Forgot Password
Pentax Camera Forums Home
 

Reply
Show Printable Version Search this Thread
03-25-2019, 01:11 PM   #31
Site Supporter
Site Supporter
BruceBanner's Avatar

Join Date: Dec 2015
Location: Blue Mountains, NSW, Australia
Posts: 3,459
Original Poster
QuoteOriginally posted by BrianR Quote
Sorry to hear that it did not work for you. I've used pdcu4 and it produces images that are pretty close to the same rendering as the in-camera processing (in most cases) and it is definitely capable of producing full resolution jpegs at high quality jpeg settings. It is certainly clunky and not a fast program.

As a side note, it is also able to rapidly extract the full resolution jpeg previews from pef or dng files out of my k5iis (note the "extract jpeg" option is distinctly different from the "save as" option).
Yeh I wrote about that in my long winded explanation that I lost. Basically 3 ways to get a Jpg;
  1. 'Extract Jpg' (pulls a ** image only instead of ***, file size is only 5mb vs 18mb like it should be from Raw Development in camera)
  2. Save As with Image Processing. Creates the largest file, can select **** (Premium, possibly even higher quality than the *** from Raw Development), however the file alters in exposure or some other way from the Jpg Preview, some kind of automatic attempt at correcting exposure (I took my shot at -0.7EV Compensation) and even without (and with) visiting Labarotory cannot stop it from applying this 'fix')
  3. Save As, can again adjust quality to **** even, but the maximum file size I got was about 8mb, 10mb shy of what In Camera or RAW+ generates.

Basically I felt it wasn't doing anything better than what FSIV was doing. It also felt slow, buggy and clunky to operate, crashed once etc.

I was just curious to see if a third party program could spit out a Jpg from a DNG to being exactly the same size, or like extremely close to it is all. I think I just need to have a play with IQ settings in FSIV, it could be that even the in camera Jpg Raw Development is overkill, and as many have pointed out having the Quality slider at 90/100 in FSIV generates a significantly smaller size and going to 100/100 doubles the file size with little to no perceived increase in IQ.

03-26-2019, 09:10 AM - 1 Like   #32
Pentaxian
rogerstg's Avatar

Join Date: Oct 2007
Location: Rhode Island
Photos: Albums
Posts: 394
Shooting a KP, I find that the camera processed image is pretty good most of the time, but like BruceB, would like the extra headroom of RAW v jpg and dislike the speed penalty of Raw+.

For now, I've settled on using RAW with mass in-camera conversion to tiff. Before downloading to my computer, I parse my photo's of obvious crap shots, then convert all to TIFF. It seems that camera settings at the time the photo was taken are used when converting, though I have not extensively tested this.

After which I transfer all to my computer. That gives me the choice of starting with unprocessed or camera processed shots while maintaining lots of headroom if needed, before saving as jpg. Usually I simply tweak the tiff file. In-camera converting to tiff can be done quickly, a folder at a time. This works for me until I learn of something better.
03-26-2019, 12:35 PM   #33
Site Supporter
Site Supporter
BruceBanner's Avatar

Join Date: Dec 2015
Location: Blue Mountains, NSW, Australia
Posts: 3,459
Original Poster
QuoteOriginally posted by rogerstg Quote
Shooting a KP, I find that the camera processed image is pretty good most of the time, but like BruceB, would like the extra headroom of RAW v jpg and dislike the speed penalty of Raw+.

For now, I've settled on using RAW with mass in-camera conversion to tiff. Before downloading to my computer, I parse my photo's of obvious crap shots, then convert all to TIFF. It seems that camera settings at the time the photo was taken are used when converting, though I have not extensively tested this.

After which I transfer all to my computer. That gives me the choice of starting with unprocessed or camera processed shots while maintaining lots of headroom if needed, before saving as jpg. Usually I simply tweak the tiff file. In-camera converting to tiff can be done quickly, a folder at a time. This works for me until I learn of something better.
Thanks rogerstg, I never thought about DNG>Tiff in camera, I'll do some testing and report back.

The only time I have taken a DNG file and produced a Tiff file from it in the early stages of editing has been when taking Pixelshifted images to RawTherapee for motion correction, the outcome is a Tiff file. I have noticed that this process we lose the Lens Profile Correction once we take that Motion Corrected Tiff file to LR, now I'm curious if the in camera RAW>Tiff loses any EXIF etc.
04-18-2019, 07:21 AM - 2 Likes   #34
Veteran Member




Join Date: Jul 2009
Location: Ottawa, Canada
Posts: 363
Sorry to chime in a little late on this thread. I haven't read every post in detail, so I apologize if I'm repeating what others have already said.

I've been using FastStone for many years, starting back when I got my K200D - around 10 years ago. I think FastStone is a fantastic tool for viewing/culling photos, and for performing simple edits of JPG/TIFF images.

With RAW files, by default it works with the embedded JPG. You can set it up to convert the RAW file (using DCRAW I believe), but this is slow, and you have no control over how the conversion is done ( AFAIK you can't adjust exposure, white balance, etc. ). IMHO that defeats the purpose of FastStone ( which is to give you a fast, lightweight viewer/editor ), and doesn't really give you any of the advantages of even the crudest RAW converter.

