Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
01-31-2015, 04:31 PM   #16
Veteran Member




Join Date: Jan 2012
Location: Slovenia
Photos: Gallery
Posts: 2,182
I did some (possibly wrong) calculations once, and came to the conclusion that 4 stars was basically compression 100, 3 stars 95, 2 stars about 85 and 1 star 50. The dropped Bestestest setting doesn't contribute as much to the image at first sight, but IMO it can be edited a bit better than the others.

01-31-2015, 06:49 PM   #17
Loyal Site Supporter
Loyal Site Supporter
Bruce Clark's Avatar

Join Date: Mar 2010
Location: Ocean Grove, Victoria
Photos: Gallery | Albums
Posts: 4,458
Interesting debate. I do not know the answer but I will contribute this. The K-7 has a 1, 2, 3 or 4 star rating for JPEG quality as well as a JPEG Recorded Pixel setting of 2, 6, 10 and 14 M. Both these settings can be adjusted independent of each other. You can have a 4 star quality with 2M pixels or a 1 star quality with 14M pixels or any combination of these. I imagine the resultant file size is a result of the interaction of these two settings and scene content as well. Too complex for me to try to understand and I doubt it would help me much anyway.
01-31-2015, 08:17 PM   #18
Banned




Join Date: Feb 2013
Posts: 390
Interesting that everyone seems to think Pentax dropped the highest setting by going from 4 stars to three. It's much more likely they dropped the lowest, after all, you can't start numbering things with two!


FWIW, similar scenes I've shot with the K20, K30, and K3 vary greatly. The K20 avgs. 10 - 11mb, the K30 = 6 - 7mb, the K3 = 12 - 13mb, all set at the highest setting. What it all means? Haven't got a clue! I really couldn't care less about the size of the files as long as I've succeeded in capturing what I was going for.


But that's just my opinion, I could be wrong!
01-31-2015, 10:58 PM   #19
Veteran Member




Join Date: Dec 2008
Location: Roodepoort, South Africa
Photos: Gallery | Albums
Posts: 3,561
QuoteOriginally posted by Al_Kahollick Quote
Interesting that everyone seems to think Pentax dropped the highest setting by going from 4 stars to three. It's much more likely they dropped the lowest, after all, you can't start numbering things with two!
I think that you did not read my earlier post

K30 3 star ('best' setting) gives the same estimated filesize as K5 3 star ('best' setting); K5 series however still has 4 star ('premium' setting) which gives a larger estimated filesize. So yes, in this case Pentax definitely removed the 'premium'.

It however is possible that 3 star on the K3 is somewhere between 3 star and 4 star of the K5.

Below table gives an overview from the K100D onwards; the 3* and 4* columns give the approximate storage capacity (number of files).

ModelMpix3*4*
K100D6660 
K10D10409 
K200D10409 
K200010469 
K-x12281 
K-r12281 
K20D14239138
K714238148
K0116251 
K5 series16214134
K30 / K5016214 
K-S0120134 
K324111 

Two possibly interesting points:
  • Only the K20D, K7 and K5 (series) have a 'premium' quality setting.
  • Difference between the K200D/K10D and the K2000; both using the same sensor; with the same amount of recorded pixels, the compression of the K2000 seems a tad higher and the quality should therefore be a tad lower.


02-01-2015, 07:27 AM   #20
Veteran Member




Join Date: Dec 2010
Location: Ontario
Photos: Gallery
Posts: 3,332
QuoteOriginally posted by Al_Kahollick Quote
Interesting that everyone seems to think Pentax dropped the highest setting by going from 4 stars to three. It's much more likely they dropped the lowest, after all, you can't start numbering things with two!
I've just done a couple comparisons of the star ratings on a k5iis, the 4 star rating is on the absurd level of extra MB for miniscule quality gain. On one of my images, the 4 star gave a 9.6MB image, 3 star gave a 6.3MB image. I stacked them in photoshop and had it take the difference of the two versions. 54.3% of the pixels matched, 87.77% were off by 1 level or less, 97.48% were off by 2 levels or less, 99.59% were off by 3 levels or less. Results may vary depending on the content but in practical working terms, there's not much difference for the image I used...

The difference between 4 and 1 star is also pretty small (31% of pixels matched, 92% were off by 3 or less, 99% were off by 7 or less), just flicking between the two at 100% it takes some staring to tell the difference. I was going to complain about the 1 star setting being called 'good' and not 'crappy', but it actually is pretty good. They don't have an in-camera jpeg setting that majorly gimps you with compression.

