'official' RAW processor. PDCU 5 may even have it's NR and other RAW development options specifically tuned for the K-3, in a way that ACR can't match (yet).

11-01-2013, 02:12 PM   #50
Site Supporter
Site Supporter
mattb123's Avatar

Join Date: Apr 2010
Location: Colorado High Country
Photos: Gallery
Posts: 10,867
I wonder if LR4 will ever support this camera specifically or if this will be the tipping point for me to upgrade to LR5...
11-01-2013, 02:19 PM   #51
Veteran Member




Join Date: Jun 2009
Posts: 11,913
QuoteOriginally posted by mattb123 Quote
I wonder if LR4 will ever support this camera specifically
I doubt it. LR 4 is frozen in time now, at v4.4.

I'm waiting for LR 5 too. When it gets updated to include the K-3, I'll upgrade.

You could work around needing LR5 (perhaps) by downloading the Adobe DNG convertor, which gets updated in sync with the latest ACR and LR, and using it to convert your K-3 RAW's into DNG's, in the hope that the latest DNG convertor will have applied some of the optimisations that Adobe may have worked out for the K-3. But that sounds like too much work to me.

Last edited by rawr; 11-01-2013 at 02:27 PM.
11-01-2013, 02:23 PM   #52
Site Supporter
Site Supporter
Miguel's Avatar

Join Date: Sep 2007
Location: Near Seattle
Photos: Gallery
Posts: 2,743
QuoteOriginally posted by mattb123 Quote
I wonder if LR4 will ever support this camera specifically or if this will be the tipping point for me to upgrade to LR5...
Even if you save your files as DNG, LR4 will only give you a generic "non-profile" that doesn't reflect any advancements in the newer versions of ACR that is the engine of LR. I've always assumed that if one gets a new model camera, the latest and greatest software has to be used to ensure feature compatibility. It's kind of a racket that way. I would think that the next dot release of LR5 will integrate the K-3 and any new Pentax lenses. You may be able to roll your own K-3 profile for use in LR4 but that involves time, discipline, and maybe some money. I'd rather pay for the upgrade.

M

11-01-2013, 02:32 PM   #53
Veteran Member




Join Date: Dec 2007
Photos: Gallery
Posts: 8,237
QuoteOriginally posted by panoguy Quote
Try "instant," and this to display part of 300 images.
Really, 1.5 sec. on the D600? That's nuts...



There is no flaw in my "analysis," but rather in your assumptions.
I use all the pixels a camera gives me, and when pushing the shadows on a K-5 raw file, there is much less noise at comparable ISOs. If merge multiple K-5 files to get the same MP as the K-3, the shadows will still have less noise. Downsampling a K-3 file to 16MP isn't something I will do. The noise is definitely related to the MP count on the sensor (photosite density), and perhaps DxO tries to equalize this in their equations. That's their analysis...
Actually audiobomber is right - when you look at it at 100% pixel view, you're putting the higher-res sensor at a comparative disadvantage, and when you say you 'won't downsample the K-3 to 16MP' you're not realizing what you actually do when you print. Unless you're printing at larger than 13''x19'', you're probably downsampling every time anyway unless you're trying to maintain more than 300 dpi in the print. Every single image you display at web sizes (even big, 1600 or 2048pixel-width) is downsampled, of course. Here is a very simple illustration of potential benefits of having more MP to downsample with (use the mouse-overs.) And here is a detailed description of how more MP can offset noise.

DxOMark equals the file sizes to 8MP for a very valid reason - because it creates a common ground where the output can be realistically compared, in the same way a photographer would compare prints.

In your examples, the NR version looks smudged, plastic, horrid, and the non-NR version actually looks pretty good - love the detail. When printed or displayed with no NR at the same sizes as the K-5 image, it will probably look just as clean, with a slightly crisper, sharper, more detailed look to it. (The raw converter could make a big difference, though.)

The K-3 looks like a potential winner to me.

.