When I was using my K200D, I had problems shooting RAW+. The camera had trouble keeping up and would often corrupt slower memory cards, so I got in the habit of just shooting RAW. That particular camera embedded very high quality JPGs in the RAW files, which I could extract using FastStone. The quality of the embedded JPGs rivaled what I could get by converting the RAW file using any other converter I tried including Lightroom, Elements, DxO, Rawtherapee, Silkypix, and PDCU ( assuming the photo had been exposed correctly to begin with ). I think this was an anomaly, and most cameras do not embed high quality JPGs in the RAW file.

When I moved up to a K30, my card issues went away, so RAW+ became a viable option again. I noticed that the DNG files generated by my K30 weren't proportionally bigger than those generated by my K200D. I also noticed that the quality of the embedded JPGs that were in the K30's RAW files were not as good as those from my K200D. You could see posterization, etc. My guess is that the K30 increased the compression on the JPGs embedded JPGs in its RAW files.

I can say with certainty that "normally", FastStone does use the JPG that's embedded in the RAW files. Why would one want to use the embedded JPG? It depends on what you use your photographs for. My photos serve 2 purposes - many of my photos are just recordings of observations. I don't need fantastic photos for this purpose - a low res JPG that I can archive and/or attach to an email will do. And I don't necessarily want to spend a lot of time generating that JPG. Shooting JPG is not an option as I don't know in advance when the opportunity for the "money shot" will occur. I can't afford to miss a shot while changing mode from JPG to RAW. My preference is to shoot RAW+, and then to cull after the fact (keep a cropped version of the JPG for pure record shots, and retain the RAWs for the "keepers"). With my K200D, RAW+ didn't work so well, so I defaulted to shooting RAW. I didn't want to spend a lot of time converting all the RAW files for my "record" shots, so I would extract and keep the embedded JPGs for them, and only keep the RAWs for the really good shots.

So I can see valid reasons for shooting RAW alone, and for RAW+. But I think you want to look very carefully at the JPGs you extract from your RAW file before you opt for using them for anything critical. Those I got from my K200D were high quality, but those I get from my K30 aren't great (but they may be good enough for some uses). I haven't looked carefully at those I can get from my KP yet. Also, the OP should note that you can't increase the quality of the embedded JPG using FastStone using "Save As". The software gives you the option of saving that image at 100%, or as a TIFF, but that isn't going to improve the image quality over what the camera stored in the RAW file. You can set it to 100%, but that just means you are storing a big file. It's the same if you open a JPG file that was originally stored at 85%. Storing it as 100% doesn't improve it.

Normally, I work almost exclusively with FastStone. Day to day, most of what I do is culling my shots and generating low res "sample" JPGs that I use for various purposes. FastStone is excellent for this kind of thing, especially if your computer is on the slower side.

It's only when I have a photo that either needs adjustment (exposure and/or white balance), or where I want to get the absolute best result I can (say, for publication) that I'll resort to using a RAW converter. I've used several over the years. I've settled on PDCU and DxO as my preferred RAW converters. In general, I really like the results I get from PDCU. It is my belief that PDCU is essentially the software we have in our cameras, with a somewhat crude UI bolted on. That is why the software understands all the camera settings, and can (easily) duplicate the JPGs produced by the camera. It's a bit clunky, but for one-off processing, it works for me. If I'm doing a bunch of landscape photos, I'll use DxO.

The OP should note that you can launch some RAW converters from FastStone (right click, "Edit with External Program"). This is very handy. I have all my RAW converters added to my external program list in FastStone. So while I'm browsing a folder of photos, if I decide I want to work on a particular RAW file, I right click, select my converter, and the image will pop up in the converter I've chosen. So basically, I navigate my photo archive using FastStone, and use it to select what I want to work on, either within FastStone, or within some other converter or editing program.

I hope this helps.

PS: you might want to try to sort out what is going wrong with PDCU on your computer. If you like the camera rendering, it's probably the best option available for duplicating that. What operating system are you using?

04-19-2019, 10:38 AM   #35
Pentaxian
rogerstg's Avatar

Join Date: Oct 2007
Location: Rhode Island
Photos: Albums
Posts: 394
QuoteOriginally posted by arkav Quote
I've been using FastStone for many years, starting back when I got my K200D - around 10 years ago. I think FastStone is a fantastic tool for viewing/culling photos, and for performing simple edits of JPG/TIFF images. . .

I hope this helps.
Your comprehensive response is very helpful and will probably change my initial workflow steps. One question: are your experiences based on using DNG or PEF as camera RAW files?
04-19-2019, 09:31 PM   #36
Site Supporter
Site Supporter
BruceBanner's Avatar

Join Date: Dec 2015
Location: Blue Mountains, NSW, Australia
Posts: 3,459
Original Poster
QuoteOriginally posted by arkav Quote
Sorry to chime in a little late on this thread. I haven't read every post in detail, so I apologize if I'm repeating what others have already said.

I've been using FastStone for many years, starting back when I got my K200D - around 10 years ago. I think FastStone is a fantastic tool for viewing/culling photos, and for performing simple edits of JPG/TIFF images.

