Puget Systems print logo

https://www.pugetsystems.com

Read this article at https://www.pugetsystems.com/guides/1571
Article Thumbnail

PugetBench for Lightroom Classic

Written on December 17, 2019 by Matt Bach
Share:

At Puget Systems, one of our primary goals is to provide fast and reliable workstations that are perfectly tailored to each customer's unique workflow. As a part of this, we publish an ongoing series of hardware articles for Lightroom Classic - as well as numerous other software packages - in order to discover what hardware configurations are optimal for these specific applications.

However, while this ensures that we are selling the right hardware, it does not give our customers a great idea of how much faster a new workstation might be compared to their existing system. In addition, with the amount of hardware that is currently available, it is impossible for us to test everything on our own, so we want to provide tools for other hardware reviewers to perform real world benchmarking in professional applications like Lightroom Classic.

To address both of these issues, we are making our Lightroom Classic benchmark available for download so that anyone can can perform the same testing we do in our hardware articles.

Download PugetBench for Lightroom Classic

PugetBench for Lightroom Classic free edition

Free Edition

Download V0.85 BETA (1.7GB)

Suitable for personal use

PugetBench for Lightroom Classic commercial edition

Commercial Edition

More Information

For commercial use
Includes email support, automation, and logging

Be sure to check out the full list of benchmarks we have available!

System Requirements

  • Adobe Lightroom Classic 2020 or Adobe Lightroom Classic 2019
  • 16GB of memory
  • Windows 10 (no MacOS support currently)
  • Lightroom Classic/OS language must be set to English

Note that this benchmark is still in BETA. Plug-ins and customized preferences in particular may prevent the benchmark from running properly. If you run into any issues, we recommend resetting your preferences either by following Adobe's guide or by using our Adobe CC Preference Backup/Restore/Clear utility (Windows only).

The benchmark requires that the Lightroom window stays in focus and that the mouse/keyboard is left alone for the duration of the test. Trying to use your system while the benchmark is running will result in the benchmark failing.

How to run the benchmark

  1. Downloading the benchmark from the link above
  2. Unzip the benchmark to a convenient location - we recommend placing it on the same drive you storage your active projects as storage performance can affect the results.
  3. Launch Lightroom Classic and open the "Benchmark Catalog.lrcat" catalog.
  4. Ensure that the proper application settings can applied.
    • If one of the following setting is not correct, you will need to import one photo in order to unlock the develop module in order to change the setting.
    • Set the Develop Module 'Spot Removal' tool is set to 'heal', radius: 75, feather: 0, opacity:100. This brush is used for the "brush lag" test and requires that it be set to this (default) value to ensure benchmark consistency
    • Panorama merge is set to: 'Spherical' with 'Auto Settings' (all other options unchecked)
    • HDR merge is set to: 'Auto Align', 'Auto Settings', 'None' deghost amount (create stack unchecked)
    • Alternatively, you can backup and reset your Lightroom Classic preferences using Adobe's guide or by using our Preference Backup/Restore/Clear utility instead of manually changing these settings as the benchmark uses the default settings.
  5. Install the benchmark plugin
    • Go to "File -> Plug-in Manager"
    • Click "Add"
    • Navigate to the benchmark folder and select the "PugetBench.lrplugin" folder
    • Click "Done"
  6. Go to "File -> Plug-in Extras" and select "Benchmark Run"

When the benchmark is complete, it will give you an "Overall Score" as well as individual scores and results for each test. A screenshot of the results is saved in the benchmark plugin folder that you can share. You can compare your scores to those in our Lightroom Classic hardware articles - just make sure they used the same benchmark version!

Troubleshooting

If you have any issues with the benchmark not properly running, firs verify that your system meets the System Requirements and that the version of Lightroom Classic you are using is supports. If there are no issues with your system or software, there are a number of troubleshooting steps you can try:

  • Make sure you are following each step listed in the How to run the benchmark section
  • Do not attempt to use the system while the benchmark is running. Defocusing the application window can break the benchmark in many situations.
  • Clear the application preferences and reset them to default.
  • Ensure that your OS and Lightroom Classic is set to English and that the path to the benchmark script does not have any special or non-English characters.
  • The benchmark does not work over an unmapped network drive (I.E. //192.168.1.2/Benchmark/). Mapped drives should work, but we highly recommend having the benchmark on a local drive as storage performance can affect the results.

