Does subjecting RAW files to a pre-set batch mode still result in better JPEGs?

Status
Not open for further replies.

New Daddy

Member
Joined
Oct 29, 2011
Messages
94
Lightroom Experience
Beginner
Let's say I want to spend the minimum amount of time in front of the computer. So I set up a batch conversion routine that pretty much does what the camera's JPEG engine would have done - e.g., use the white balance in the metadata, use automatic exposure, and adjust contrast/sharpness/saturation based on my pre-set preferences. Is the result still going to be better than the out-of-camera JPEG?

In other words, is RAW-to-JPEG going to produce better result even when the development is mechanical without any individual post processing?
 
Impossible to say anything in general. It depends on a lot of factors. Some photos may end up better than the out-of-camera JPG's. Some may be worse. And what is "better" really?

RAW gives you the ability to create better images than the in-camera JPG-processing, but it also require that you do some work.

That said: creating your own preset (and perhaps applying that as a default) may be good enough for the majority of your photos? That's the way I do it.
 
There are some camera profiles that mimic the look certain cameras produce. For example many felt that the out of the camera 'look' produced by the Nikon D2X was both very distinctive and very desired. There are a set of three D2X profiles for Landscape, Normal and Portrait that do a very good job of mimicking that iconic 'look'.. I have a variation of these in my LR Camera Calibration Panel. I find they can quickly provide a very good starting point for further edits. So much so that I sometimes make no further edits at all.

It might be worth checking out some calibration presets to find your own preferred 'look'.

A few links
http://peterliuphoto.com/lightroom-camera-calibration-and-user-presets/
http://www.presetsheaven.com/2011/04/19/lightroom-camera-profile-explanation/
 
There are some effects that Lightroom can probably do better than the embedded software in the camera, such as sharpening, noise reduction, and lens corrections, largely because Lightroom has a vastly more powerful computer under it. There are also some things that the camera can do better than Lightroom, such as dynamic range enhancement through Active D-Lighting (Nikon) or Highlight Tone Priority (Canon (do I have the name right?)) or equivalents from other manufacturers.

Like everything else it's a tradeoff, and "better" is in the eye of the beholder.
 
There are some effects that Lightroom can probably do better than the embedded software in the camera...
Like everything else it's a tradeoff, and "better" is in the eye of the beholder.
Better or not, LR will always be working with 12-14 bit RAW data whereas the embedded software in the camera is going to produce an 8 bit JPEG (or possibly TIFF)
 
... Active D-Lighting (Nikon) or Highlight Tone Priority (Canon (do I have the name right?)) or equivalents from other manufacturers.

There is ALO (Auto Lighting Optimizer) and Highlight tone priority in recent Canons.
 
Better or not, LR will always be working with 12-14 bit RAW data whereas the embedded software in the camera is going to produce an 8 bit JPEG (or possibly TIFF)

That's something I've had and still have hard time understanding. The end-product from LR will still be an 8 bit JPEG, right? So, what's the difference?

Also, the camera isn't writing a 12-14 bit RAW file, but isn't the camera still using the 12-14 bit data to produce a 8-bit JPEG file?

The engine may be more robust for LR (a computer CPU vs. a camera CPU), but I don't see any difference in the raw data (12-14 bit) that each engine has to work with nor the format of the final product (8-bit JPEG).
 
First, the end product fro LR is not limited to 8-bit JPEGS Second, if your output SOOC is a compressed 8 bit JPEG, you can not further manipulate that JPEG without decompressing it and saving a new compressed JPEG. Data is lost in each lossy compression generation.

Assume that you recover 80% of the original data when you open a SOOC JPEG for further processing. That SOOC JPEG is already a 2nd generation lossy image. Import that into LR or any pixel altering editor if just for a crop or to tweak the NR, the output is now a 3rd generation removed from the original data captured. Open this derivative file up to print and if you are again able to recover 80% of the data quality that went into the 2nd generation JPEG and you are looking at 64% of the data quality of the data captured by the camera.

Let's say you don't want to do your own post processing and your camera can output 16 bit processed TIFF instead of 8 bit JPEG. While all of your in camera adjustments are 'baked in', your output is still a 16 bit lossless file that can be further manipulated , cropped or tweaked without further loss to your data quality. The only thing you lose in this case is that your WB and contrast get 'baked in' along with all the other adjustments and you have limited ability to recover mistakes in lighting and exposure.
 
Cletus, I think you may have missed the point of the original question.

The comparison is between unprocessed SOOC JPEGs and raw files batch-processed into JPEGs with no individual processing.
 
Let's try not to turn this into another meaningless RAW vs. JPEG debate.

If you read my OP, I was asking whether a streamlined, batch-processed JPEG from LR would still have better quality than OOC JPEG, because I'm not willing to spend any significant amount of time in front of a computer even if I decide to adopt LR in my workflow.