With RAW files, by default it works with the embedded JPG. You can set it up to convert the RAW file (using DCRAW I believe), but this is slow, and you have no control over how the conversion is done ( AFAIK you can't adjust exposure, white balance, etc. ). IMHO that defeats the purpose of FastStone ( which is to give you a fast, lightweight viewer/editor ), and doesn't really give you any of the advantages of even the crudest RAW converter.

When I was using my K200D, I had problems shooting RAW+. The camera had trouble keeping up and would often corrupt slower memory cards, so I got in the habit of just shooting RAW. That particular camera embedded very high quality JPGs in the RAW files, which I could extract using FastStone. The quality of the embedded JPGs rivaled what I could get by converting the RAW file using any other converter I tried including Lightroom, Elements, DxO, Rawtherapee, Silkypix, and PDCU ( assuming the photo had been exposed correctly to begin with ). I think this was an anomaly, and most cameras do not embed high quality JPGs in the RAW file.

When I moved up to a K30, my card issues went away, so RAW+ became a viable option again. I noticed that the DNG files generated by my K30 weren't proportionally bigger than those generated by my K200D. I also noticed that the quality of the embedded JPGs that were in the K30's RAW files were not as good as those from my K200D. You could see posterization, etc. My guess is that the K30 increased the compression on the JPGs embedded JPGs in its RAW files.

I can say with certainty that "normally", FastStone does use the JPG that's embedded in the RAW files. Why would one want to use the embedded JPG? It depends on what you use your photographs for. My photos serve 2 purposes - many of my photos are just recordings of observations. I don't need fantastic photos for this purpose - a low res JPG that I can archive and/or attach to an email will do. And I don't necessarily want to spend a lot of time generating that JPG. Shooting JPG is not an option as I don't know in advance when the opportunity for the "money shot" will occur. I can't afford to miss a shot while changing mode from JPG to RAW. My preference is to shoot RAW+, and then to cull after the fact (keep a cropped version of the JPG for pure record shots, and retain the RAWs for the "keepers"). With my K200D, RAW+ didn't work so well, so I defaulted to shooting RAW. I didn't want to spend a lot of time converting all the RAW files for my "record" shots, so I would extract and keep the embedded JPGs for them, and only keep the RAWs for the really good shots.

So I can see valid reasons for shooting RAW alone, and for RAW+. But I think you want to look very carefully at the JPGs you extract from your RAW file before you opt for using them for anything critical. Those I got from my K200D were high quality, but those I get from my K30 aren't great (but they may be good enough for some uses). I haven't looked carefully at those I can get from my KP yet. Also, the OP should note that you can't increase the quality of the embedded JPG using FastStone using "Save As". The software gives you the option of saving that image at 100%, or as a TIFF, but that isn't going to improve the image quality over what the camera stored in the RAW file. You can set it to 100%, but that just means you are storing a big file. It's the same if you open a JPG file that was originally stored at 85%. Storing it as 100% doesn't improve it.

Normally, I work almost exclusively with FastStone. Day to day, most of what I do is culling my shots and generating low res "sample" JPGs that I use for various purposes. FastStone is excellent for this kind of thing, especially if your computer is on the slower side.

It's only when I have a photo that either needs adjustment (exposure and/or white balance), or where I want to get the absolute best result I can (say, for publication) that I'll resort to using a RAW converter. I've used several over the years. I've settled on PDCU and DxO as my preferred RAW converters. In general, I really like the results I get from PDCU. It is my belief that PDCU is essentially the software we have in our cameras, with a somewhat crude UI bolted on. That is why the software understands all the camera settings, and can (easily) duplicate the JPGs produced by the camera. It's a bit clunky, but for one-off processing, it works for me. If I'm doing a bunch of landscape photos, I'll use DxO.

The OP should note that you can launch some RAW converters from FastStone (right click, "Edit with External Program"). This is very handy. I have all my RAW converters added to my external program list in FastStone. So while I'm browsing a folder of photos, if I decide I want to work on a particular RAW file, I right click, select my converter, and the image will pop up in the converter I've chosen. So basically, I navigate my photo archive using FastStone, and use it to select what I want to work on, either within FastStone, or within some other converter or editing program.

I hope this helps.

PS: you might want to try to sort out what is going wrong with PDCU on your computer. If you like the camera rendering, it's probably the best option available for duplicating that. What operating system are you using?
I love chime ins, and I love your response, so thanks.

If I may I might point you to the post I made at the top of this page; FastStone Image Viewer and extracting Jpg's from RAW DNG's - Page 3 - PentaxForums.com

What I thought interesting was fact that PDCU offers **** Premium setting, whereas in camera development cuts out at ***. Does that mean the camera's *** = **** for PDCU? Does it mean anything? Does it mean the camera itself is not actually recording Jpgs as good as what they could be? Does that mean we can get a better Jpg from a RAW and selecting ****?!?

Odd.

But I understand what you are saying that if when taking a RAW file it keeps a Jpg for later extracting and that Jpg has 'a limit', what I don't understand then is when playing around with ***/**** quantities or other extraction settings the file size does change.