Lower than expected performance can be caused by a number of factors, but the most common are:

  • Outdated BIOS/drivers/software
  • Running background processes (Antivirus, Windows Updates, etc.)
  • Windows power profile (High Performance is recommended for the best results)
  • Adobe application settings are not correct (GPU acceleration not enabled, etc.)

If you continue to have issues, we recommend posting in the comment section below - ideally with a screenshot of the problem and a copy of any log files (inside the "PugetBench.lrplugin" folder) that are generated by the benchmark.

What is tested?

Our Lightroom Classic benchmark looks at performance with three sets of images for both "active" and "passive" tasks. The image sets include 500x images from the following cameras:

  • Canon EOS 5D Mark III - 22MP .CR2
  • Sony a7r III - 42MP .ARW
  • Nikon D850 - 45MP .NEF

The following tasks are tested:

Active Tasks:

  • Library Module Loupe Scroll (The time it takes between hitting the right arrow ey and the next image displaying on screen)
  • Develop Module Loupe Scroll (The time it takes between hitting the right arrow ey and the next image displaying on screen)
  • Library to Develop Switch (The time between hitting the Devlope panel button and when Lightroom actually switches)
  • Develop Module Auto WB & Tone (How long the image takes to update after applying the Auto Tone and White Balance)
  • Develop Module Brush Lag (When using the spot heal tool, how quickly can you move the brush before it significantly lags)

Passive Tasks:

  • Import 500x Images (Add from location, minimal previews, do not generate Smart Previews)
  • Build 500x Smart Previews (Default settings)
  • Photo Merge Panorama (Time to create a panorama of 6 images using default settings [Ctrl-M])
  • Photo Merge HDR (Time to create an HDR of 6 images using default settings [Ctrl-H])
  • Export 50x JPEG (Time to export 50 images to JPEG [60 quality] with the longest side being 3840 pixels. Auto Tone and W/B applied)
  • Convert to DNG 50x Images (Time to convert to DNG [dng, Camera Raw 11.2 and later, medium size, embed fast load data])

How does the scoring work?

The scoring system used in our benchmark is based on the performance relative to a reference system with the following specifications:

  • Intel Core i9 9900K 8 Core
  • NVIDIA GeForce RTX 2080 8GB
  • 64GB of RAM
  • Samsung 960 Pro 1TB
  • Windows 10 (1903)
  • Adobe Lighroom Classic CC 2019 (ver. 8.4)
  • Overall Score: 1000
  • Active Tasks Score: 100
  • Passive Tasks Score: 100

The Overall Score is simply the average of the active and passive scores and multiplied by 10 since bigger score = more important.

PugetBench for Lightroom Classic sample results

Sample result screen

Update Log

Version .85 BETA

  • Added functional progress bar to DNG and export image deletion during cleanup (this can sometimes take longer than you would expect)
  • Additional logging for plugin and catalog path (useful for troubleshooting)

Version .8 BETA

  • Rename to "PugetBench for Lightroom Classic"
  • Improved timing accuracy for the Active tests
  • There is now a "Benchmark Results" screen that comes up at the end of the benchmark that displays a bunch of useful information including: benchmark version, ccores, results for each individual test, and system information like CPU, RAM, OS, GPU, and Photoshop version
  • The benchmark now also makes a PNG of the "Benchmark Results" screen for easy sharing
  • Removed .csv log file support in the free edition (log files will be a feature in the commercial use version)

Version 0.2 BETA

  • First public release.

Looking for a Lightroom Classic Workstation?

Puget Systems offers a range of workstations that are tailor-made for your unique workflow. Our goal is to provide the most effective and reliable system possible so you can concentrate on your work and not worry about your computer.

Configure a System!

