Fast Raw Viewer

[fdisilvestro]

Why would looking at the out of camera JPEG not be a better starting point for triage?

You are kidding, right?

[eliedinur]

Iliah, Will the RC overwrite 1.0.5 or will it install itself in a different folder?

[alextutubalin]

Iliah, Will the RC overwrite 1.0.5 or will it install itself in a different folder?

By default, it will install over 1.0.x

You may choose different installation folder on install ('Select Destination Location' step in install wizard)

[Iliah]

Release Candidate 4 is here: http://www.fastrawviewer.com/forum/fast-raw-viewer-110-RC4

What's new in Release Candidate 4 (compared to RC2):
- Filmstrip Off/On button on the right of status (bottom) bar.
- Filmstrip/Thumbnails performance tune: Preferences - Performance - Thumbnail cache - Thumbnail decoder thread count. You can raise the value for fast SSD drives (and ultra-fast flash cards, like CFast or UHS-II in fast readers), and decrease the value for slow HDD drives/slow flash cards/slow card readers.

[bjanes]

Release Candidate 4 is here: http://www.fastrawviewer.com/forum/fast-raw-viewer-110-RC4

What's new in Release Candidate 4 (compared to RC2):
- Filmstrip Off/On button on the right of status (bottom) bar.
- Filmstrip/Thumbnails performance tune: Preferences - Performance - Thumbnail cache - Thumbnail decoder thread count. You can raise the value for fast SSD drives (and ultra-fast flash cards, like CFast or UHS-II in fast readers), and decrease the value for slow HDD drives/slow flash cards/slow card readers.

Iliah,

I am a long time and very satisfied user of Rawdigger but am late with Fast Raw Viewer, since I thought Rawdigger did most of what I wanted to do. I just downloaded it and immediately purchased it after trying it out for a few minutes. I then downloaded RC4 and was impressed with the improvements. FRV is a valuable addition to my toolkit. It is amazingly fast with my 36MP D800e NEFs.

There is one thing I would like to do with FRV and have not figured out if it can be done. I have a series of test exposures of a saturated red flower at various exposures short of clipping in the red channel of the raw file. With the camera space set to AdobeRGB FRV does show the clipping in the red channel internal histogram with increasing exposure. This could be saturation clipping from rendering into the relatively small AdobeRGB space and this can be verified in some cases by rendering into a ProPhotoRGB JPEG with ACR with no resulting clipping of the reds. I can view the external JPEG histogram in FRV, but it would be helpful for FRV to show a histogram of the raw file rendered into ProPhotoRGB without the need to render a JPEG in ACR. Is there any way to do this?

With increasing exposure, the red channel can be clipped ProPhoto due to white balance. In this case the clipping can be eliminated by negative exposure in the raw developer as shown by rendering with ACR. Can the same be accomplished with FRV?

In any case, hats off to you and Alex for a great program.

Bill Janes

[Iliah]

Dear Bill,

We have 2 histograms, internal/external JPEG and raw. What you are looking for is the third one, call it "raw converted to ProPhoto"? This is actually doable, but will not be a "match" - as we may use one conversion method and one set of colour transforms while  some other program (like ACR) will be using a different one.

Thank you for the kind words, and always happy to hear your suggestions.

[AlterEgo]

as we may use one conversion method and one set of colour transforms while  some other program (like ACR) will be using a different one.

if you pick & use a dcp profile from the set available to Adobe (and apparently by FRV user intending to use ACR/LR later down the pipeline like bjanes) then you shall be able to match, no ? that is quite some code to write though, albeit it is documented by Adobe.

[Iliah]

Unfortunately, picking a dcp profile is not enough, there are more components to that - like hard-coded baseline exposure compensation. There are also several versions of the "process", and several dcp profiles to choose. There is interpolation between 2800 and 6500; and extrapolation. Need to think about it. It is actually easier to repair false blow-out caused by white balance.

[bjanes]

Unfortunately, picking a dcp profile is not enough, there are more components to that - like hard-coded baseline exposure compensation. There are also several versions of the "process", and several dcp profiles to choose. There is interpolation between 2800 and 6500; and extrapolation. Need to think about it. It is actually easier to repair false blow-out caused by white balance.

Iliah,

You know the complications much better than I, but what I am interested in is a rendering of the camera XYZ data before white balance into ProPhotoRGB going from the 3x3 camera matrix (of which there are many, but I would image that Adobe's as used in DCRaw would be suitable) to ProPhotoRGB. I see that you have already taken the Adobe baseline offset into account.

Adobe PV2012 is not suitable for this purpose, since it is image adaptive. I have done quite a bit of work with a Stouffer wedge exposed with my Nikon D800e and have found that PV2010 with a linear tone curve and all the sliders set to zero does a pretty good job when rendered in to ProPhotoRGB with ACR. I use the Adobe Standard profile for this purpose. The data are still gamma encoded, but this is no problem in a color managed workflow. If desired, one can convert to linear with the linear_RIMM-RGB_v4.icc profile using the colorimetric intent as described in this ICC white paper.