Example. You have a RAW 40mb file, you use the in camera menu and use all the High Quality settings (*** etc). It gives you say a 15mb Jpg. We can imagine then this is the BEST Jpg inside the RAW. How then can we take that RAW file to various other programs (including PDCU) and generate a Jpg that is larger than this? If there is no higher quality possible, what explains the size difference? If say you pushed a slider up to 100% or whatever, thinking it would be higher quality, wouldn't the extraction just dump a file exactly the same as what the in camera extract can manage?

Either way, RAW+ is not nice, especially when writing to two sd cards, that is buffer hell for us pro event shooters. I enjoy the fact that I can shoot RAW (writing to two cards) and somehow get a decent Jpg from the RAW files at a later time if I really want it.

Right now my curve is whether shooting Jpg vs RAW (and extracting Jpg) actually has a significant difference in quality. The editing restrictions of Jpg vs Raw I think I can live with, I'm just talking about general stuff here like noise and sharpness. I do note for example when I use FS for my cull, I do use Jpgs of course for swiftness (this is just my first stage of culling), I get a good idea of whether the image is worth working on but often I can feel like the image is not good, missed focus, sharpness or noise bad, but fire that baby up in LR (using the aforementioned Edit in External Program) and everything changes, things are now fine (albeit flat and boring ha!). It could be my rendering settings for FS for the RAW Jpg previews are just set to being pretty bad.
04-19-2019, 09:41 PM   #37
Site Supporter
Site Supporter




Join Date: Mar 2008
Location: Prince George, BC
Photos: Gallery | Albums
Posts: 2,778
I think getting identical jpegs from different jpeg engines is asking the impossible.
04-19-2019, 11:14 PM   #38
Site Supporter
Site Supporter
BruceBanner's Avatar

Join Date: Dec 2015
Location: Blue Mountains, NSW, Australia
Posts: 3,459
Original Poster
QuoteOriginally posted by jbinpg Quote
I think getting identical jpegs from different jpeg engines is asking the impossible.
I think you're right, what is curious however is how PDCU (Pentax Digital Camera Utility) seems to very closely mimic the options of 'in camera' options, so I was curious how in specific this software works. So you can imagine my surprise when it offers a **** 'Premium' extract vs a *** which is offered in camera (and also as the maximum Jpg only shooting if that's what you want to do...).

Does it mean that you get higher quality Jpgs from shooting RAW and then using PDCU for a **** Jpg extracting process than actually turning off RAW and shooting Jpg only in camera at the maximum *** option given?!

Yeah...

04-20-2019, 09:58 AM   #39
Otis Memorial Pentaxian
Loyal Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 32,479
QuoteOriginally posted by BruceBanner Quote
Does it mean anything?
I doubt it, though you are welcome to do the testing. JPEG quality level is simply a measure of compression with the highest being the minimum allowable (there is always some).


Steve

Last edited by stevebrot; 04-20-2019 at 10:07 AM.
04-20-2019, 10:16 AM   #40
Otis Memorial Pentaxian
Loyal Site Supporter
stevebrot's Avatar

Join Date: Mar 2007
Location: Vancouver (USA)
Photos: Gallery | Albums
Posts: 32,479
QuoteOriginally posted by BruceBanner Quote
Does it mean that you get higher quality Jpgs from shooting RAW and then using PDCU for a **** Jpg extracting process than actually turning off RAW and shooting Jpg only in camera at the maximum *** option given?!
Applying **** in PDCU is not JPEG extraction, at least not in my copy. It is RAW development to JPEG and you get what you get. A rough measure might be to compare file size *** from in-camera with **** for PDCU.


Steve

Last edited by stevebrot; 04-20-2019 at 10:22 AM.
04-21-2019, 04:04 AM - 2 Likes   #41
Veteran Member




Join Date: Jul 2009
Location: Ottawa, Canada
Posts: 363
QuoteOriginally posted by rogerstg Quote
One question: are your experiences based on using DNG or PEF as camera RAW files?
I've been using DNG. I used PEF only briefly when I first got my K200D. I decided that compatibility with different RAW converters outweighed the small savings in storage. I can't remember now, but I may have been experimenting with a RAW converter that didn't accept PEF at the time. I can't remember if my K30 even supports PEF.

---------- Post added 04-21-2019 at 07:53 AM ----------

QuoteOriginally posted by BruceBanner Quote
What I thought interesting was fact that PDCU offers **** Premium setting, whereas in camera development cuts out at ***. Does that mean the camera's *** = **** for PDCU? Does it mean anything? Does it mean the camera itself is not actually recording Jpgs as good as what they could be? Does that mean we can get a better Jpg from a RAW and selecting ****?!?

Odd.
AFAIK, you have no control over the size/quality of the JPG that the camera embeds within the RAW file. It is there for preview purposes, and it isn't really intended to be "extracted" and used as an image. It doesn't matter how you set the JPG quality on the camera - the embedded JPG is of a fixed size/quality (and that appears to vary from camera to camera ). The designers probably tweak the size/quality of the embedded JPG to hit certain performance criteria (frame rate, etc.). The embedded JPG only needs to be good enough for viewing on the rear screen of the camera as far as Pentax is concerned. It could be that as pixel counts get higher, the designers have dropped the quality of the embedded JPG somewhat to keep the RAW files from inflating too much (which would slow down the write speed).

