Lightroom STILL doesn't make images unique

[dkperez]

Lightroom Version Number

9.4

Operating System

1. Windows 10

I've asked this over in the Lightroom Forum in the past, and I"ve never gotten a useful answer....

I take a set of images and run them through Helicon Focus. I render them with a specific setting and get back a .dng file with a unique name. I render a second time with a DIFFERENT set of parameters and get bad a DIFFERENT .dng file with a unique name. They're not the same, they don't look the same.

In Lightroom, in Library or in Develop, they show up THE SAME. At 1:1 they show up the same. Does anyone here have a way to make Lightroom display them as the actually are? It doesn't happen when I save as jpegs, and it doesn't happen when I save as tiffs, but with dng files I keep having to open them in Adobe Bridge to see the differences...

 

[Johan Elzenga]

Actually, I don't think you do not need to be that drastic. Select the images in Lightroom Classic.
From the menu {Library}{Previews} {Discard 1:1 Previews} and then {Library}{Previews} {Build 1:1 Previews}

That may work, and then again it may not. The preview cache could be corrupted.

 

[Johan Elzenga]

I thought for sure the nuclear option of deleting 36GB of previews would make these things unique... It certainly wouldn't be a fix, and definitely not a reasonable thing to do, but in this case it was worth a try.
Nope. Restarted Lightroom, opened the three images in loupe view, still the same. Created 1:1 previews to see if that would matter. Nope, still the same.

I also opened them in Adobe Bridge. On the first view they were different. After that, viewing them a second time, all the same.

Opened Faststone viewer to make sure they WERE different. Yup, different in there.
I even opened Luminar and opened the three images... Different.

I even ran Helicon Focus, rendered the images saved, exited back to Lightroom. Then took 8 of the 19 images exported back to Helicon and rendered those. Completely different executions.
Back to Lightroom, I now have 3 different images from 2 days ago, and 2 images created in separate executions of Helicon Focus this morning with unique names. 5 .dng files, days apart, at least 4 that are perceptibly different, all with unique names. ALL with "previews" created this morning after deleting the previews.
In Lightroom, they're all identical.

I have no idea what's different between the two computers I've tried this on here and what y'all are doing, but Lightroom here doesn't show the difference between the images. I figure it's going to be something stupid as soon as I know what it is. Preference? Something else?

Could be the catalog itself somehow. Create a brand new catalog and import the three images. See if they are different in this catalog.

 

[Johan Elzenga]

I also opened them in Adobe Bridge. On the first view they were different. After that, viewing them a second time, all the same.

So now the same problem can be seen in Bridge? In that case it could be the Camera Raw cache. You can purge that from the Lightroom preferences.

 

[dkperez]

Is there any way a preference can be causing it…
On General I have

• Ignore camera-generated folder names NOT checked
• Treat JPEG files as separate checked
• Replace embedded previews with standard checked

I presume there’s nothing in presets or external editing that can cause this?

File handling

• File extension dng
• JPEG preview is medium
• Embed Fast Load Data checked

Nothing I can see in Interface

Performance

• Have had graphics process off, auto and custom, doesn’t make any difference.
• Big cache file on it’s own partition.
• Generate previews in parallel checked.

In the catalog settings

File Handling

• Standard Preview Size 1680 px
• Preview quality medium

Metadata

• Include develop settings in metada… checked

Nothing else

---------------------------

I was wrong about Bridge. I just went over and tried again, and realized I was looking at different versions of the same rendering. when I look at all 5 .dngs that are sitting there, the ones that SHOULD be different are. I purged cache in Lightroom and I had Bridge purge local cache. BTW: The two caches are on completely different drives...
Anyway, Bridge IS still showing them different.

I think I'm starting to get punchy looking at this stupid thing...

 

[clee01l]

So now the same problem can be seen in Bridge? In that case it could be the Camera Raw cache. You can purge that from the Lightroom preferences.

DNGs creates as focus stacked composites will not be RAW images and will not have been processed through ACR Helicon Focus has to be sending back RGB images.
The fact that several of us are seeing differences in the images and the OP on one computer is not suggests that Windows 10 may be caching images somehow (I know this can happen for images in a web browser)

Look for a cache folder similar to

C:\Users\username\AppData\Local\Microsoft\Windows\INetCache

 

[dkperez]

Seems unlikely MY Windows 10 would be the only one doing that. Or are all of you looking at the on Macs? In any case, it happens on at least 2 systems.