Tags: Benchmark, Lightroom CLassic, Public, Download
Phil Erb

I am attempting to run the Lightroom Classic script on my old (but current daily user) i7-2700k system, but I am getting the following error about 3 seconds in:

Error:
An internal error has occured: [string "BenchCode.lua"]:301: attempt to call upvalue
'writelog' (a boolean value)

I do have the folder on the SSD disk I use to import and work all my LR projects.
I also get the same error if I launch lightroom.exe as administrator.

Posted on 2019-12-18 14:54:33
MM

same here, pops up instantly

Posted on 2019-12-18 17:32:09
Fred

Same problem on my i7-4790K, Z97, 16GB RAM, 1TB MX500 SSD, Win 10 Pro.

P.S. Many thanks for sharing the benchmark. I'm very eager to compare i7-4790K with the new 3600/3700X Ryzens.

Posted on 2019-12-18 20:24:31

What version of Lightroom Classic are you running?

Posted on 2019-12-18 22:37:41

Nevermind actually, I think I figured it out. Bit of broken code that only happens if you don't have the spot removal tool set to the correct settings. I'll have a new version uploaded in a few minutes, or just make sure to follow step #4 of "How to run the benchmark"

Edit: new version is up. You can either re-download the entire benchmark, or just make sure the Develop Module 'Spot Removal' tool is set to 'heal', radius: 75, feather: 0, opacity:100 before running the previous version of the benchmark. If you do re-download, make sure to remove the old plugin before adding the new one just in case there is some conflict.

Posted on 2019-12-18 22:41:41
Phil Erb

Matt, thanks for the quick reply and fix production!
I had set that values to 75 / 0 / 100 before and had it set to heal, but I still had the new popup stating to change the values.
I used continue and it went through process for about 100 minutes or so (remmeberm i7-2700k non-OC, so a bit slow), but then an error popped up while trying to create a panorama.
From what I can see, it grabbed the wrong images. It looks like it should have grabbed the 6 from 1477 to 1482, but instead had 1483 to 1488 which are not suited for panorma creation.
https://uploads.disquscdn.c...

There were a couple of other errors (on during initial import. I will try to get screen shots next time.

Posted on 2019-12-19 03:10:00
Phil Erb

https://uploads.disquscdn.c... Second run through, same error.

Posted on 2019-12-19 11:54:50
Phil Erb

I will test with changing projection to spherical when I have some time.

Posted on 2019-12-19 14:30:15

Oh, interesting. I was under the impression that automatically generating HDR/Panorama would use the default settings, but I guess it must be pulling from whatever the last used settings was. Hmmm, I'll have to think about how to get around that, because I don't think there is a way to detect or change those settings from within the Lightroom API.

Thanks for all the information about this bug!

Posted on 2019-12-19 20:18:53

Just an update, I did confirm that if you have non-default panorama or HDR settings that the benchmark can break (or give inaccurate results). I updated the instructions to show which settings to check, but an alternative is just to reset Lightroom back to default preferences. I have had it in mind to make a little backup/reset/restore utility for Adobe application preferences, and this might be the little nudge I need to go ahead and do that. That way, step 1 for all our benchmarks can be "backup your preferences and reset to default" in order to avoid these little issues.

Posted on 2019-12-19 23:27:48
Phil Erb

Also confirmed. Spherical works correctly.. my poor old i7-2600k with newish 500 GB SSD's stumbled through it in 2 hours and 20 minutes or so
Specs:
ASRock Fatal1ty Gen 3 z68
i7-2600k @ 3.4 GHz
OS drive - Samsung 860 EVO 500 GB -- Appdrive - Crucial MX300 525GB
Memory - 4 x 4 GB DDR3 (Corsair Vengeance 1600 Mhz 9-9-9-24)
https://uploads.disquscdn.c...

Posted on 2019-12-20 00:53:49

Hey, that's not too bad for a CPU that was released almost 9 years ago! It is always interesting to see what older systems get since I usually am only dealing with the latest 1-2 generations of hardware in my testing. A 2.5x increase in performance over 9 years is really cool to see.

Thanks for sharing the results, and helping me work out the issue with the benchmark re-using the "last used" settings for panorama and HDR!