FastStone uses these preview JPGs because it is the fastest and least CPU intensive way of displaying the contents of a folder full of images. Most RAW converters don't do this, and that's why there is so much confusion around how FastStone works.

My current desktop isn't particularly high powered by today's standards, but I used to use an even slower machine. Back then, it would take forever for programs like Lightroom or Rawtherapee to open a folder of RAW files because the program had to generate the JPGs from the RAW files. I found it very frustrating because I might want to just go into the folder to get a quick JPG version of one photo, and I would have to wait for 10 minutes while the program would grind away generating previews of the entire folder.

FastStone would open the folder and display previews in a few seconds - allowing me to locate the image I was after and generate a quick JPG with minimal fuss. If I wanted a high quality image, I would then fire up a RAW converter to create one. But my preference was for RAW converters that would only process the selected file (eg. PDCU, DxO, SilkyPix, PSE). I lost patience with converters that would have to process the entire folder when I only wanted to look at a single image (eg. Lightroom).

As far as the camera topping out at *** while PDCU goes to ****, I must confess that I have never noticed this. If true, it was probably a compromise made by the camera designers in order to maintain the frame rate, save storage, etc.. The thinking was probably "If the user needs a super high quality image, they will likely generate it in post". Because PDCU is running on your desktop, it doesn't have to keep up with your frame rate, so it can offer the option of generating a premium JPG that will be larger and would take too long to write to an SD card. Similarly, PDCU allows you to save an image as a TIFF, but your camera doesn't. There's no technical reason it couldn't save a TIFF, but what's the point if you can save RAW? That's my guess anyways.

When shooting RAW+ with my K30, I noticed that when I open the camera JPGs in FastStone, the numeric value of the JPG quality varies a bit from image to image ( eg. one might be 99%, another might be 96%). I suspect that when I'm shooting rapidly, and the camera buffer is filing up, the camera drops the JPG quality a bit to speed up the write speed so it doesn't fall too far behind. Note that I never shoot in burst mode, so I have limited experience with pushing my cameras to their limits.

I would recommend that you do some simple experiments with shooting RAW+, with the JPG quality set to the highest level that the camera allows. Then open the resulting images in FastStone. You will see that for a particular image, the thumbnails of the RAW/JPG look identical ( which proves that for the RAW file, FastStone is showing you the embedded JPG). If you use the image comparator tool (which shows two images side by side on the monitor) to compare the RAW vs JPG, and you zoom to 100% (you can configure FastStone to do that when you right click), you should see that there is a difference in the quality of the two images. The RAW image (really, the embedded JPG) should have more JPG artifact, posterization, etc.

Then exit the folder, and change the settings so that FastStone doesn't use the embedded JPG (on my version, you press F12, go to the RAW tab, and it's the option at the top of the menu). Now do the side by side comparison of RAW vs JPG - you will see a slight difference in the images (depending on camera JPG settings). I just did this with some of my KP images. It's REALLY slow as FastStone tries to convert the RAW file ( does not affect the thumbnail previews - I guess it still uses the embedded JPGs for that in spite of the setting ). When I did the side by side comparison on the embedded JPG from the RAW file with the JPG saved by the camera, I can see a slight difference. I have to pixel peak on the selected image to see a difference. But that's the JPG generated by the camera. It might be easier to see the difference between the embedded JPG and a JPG (or TIFF) generated using PDCU (or other RAW converter).

The extracted JPG might be OK if all you need is a quick preview of your image, but only you can make that call. I don't know how you use your images.

For my shooting RAW+ works, and as I explained I cull after the fact - keeping the RAW files for only those images that I think might be useful in the future. If I was hard pressed for card storage and/or write speed, I could shoot RAW and work with extracted JPGs (as I did with my K200D), but it would involve more fiddling in post.

If you can't shoot RAW+, then I would suggest you shoot RAW, and use "extracted" JPGs for sample/preview purposes, and when you want a really good version of selected images, you generate them using a RAW processor. If you really like the camera rendering, then PDCU is your best choice for a RAW converter (and it's free). If you only have to generate a few high quality images from any given shoot, it may be that you can live with the shortcomings of PDCU. If you nail your exposures/white balance in camera, you may only need to make minimal adjustments within PDCU, and so you don't have to struggle with the UI very much, or you can store adjustment parameters from one image and apply them to others.

If you find that PDCU is unworkable for you, then you'll have to find a different RAW converter. Everyone has a favourite. Personally, I prefer DxO over all the others I've tried. I really liked the workflow in Silkypix, but it back when I was trialing it, the price for that program was on the high side. You will probably want to experiment with a number of them until you find the one that best suits your work flow.

Bottom line - if you shoot RAW and extract the embedded JPGs using FastStone, you should probably only use them as low res samples. For quality work, you will probably want to generate high quality JPGs or TIFFs from the RAW files using a RAW converter.

Last edited by arkav; 04-21-2019 at 07:32 AM.
04-22-2019, 12:23 AM   #42
Site Supporter
Site Supporter
BruceBanner's Avatar

Join Date: Dec 2015
Location: Blue Mountains, NSW, Australia
Posts: 3,459
Original Poster
QuoteOriginally posted by arkav Quote
I've been using DNG. I used PEF only briefly when I first got my K200D. I decided that compatibility with different RAW converters outweighed the small savings in storage. I can't remember now, but I may have been experimenting with a RAW converter that didn't accept PEF at the time. I can't remember if my K30 even supports PEF.

