Topic for feature requests

 

Please post feature requests in this forum thread.

Assuming you're running Windows, create one-line .cmd-file like this:

exiftool -s -k %*

And add it to External programs list in Preferences

What about on Mac?

On Mac this is not that simple, because running a command-line tool will not open (terminal) window.

There are two possible solutions:

#1: something like this: https://stackoverflow.com/questions/4404242/programmatically-launch-term...

This is not final solution, but just a sample how to run Terminal app with specified command started inside.

#2: Use PyExifToolGui  ( https://github.com/hvdwolf/pyExifToolGUI/releases ), or other Exiftool GUI variants (not many for Mac?)

I've never used #2, I'll play with it later today to see is it working or not.

Iliah, Lexa et al,
Thank you for a wonderful program and for how well it works in conjunction with RawDigger (which I recently bought). Here's something that would be a hybrid idea.
I would find the histogram in FRV a lot more useful if it contained more detail on the scale (perhaps a bit closer to what I see in RawPhotoProcessor 64). I would also find the FRV evaluation of an image more helpful if it would support area selection such as in RawDigger and other programs. It would be great if the OE/UE stats would optionally apply to a selected area. Lastly, it would be helpful if I could see RGB and/or L* values for where the cursor is pointing or even averaged for a selected area. 
-------------------------------------------------------------------
I also have a "tip" for general interest:
Since my ratio of images to work on vs. images to not work on is low, I use the "Reject" folder as a "workonit" folder and copy the good ones to there, leaving the entire set of originals where they are stored after copying from the camera. This does not require any software change, just a change to how the user approaches the culling activity.

Dear Sir:

1st: Histogram calculation is already the slowest part of FRV image processing pipeline (with GPU debayering, introduced in FRV 1.4.6, we usually display RAW rendering faster than a  JPEG of the same size), so adding details (more histogram bins), or adding histogram on selected part will slow down FRV a lot. We provide RawDigger for detailed raw data examination, and FRV for 'techical image evaluation in 1-3 seconds'.



2nd: Although _Rejected may be used as 'selected', we provide several tools to quickly move files to different (sub)folders:

  - Move/Copy to subfolder feature (Alt-M/Alt-C) accompanied by a drop-down list of recently used subfolders

  - Shortcuts for Move/Copy to last used folder (Shift-M/Shift-C) and also to '2nd...5th' used folder (unassigned, but you may assign shortcuts for that via standard method).

So, it is easy not to move keepers to a single (_Rejected) subfolder, but quickly sort files placing them to multiple destination (sub-)folders.

Dear sirs, I just noticed an inconsistent behavior.
Test case: two exposures of a blank white target, center metered, looking for overexposure. One shows as OE, the other does not. They are 1/3 stop apart in original exposure.
When I use the exposure up/down adjuster on the not-OE file the OE stats eventually show the OE and the purple spot appears. This tells me how much below OE  threshold the original exposure was.
However when I do the reverse with the OE image, the stats don't change and the purple spot does not vanish. So I'm unable to see how much over exposed it was.
This is not a big problem but it's a bit puzzling because the image does get darker with each incremental adjustment downward.
 
Thanks.

Dear Sir:

OE display takes positive exposure correction (including 'Adobe hidden correction, if enabled) into account by default (to turn this off, uncheck Preferences - Exposure - Exposure adjustments affect Overexposure display).



So, when you increase exposure on a non-overexposed shot, OE display will appear.

On already overexposed (so detail-less) areas, the details  will not appear on exposure (really brightness) decrease. FRV OE display reflects this.

Followup:

Any (e.g. +6 stop and +10 stop) overexposed areas looks very similar (flat area at maximim possible pixel values), there is no way to determine exact overexposure using only RAW file.

i know the way of FRV that it not mess with inner metadata. BUT.
And thats big but.
When i switched form LR to C1 (oh yeah) i lack there two very important functions.
Adjusting capture time and fast renaming based on presets.