Posted on 2019-12-20 00:58:18
Phil Erb

Hey Matt,
Thanks for doing this.. I now have ammo to assist in the replacement of the old girl. I have updated the specs of my system above in case you or anyone else is interested.
FYI, even though I had the spot removal tool set to 75/0/100 and heal, I still got the pop up and had to use the continue button.
I'm amazed at the Ryzen 3900 below where the smart previews are 33 seconds to my 4 minutes, but his import of the CR's was actually longer by a smidge....(11.85 vs 12.37)

Posted on 2019-12-20 13:39:21
Jean Du Bois

Hi Matt. It managed to finish the benchmark but no results where displayed. How do i know what the scores are?

Posted on 2019-12-19 09:18:19
vfeast

Scored 1268.5 w/ this 3900x clocked at 4100. SSD is a Samsung 860 EVO 500GB. Nvidia studio driver 441.28.

https://uploads.disquscdn.c...

Posted on 2019-12-20 06:30:07
Jim Kiefer

Check out that brush lag!

Posted on 2019-12-21 00:25:05
FX-8350

Very impressive result, gratulations! But, how comes? 121,1 in active tasks score, this is 21% more then the fasteast tested CPU from PugetSystems. Is your RAM overclocked?

Posted on 2019-12-27 08:29:00
vfeast

The 3600MHz ram is only at 3666MHz, but more importantly, the timings have been tweaked with the Ryzen memory calculator, and the fabric clock is running full speed with the RAM. The 4100MHz CPU clock probably helps quite a bit as well. I had everything running much faster but have backed it down because I need my system to be 100% stable for work.

Posted on 2019-12-27 16:10:46
dogwalker350 .

After the benchmark utility was no longer showing any images remaining in the benchmark catalog and no longer showing any sign of activity, I noticed the upper left hand corner had a progress bar about removing images (that was not showing any progress in the thermometer area and appeared either frozen or completed), After waiting perhaps 5 minutes more, I minimized Lightroom and looked around the benchmark folders for the expected summary score screen shot without success. A txt file was present with individual process results but without summary scores..

I then navigated to this forum, reviewed comments,and prepared to voice concern about not getting the result score. While getting ready to type, a window sprang open with the test results - perhaps 20 to 30 minutes after the test seemed completed.Then the screen flashed briefly, capturing a screen shot of part of the results (now saved in the expected place), but with the screen shot having the final scores on the right hidden behind the forum window I had open.

On re-expanding the Lightroom application, the progress bar was no longer present in the upper left hand corner. So some aspect of the utility had not fully finished even though I could see no evidence of activity. It could be useful for the instructions to be more explicit about what we should look for to know the utility is done. Also, it looks like the result reporting process could use some work.

I have i7-4930K @ 3.40GHz, 32 GB RAM, and got benchmark result of 661.5 overall, active tasks 74, passive tasks 58.3. Would be good to have a fuller explanation of the scoring when possible.

Posted on 2019-12-21 02:22:16
Dragon

@ Matt Bach. On a related note, has Adobe fixed the hyperthreading problem in LR yet or do we still need do run with HT off for best performance?

Posted on 2019-12-21 19:43:31

Smart previews and exporting is still much faster with HT off, but the most recent version of Lightroom Classic saw a performance boost for pretty much everything else with HT on. So personally, I would keep it on unless you export a huge amount of images. And if you do that, you might consider switching to Ryzen for your next system since (at least for now) it is way faster than Intel even with HT off.

Posted on 2019-12-23 07:20:12
Dragon

Thanks, Matt. I just built (about 6 months ago) a really nice system with a 9900k, so will live with that at least until the 7nm EUV stuff starts flowing and then see who is in the lead (and where the software has evolved).

Posted on 2019-12-23 16:51:24
OuiOuiPhoto

Hello. The Tool copy the files but at the import stage a windows appear asking me to select the path. i have try to select PugetBench.lrplugin\Images and click import. But after importing i see nothing mooving. I have reset my pref and put LR in English.