---------- Post added 04-21-2019 at 07:53 AM ----------



AFAIK, you have no control over the size/quality of the JPG that the camera embeds within the RAW file. It is there for preview purposes, and it isn't really intended to be "extracted" and used as an image. It doesn't matter how you set the JPG quality on the camera - the embedded JPG is of a fixed size/quality (and that appears to vary from camera to camera ). The designers probably tweak the size/quality of the embedded JPG to hit certain performance criteria (frame rate, etc.). The embedded JPG only needs to be good enough for viewing on the rear screen of the camera as far as Pentax is concerned. It could be that as pixel counts get higher, the designers have dropped the quality of the embedded JPG somewhat to keep the RAW files from inflating too much (which would slow down the write speed).

FastStone uses these preview JPGs because it is the fastest and least CPU intensive way of displaying the contents of a folder full of images. Most RAW converters don't do this, and that's why there is so much confusion around how FastStone works.

My current desktop isn't particularly high powered by today's standards, but I used to use an even slower machine. Back then, it would take forever for programs like Lightroom or Rawtherapee to open a folder of RAW files because the program had to generate the JPGs from the RAW files. I found it very frustrating because I might want to just go into the folder to get a quick JPG version of one photo, and I would have to wait for 10 minutes while the program would grind away generating previews of the entire folder.

FastStone would open the folder and display previews in a few seconds - allowing me to locate the image I was after and generate a quick JPG with minimal fuss. If I wanted a high quality image, I would then fire up a RAW converter to create one. But my preference was for RAW converters that would only process the selected file (eg. PDCU, DxO, SilkyPix, PSE). I lost patience with converters that would have to process the entire folder when I only wanted to look at a single image (eg. Lightroom).

As far as the camera topping out at *** while PDCU goes to ****, I must confess that I have never noticed this. If true, it was probably a compromise made by the camera designers in order to maintain the frame rate, save storage, etc.. The thinking was probably "If the user needs a super high quality image, they will likely generate it in post". Because PDCU is running on your desktop, it doesn't have to keep up with your frame rate, so it can offer the option of generating a premium JPG that will be larger and would take too long to write to an SD card. Similarly, PDCU allows you to save an image as a TIFF, but your camera doesn't. There's no technical reason it couldn't save a TIFF, but what's the point if you can save RAW? That's my guess anyways.

When shooting RAW+ with my K30, I noticed that when I open the camera JPGs in FastStone, the numeric value of the JPG quality varies a bit from image to image ( eg. one might be 99%, another might be 96%). I suspect that when I'm shooting rapidly, and the camera buffer is filing up, the camera drops the JPG quality a bit to speed up the write speed so it doesn't fall too far behind. Note that I never shoot in burst mode, so I have limited experience with pushing my cameras to their limits.

I would recommend that you do some simple experiments with shooting RAW+, with the JPG quality set to the highest level that the camera allows. Then open the resulting images in FastStone. You will see that for a particular image, the thumbnails of the RAW/JPG look identical ( which proves that for the RAW file, FastStone is showing you the embedded JPG). If you use the image comparator tool (which shows two images side by side on the monitor) to compare the RAW vs JPG, and you zoom to 100% (you can configure FastStone to do that when you right click), you should see that there is a difference in the quality of the two images. The RAW image (really, the embedded JPG) should have more JPG artifact, posterization, etc.

Then exit the folder, and change the settings so that FastStone doesn't use the embedded JPG (on my version, you press F12, go to the RAW tab, and it's the option at the top of the menu). Now do the side by side comparison of RAW vs JPG - you will see a slight difference in the images (depending on camera JPG settings). I just did this with some of my KP images. It's REALLY slow as FastStone tries to convert the RAW file ( does not affect the thumbnail previews - I guess it still uses the embedded JPGs for that in spite of the setting ). When I did the side by side comparison on the embedded JPG from the RAW file with the JPG saved by the camera, I can see a slight difference. I have to pixel peak on the selected image to see a difference. But that's the JPG generated by the camera. It might be easier to see the difference between the embedded JPG and a JPG (or TIFF) generated using PDCU (or other RAW converter).

The extracted JPG might be OK if all you need is a quick preview of your image, but only you can make that call. I don't know how you use your images.

For my shooting RAW+ works, and as I explained I cull after the fact - keeping the RAW files for only those images that I think might be useful in the future. If I was hard pressed for card storage and/or write speed, I could shoot RAW and work with extracted JPGs (as I did with my K200D), but it would involve more fiddling in post.

If you can't shoot RAW+, then I would suggest you shoot RAW, and use "extracted" JPGs for sample/preview purposes, and when you want a really good version of selected images, you generate them using a RAW processor. If you really like the camera rendering, then PDCU is your best choice for a RAW converter (and it's free). If you only have to generate a few high quality images from any given shoot, it may be that you can live with the shortcomings of PDCU. If you nail your exposures/white balance in camera, you may only need to make minimal adjustments within PDCU, and so you don't have to struggle with the UI very much, or you can store adjustment parameters from one image and apply them to others.