Could FRV do this? I have to import every session into LR just to make those 2....

ugrandolini's picture

Thank you so much for this great software!
After almost one year of usage I guess I saved quite a lot of time when culling my images :-)
Last may I went to a workshop ran by a famous photographer and he suggester PhotoMechanic to cull images, so I gave it a try and at the end decided to stay with FastRawViewer.
However I now have some feature requests to make FRV even better:

  1. in my workflow when I get back to the studio I copy all the images of the shooting in a NAS then I then cull the images from the NAS on a Mac connected trough Wifi and I must say it takes time to move from an image to the next; it will be nice to have the option to create a local temporary cache of all the images so moving between images will take less time – there is some more time to wait while creating the cache but I can do other things meanwhile and be more productive
  2. it would be nice to have the option to export a JPG version of the selected images even better if I can add a watermark; this way I can send unretouched versions to the customers without the need to open Capture One after the culling just to export the JPGs

I'm ready to pay for an upgrade just to get these two features implemented!
Thank you,
Ugo
 

Dear Sir:

Thank you for your kind words.

WiFi is hardly suitable for performing massive file copy operations in a fast manner, and the size of files tend to grow faster than the speed of WiFi. Please have a look at https://www.lifewire.com/how-fast-is-a-wifi-network-816543 To quote, "Even though 802.11n rates at 600 Mbps compared to wired Fast Ethernet at 100 Mbps, the Ethernet connection can often outperform 802.11n in real-world usage." From my personal experience, even a 1 Gbps network is slow for culling a day's work. Because of WiFi speed issues, making a local copy will be slow too, making it not-so-acceptable as a solution.

Yet another possible bottleneck is the NAS speed, ready-made WiFi-capable NAS systems are often slow.

If you absolutely must use WiFi, please look at the settings we recommend in "Flash Cards and Network Volumes" chapter of the Manual (it is p.104 of the current version of the Manual); further explanations are in "Performance" chapter (p.142).

For your convenience, FastRawViewer Manual comes with the installation (main Menu - Help - "PDF Manual") and is fully searchable. You can also download it separately from https://www.fastrawviewer.com/download - direct link is http://updates.fastrawviewer.com/data/FastRawViewer-Manual-ENG.pdf

Local copy can save the day only if images are already on the computer, but because of the WiFi / network speed limitations and possible NAS bottleneck it is hardly the case in your scenario - you need to first copy the images to the local computer, and that takes time. Maybe you can consider an adjustment to workflow, culling from the cards to the local computer, and after that to copy the culled images to NAS? - that's how I do it. It is worth mentioning that FastRawViewer caches the files in memory, and it needs the whole raw file, not just the 10% or less the embedded JPEG takes.

ugrandolini's picture

Ok I can use a temporary hardi disk to do the culling.
What about having the ability to export JPG directly from FRV?

We have JPEG and/or TIFF export in our TODO list, but this is not so easy as it sounds (export without cropping/rotation is useless in many cases).
ugrandolini's picture

hey!
For Network volumes:
When using a Wi-Fi network:
Lower the preferences values
Preferences – File Handling – Removable Media – Prefetch depth on network volumes to 2-3

When using fast network/NAS (1Gb/s and faster, in NAS with 4 and more disks in RAID):
Raise the preferences values
Preferences – File Handling – Removable Media – Prefetch depth on network volumes to 6-8
 
this was really useful, it's much better now.
Please add the ability to export JPG!

Your application is great in many ways and I am impressed how fast and smooth working with it is. However, it is lacking a feature that my workflow simple requires: tagging. I understand I can use labels and rating but having thousands of photos and working on a couple of projects at the same time, my library of photos requires searching for them by tags and/or categories. I would really like to be able to not only search by filename but also assign custom tags (e.g. "wedding", "documentary", etc.) and search by those. It would also be great to be able to search by title and description.

