Reading mode: Light Dark

JPG vs RAW is there any software that will compare images side by side

Go to last post
S

solo1

Leading Member
Messages
612
Reaction score
0
Location
Ballarat, AU
I have been undergoing a major shift in my photography at present where I'm becoming more and more critical abut the images I produce.

I talk to comitted amatures and somewhat disinterested pro's and two themes present. "Are you using a tripod" and "Are you shooting RAW"

In answer to the first question I say no, so I therefore did some tests that were conclusive to me.

A tripod (or a well braced monopod) does make a difference up to about f5.6 from there to f 8 its either or, but after f11 there appears no difference and my conclusion was "when the apertures get smaller the speed gets slower and the antishake on the K10D kicks in and negates the use of the tripod in favour of a well braced hand held.

I used ACDSee version 8 where you can select and compare a number of images side by side on the screen, hence making a very good comparison.
All these tests were for my own benefit and conducted in a well lit garden.

However when it comes to JPG vs RAW I cannot find a piece of software that will allow me to view a JPG and a RAW side by side.

My limited viewing of identical images taken using the RAW button gives me to believe the unprocessed RAW image is slightly sharper, but its very difficult when you have to swap different viewers.

I dont want to waste a lot of paper printing the images and I hope someone knows of a piece of software that could help in this area.
 
solo1 said:
I have been undergoing a major shift in my photography at present
where I'm becoming more and more critical abut the images I produce.
I talk to comitted amatures and somewhat disinterested pro's and two
themes present. "Are you using a tripod" and "Are you shooting RAW"
In answer to the first question I say no, so I therefore did some
tests that were conclusive to me.
A tripod (or a well braced monopod) does make a difference up to
about f5.6 from there to f 8 its either or, but after f11 there
appears no difference and my conclusion was "when the apertures get
smaller the speed gets slower and the antishake on the K10D kicks in
and negates the use of the tripod in favour of a well braced hand
held.
Click to expand...
The aperture makes no difference to the amount of shake in an image. Thats totally down to shutter speed. However as you shut down the lens it gets softer from diffraction so you wont see much difference.

SR works at all speeds. Whether SR is used or not, faster shutter speeds induce less shake.

A tripod is not strictly necessary for sharp results if you can keep the shutter speed up to reasonable levels. Whats more a tripod is not much use unless you also use mirror lockup. The vibration from the mirror will negate much of the advantage of a tripod.

To compare RAW and JPEG at the same time, get a copy of PS elements or lightroom with the Adobe RAW converter.

--
Steve
Pixel peepers miss the big picture.
http://www.pbase.com/steve_jacob
 
Hi

The anti shake works very well, but cannot be better than a sturdy tripod using proper technique. Technique being to switch off the anti shake and use a remote release, ideally with the 2 second timer. This method enables the mirror lock up.

You use a tripod to eliminate camera shake caused by using a slow shutter speed, the aperture itself is irrelevant, except of course as you have noted, a smaller aperture neccessitates a slower shutter speed.

The raw file will always require post processing before it becomes a satisfactory image. Ususally viewing an unprocessed raw file and a jpeg from the camera of the same image it is the raw file which will look flat.

One of the pp steps is almost always to apply some sharpening to the final image after it has been re-sized, so it is unlikely that your raw file is sharper than a jpeg from the camera unless you have turned down sharpening and/or contrast in the camera jpeg settings.

--
Peter Schluter

http://www.flickr.com/photos/peter_schluter/

 
solo1 said:
A tripod (or a well braced monopod) does make a difference up to
about f5.6 from there to f 8 its either or, but after f11 there
appears no difference and my conclusion was "when the apertures get
smaller the speed gets slower and the antishake on the K10D kicks in
and negates the use of the tripod in favour of a well braced hand
held.
Click to expand...
The use of a tripod is not directly depending on the aperture you use, but it is depending on shuttertime and focuslenght and how steady you can hold your hands. Normaly (without SR) the shuttertime should be at least 1.5 times faster as the the focuslenght in milimeters is (on APS-C). If you have really steady hands, you might be able to get sharp images at a shuttertime half of the focuslenght. With SR you can use even a slower shuttertime. Try find out your own limits.
solo1 said:
However when it comes to JPG vs RAW I cannot find a piece of software
that will allow me to view a JPG and a RAW side by side.
My limited viewing of identical images taken using the RAW button
gives me to believe the unprocessed RAW image is slightly sharper,
but its very difficult when you have to swap different viewers.
I dont want to waste a lot of paper printing the images and I hope
someone knows of a piece of software that could help in this area.
Click to expand...
With RAW, there is no PP inside the camera, so you won't lose anything and you are able to PP to your own liking. With JPG, the camera does some PP on it's own. If you like the JPG's enough, it is good to use them (and save memory space), but if you want to do more PP by yourself It's perhaps better to use RAW.

