Forgot Password
Pentax Camera Forums Home
 

Closed Thread
Show Printable Version Search this Thread
01-07-2020, 12:45 PM   #1306
Pentaxian




Join Date: May 2015
Photos: Albums
Posts: 3,304
QuoteOriginally posted by Kunzite Quote
Oh, boy. Why do you keep assuming the KP RAWs are unprocessed?
I'm sorry I don't follow you here.

QuoteOriginally posted by Kunzite Quote
And it doesn't work as you think, except on uniform surfaces like test charts. Try this "pulling away" thing while retaining a high image quality...
I remember seeing a while ago two images equally ruined by too much NR as a "proof" that the accelerator can be replicated in software. Sigh..,
We don't know if the accelerator is better or worse than any pp noise reduction that's absolutely true. It could well do tricks that you can't replicate as successfully at other points in the pipeline. We do know that you can raise DR significantly by running nr. Those graphs will pull away for the other cameras if NR is applied, the amount can vary depending on pp. This means that those graphs aren't showing what reh321 implies.

The KP could however retain more detail or have otherwise more pleasing results with the acellerator but that's not what is being discussed. The accelerator can well be a great trade off. That's not the issue here.

The following statements remain.

1. Noise reduction will increase dynamic range
2. Noise reduction will hamper further noise reduction down the pipe. The fact the bclaff can detect the nr in the raw data means it will affect noise reduction down the pipe.

---------- Post added 01-07-20 at 12:51 PM ----------

QuoteOriginally posted by reh321 Quote
These 'raw' files can still be changed. Colors can be changed, and noise can be changed. That is all that matters.

And by the way, has anyone actually looked at these graphs? The D500 has kinks in its {blue} lines too; they are smaller and earlier than those in the {green} KP lines, but they are there. I don't hear anything about Nikon "cooking" its 'raw' files, but the lines seem to say that Nikon starts processing earlier.

AND REMEMBER - the only reason I posted these two charts originally was to show in them what Nikon users have told me: the D750 is no slouch either; the D500 and KP were there initially just for reference.

Those graphs show that *all three* cameras are cooking the raws! It's indicated by the triangles. The difference is that the KP is doing it at relatively low iso whereas the others are doing it way up the scale.

Your point about the two Nikons is great and shows how much camera lore should be taken with a grain of salt. There could however be other factors not shown in these diagrams that make the 750 a great low light camera.


Last edited by house; 01-07-2020 at 12:51 PM.
01-07-2020, 12:56 PM - 2 Likes   #1307
Pentaxian
reh321's Avatar

Join Date: Nov 2014
Location: South Bend, IN, USA
Photos: Gallery | Albums
Posts: 23,125
QuoteOriginally posted by house Quote
We don't know if the accelerator is better or worse than any pp noise reduction that's absolutely true. It could well do tricks that you can't replicate as successfully at other points in the pipeline. We do know that you can raise DR significantly by running nr. Those graphs will pull away for the other cameras if NR is applied, the amount can vary depending on pp. This means that those graphs aren't showing what reh321 implies.

The KP could however retain more detail or have otherwise more pleasing results with the acellerator but that's not what is being discussed. The accelerator can well be a great trade off. That's not the issue here.

The following statements remain.

1. Noise reduction will increase dynamic range
2. Noise reduction will hamper further noise reduction down the pipe. The fact the bclaff can detect the nr in the raw data means it will affect noise reduction down the pipe.
There is no reason to believe that other cameras will "pull away" from the KP when "PP" is applied; if two cameras are the same on the charts, there is a priori no reason to assume that they will not react to the same "PP" the same way. There is no reason to believe that Nikon processing leaves a D500 'raw' file more open to a "PP" program than Pentax processing does.

REMEMBER: the reason for showing the graphs was to show the D750; the D500 and KP were there only as a reference

Last edited by reh321; 01-07-2020 at 01:02 PM. Reason: word selection
01-07-2020, 01:18 PM - 1 Like   #1308
Pentaxian




Join Date: Apr 2007
Location: Romania
Posts: 15,132
QuoteOriginally posted by house Quote
I'm sorry I don't follow you here.
If you process the other camera's RAWs, you should also process the KP's.