Last edited by jsherman999; 11-01-2013 at 02:49 PM.
11-01-2013, 04:07 PM   #54
Pentaxian
panoguy's Avatar

Join Date: Sep 2007
Location: Washington, D.C.
Photos: Gallery | Albums
Posts: 3,327
Original Poster
QuoteOriginally posted by audiobomber Quote
DXO equalize all results to 8mp in their "Print" results. This is for comparing noise at the same resolution on a computer screen, and at the same print size (8"x10"). That makes sense to me. At 8mp, the K-3 and K-5 II will show the same amount of noise. At 16mp, they will show the same amount of noise.
QuoteOriginally posted by jsherman999 Quote
Actually audiobomber is right - when you look at it at 100% pixel view, you're putting the higher-res sensor at a comparative disadvantage, and when you say you 'won't downsample the K-3 to 16MP' you're not realizing what you actually do when you print.
Wow... everybody knows what I do with my images better than I do. Maybe I should stop selling the digital files and start printing my stitched, panoramic HDR (32-bits per channel) files.

Oh wait...

Let me say that my discussion of the noise in the K-3 images is based on a camera that I've only had for two days, with firmware 1.00, and using software (ACR 8.1) that doesn't know the camera model exists beyond what's in the EXIF. This is compared to a K-5 which I've used for my professional work for two years and 40,000+ images (on the main body), so I definitely know my way around it's raw files!

The K-3 is certainly a great camera, but so far I am not seeing the kind of "in the raw file" dynamic range that the K-5 has (and I presume is in the K-5II and IIs). By dynamic range, I mean recognizable detail that can be pulled back from highlights and shadows without as much noise as I see here at the pixel level. And of course I'm looking at the files at 100%, that's where you see noise or other artefacts!

I've installed the Ricoh DCU 5 software (Silkypix) that came with the camera, and am trying to find my way around it. I'll post what I see.
11-01-2013, 04:33 PM   #55
Veteran Member




Join Date: Jun 2009
Posts: 11,913
QuoteOriginally posted by panoguy Quote
I've installed the Ricoh DCU 5 software (Silkypix) that came with the camera, and am trying to find my way around it.
If it is anything like DCU 4, using it may be disorienting at first. But stick with it!
DCU is traditionally very tightly integrated with all the Pentax camera settings, down to all the image tone settings and scene modes, and has some novel features too - like it's multi-point AWB selection tool. More to the point, it's noise reduction can also be surprisingly effective.
11-01-2013, 04:48 PM   #56
Senior Member




Join Date: Dec 2009
Location: United States
Posts: 156
What does this mean, actually?

QuoteOriginally posted by jsherman999 Quote
Actually audiobomber is right - when you look at it at 100% pixel view, you're putting the higher-res sensor at a comparative disadvantage, and when you say you 'won't downsample the K-3 to 16MP' you're not realizing what you actually do when you print. Unless you're printing at larger than 13''x19'', you're probably downsampling every time anyway unless you're trying to maintain more than 300 dpi in the print. Every single image you display at web sizes (even big, 1600 or 2048pixel-width) is downsampled, of course. Here is a very simple illustration of potential benefits of having more MP to downsample with (use the mouse-overs.) And here is a detailed description of how more MP can offset noise.

DxOMark equals the file sizes to 8MP for a very valid reason - because it creates a common ground where the output can be realistically compared, in the same way a photographer would compare prints.

In your examples, the NR version looks smudged, plastic, horrid, and the non-NR version actually looks pretty good - love the detail. When printed or displayed with no NR at the same sizes as the K-5 image, it will probably look just as clean, with a slightly crisper, sharper, more detailed look to it. (The raw converter could make a big difference, though.)

The K-3 looks like a potential winner to me.

.
So I'm wondering what the practical implications are for the modern day, professional digital photographer? And for that matter, anyone who enjoys digital photography? The amount of printing done these days is miniscule compared to the number of digital images produced. Even brides don't print their wedding photos that much anymore. But increasing the magnification of a photo on one's computer is certainly done proportionately higher statistically than printing the photo. The question of increased noise vs increased megapixels is part of the digital photographic landscape. I think we all hoped/assumed that advances in technology would allow for higher resolution without the compromise of increased noise, and this has been true to some extent. But perhaps the megapixel race is still ahead of the ability to control noise? Which would support panoguy's suggestion that if lower noise photographs (as viewed on a computer screen) are more advantageous to the photographer's work, then perhaps the lower MP camera is a better choice?
11-01-2013, 05:00 PM - 1 Like   #57
Senior Member