Just my two cents, others might think quite differently.
With kind regards,

Jos de Bont



http://www.xs4all.nl/~josdeb/
 
Oh, there's many solutions out there to view several images at the same time. It is even possible doing it in Fider by having "view as images" selected in preferencies instead of "view as symbols".

But, there is one thing to know about RAW and JPEG's.
RAW is an unprocessed file. It is meant to be post-processed.

JPEG is a post processed file. So, a JPEG should actually look better than an unprocessed RAW. An unprocessed RAW has no upsharpening applied, but a JPEG has.

If your RAW is looking good, then you can't be looking at an untouched RAW. It is either the embedded JPEG you are seeing, or the software you use for viewing it has applied some post processing (many does this by default).

An unprocessed RAW is unsharp, greyish and bland looking. That is why it needs to be processed ("developed") in a RAW converter, and RAW converters has the ability to use more tools and settings than the JPEG-engine in the camera has.

Oh, and RAW are 12 bit files and JPEG's are 8-bit. If you process the RAW in 16-bit mode and then save to 16 bit TIFF's, the result is images with more data and information than the 8-bit JPEG and those do gives a difference in quality.

If you process in 16 bit and saves as 8 bit, you are not getting that much in advance over JPEG, but with some clever post processing you will still be able to squeese out more from the RAW.
--
Take care
R
http://www.flickr.com/photos/raphaelmabo
 
Faststone is an image browser (and more) that will allow you to view PEF, JPG, and many other types of image files side-by-side. Free and highly recommended.
 
Thank you all for your helpful comments, I have tried Phototshop elements 5 as suggested and it does give me a great side by side view, I also downloaded Faststone as suggested and it looked the goods except I cannot select more than one image at a time even though it says hold down cntrl key and click on images required, a la windows convention, but it doesn't work on my computer. I have tried an unistall and reinstall but to no avail.

I will try Roland Mabo suggestion of processing RAW in 16 bit and saving as a 16 bit TIFF to see if it improves the situation.

Looking at GaryDems two views, like him I cannot see too much difference between the processed and converted RAW and JPEG on screen but I wonder what the print at 16X20 would look like.
Once again thanks for caring
 
solo1 said:
I will try Roland Mabo suggestion of processing RAW in 16 bit and
saving as a 16 bit TIFF to see if it improves the situation.
Click to expand...
I also suggest in selecting Adobe RGB, and printing out on a printer capable of handling Adobe RGB. But to see the difference on screen, you need a calibrated setup with software that handles colour profiles correctly.

The difference in 16 bit vs 8 bit comes more clear the more processing you do.

Image processing destroyes image data, you see this by looking at the histogram. The more "tagged" it looks, the more data is destroyed. What 16 bit gives you is more headroom in processing.

--
Take care
R
http://www.flickr.com/photos/raphaelmabo
 
Hi Solo,

For me unique FREE photo software which NEVER will destroy your "Originally" snapped photos is the 100% FREE naned Picasa photo organizer from Goggle...takes ALL the RAW/JPG...etc files & organizer it for you in a simpler easy to comprehend folders...I think that cam be use in various languages...I use it in English...100% known that exist also in Spanish...whatever...is to my eruditely THE VERY BEST ONE!

Good luck,

Peace,
--
Alex 007!
 