If you find that PDCU is unworkable for you, then you'll have to find a different RAW converter. Everyone has a favourite. Personally, I prefer DxO over all the others I've tried. I really liked the workflow in Silkypix, but it back when I was trialing it, the price for that program was on the high side. You will probably want to experiment with a number of them until you find the one that best suits your work flow.

Bottom line - if you shoot RAW and extract the embedded JPGs using FastStone, you should probably only use them as low res samples. For quality work, you will probably want to generate high quality JPGs or TIFFs from the RAW files using a RAW converter.

I agree with much of what you say, but I think this may all be camera specific where we're getting some cross wires. Let me explain;

I just pointed my camera at a dull scene (image not really worth posting) and made it do a RAW+ with my K-1, it gave these files;

RAW 40mb
Jpg 12mb

The Jpg settings were srgb *** Large.

I then took another shot at the same scene, this time toggling off Raw+ and just doing RAW, I then used the in camera development and extracted a jpg from that RAW file and it too was 12mb (with the in camera development on the K-1 it actually lets you choose the *** quality of Jpg to extract from the RAW as well as L or sRGB etc).
The extracted jpg was also 12mb.

So you can shoot RAW+ with the K-1 and get;

40mb RAW
12mb Jpg

or... shoot just RAW (40mb) and then (even a batch process) extract the jpgs which by all accounts seem to be exactly the same in terms of menu choices and file size to what a RAW+ would give. So there seems to be no advantage to shooting RAW+ with a K-1 unless you somehow need a Jpg quickly. Writing RAW+ to two sd cards at the same time (as the K-1 is dual slot) is not a fun experience at all, pixelshifted shots the funniest wait times ever

So whilst you might be correct, it seems at least with the K-1 (and I bet the KP as well as it's their latest camera) it seems RAW+ is pretty redundant really, a bit pointless like I say. I just can't imagine many times a shooter needs a Jpg right there and then, and if he can wait for a second for a Raw Development in camera he can have it... I'm not sure what shooting session requires a handover of Jpgs immediately.

I started this thread as I was curious if we could do what the camera does in it's raw development but externally and quicker via software (such as FS). The only gripe is the in camera extraction of Jpgs from RAW can be quite a long time, and when I started looking into things deeper I was very confused over what actually might be going on. I too suspected RAWs had naffish Jpgs... but now I'm really not so sure... Maybe the K-1's RAW files are larger than they need be because they are actually keeping full Jpgs 'just in case' :/

I no longer have PDCU installed, but if you look back through the forums I do write about my findings in greater detail, regarding the **** extraction and whether the file size is larger or smaller etc (I honestly cannot recall).
04-22-2019, 12:43 AM   #43
Site Supporter
Site Supporter




Join Date: Mar 2008
Location: Prince George, BC
Photos: Gallery | Albums
Posts: 2,778
Sports event shooters working for big events such as the Olympics need to shoot jpegs since the processing pipeline from camera to viewer is very short compared to raw files. But then you knew this, Eddie.
04-22-2019, 01:25 AM   #44
Site Supporter
Site Supporter
BruceBanner's Avatar

Join Date: Dec 2015
Location: Blue Mountains, NSW, Australia
Posts: 3,459
Original Poster
QuoteOriginally posted by jbinpg Quote
Sports event shooters working for big events such as the Olympics need to shoot jpegs since the processing pipeline from camera to viewer is very short compared to raw files. But then you knew this, Eddie.
Yeah I think a lot of Sport shooters just shoot Jpg and don't even bother with the RAW, my argument is not with Jpgs, it's with the point of RAW+ (at least with the recent Pentax releases), it just seems a little meaningless.
04-22-2019, 04:56 AM - 2 Likes   #45
Veteran Member




Join Date: Jul 2009
Location: Ottawa, Canada
Posts: 363
QuoteOriginally posted by BruceBanner Quote
I then took another shot at the same scene, this time toggling off Raw+ and just doing RAW, I then used the in camera development and extracted a jpg from that RAW file and it too was 12mb (with the in camera development on the K-1 it actually lets you choose the *** quality of Jpg to extract from the RAW as well as L or sRGB etc).
The extracted jpg was also 12mb.
I think there may be some confusion here - partly due to the vocabulary we've invented around this. Maybe we should step back and define some terminology.

When I say that FastStone "extracts" an embedded JPG from a RAW file, I mean that it takes a preview image that's inside the RAW file and presents it to you. This is not the "normal" way to do things - it's a shortcut that FastStone uses. Most programs "process" or "develop" the RAW image and generate a JPG. This takes more time and is computationally more expensive, but it gives you far more latitude for adjusting exposure, white balance, etc. etc. That's why people save RAW images.

When you used your camera development feature, you were not "extracting" the embedded JPG. The camera was "developing" the RAW image, so it's not surprising that you got the exact same sized JPG file that you got when you shot RAW+. You were essentially performing the same operation in both scenarios. When you shoot RAW+, the camera captures the RAW image and develops it into a JPG image - it embeds a somewhat lower quality version of the JPG image inside the RAW file, as well as a higher quality version along side it (quality of that image set according to the camera settings). When you shoot RAW, and then develop the image in camera, the camera processor takes the RAW image out of the RAW file, and develops it a second time (according to your current settings). If there was no change in the camera settings, you should get the same image JPG image that you would have got if you shot RAW+.