Keywords (tagging) is in TODO for some future version (not for FRV 1.5), sorry no ETA

XMP Rating Filter suggestion: Only a small probem: If I rate my images with stars *, **, *** and so forth, I always have to do the following: I go through all the images and make a first large selection with *. After that, I do a smaller one with **, the best out of the *. To do this, I enable a filter with *, to see only the first large selection. Then I apply the first **. This image vanishes, when I apply the **, because the filter only * is selected. To see it anyway, I have to adjust the filter with a hook at ** too. It is always the same procedure: I can't set * and ** in the filter without setting a first ** rating to an image, because all fields, that do not have images applied, are great out. In my opinion, it would be better, to make the check mark field at all stars in the filter section always available, independant from the count of ratings set to the images with this rating. Then I would be able to set the rating for * and more stars before my selection starts. The number "0" behind the star rating field in the filter shows me the count. To my eyes, there is no need to grey the check box out, if the rating is not applied in the images.
 
All other programs like Lightroom or Capture One allow to do this and I am used to this procedure.
 
But anyway, Fast Raw Viewer is my favorite for selections of great amounts of photos. A really great tool!!!

Dear Sir: Thank you for kind words. Enabling, say, '2-star' checkbox if no files in folder have this rating will result in another interface problem: if someone will check this chekbox, than all files in folder will be filtered out. This may confuse (new) users. So, we'll consider 'all filter checkboxes are always enabled' as possible option (settable via Preferences), while default behavior will not change. Added to TODO list for version 'next after 1.5.0' (or, may be, we'll add it during 1.5.0 beta-test phase).

Hi, when browsing through photo bursts with 100s of almost identical pics I find it difficult to see from the thumbnails in the filmstrip panel which thumbnail is selected/active.
In my actual configuration the selected thumbnail has no frame, all other thumbnails have a thin grey outline. This is the only visual difference to distinguish the selected pic.
Is there a way the configure the highlighting / framing of the selected thumbnail so I can detect it more easily?

Thanks for any help,
Ralf

Dear Sir:

Is FastRawViewer Preferences -> Interface -> "Active (current) file border" and "Selected files background" what you are looking for?

You may find an example in the Manual, "Operations with Multiple Files" section, useful. The settings for the border and background are described in "Interface" section.

For your convenience, FastRawViewer Manual comes with the installation (main Menu - Help - "PDF Manual") and is fully searchable. You can also download it separately from https://www.fastrawviewer.com/download - direct link is http://updates.fastrawviewer.com/data/FastRawViewer-Manual-ENG.pdf

Hi, when browsing through photo bursts with 100s of almost identical pics I find it difficult to see from the thumbnails in the filmstrip panel which thumbnail is selected/active.
In my actual configuration the selected thumbnail has no frame, all other thumbnails have a thin grey outline. This is the only  visual difference to distinguish the  selected pic.
Is there a way the configure the highlighting / framing of the selected thumbnail so I can detect it more easily?

Thanks for any help,
Ralf
 

Hi,
A right-click option to rename individual files would be very helpful.

Suggestions: RGB display for point under cursor; chromatic abberation correction

I'm very happy with FRV and the rapid pace of improvements. So consider these "suggestions" rather than requests.
First, it would be nice if FRV could display the RGB values of the point under the cursor. Ideally I'd like to see both raw values, and values after color temp., curves, etc. have been processed.
Second, I'm spoiled by so much photo software that automatically adjusts for chromatic abberation. I would like to see FRV do this, under user control of course. It would help with checking focus in the edges and corners of the frame. I expect this will take some work so I won't hold my breath.

Last, have you considered working with the Darktable and Raw Therapee teams on XMP compatibility?
Thanks again for an absolutely essential tool.
 

I love this software, it saves me untold amounts of time.  I tend to go through large amounts of files (1000s) at a time, building up a selection and rejecting others.  I typically move the selection to a _Process folder every ~50 images but once in awhile I mis-click or otherwise do something that causes me to lose a large selection.  Maybe it would be possible to add a feature to re-select the previous selection?  :)

FastRawViewer default selection mode  is  'Advanced selection mode: accidental click does not ruin selection', so single click should not break entire selection. Only excplicit 'Deselect All' action (Ctrl-D in default key settings) will do that.

If you use 'standard' selection mode, consider switch to advanced one: it was designed with 'accidental clicks' in mind.

Recording each action/selection change in the undo stack will result in fast undo stack overflow (and removal of really dangerous actions such as batch move to trash subfolder).