Here's an example of a JPEG processed by my K10D and the same image processed from the RAW. The EXIF data applies to both images. The only post processing of the JPEG image was to resize 50% for presentation here. The RAW was processed in PhotoShop CS3 with chromatic aberration removal, white balance and highlight recovery. Sharpening was done with a high pass filter applied to a control mask. The RAW image was output to a TIFF file to maintain 12 bit per channels. The two images were compared side by side in FASTSTONE and a PNG cropped capture was made for display here.



I picked this example to show how you can recover highlight information from the RAW that was lost in the camera processed JPEG. Aside from the obvious loss of detail in the t-shirt look at the skin tones in the arm.

Here's the full image processed from RAW resized to 25% - It's large 16 bit/channel PNG - (2.72mb) so I have posted the link.

http://clanstirling.net/images/fullsize.png

Here's the EXIF ...

[PhotoME]
PhotoME version: 0.78 (Build 565)

[Overview]
File name: 2007-04-14_11.27.04.pef
File type: Pentax Camera RAW
File size: 10,097 KB
Creation date: 4/14/2007 11:27
Last modification: 4/14/2007 10:27
Make: PENTAX Corporation ( http://www.pentax.com )
Camera: PENTAX K10D
Lens: smc PENTAX-DA 18-55mm F3.5-5.6 AL
Software: K10D Ver 1.10
Dimension: 3936 x 2624 px (10.3 MP, 3:2)
Focal length: 39 mm KB (26 mm Real)
Aperture: F5
Exposure time: 1/1000"
ISO speed rating: 100/21°
Program: Not defined
Metering Mode: Pattern
White Balance: User-Selected
Flash: Flash did not fire, compulsory flash mode

[Camera]
Exposure time: 1/1000"
F number: F5
Exposure program: Not defined
ISO speed rating: 100/21°
Date and time of original data generation: 2007-04-14 11:27:04
Date and time of digital data generation: 2007-04-14 11:27:04
Exposure bias: ±0 EV
Metering mode: Pattern
Flash: Flash did not fire, compulsory flash mode
Lens focal length: 26 mm
Manufacturer notes: 0x000002AC

[Manufacturer notes]
Pentax Version: 30.0.0
Pentax Mode: Auto
Preview Image Size: 640 x 480
Preview Image Length: 28290 bytes
Preview Image Start: 0x00007F08
Pentax Model: K10D
Date: 2007-04-14
Time: 11:27:04
Quality: RAW
Flash Mode: Off
Focus Mode: AF-S
AF Point Selected: Fixed Center
Exposure Time: 1/1000"
F Number: F5
ISO: 100
Exposure Compensation: ±0
Metering Mode: Multi Segment
Auto Bracketing: 0 EV
White Balance: Daylight
White Balance Mode: User-Selected
Focal Length: 26 mm
Saturation: Med High
Constrast: Med High
Sharpness: Very Hard
World Time Location: Hometown
Hometown City: Montpelier
----------------
cheers,
Rick
 
i shoot only jpeg though i have tried raw. i have enlarged my jpegs to 16x20 and 20x30 with no problems.

for enlargement it is not jpeg vs raw, but the quality of the image to be enlarged that counts. if the pixel count gets too small on a side it limits the enlargement size.

i am using a 6mp dslr.
this printed to 16x36.



this is 20x30 and is on our bedroom wall.



on wall.

 
Roland Mabo said that unprocessed RAW looke flat unsharp and generally not nice but when I looked at the RAW vs the JPG in faststone I cannot tell the difference, even though the Histograms are slightly different.

I processed one RAW file in Lightroom and saved as 16 bit tiff and then processed the equivalent JPG in photoshop CS2, looked at the 4 files on the same screen

original RAW
16 bit Tif
original JPG
enhanced JPG

and my preference was for the enhanced JPG.

I may have slightly over done the exposure compensation in Lightroom but not the sharpening bit but the enhanced JPG still looked better.

Anyhow in answer to my original question, re a viewer for both RAW and JPG Faststone looks the bees' knee's. As a general viewer I like it better than the ACDsee ver 8 I've been using just for JPG and TIFFs and will hang with it for a while.