I'm about to throw this one, which unfortunately is a major PITA, on the same garbage pile with all the other stuff over the years.

 

[Hal P Anderson]

I'm on Windows 10, LRC 9.4, and they all look different.

When you select each image in Library, do all of their file sizes differ, or are they the same?

 

[dkperez]

All the files are different sizes. Different names, different sizes, even different dates.

 

[dkperez]

I have no idea what this means or if it'll help or not...
Ran the SAME 19 image stack and rendered it with Method B. Stayed in Helicon and rendered it with Method C. THEN I unchecked the FIRST 4 images and the last 6 images so it would only render images from the middle of the stack. Rendered those and saved.

The first 2 are still the same BUT, the rendered images from the middle of the stack are different. This does not happen when I render a subset from the front of the stack...

 

[Linwood Ferguson]

You probably don't need another "me too" but I downloaded, added to LR, and all three look different for me also.

One other thing to try if you are still in an experimenting mode -- when you import into LR set the preview size to "embedded". This will cause LR to use the preview contained inside and not build it again. When you import make sure you are not setting any develop preset that will force a preview build; you should get the little embedded icon on the top right.

I'm curious if it will get the right one before it builds its own preview (to be more precise, what LR does is read the preview from the file and substitute it for its own preview, in the preview cache; but this will tell if the issue is in the preview creation (applying LR develop) or in the preview access itself.

 

[GregJ]

Thanks for the reply Greg... Take one of your stacks - I don't know what a GFX 100 is, but presuming you're doing focus stacks by changing focus.
Take one of your stacks - maybe a small one with 20 or 30 images. Presumably they're shot in RAW, imported into Lightroom, then sent in RAW to Helicon Focus.
Once there, render the stack in Method B, using the default radius (8) and smoothing (4). Save the output back to Lightroom as a .dng file. STAY in Helicon and change the method to C, set smoothing to 2 and render again. As you're likely aware, it's a LOT faster to do this than to export from Lightroom every time. Again, save the Method C render as a .dng.
STILL STAYING in Helicon Focus, select all the images but the first 8, and UNCHECK them. Switch back to Method B, render again. Save as a .dng.
Exit from Helicon Focus back to Lightroom and you'll now import 3 images, 1 with ALL the images rendered in Method B, one with ALL the images rendered in Method C, which will look VERY different than Method B, and one with only part of the stack (8 images worth) that is sharp, rendered in Method B, which will look extremely different than either of the others.
In Library, open the 3 images in Loupe view. Do they look as different as they actually are, or do all three look the same?
On all THREE of the Windows 10 systems I have done this on, across every version of Lightroom I've ever tried it on, they look exactly the same. With .dng files. In Lightroom. Tiffs look "normal (different)", JPEGS look "normal (different)", .dngs all look the same.

I will do that next time I shoot a stack. Believe it not, I have not kept the stacks but only the resulting DNG that I'm happy with. I always shoot raw, then export to Helicon Focus as DNG files. But I did learn something from your post. I didn't know you could render the stack various ways without starting over. I always like to compare Method B and C. I think you need to repeat this issue on the Helicon Board. Something is not right because when I render one at a time with different methods the files are clearly different.
Also, someone told me that you had to make sure that you have the latest version of DNG Converter downloaded from Adobe. Apparently it does not update automatically with the cloud service along with LR, PS, Bridge, etc... I wonder if that is really true? I need to ask Cletus or Jim.... It seems Adobe would do that.

 

[dkperez]

I don't import the output from Helicon Focus into Lightroom, it comes in automatically. I don't know what settings it uses, though I can set Lightroom to Embedded and see if anything changes.

This has been brought up in other forums (I don't recall if I"m asked at Helicon 'cause it's rare to get a useful answer - EVERYTHING is always something else's fault), but the only "answer" I ever got was something to the effect that images coming back from Helicon get put in Lightroom SO FAST that LR doesn't have time to build previews and that's why...

As far as the dng converter... Helicon doesn't care as long as it works. With 7.6.4 Pro on, I got errors with raw images as input, complaining that the .dng couldn't be created. My .dng converter was a version old, updated, and it was fine. So, it's BEST if you keep it current...

 

[clee01l]

Seems unlikely MY Windows 10 would be the only one doing that. Or are all of you looking at the on Macs? In any case, it happens on at least 2 systems.