 

Auto-Select new image folder request

Hi FRV Team!
I'm very happy with my FRV Tethering work-around using Sony Remote Tethering Software set to copy RAW images to a specific location and FRV open and watching that location. This proves a great tool for my assistant to check exposure and focus!

My feature request (if this does not already exist - i've scoured the option menus) is for an option to have FRV auto-select the newest image in a folder. This will save having to arrow key or click the lastest image and essentially make FRV a solid tethering solution.

Many Thanks!

Dear Sir:

The problem is not as simple as it sounds. There are a lot of corner cases that are not easy to handle in (already complex) FastRawViewer application:

  • file could be not complete when FRV first detects it
  • file could be created with temp. name, than renamed
  • file could be locked by tethering app and not possible to open for some time.

We discussed this internally. Most likely we'll provide small separate app (hotfolder.exe/hotfolder.app Win/Mac) that will deal with (most) corner cases and submit the new file to FastRawViewer only if that file is not corrupted and not locked.  This small app is scheduled to FastRawViewer 1.6.

Page 4 (by SVChucko) has a request for the sRGB value to be shown under the cursor.  I'd like to cast another vote for that feature to be added.  Shoiwng a 0-255 there would allow me to apply it to another editor or get an idea if a neutral gray card of sRBG of 118,118,188 is really well white balanced or needs a correction.
 

FastRawViewer doesn't have sRGB data available on CPU. To add it means to slow the application down, and to add a lot of complexity to it. Plus, the render is not an absolute, different raw converters will interpret the raw data differently.

Sharpness Statistics
Just like the exposure statistics, show sum of edge intensities in the focus peaking modes: the overlay is dark if there is no edge (0) and bright green if there is (255), just add all these numbers up. This makes it easy to select the sharpest image.

This is  not possible: focus peaking is calculated completely on GPU side (while displaying the image), there is no data available to display.

Generally speaking, it is (theoretically) possible to transfer calculated overlays on CPU side and perform requested numbers calculation. This will slow down things a lot (several orders of magnitude), so it does not looks like real solution.

You don't have to take the data from the overlay. You have the raw data on the CPU side, and just like you do for the exposure stats you can sample, say, 10000 random pixels, calculating the difference in intensity for all three channels to neighboring pixels and sum the absolute values of the differences.
Like:
sum = 0
for p in N random (x,y): // let p be a random image coordinate, repeat until convergence or whatever
    diff = abs(red(x,y) - red(x - 1, y)) + abs(red(x,y) - red(x, y - 1)) +
        abs(green(x,y) - green(x - 1, y)) + abs(green(x,y) - green(x, y - 1)) +
        abs(blue(x,y) - blue(x - 1, y)) + abs(blue(x,y) - blue(x, y - 1));
    sum +=  diff
 

Your code assumes that the demosaiced and linearized data is present on CPU side. That is not so if RAW GPU processing is used.

Ok, but my point is this: you can seemingly do computations that get you the exposure stats.

Yes. But sharpnes stats are very different.

Exposure stats are based on histogram calculation. Fortunately, it is possible to calculated histogram on unaltered/unprocessed RAW values (not linearized, not white balanced, etc, etc), than adjust histogram same way as single pixel. Also, it is possible to calculate histogram in advance (on raw data prefetch).

And, yes, histogram calculation is slowest part of FRV image processing pipeline.

Laplacian calculation:

a) ~order of magnitude slower (9 pixel fetches per output value, compared to single pixel sampling for histogram)

b) requires complete image processing on CPU side to get data for laplacian: another 2-3 times slow down (compared to GPU processing).

So, yes, sharpness stats are possible, but will slow down FRV a lot.

Checksum in XMP
Please consider adding raw file checksum to the xmp sidecar file in order to verify raw file has not been changed or corrupted in future operations. 

thanks...jeff

Several raw converters and some other applications do write into raw files (EXIF blocks, GPS blocks, XMP blocks, VRD blocks, etc.), checksum will be invalidated by such writes.



We have a different application, not FastRawViewer, in mind to sort this out.

Pages

Add new comment