When you extract a JPG image from a RAW file with FastStone, I don't think it's easy to be sure what the size of the embedded file was before you extracted it - so I don't think comparing file sizes is valid in that scenario. I haven't updated FastStone in a while, but in the version I'm using, it gives you the option of saving the file with the original JPG quality, but because the JPG was extracted from a DNG, FastStone doesn't have an "original" JPG quality (or at least, it doesn't tell you what it was), so it uses whatever value you have set as your default. If your default is high, you'll end up with a large JPG. If it's low, the JPG will be smaller. All you can see is the size of the file that FastStone will create from the extracted JPG, not the size of the file that was embedded in the RAW file.

As I mentioned previously, I looked at some images where I shot RAW+, and I did a side by side comparison of the JPG extracted from the DNG file and the JPG generated/saved by the camera. I can see a slight difference in quality - there are more artifacts in the JPG extracted from the RAW file - it's not quite as sharp. I have to pixel peep to see the differences, but they are there. I could try to save 100% crops of the two versions of the same image and post them here, but the compression applied by the website might mask the subtle difference between the two. I haven't posted an image to the forum in years, so I'm not sure how this will look. The image was shot RAW+ using my KP. The first - "camera_jpg_crop" - is a crop from the JPG generated by the camera. The second - "extracted_jpg_crop" is a crop from the embedded JPG that FastStone extracted from the DNG file. I zoomed in on a detail from the image, then upsized it so that the JPG artifacts can be seen. I performed the same sequence of operations on both images, in the same order. For the camera JPG, I saved it at 98% quality, which is what FastStone says in the quality saved by the camera. For the DNG crop, I saved it at 100%. I can see a difference in the posted images, but I concede that this is pixel peeping. It's just to demonstrate that there is a difference in the quality of the JPG image that is embedded in the RAW file, and the one saved by the camera when you shoot RAW+. Is the difference important? That depends on what you are going to do with the image.

As I said, I would suggest you do this test for yourself. Shoot an image RAW+ . Make sure it's an image that has some strong light/dark contrast, a fair amount of detail, and perhaps a swath of sky in it. Then, in FastStone, click on both the DNG/PEF and the JPG, and use the image comparator to bring up both images side by side. Then zoom in to 100% (or higher) and scroll around the images. You should see more JPG artifact in the image extracted from the RAW file. The difference is quite noticeable in my K30 images, and a little less noticeable with my KP images. Only you can decide whether that difference is important or not.

Assuming you want to save RAW files, I maintain that there are 3 options for obtaining JPG images from them. In order of increasing image quality they are:

1) shooting RAW and extract the JPG that's embedded in the RAW file (using FastStone - there may be other programs that will allow you to do this but I don't know of any off the top of my head)
2) shooting RAW+ and use the JPG generated by the camera
3) processing/developing your RAW images to generate JPGs using a RAW converter (eg. PDCU, DxO, Lightroom, Rawtherapee, Darktable, etc. etc.)

The last option is where you have the most flexibility to adjust exposure, white balance, etc., and where you can increase the JPG quality to the maximum, or save as TIFF.

PS: for anyone curious - the crop is taken from a shot of someone standing in a winter scene. You're looking at one lens of the person's sunglasses, where you can see reflections of a shadow, and footprints in the snow.

---------- Post added 04-22-2019 at 08:33 AM ----------

Another test you can do: Set your camera to shoot RAW. Take 3 shots of the same scene with the JPG quality setting at *, **, and ***. I bet there won't be much of a difference in the size of the resulting RAW files, which tells you that the quality of the JPG that is embedded in the RAW file is not under your control.
Attached Images
View Picture EXIF
PENTAX KP  Photo 
View Picture EXIF
PENTAX KP  Photo 

Last edited by arkav; 04-22-2019 at 05:14 AM.
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, crop, dng, faststone, file, files, folder, fs, image, images, jpg, jpgs, mode, pdcu, photography, photoshop, quality, res, shot, size
Thread Tools Search this Thread
Search this Thread:

Advanced Search


Similar Threads
Thread Thread Starter Forum Replies Last Post
K-70 "This image cannot be displayed" when DNG converted to JPG in Photoshop CS5 Wolfeye Pentax K-70 3 12-08-2016 02:40 PM
Pixel Shift single image extraction from RAW using Fast Stone Image Viewer rgknief60 Pentax K-1 3 08-18-2016 02:37 PM
K-S2 JPG's versus K5ll JPG'S and K50 JPG's LoneWolf Pentax DSLR Discussion 22 03-28-2015 12:58 PM
Display both RAW/DNG and the JPG DavidSKAF3 Troubleshooting and Beginner Help 2 02-04-2015 02:34 AM
Problem comparing images in FastStone Image Viewer altopiet Digital Processing, Software, and Printing 2 10-18-2012 11:34 PM



All times are GMT -7. The time now is 06:42 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