QuoteOriginally posted by house Quote
We don't know if the accelerator is better or worse than any pp noise reduction that's absolutely true. It could well do tricks that you can't replicate as successfully at other points in the pipeline. We do know that you can raise DR significantly by running nr. Those graphs will pull away for the other cameras if NR is applied, the amount can vary depending on pp. This means that those graphs aren't showing what reh321 implies.

The KP could however retain more detail or have otherwise more pleasing results with the acellerator but that's not what is being discussed. The accelerator can well be a great trade off. That's not the issue here.

The following statements remain.

1. Noise reduction will increase dynamic range
2. Noise reduction will hamper further noise reduction down the pipe. The fact the bclaff can detect the nr in the raw data means it will affect noise reduction down the pipe.
Those graphs are showing that the KP applies a form of processing, on uniform image areas, that can be detected by the analysis done by bclaff. An analysis incomplete at best, and much misused.

On the contrary, the question should always be the real world effect of the accelerator.

So far the issues are all theoretical or imaginary.
The big problem is the post in which you were basically saying the accelerator processing prevents any processing to be done. Now you're talking about hampering this, affecting that - did you actually try, real world scenarios in which the accelerator did more harm than good?
01-07-2020, 01:31 PM - 8 Likes   #1309
Site Supporter
Site Supporter
Pål Jensen's Avatar

Join Date: Dec 2007
Location: Norway
Photos: Albums
Posts: 4,371
This thread should be renamed " No new information about new flagship selling date"...

01-07-2020, 01:49 PM   #1310
Loyal Site Supporter
Loyal Site Supporter




Join Date: May 2016
Location: East Coast
Posts: 2,903
QuoteOriginally posted by reh321 Quote
I normally hear of the 'accelerator' in the context of Noise Reduction. Testing shows cameras with it also have an increase in Dynamic Range; most likely the color fidelity I've noticed {but is not tested, at least not in the lab studies I've seen} is also connected to the same 'accelerator' operation. In general, no matter what you look at, the 'accelerator' just improves higher ISO performance.
Yep! And it really makes high ISO easy. I used to top out around 3200 on my K-1, but now I go much higher on my K-1 II. The files are pretty darn good right out of the camera. I always shoot RAW+, but the OOC jpegs are pretty solid, and the level of effort is nil. Like that!
01-07-2020, 01:58 PM - 1 Like   #1311
Pentaxian
reh321's Avatar

Join Date: Nov 2014
Location: South Bend, IN, USA
Photos: Gallery | Albums
Posts: 23,125
QuoteOriginally posted by clickclick Quote
Yep! And it really makes high ISO easy. I used to top out around 3200 on my K-1, but now I go much higher on my K-1 II. The files are pretty darn good right out of the camera. I always shoot RAW+, but the OOC jpegs are pretty solid, and the level of effort is nil. Like that!
Then I just noticed this photo submitted by a happy user - bringing us to reality {which is all that ultimately matters}
Amsterdam Central Station - PentaxForums.com

Last edited by reh321; 01-07-2020 at 02:05 PM. Reason: continuing thought
01-07-2020, 03:17 PM   #1312
Senior Member
xmeda's Avatar

Join Date: Feb 2011
Location: Brno
Posts: 295
Well if you have good 60mpix FF sensor output and you resize it to 24mpix.. you'll clean out a lot of noise. Also there is the advantage in image crops. Even if you cut the photo a lot, it retains a lot of details.

But yes, the hard-drive consumption is huge

01-07-2020, 03:26 PM   #1313
Pentaxian
photoptimist's Avatar

Join Date: Jul 2016
Photos: Albums
Posts: 5,113
QuoteOriginally posted by xmeda Quote
Well if you have good 60mpix FF sensor output and you resize it to 24mpix.. you'll clean out a lot of noise. Also there is the advantage in image crops. Even if you cut the photo a lot, it retains a lot of details.