. But it does seem unique to only your two machines. At least Linwood is on Windows. So, I think that rules out a Windows only issue.

 

[dkperez]

I changed the Lightroom import to embedded and imported a couple images to it hopefully became the default. Ran a stack, exported .dng to Lightroom. Exactly the same.

I even went into the Raw development settings - there aren't any. BUT, I had installed the Nikon WIC codecs a while back, so I uninstalled those and rebooted. Ran a stack, sent .dng back to Lightroom. No difference.

It appears I have the only two Windows 10 systems in the universe that are doing this. One is Windows 10 I installed on the desktop, the other is still the completely standard Windows 10 that came on it from Dell on the laptop. At the moment I have 5 .dng files of the stack, each with a unique name, generated in three separate sessions of Helicon focus, 1 of 19 images, 2 of the first 8, and 1 of the first 9. And in Lightroom Library, they ALL LOOK IDENTICAL.

 

[Hal P Anderson]

Which picture is the one that they all look like?

I could have missed it in the thread: did you try creating a brand new catalogue and importing just those weird DNGs?

 

[dkperez]

It APPEARS they all look like whichever image was the active one when I open it in Loupe view. I've done it with one of the 19 image stacks active and they look like that. I've done the same thing with one of the 8 image subsets active and they all look like THAT one.

In Develop, as I move through the (however many) outputs from Helicon, the images look like what they are for a split second, they change to the originally active one. So if I have a 19 image output and an 8 image subset output, and start with the 19 image output active, when I move to the 8 image one, it shows the 8 image one for a VERY short time, less than a second, then switches to the one that was originally active...

I'm going to see if there's a way to have something record my Lightroom loupe window whhile I display the images. The image name changes but the images don't.

And yes, I've created a test catalog and only used those image in it, and they look the same.

 

[John Little]

Have you tried resetting the preferences? The experts here often suggest that when LR starts acting strangely.

https://www.lightroomqueen.com/comm...the-lightroom-preferences-file-updated.25197/

 

[dkperez]

I tried that early on. One thing that has me curious is that if I create a stack from image 1 - 19 I get an output. If I create a stack with image 1 - 8 I get a completely different output but they look the same in Lightroom. But if I create a stack with image 2 - 19 Lightroom sees it as different. So, what piece of information, even though the filenames are different, is Lightroom looking at that tells it to show the same preview?

Buried way down in the raw-in-dng-out part of Helicon I DID come across a piece of information on an image. It said "iso 22028-2 ROMM RGB profile". Is this how the outgoing .dng file is encoded by Helicon Focus? Either way, while interesting, I presume EVERYBODY'S Windows 10 system, running Helicon Focus, is going to use this same encoding. So WHY do the images look the same on these two computers and not others? WHAT's different?

 

[John Little]

Have you tried Zerene Stacker? That might help narrow down the range of possibilities (if Zerene gives the expected behavior instead of what you're seeing, it suggests the problem is in Helicon, or in the way Helicon sends its output to LR).

 

[dkperez]

Unfortunately, I tried Zerene some months ago and it hasn't changed since, so my trial expired. I MAY be able to try it on the laptop tomorrow.

 

[clee01l]

Have you tried Zerene Stacker? That might help narrow down the range of possibilities (if Zerene gives the expected behavior instead of what you're seeing, it suggests the problem is in Helicon, or in the way Helicon sends its output to LR).

Everyone looking at copies of output images on their computer see. Differences in the image. So, what ever is happening is happening with the OPs Windows machine and Lightroom. Helicon is producing different images and everyone else can see the differences.


Sent from my iPad using Tapatalk

 

[dkperez]

Now I just need SOMEONE with a lot more knowledge than I have to tell me WHAT's different. Helicon support has been silent so far.

 

[Linwood Ferguson]

One other possiubil

Everyone looking at copies of output images on their computer see. Differences in the image. So, what ever is happening is happening with the OPs Windows machine and Lightroom. Helicon is producing different images and everyone else can see the differences.

Except all of us who tried it with the provided files are doing an import, whereas Helicon is doing something to return a file(s) back to LR as a plugin. The mechanics are apparently different.

@dkperez, can you save the files from Helicon somewhere else, and NOT return them at all to LR. Cancel out or something similar. Then import the files instead with the regular import dialog, as we are doing.