To my OP, you responded "LR will always be working with 12-14 bit RAW data whereas the embedded software in the camera is going to produce an 8 bit JPEG". I was trying to break down your rationale for giving LR an upper hand. To reiterate, (a) doesn't the camera also work with 12-14 bit data when it produces a JPEG through its engine, and (b) doesn't LR also produce an 8 bit JPEG?

I'd like to compare (a) batch-processed LR JPEG vs. (b) OOC JPEG,
NOT, (a) batch-processed LR JPEG vs. (b) OOC-JPEG-further-processed-in-LR.

Post-processing of OOC JPEG, which would produce a "3rd generation" product like you said, is irrelevant in the context of my question.

First, the end product fro LR is not limited to 8-bit JPEGS Second, if your output SOOC is a compressed 8 bit JPEG, you can not further manipulate that JPEG without decompressing it and saving a new compressed JPEG. Data is lost in each lossy compression generation.

Assume that you recover 80% of the original data when you open a SOOC JPEG for further processing. That SOOC JPEG is already a 2nd generation lossy image. Import that into LR or any pixel altering editor if just for a crop or to tweak the NR, the output is now a 3rd generation removed from the original data captured. Open this derivative file up to print and if you are again able to recover 80% of the data quality that went into the 2nd generation JPEG and you are looking at 64% of the data quality of the data captured by the camera.

Let's say you don't want to do your own post processing and your camera can output 16 bit processed TIFF instead of 8 bit JPEG. While all of your in camera adjustments are 'baked in', your output is still a 16 bit lossless file that can be further manipulated , cropped or tweaked without further loss to your data quality. The only thing you lose in this case is that your WB and contrast get 'baked in' along with all the other adjustments and you have limited ability to recover mistakes in lighting and exposure.
 
This is very interesting. Art imitating life, or life imitating art.

So if people with actual Nikon bodies were to use these profiles, then there is going to be still less difference between batch-processed LR JPEG and SOOC JPEG. But for non-Nikon shooters, these profiles will come in very handy.

(a) Is there a gallery of these profiles for preview?
(b) Are these profiles free to download, or do you purchase them at extra like plug-ins?

Believe it or not, I like the Canon look for people photography. It seems to retain the right amount of saturation and warmth in white balance for pleasant look. I can't seem to re-produce that Canon-look with my micro four thirds even with PP. If a Canon-look profile is readily available for batch processing, maybe that's a good reason for me to adopt LR in my workflow.


There are some camera profiles that mimic the look certain cameras produce. For example many felt that the out of the camera 'look' produced by the Nikon D2X was both very distinctive and very desired. There are a set of three D2X profiles for Landscape, Normal and Portrait that do a very good job of mimicking that iconic 'look'.. I have a variation of these in my LR Camera Calibration Panel. I find they can quickly provide a very good starting point for further edits. So much so that I sometimes make no further edits at all.

It might be worth checking out some calibration presets to find your own preferred 'look'.

A few links
http://peterliuphoto.com/lightroom-camera-calibration-and-user-presets/
http://www.presetsheaven.com/2011/04/19/lightroom-camera-profile-explanation/
 
The only profiles available from Adobe are ones that mimic that brand's looks. That is, there are Nikon profiles for Nikon cameras and Canon profiles for Canon cameras. There are no cross-brand looks available from Adobe, but there might be from third parties, or you can try creating one yourself using the free DNG Profile Editor.
 
Is the result still going to be better than the out-of-camera JPEG?

It all depends on what you mean by "better". What is better for one person may be worse for another.
In LR you are not currently able to exactly replicate the camera produced Jpeg, even for one photo, let alone a whole batch of photos using a preset.
 
Let's try not to turn this into another meaningless RAW vs. JPEG debate.

If you read my OP, I was asking whether a streamlined, batch-processed JPEG from LR would still have better quality than OOC JPEG, because I'm not willing to spend any significant amount of time in front of a computer even if I decide to adopt LR in my workflow...
I'm not trying to debate one policy over another. But I do think it is important that you recognize the shortcomings of using a lossy file format over a Lossless one. Many of my photos SOOC are "Good Enough" without individual attention except for a judicious crop. I can't crop a JPEG that I have imported without compromising the data quality further. One option is to set the camera to shoot RAW+JPEG. this gives you the best of both. JPEGs for those that are "good enough" and RAW images that can be massaged to produce that once in a lifetime prize winning photo. For quite a while until I gained confidence in my LR post processing abilities, I shot RAW+JPEG with my camera. Many people shoot with their camera configured this way. You can too is that helps solve your issue. If your camera will let you shoot TIFF you get a lossless fully processed image SOOC.

My Pentax does not have LR camera profiles like those that mimic Canon or Nikon. But I do have a Develop Preset that I apply on import to every photo taken by that camera. Sometimes the results after import are "good enough". But I always have the latitude to improve what Pentax can do with LR.
 
Status
Not open for further replies.
Back
Top