Join Date: Dec 2009
Location: United States
Posts: 156
QuoteOriginally posted by The K-3 is certainly a great camera, but so far I am not seeing the kind of "in the raw file" dynamic range that the K-5 has (and I presume is in the K-5II and IIs). By dynamic range, I mean recognizable detail that can be pulled back from highlights and shadows without as much noise as I see. here [I:
at the pixel level[/I]. And of course I'm looking at the files at 100%, that's where you see noise or other artefacts!
The K5/K5II's raw file dynamic range capability seems almost unprecedented. I'm not sure if I will buy the K3, but even if I do, I want to have the K5 as well. I think your perceptive comments have demonstrated the reason for having both. After all, one buys/changes lenses to gain certain attributes. Why shouldn't that also be the case with camera bodies? Imagine what it would be like if all lenses and cameras could communicate effectively with each other? I don't think it would harm photography.
11-01-2013, 06:01 PM   #58
Veteran Member
audiobomber's Avatar

Join Date: May 2008
Location: Sudbury, Ontario
Photos: Albums
Posts: 6,806
The K-3 will score slightly higher than the K-5 IIs on DXOMark and some of you will wonder how. Oh well.
11-01-2013, 06:20 PM   #59
Pentaxian
panoguy's Avatar

Join Date: Sep 2007
Location: Washington, D.C.
Photos: Gallery | Albums
Posts: 3,327
Original Poster
Another nice thing about the K-3 that was made apparent while messing around: While the actual focusing speed for my lone SDM lens (the ultra-pokey DA*50-135) was not improved at all by the K-3's new AF... my Sigma 50-500 OS lens with Sigma's HSM is like bottled lightning! I can't say if it's the lack of hesitation in the AF algorithm, the increase in the number of points, or some other difference from the K-5, but this lens truly zips from point to point. Sadly, the wind was so high today the birds never stopped by the backyard, so I didn't get subjects to try the AF.C out on.

QuoteOriginally posted by audiobomber Quote
The K-3 will score slightly higher than the K-5 IIs on DXOMark and some of you will wonder how. Oh well.
I won't wonder, since I don't read the DXOMark scores unless someone points to them. Oh well.
11-01-2013, 07:09 PM   #60
Veteran Member




Join Date: Mar 2010
Location: SW Washington
Posts: 833
QuoteOriginally posted by panoguy Quote
f I stitch together multiple K-5 files to get the same MP as the K-3, the shadows will still have less noise.
Of course you do. If you combine two images, you are using twice the light (1 stop) and thus twice the information to produce a single final image. This works with any camera, even a crappy phone.
QuoteQuote:
The K-3 is certainly a great camera, but so far I am not seeing the kind of "in the raw file" dynamic range that the K-5 has (and I presume is in the K-5II and IIs). By dynamic range, I mean recognizable detail that can be pulled back from highlights and shadows without as much noise as I see here at the pixel level. And of course I'm looking at the files at 100%, that's where you see noise or other artefacts!
Why? Do you show people images or do you show them individual pixels?

Here's a real simple, objective test. Forget DXO, theoretical science, numbers, whatnot. Take a picture with a K-5IIs, and take the same picture with the K-3 (same lens, same spot, same settings). Can you point out anything, actually specifically identify something (e.g. "that line", "that pattern", "those spots", etc.) that can be seen on the K-5IIs, but cannot be resolved on the K-3 as a result of noise? Is the noise over the entire area of that specific detail significantly different when viewed at the same size?

Of course it has to be viewed at the same size for a fair comparison. View it at 100% individual pixel level, and you are magnifiying and potentially seeing finer details (and finer noise) on the K-3 that are not physically possible to see on the K-5IIs due to lower pixel count.

Last edited by Cannikin; 11-01-2013 at 07:19 PM.
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!
camera, card, dslr, k-3, k-5, k3, mode, pentax k-3, sensor, time, video, view
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
The best Pentax ad I've seen so far bwDraco Photographic Industry and Professionals 19 11-03-2012 04:39 AM
Here is the crasiest person I've seen so far :) benhai Pentax DSLR Discussion 23 06-15-2012 08:22 AM
Something I've noticed about Prius drivers Damn Brit General Talk 146 04-29-2009 09:45 PM
One thing I've noticed about the K20D Rickster Digital Processing, Software, and Printing 13 04-03-2008 06:49 PM
My April angst and what I've done about it so far... photo_mom Pentax SLR Lens Discussion 13 04-29-2007 08:54 PM



All times are GMT -7. The time now is 11:08 PM. | 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