I'd suggest trying existing files but who knows if they are remembered in some way, that's why I am suggesting doing it with a new trio.

Which actually begs the question --is Helicon actually brining all three back without you doing an import? As one operation? Or are you doing three separate plugin invocations? You probably said but I'm getting confused now also what's been tried.

Have you considered stronger liquor than you are using now?

 

[dkperez]

You and me both, Ferguson!

I ran a small stack through 2 rendering methods and saved them to a completely separate hard drive. When I exited Helicon Focus they still got imported into Lightroom and still displayed the same.
SO, I did the same thing again, saved them to a DIFFERENT hard drive, then MOVED them to yet ANOTHER drive before I exited from Helicon. Lightroom didn't find them so I imported them. Exactly the same.

I THOUGHT MAYBE Lightroom was using the embedded jpegs to view in the Loupe (I know that wouldn't answer the question in Develop or when viewed 1:1, but at this point we're WAY past any "rational" answer and deeply into the bizarre). And that if you run multiple renderings that Helicon Focus was using the SAME jpeg (the first image of the stack) for all of them - which might follow from the fact that when I used a middle subset of the set I DID see a difference (the first image of the subset stack was different than the first image of the full stack). SO, I sent the stack to Helicon and ran method B. EXITED from Helicon and sent the stack again from Lightroom and rendered with Method C. In loupe , EXACTLY the same.

One MORE thought... BOTH my systems, the desktop AND the laptop are Windows 10 Pro. I just had a friend look at the 3 images from the website and run a stack of his own. Both my three and HIS output from one of his stacks are identical when viewed in loupe. To confirm his actually ARE different, he sent them to Photoshop where they ARE different. SO, ANY chance it's Windows 10 Pro - I don't know what O/S you folks that see them as NOT the same are using...

BTW: When I exited Helicon Focus back to Lightroom after MOVING the images I got the following error in Lightroom:
An internal error has occurred:
AgImportSession.addOnePhotoToLibrary: failed to import photo

which I would expect something like, since Lightroom is being told to import images that aren't there.

So if it's NOT a Lightroom preference (I listed mine above), or a Catalog preference (also above), unless it some Windows 10 Pro thing and y'all are using something else, I'm out of ideas.

I've started TWO topics at Helicon Support, one for the the non-distinct images. The OTHER is because Helicon Focus has a "Save All" to save all the images at one time instead of sending each one back. I THOUGHT that might give different results. Unfortunately, when I tried that one, it failed with some gibberish error, so I CAN'T find out if that would give different results. And, of course, it's only been a day, so there's been no response from Helicon.

 

[Linwood Ferguson]

FWIW I am on Windows 10 Pro also. Not to mention I do not see how that would affect what LR sees in loupe.

I'm sorry. I am at a loss.

So let me think outloud just a moment... if LR shows the same wrong rendering in both Develop and Loupe (which use different preview mechanisms), and if we give it credit it is rendering the two separately, it must be pointing somehow to the wrong DNG, or the DNG's are not actually different.

But the DNG's saved to disk are different, because other programs see them as different, only LR is seeing (or at least rendering them) as the same. Right?

Wait... are you using Smart Previews? If so, turn them off and try? But assuming not...

So if it's getting its wires crossed and pointing to the wrong file, can we do something to prove that? I think you have already done things like import separately and move around first to no available, right? Seeming to indicate it relates to file name or contents, that it is still connecting them together somehow.

When you exit from Helicon and LR does the import, is there a time when you can stop? I.e. does Helicon save the files and then you exit, so you could pause there? And get to the files before LR does?

What happens if you substitute your own DNG for one or two of the files, some completely different subject but exactly in the same place and same file name, also a DNG. A little slight of hand before LR gets it. What happens then?

Or... if you can do that... what happens if before LR imports but after Helicon saves it, you open them in Photoshop and edit it somehow, write a big text 1, 2, 3 on each or similar, and resave in place over that DNG. This will rewrite the DNG, and maybe restructure it. If LR then works correctly, it seems to point to an issue in how Helicon wrote it.

Something is happening in LR that is confusing the files, somewhere. The only way I can see to find the problem is to isolate what you can do to stop it, even if it is not practical normally. Find a way to get them all imported correctly, then try changing one thing at a time to recreate the problem. Maybe even create an empty catalog and try it there.

Or, have you considered astrophotography instead of Macrophotography.