From: Eric Stevens <>
Subject: Re: A lightroom question
Full headers:
From: Eric Stevens <>
Subject: Re: A lightroom question
Message-ID: <>
References: <> <> <> <> <ot5giq$1mtk$> <> <ot5lh5$1tq5$> <> <ot5rs9$60q$> <>
User-Agent: ForteAgent/
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 8bit
Lines: 150
Organization: Forte -
X-Complaints-Info: Please be sure to forward a copy of ALL headers otherwise we will be unable to process your complaint properly.
Date: Mon, 30 Oct 2017 17:50:42 +1300
X-Received-Body-CRC: 3796023202
X-Received-Bytes: 8316
Print Article
Forward Article
On Sun, 29 Oct 2017 19:06:09 -0700, Savageduck
<savageduck1@{REMOVESPAM}> wrote:

>On Oct 29, 2017, Mayayana wrote
>(in article <ot5rs9$60q$>):
>> "Savageduck"<savageduck1@{REMOVESPAM}>  wrote
>> > The original JPEG has lost all it is ever going to lose, there is no
>> "final
>> > save" in LR.
>> You take a picture as JPG.
>Yup! That sets your base original.
>> That's loss #1.
>No loss, it is converted in-camera from the RAW data to an original JPEG 
>which is imported into LR.
>> You then edit it, say 5 times.
>You can edit a 100 virtual copies, you are not editing the originally 
>imported JPEG.
>> Each time it’s saved in LR and there's no loss to the original.
>It isn’t saved after each edit, each edit is written to XMP sidecar files.
>> That's good. LR is saving the bitmap image along with a history of changes you've made.

This is the area where Mayayana's understanding goes off the rails.

When you load in amge into LR the first thing LR does is create a
screen display which (according to the preference you have selected)
is a more compact and less precise version of the image in the file.
It might be less precise but it's usually more than good enough to
enable you see what the effects of your edits may be.

Once you start editing each setting is saved in the XMP file. It's not
the effect of each edit that is saved: it's the instructions and
settings for each edit that is saved. The original file is not
affected by any of this: only the simplified version you see in the
screen. At the end of the edit session the XMP file has accumulated a
long list of edit data.

When you leave LR the original image file remains untouched. If you
come back again later the list of edits is reloaded and works on the
simplified version of the image file to show you on the screen where
you got to last time.

The XMP file only gets to work on the original image file only when
you go to export it in some way. Either by Saving the file, or Save As
or printing. The the XMP gets to work on the original image file (in
full precision) and creates whatever is required for export.

You can if you are so inclined edit a JPG and eventually save it on
top of itself but it is better to (for instance) as I do save JPGs in
a subfolder so that there is no confusion.

This way you can go on editing and saving JPGs until you are blue in
the face without accumulating corruptions of the original JPG.

>Nope. That is not how LR works with virtual copies. You are speculating on 
>how LR works without any actual knowledge, or experience. Don’t project 
>your ignorance into a thread where you are just guessing. I have been using 
>LR since the first beta.
>> Each time you work on it, you're really working on the bitmap and LR is saving that, along with
the edit history data.
>> What's different is that LR is hiding that complication and you don't need to keep track of
various saved files.
>You are guessing that is what is happening, but you are wrong.
>> LR is doing that for you. But once you decide to export it as an edited image in JPG that’s
loss #2. You can't edit the image and then save a new image as 
>JPG without a second loss. You can do 5 lossless edits inside LR, but the 
>final save will be lossy.
>Here you are close. The edited image is exported, and the export criterea for 
>resizing, file type, compression if the file type is lossy. If it is a JPEG 
>the loss will occur with the file at the export location. That file never 
>makes it back to LR, and other than posting it using whatever method to 
>recipients it will not be present on LR for any further editing, the degree 
>of loss is deliberate and planned.
>> I don't mean to complicate things. It's just that most people are not familiar with the
>> in file formats, so I think it's worth reiterating that JPG is lossy.
>Why do you think that I might not be familiar with JPEGs? I certainly am well 
>aware that JPG is a lossy format.
>> Otherwise it's very easy to drop out data unnecessarily. The LR feature is nice, but
>> it's still a process that drops out data twice if you edit the photo.
>Again, your knowledge and understanding of the function of LR is quite wrong.
>> So you take a JPG, put it into LR, edit as you like, and eventually save a version as JPG.
>No, I edit a virtual copy of the JPEG, or RAW file as I like. I don’t save 
>a version as a JPEG within LR. However, if I choose to export an edited 
>version of that JPEG, or RAW file I can export it to the export location of 
>my choice, as the file type of my choice () all without reintroducing it into 
>> I take a photo as JPG, save my first edit as BMP or TIF, then save all other versions the same
way. I end
>> up with a folder containing numerous versions of the photo. You end up with a history in LR. If I
edit it 5 times and maybe save 5 versions there’s no 
>loss. If necessary I might eventually convert one of those to JPG to send to 
>Well if that works for you, go ahead. You are probably never going to use any 
>Adobe applications, so I don’t see how you have done anything to solve 
>Peter’s original LR issue, or if you even understood it.
>> We both then end up with 2 lossy steps: The original JPG photo and the final JPG save of an
edited image.
>> The only difference is that LR is managing the file storage for you so you don't need to save
TIFs or
>> maintain systematic file storage.
>You really don’t understand anything about LR.
>> > > That's basically what LR is doing -- saving some kind of backup bitmap
>> image.
>> >
>> > Nope, an XMP sidecar file is not some kind of backup bitmap image. It is a
>> > set of instructions detailing the edits and adjustments.
>> >
>> > <>
>> > <>
>> We discussed this once before. It's all bitmaps. Any raster image is essentially a bitmap. A JPG
>> is a compressed bitmap with some data dropped out. A TIF is usually just a compressed bitmap.
>> A GIF is a bitmap. A PNG is a bitmap. Those are all just different ways to store the image data.
>> Proprietary formats, like Paint Shop Pro's PSP or the PS PSD, store the image plus editing
>> history, unmerged layers, etc. But the image is still going to be a bitmap -- a grid of pixel
>> values. That's what goes to the printer or the computer screen. That's what you're applying
>> sharpening, etc to. Those are all just mathematical formulae applied to bitmaps. Increase the
>> values and you've lightened. Increase the difference between contiguous pixels and you've
>> Of course it gets very sophisticated when it can do things like remove a chain link fence from
the image,
>> but it's still essentially the same thing.
>You are obviously trapped in bitmap theory.


Eric Stevens