I believe the default setting on the k5iis was 3 stars. There's not much to be gained going to 4 (that's not to say some people might find it worth it). I wouldn't be surprised if they chopped 4 stars, but as sterretje and I have conjectured they could easily have moved the other star ratings around. It's tough to compare, but a k3 saving jpegs at 14mp compared to the k20 might be illuminating.
02-01-2015, 09:01 AM   #21
Banned




Join Date: Feb 2013
Posts: 390
QuoteOriginally posted by sterretje Quote
I think that you did not read my earlier post

K30 3 star ('best' setting) gives the same estimated filesize as K5 3 star ('best' setting); K5 series however still has 4 star ('premium' setting) which gives a larger estimated filesize. So yes, in this case Pentax definitely removed the 'premium'.

It however is possible that 3 star on the K3 is somewhere between 3 star and 4 star of the K5.

Below table gives an overview from the K100D onwards; the 3* and 4* columns give the approximate storage capacity (number of files).

ModelMpix3*4*
K100D6660 
K10D10409 
K200D10409 
K200010469 
K-x12281 
K-r12281 
K20D14239138
K714238148
K0116251 
K5 series16214134
K30 / K5016214 
K-S0120134 
K324111 
Two possibly interesting points:
  • Only the K20D, K7 and K5 (series) have a 'premium' quality setting.
  • Difference between the K200D/K10D and the K2000; both using the same sensor; with the same amount of recorded pixels, the compression of the K2000 seems a tad higher and the quality should therefore be a tad lower.

You're certainly entitled to your opinion, but why would you do away with your "best" setting and leave the "worst"?? Seeing as how each camera has it's own/different processing engine, and uses differing algorithms, that table you show is meaningless.
02-01-2015, 09:18 AM   #22
Veteran Member




Join Date: Apr 2013
Location: Northern Minnesota
Photos: Gallery | Albums
Posts: 1,812
QuoteOriginally posted by sterretje Quote
I think that you did not read my earlier post

K30 3 star ('best' setting) gives the same estimated filesize as K5 3 star ('best' setting); K5 series however still has 4 star ('premium' setting) which gives a larger estimated filesize. So yes, in this case Pentax definitely removed the 'premium'.

It however is possible that 3 star on the K3 is somewhere between 3 star and 4 star of the K5.

Below table gives an overview from the K100D onwards; the 3* and 4* columns give the approximate storage capacity (number of files).

ModelMpix3*4*
K100D6660 
K10D10409 
K200D10409 
K200010469 
K-x12281 
K-r12281 
K20D14239138
K714238148
K0116251 
K5 series16214134
K30 / K5016214 
K-S0120134 
K324111 

Two possibly interesting points:
  • Only the K20D, K7 and K5 (series) have a 'premium' quality setting.
  • Difference between the K200D/K10D and the K2000; both using the same sensor; with the same amount of recorded pixels, the compression of the K2000 seems a tad higher and the quality should therefore be a tad lower.
Just want to point out that the file size is inversely proportional to the number of files that fit on the example SDHC card. The K-3 at 111 photos at 3 star has larger JPEG files than the K-5's 4 star 134.

02-01-2015, 09:21 AM   #23
Veteran Member




Join Date: Dec 2010
Location: Ontario
Photos: Gallery
Posts: 3,332
QuoteOriginally posted by Al_Kahollick Quote
You're certainly entitled to your opinion, but why would you do away with your "best" setting and leave the "worst"??
For reasons I've explained- the higher quality levels of jpeg compression become very inefficient in terms of file size costs vs gain in quality (or rather 'lack of loss' in quality) and at some point are just not worth it. It looks to me that the 4 options they had for several models pushed the upper level of what I'd personally call 'worth it', and not inconceivable that they would remove the extreme on the high end (I'm not ruling out re-positioning the lower options). You can draw your own conclusions from the available evidence.