But yes, the hard-drive consumption is huge
Indeed!

And yet hard disks are so cheap, about 3 cents per gigabyte. These days, a $1 will buy enough space to store 266 images with 60MPix and 16 bits depth.
01-07-2020, 03:39 PM   #1314
Senior Member
xmeda's Avatar

Join Date: Feb 2011
Location: Brno
Posts: 295
Oh yes.. still keeping my first MMC 64MB card bought for my first 4Mpix Casio QV-R40.. Memory card for about 35-40 fine JPGs with the price same as 1TB HDD cost today
01-07-2020, 03:47 PM - 4 Likes   #1315
Digitiser of Film
Loyal Site Supporter
BigMackCam's Avatar

Join Date: Mar 2010
Location: North East of England
Posts: 20,572
QuoteOriginally posted by xmeda Quote
Well if you have good 60mpix FF sensor output and you resize it to 24mpix..
...
But yes, the hard-drive consumption is huge
This is a good point worth touching on a little further...

It's not just hard drive consumption that's a concern... It's processing power too. With 50 - 60MP (and bigger) files, you need a current (or, at least, very recent) generation CPU, a good GPU, lots of memory for both, plus - ideally - fast SSD working storage to edit and process smoothly in real time and export collections of images with many adjustments in acceptable time-frames. Then, as you indicate, significant hard drive storage is required... not only for the working raw and final output files, but for immediate backups and archives too.

Of course, for a certain very narrow range of working professionals who actually need the very highest resolution and frequently print at very large dimensions for close-in viewing, the costs of such a technology setup in addition to the photographic equipment will be justifiable. Most of those folks don't even post here, though. The vast majority of our members range from beginners to advanced hobbyists, plus a few professionals at varying levels of technical ability and customer requirement. In this group, I'd argue that 24MP APS-C and 36MP full frame is more than ample. At these "lower" resolutions, most serious amateur applications are more than adequately covered for print sizes, and sensor performance in other areas is better whilst keeping the cost - both for the photographic equipment itself and the computer technology required to process and store images - at realistic levels.

I've found 24MP files to be a nice size to work with, regardless of format. Both my Pentax APS-C and Hasselblad / Sony full-frame gear produce 24MP files. They give me all of the image quality I need for reproduction sizes and purposes I'm ever likely to require, and they don't need a great deal of processing power or storage to work with. For all the talk on forums such as these by a very limited number of more demanding users (few of whom I know to actually need more, as opposed to talking about it), I suspect I'm part of the vast majority...

Last edited by BigMackCam; 01-07-2020 at 05:05 PM.
01-07-2020, 03:52 PM - 6 Likes   #1316
Site Supporter
Site Supporter
Pål Jensen's Avatar

Join Date: Dec 2007
Location: Norway
Photos: Albums
Posts: 4,371
The concept of "good enough" is underrated...
01-07-2020, 04:02 PM   #1317
Senior Member
xmeda's Avatar

Join Date: Feb 2011
Location: Brno
Posts: 295
Some years ago people were saying that 6 Mpix is plenty enough and that 12 Mpix EOS 5D is only for hi-end lanscapes..
Now my 5 years old compact camera has 12 Mpix CMOS and my phablet has 16 Mpix

I'm pretty sure that next decade we will be talking about 100 Mpix FF cameras as standard or at least cameras with smaller sensor capable to do multi-frame-stack even in short time like 1/200s to gain huge resolution output.



Same with video.. my first compact camera had 320x240 30sec videos without audio and now we have tools for 8K capture...
01-07-2020, 04:14 PM - 2 Likes   #1318
Digitiser of Film
Loyal Site Supporter
BigMackCam's Avatar

Join Date: Mar 2010
Location: North East of England
Posts: 20,572
QuoteOriginally posted by xmeda Quote
Some years ago people were saying that 6 Mpix is plenty enough and that 12 Mpix EOS 5D is only for hi-end lanscapes..
What's interesting about that observation is, depending on the final output size and media, it's still more-or-less as true as ever. See my previous post... Plenty of folks talk about needing more, but the number that actually need more is far, far less, IMHO. Many hobbyists can still get by just fine with a 6 or 10MP camera (so long as they're not cropping heavily or printing big) or a somewhat more recent 16MP camera like your own. With a recent 24 - 36MP camera, regardless of format, most of us have more to work with than we'll ever need...