As for RAW versus JPG my jury is out on that question, the reality is what works for you and since I'm a old dog and new tricks are hard to comprehend it will take some time for it to show its relativity to my situation but I will keep on tryimg and maybe, just maybe it will finally sink in.
I have gotten my workflow in CS2 down pat, its very quickand effective
1. Straighten and Crop if necessary
2 do any cloning out of unwanted objects if required.
3. levels
4. Increase saturation
4, Unsharp mask
5 save as and print (if required)

and thats it, takes between 2 to 5 minutes for an old guy like me which is more time than I like spending on post processing but often I do no more than straighten and crop.

Once again thanks for all your comments and interest
 
If you want info and comparisons between jpeg and Raw using the K10d, the info is here on dpreview.
Go to " Camera Database" and click on "Pentax"

When all the Pentax camers appear, look for the K10d and click on "Our Full Review"

There is 26 pages of information on the k10d. Near the top of the page you will see "introduction" in a white box. Click on the down arrow beside it. That will allow you to access all the pages. Read the whole thing its great information.

The main difference is the jpeg photos are not as crisp at the edges as they are in a RAW. The PRIME processor in the k10d smooths out the edges in JPEG. Dont worry about it if your prints will be less than A3( 11x14) size.
 
Rick, your efforts did not go unheeded, your example is certainly dramatic proof of the benefit of RAW, I dont think there is any doubt about the advatages of RAW if the shot is a bit off, there is certainly more scope for recovery but if the shot out of the camera is good then to my mind theres not that much benefit however as I mentioned before I will plug on with it and it could be that I will eventually see the benefits.
Thanks for putting in the time and effort on my behalf
 
The question is if you really looked at the unprocessed RAW or if the software you used display the embedded JPEG inside the RAW (yes, every PEF also contains a JPG) or if the viewing software you used had settings applied - as most software has.

If the JPEG and RAW looks the same, then you are not viewing the unprocessed RAW.

What the camera JPEG engine does, is that JPEG adds sharpening, pumps up the contrast and gives a higher saturation. Even if you have the JPEG setings in camera set to '0' this is what happens. RAW in itself does not have saturation, contrast or sharpening applied. It happens in the RAW converter or viewer. RAW is the image data without any processing.

I can clearly tell the difference between unprocessed PEF's from the camera and JPEG of the same image, on my Mac.

If I use the standard tone curve setting in LightZone, I get an image similar to the JPEG. If I skip the tone curve completely, meaning no processing of whatsoever is done, I get a greyish and soft image.

Here is my evidence to back up my claims:

Unprocessed RAW image, just saved as JPEG. No sharpening and no tone curve applied:



Same image, now processed with standard tone curve, no sharpening applied.



Both images resized to take upp less space, that is all.

If I use PhotoBrowser, I am not looking at the unprocessed RAW since PhotoBrowser does not show the RAW, it shows the JPEG image that is inside the RAW.

When manipulating images in 16 bit, it destroys the image less. I can clearly see this on the Histogram which is more "tagged" (not a smooth curve) in 8-bit after heavy post processing than in 16-bit. With heavy post-processing I mean resizing, using different sharpening algorithms, changing the colours, play with saturation, brightness, levels, highlights and shadows. The histogram looks smoother when work has been done in 16 bit compared to 8 bit. It even is noticeable when converting the 16 bit image back to 8 bit before saving.

When processing is done using more bits, it is less destructive when doing the same thing with fewer bits.

To get the best out of RAW, you need to process it.
I notice that you write "Lightroom", and perhaps this is where your problem is.

Adobes product has a Pentax-profile that is no good. It limits the dynamic range, it clips highlights and shadows and it shows the colours wrong. If you are using Adobes products, you need to re-calibrate them to get the most out of the Pentax RAW's. We have had many postings on this subject in this forum, with new profiles in XMP-format for Adobe.

Because of Adobes software not working good with Pentax "straight out of the box", I am not using them. I use LightZone (I had to tweak the profile a little there too, but it was much easier than the controls in Adobe) in combination with Pixelmator.

Few softwares has good default processing. If you are using Pentax PhotoLab, then Pentax are processing RAW the same way as the JPEG's if you are using AutoProcess - you need to use Custom processing to get the best. Same thing with Adobe, don't use the auto functions - use manual tweaking.
--
Take care
R
http://www.flickr.com/photos/raphaelmabo
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top