---------- Post added 02-01-15 at 11:48 AM ----------

QuoteOriginally posted by waterfall Quote
Just want to point out that the file size is inversely proportional to the number of files that fit on the example SDHC card. The K-3 at 111 photos at 3 star has larger JPEG files than the K-5's 4 star 134.
jpeg filesize seems to scale roughly linearly with the total number of pixels, that is given the same level of compression a 24 mp image should be roughly 1.5 times as many megabytes as a 16mp image. So you'd expect about 166 k-5 4-stars to take the same space as 111 k-3 3-star if they used the same level of compression. I don't know the nitty-gritty of the jpeg protocols, but this looks suspiciously like the k-3 3-star is lower quality than the k-5 4-star (though still very high in the grand scheme of jpeg compression).
02-01-2015, 10:51 AM   #24
Banned




Join Date: Feb 2013
Posts: 390
QuoteOriginally posted by BrianR Quote
the available evidence

What evidence? Do you mean the fact that each model has a different processor and thus uses different schemes for jpeg compression, meaning file size has nothing to do with image quality across models? Would it really be in a manufacturers best interest to dump the best quality a given camera could produce, and keep the worst??? I think not!




QuoteOriginally posted by BrianR Quote
given the same level of compression
Again, each camera has it's own processor / compression scheme, They're NOT using the same level of compression, it's apples and oranges! Equating file size with quality is futile.


Quite simply, regardless of whether it's called 3 star, 4 star, premium, bestest, or pixeliscious, the top rate will undoubtedly produce the best image for that particular sensor / processor combo. File size can only be used to guage how many photos will fit on a card of a given size at a given setting, and that's merely an estimate, not a hard and fast number.
02-01-2015, 11:12 AM   #25
Veteran Member




Join Date: Jan 2012
Location: Slovenia
Photos: Gallery
Posts: 2,182
This might be of interest:
Jeffrey Friedl's Blog
02-01-2015, 11:22 AM   #26
Veteran Member




Join Date: Apr 2013
Location: Northern Minnesota
Photos: Gallery | Albums
Posts: 1,812
QuoteOriginally posted by BrianR Quote
For reasons I've explained- the higher quality levels of jpeg compression become very inefficient in terms of file size costs vs gain in quality (or rather 'lack of loss' in quality) and at some point are just not worth it. It looks to me that the 4 options they had for several models pushed the upper level of what I'd personally call 'worth it', and not inconceivable that they would remove the extreme on the high end (I'm not ruling out re-positioning the lower options). You can draw your own conclusions from the available evidence.

---------- Post added 02-01-15 at 11:48 AM ----------



jpeg filesize seems to scale roughly linearly with the total number of pixels, that is given the same level of compression a 24 mp image should be roughly 1.5 times as many megabytes as a 16mp image. So you'd expect about 166 k-5 4-stars to take the same space as 111 k-3 3-star if they used the same level of compression. I don't know the nitty-gritty of the jpeg protocols, but this looks suspiciously like the k-3 3-star is lower quality than the k-5 4-star (though still very high in the grand scheme of jpeg compression).
Interestingly, the in-camera RAW developer will output 4 grades of JPEG, with lowest labelled XS for extra small (source K-3 manual).
02-01-2015, 11:27 AM   #27
Veteran Member




Join Date: Dec 2010
Location: Ontario
Photos: Gallery
Posts: 3,332
QuoteOriginally posted by Al_Kahollick Quote
What evidence? Do you mean the fact that each model has a different processor and thus uses different schemes for jpeg compression, meaning file size has nothing to do with image quality across models? Would it really be in a manufacturers best interest to dump the best quality a given camera could produce, and keep the worst??? I think not!
I suggest you get a camera with 4 stars and compare the same picture at different settings, or even try the 3 different stars on the ones you have. Or use your favorite photo editor and compare the jpeg outputs at different settings. Compare the difference in image quality and filesize of the top quality levels of jpeg compression. Seriously, it's worth your time if you think there's some huge difference in the top levels. There becomes little reason to include the very highest jpeg options unless you really love to bloat up your files for almost immeasurable gains or you want to declare your camera has a "Premium" setting for marketing purposes.

The jpeg compression scheme is a compromise between file size and image degradation. Which compromise you choose to use is subjective (<-that's important!), and dependent on many factors (available memory, final use of the image, etc.) and imo it's reasonable to think Pentax engineers would reevaluate the priorities as camera technology advances.

Take a look at Giklab's link. Do you think users would notice a difference in image quality if the top star corresponded to lightrooms 93-100 bin or it's 85-92 bin (or even the 77-84 bin)? Enough to be worth the massive hit in filesize?

QuoteOriginally posted by Al_Kahollick Quote
Again, each camera has it's own processor / compression scheme, They're NOT using the same level of compression, it's apples and oranges! Equating file size with quality is futile.
They're still using jpeg compression. You can't mess with it outside the jpeg parameters if you hope to be compatible with programs that read jpegs.