Last edited by BigMackCam; 01-07-2020 at 04:24 PM.
01-07-2020, 04:24 PM - 1 Like   #1319
Senior Member
xmeda's Avatar

Join Date: Feb 2011
Location: Brno
Posts: 295
QuoteOriginally posted by BigMackCam Quote
What's interesting about that observation is, depending on the final output size and media, it's still more-or-less as true as ever. See my previous post... Plenty of folks talk about needing more, but the number that actually need more is far, far less, IMHO. Many hobbyists can still get by just fine with a 6 or 10MP camera (so long as they're not cropping heavily or printing big) or a somewhat more recent 16MP camera like your own. With a 24 - 36MP camera, regardless of format, most of us have more to work with than we'll ever need...

16 Mpix mostly does the job.. but when I'm shooting some planes or birds, then cropping option comes very handy (in case lens quality allows such thing).

After playing recently a lot with colleague's D850... hmm.. food makes you hungry
01-07-2020, 04:29 PM - 2 Likes   #1320
Site Supporter
Site Supporter
Larrymc's Avatar

Join Date: Jun 2014
Location: Mississippi, USA
Posts: 5,245
QuoteOriginally posted by BigMackCam Quote
This is a good point worth touching on a little further...

It's not just hard drive consumption that's a concern... It's processing power too. With 50 - 60MP (and bigger) files, you need a current (or, at least, very recent) generation CPU, a good GPU, lots of memory for both, plus - ideally - fast SSD working storage to edit and process smoothly in real time and export collections of images with many adjustments in acceptable time-frames. Then, as you indicate, significant hard drive storage is required... not only for the working raw and final output files, but for immediate backups and archives too.

Of course, for a certain very narrow range of working professionals who actually need the very highest resolution and frequently print at very large dimensions for close-in viewing, the costs of such a technology setup in addition to the photographic equipment will be justifiable. Most of those folks don't even post here, though. The vast majority of our members range from beginners to advanced hobbyists, plus a few professionals at varying levels of technical ability and customer requirement. In this group, I'd argue that 24MP APS-C and 36MP full frame is more than ample. At these "lower" resolutions, most serious amateur applications are more than adequately covered for print sizes, and sensor performance in other areas is better whilst keeping the cost - both for the photographic equipment itself and the computer technology required to process and store images - at realistic levels.

I've found 24MP files to be a nice level to work with, regardless of format. Both my Pentax APS-C and Hasselblad / Sony full-frame gear produce 24MP files. They give me all of the image quality I need for reproduction sizes and purposes I'm ever likely to require, and they don't require a great deal of processing power or storage to work with. For all the talk on forums such as these by a very limited number of more demanding users (few of whom I know to actually need more, as opposed to talking about it), I suspect I'm part of the vast majority...
Well said, Mike! I fully agree.
Closed Thread

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!
battery, blog, camera, card, discussion, features, firmware, google, information, issue, joke, k1, k3, lcd, model, pentax, pentax news, pentax rumors, people, pm, post, product, production, prototype, ricoh, screen, shots, view
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
Who knows the actually information about new aps-c flagship? Karen the Star Pentax DSLR Discussion 48 06-28-2019 11:15 AM
Photo Taken Date vs Making Date RoundWharf General Photography 16 02-12-2019 02:53 AM
Need information how to get focus pulse / distance information from camera sliawatimena Welcomes and Introductions 12 01-09-2019 07:31 AM
K-1 not releasing with D-FA* 50mm f/1.4 K David Pentax SLR Lens Discussion 10 12-11-2018 12:25 PM
Information overflow: how to keep up-to-date? bymy141 Pentax News and Rumors 13 05-17-2009 04:00 AM



All times are GMT -7. The time now is 02:07 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