If this is not feasible in FRV, I will continue to render into a ProPhotoRGB JPEG with ACR and PV2010 as described. I can then use FRV to review all the histograms.

Regards,

Bill

[Iliah]

Dear Bill,

As I said, need to think about it. Mostly, it is a documentation challenge - how to make clear what we are doing, avoid misinterpretation and not to raise false expectations.

[bjanes]

Dear Bill,

As I said, need to think about it. Mostly, it is a documentation challenge - how to make clear what we are doing, avoid misinterpretation and not to raise false expectations.

Thanks, Iliah. No problem. I realize that FRV can not be all things for all users.

Bill

[Iliah]

Dear Bill,

Thing is - we are just starting with FRV. We will be adding features to it for sure. Right now the primary goal is to release 1.1; and after that we will be turning the page of the ToDo list.

[Iliah]

FastRawViewer 1.1 is released. Starting new ToDo list, please tell us what you feel is missing (keywords, sorting, filters are already in the new list).

[m_rouleau]

On sale for $15?! Bought without thought! It really is a no brainer.

One thing I would like to do, and this is a nit, is to be able to modify the IPTC/XMP description without having to enter a description, close, move to the next image, reopen the window, type in the new name.

It'd be awesome if I could just enter a name and then tab or arrow to the next image, without having to close the dialog box and reopen it, if that makes sense.

[Iliah]

to be able to modify the IPTC/XMP description without having to enter a description, close, move to the next image, reopen the window, type in the new name.

It'd be awesome if I could just enter a name and then tab or arrow to the next image, without having to close the dialog box and reopen it, if that makes sense.

Yes, it does make sense, thank you, have it in ToDo list.

[digitaldog]

On sale for $15?! Bought without thought! It really is a no brainer.

It is. The only area I HATE about this product is going back to Lightroom and seeing it's pathetic speed to preview images in Grid. I'm now so spoiled.

[jrp]

I'm giving this a go, but getting fairly regular random crashes such as the following:

Crashed Thread:        0  Dispatch queue: com.apple.main-thread

Exception Type:        EXC_BAD_ACCESS (SIGSEGV)
Exception Codes:       KERN_INVALID_ADDRESS at 0x000000017e7b400d

VM Regions Near 0x17e7b400d:
    MALLOC_LARGE           000000016771c000-000000017c13a000 [330.1M] rw-/rwx SM=PRV 
-->
    __TEXT                 0000123480000000-000012348043b000 [ 4332K] r-x/rwx SM=COW  /System/Library/Extensions/AMDRadeonX4000GLDriver.bundle/Contents/MacOS/AMDRadeonX4000GLDriver

Thread 0 Crashed:: Dispatch queue: com.apple.main-thread
0   com.libraw-llc.FastRawViewer  0x0000000100187422 EXIFInfo::parseIFEntry(unsigned char const*, unsigned int, bool, unsigned int, unsigned int, unsigned int*, unsigned int) + 546
1   com.libraw-llc.FastRawViewer  0x0000000100188ace EXIFInfo::parseFromEXIFSegment(unsigned char const*, unsigned int) + 1166
2   com.libraw-llc.FastRawViewer  0x000000010016a0e4 FRV_RawProcessor::readJpegData(unsigned char*, unsigned int, int, QString const&, int, float) + 404
3   com.libraw-llc.FastRawViewer  0x000000010016bef5 FRV_RawProcessor::processFile(LibRawData_wrapper*, JpegItem*, bool, int, int, bool, QVector<float> const&) + 2053
4   com.libraw-llc.FastRawViewer  0x000000010007794c FRV_MainWindow::openFileAndDisplay(QString, LibRawData_wrapper*, JpegItem*, int, int, int) + 4780
5   com.libraw-llc.FastRawViewer  0x000000010007e960 FRV_MainWindow::openFileLibRawAndJpeg(QString, LibRawData_wrapper*, JpegItem*, int, int, int) + 48
6   com.libraw-llc.FastRawViewer  0x00000001001dc55b FRV_MainWindow::qt_static_metacall(QObject*, QMetaObject::Call, int, void**) + 2907

[Iliah]

Thank you. The crash is caused by some irregularities in EXIF data. Please make available the raw file that causes the crash (and JPEG, if there is an external JPEG for this file too in the same directory); you can reach us at support@fastrawviewer.com

[jrp]

Thanks.  Co-incidentally I came across a post elsewhere pointing out that Leica did not terminate all its Exif fields.  So that could be the issue. Equally, I use photomechanic to ingest my photos and apply metadata, including xmp creation.  So it may be producing encodings that you don't like. Either way, it is obviously not good to have a program apparently reading its way through memory in an uncontrolled way.

Let me know if this is enough or if you need more.

[Iliah]

Dear Sir,

But we really need to look at the files to fix it. Please...