---------- Post added 02-01-15 at 01:31 PM ----------

QuoteOriginally posted by waterfall Quote
Interestingly, the in-camera RAW developer will output 4 grades of JPEG, with lowest labelled XS for extra small (source K-3 manual).
That's 4 levels for number of pixels, not the "star quality".
02-01-2015, 01:00 PM   #28
Veteran Member




Join Date: Apr 2013
Location: Northern Minnesota
Photos: Gallery | Albums
Posts: 1,812
QuoteOriginally posted by BrianR Quote
I suggest you get a camera with 4 stars and compare the same picture at different settings, or even try the 3 different stars on the ones you have. Or use your favorite photo editor and compare the jpeg outputs at different settings. Compare the difference in image quality and filesize of the top quality levels of jpeg compression. Seriously, it's worth your time if you think there's some huge difference in the top levels. There becomes little reason to include the very highest jpeg options unless you really love to bloat up your files for almost immeasurable gains or you want to declare your camera has a "Premium" setting for marketing purposes.

The jpeg compression scheme is a compromise between file size and image degradation. Which compromise you choose to use is subjective (<-that's important!), and dependent on many factors (available memory, final use of the image, etc.) and imo it's reasonable to think Pentax engineers would reevaluate the priorities as camera technology advances.

Take a look at Giklab's link. Do you think users would notice a difference in image quality if the top star corresponded to lightrooms 93-100 bin or it's 85-92 bin (or even the 77-84 bin)? Enough to be worth the massive hit in filesize?



They're still using jpeg compression. You can't mess with it outside the jpeg parameters if you hope to be compatible with programs that read jpegs.

---------- Post added 02-01-15 at 01:31 PM ----------



That's 4 levels for number of pixels, not the "star quality".
I know that. Just interesting that there a 4 levels in camera raw development and 3 levels on JPEG setting.
02-01-2015, 01:25 PM   #29
Veteran Member




Join Date: Dec 2010
Location: Ontario
Photos: Gallery
Posts: 3,332
Something concrete, I compared the k-5 and k-3 jpegs from Digital Cameras, Pentax K-5 Digital Camera Test Image and here Digital Cameras, Pentax K-3 Digital Camera Test Image

Both use the same Quantization tables for this part of the lossy jpeg compression (table is filled with 1's), so this part is the same.

So, jpeg gurus, why is the k-3 file not as large as you'd expect from the increased resolution? From what I've found there isn't much to be gained in the lossless portion of the jpeg compression (such as custom Huffman tables), but maybe there's enough to make a difference. There's also the conversion to YCbCr mode that might be handled differently. I dunno.
02-01-2015, 01:38 PM   #30
Veteran Member




Join Date: Apr 2013
Location: Northern Minnesota
Photos: Gallery | Albums
Posts: 1,812
QuoteOriginally posted by BrianR Quote
Something concrete, I compared the k-5 and k-3 jpegs from Digital Cameras, Pentax K-5 Digital Camera Test Image and here Digital Cameras, Pentax K-3 Digital Camera Test Image

Both use the same Quantization tables for this part of the lossy jpeg compression (table is filled with 1's), so this part is the same.

So, jpeg gurus, why is the k-3 file not as large as you'd expect from the increased resolution? From what I've found there isn't much to be gained in the lossless portion of the jpeg compression (such as custom Huffman tables), but maybe there's enough to make a difference. There's also the conversion to YCbCr mode that might be handled differently. I dunno.
Oh man, your question gave me a headache! Ask an electrical engineer, shall we?
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!
amount, camera, cameras, compression, compromise, course, dslr, file, image, jpeg, k-3, k-5, k3, level, levels, parameter, parameters, photography, quality, settings, size, star, stars
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
One-Push File format JPEG to RAW+ timmijo Pentax K-30 & K-50 8 09-08-2015 06:50 PM
File size for slideshows? frogoutofwater Digital Processing, Software, and Printing 4 08-25-2014 06:43 AM
K-x settings for file size noratorius Troubleshooting and Beginner Help 7 06-26-2013 06:47 AM
K20d jpeg file size rustynail925 Pentax DSLR Discussion 8 09-05-2009 06:00 PM
JPEG file size Mike Moffet Film SLRs and Compact Film Cameras 4 10-28-2007 03:20 PM



All times are GMT -7. The time now is 12:52 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