Posted on 2019-12-22 16:32:17
OuiOuiPhoto

And in the log i have

-----Starting Benchmark Run-----

Verifying Images
Getting System Information
Changing to Library view and clearing photos
Removing existing photos
Creating copies of images
Starting Import Benchmark
Importing 22MP photos

Posted on 2019-12-22 16:47:19
OuiOuiPhoto

I have dig a bit in the LUA in function benchmark_Import_loop(value) and add a write status to see the command line executed

writestatus("Command Line Is "..commandLine)
LrTasks.execute (commandLine)

The result of the write status is

Command Line Is "F:\PugetBench.lrplugin\Command\PScommand.exe" importprep Images\5Dmkiii

and i have try to execute it with CMD
F:\PugetBench.lrplugin\Command\PScommand.exe importprep Images\5Dmkiii
The result is the same i have a windows asking for a path that show up.

I have try with adding the full path
F:\PugetBench.lrplugin\Command\PScommand.exe importprep F:\PugetBench.lrplugin\Images\5Dmkiii
The result is the same i have a windows asking for a path that show up.

So for a reason i don't know i have always a windows "Chose sorce folder" that appear and stop the code

Posted on 2019-12-22 17:31:55
OuiOuiPhoto

Hello

First try and at the import phase it was stopping on the chose directory. I have solved it by installing English language on W10. But now it stop without error message at this stage

Starting Library to Develop Switch Benchmark
Library to Develop Switch 22MP

Posted on 2019-12-22 20:58:58
FX-8350

Same here. Is there a solution for Windows multilanguage support?

Big THX for your great Work!

Posted on 2019-12-23 15:38:36
OuiOuiPhoto

Hello FX-8350. You have the import problem or the other one ?

Posted on 2019-12-23 16:22:38
FX-8350

Hi OuiOuiPhoto,

the import problem is my problem :-)

(on Windows 10 german language). I tried also to extract PugetsysSytem's zip file in the root of C:\ or in the Pictures folder but in both cases the Benchmark ask me the location for importing files.

CU

Posted on 2019-12-23 17:21:43
OuiOuiPhoto

So for the import problem just install english language in your windows 10

Posted on 2019-12-23 18:05:21
FX-8350

Thanks for your suggestion! I installed english language, reseted Lightroom and everything work very well (about 40min.) till first panorama stitching. On this point benchmark stops.

Posted on 2019-12-23 21:08:41
OuiOuiPhoto

So seems we have the same behaviour

Posted on 2019-12-24 11:02:08
Fred

I had the same problem, but it was due to the fact that I had the import settings carried over from with my previous imports and not Lr default.

Check if your import settings are:
-Add pictures in the current location
-Minimal previews
-Smart Previews unchecked

Posted on 2019-12-23 16:16:00

That should be tied to the catalog file - so as long as you are opening the benchmark catalog and not changing things, it should have the right import settings applied.

If that isn't the case though, then we'll have to find some other way to get those settings set (it can't be done in the API or an external script since those checkboxes are actually technically just buttons)

Posted on 2019-12-23 16:18:36
StefanUrkel

Hi Matt regarding yout article on tesilve and radeon vii, i tried running your benchmark program on 1950x with 128gb ecc and 2 x radeon vii on a 1000w seasonoc prime psu. power draw per gpu max out at around 250w. at sone point into the full bench my system will power off suddenly without warning. temps seem fine for cpu and gpu.. 70c max. when i unplug one radeon vii and rerun it finishes fine the entire full bench.. when i pluf the power back to the 2nd radeon vii and rerun it either powers off my system at sone point or the bench freezes on a test but my system is still on.. sometimes i can kill the test peocesses and sometimes my desktop is unresponsove/frozen/hung. habe you encountered this? unchecking use display gpu for compute or unchecking the other 2 options dont make a difference.

Posted on 2019-12-30 04:26:21
OuiOuiPhoto

Hello Fred. Your problem was at import ? If yes my problem is no more at import. I have reset the prefs and it was because of W10 language. My problem It is after 10 Minutes and the Status log is

-----Starting Benchmark Run-----

Verifying Images
Getting System Information
Changing to Library view and clearing photos
Removing existing photos
Creating copies of images
Starting Import Benchmark
Importing 22MP photos
Import 500x Images - 22MP .CR2 done in 7.85 seconds
Importing 42MP photos
Import 500x Images - 42MP .ARW done in 11.1 seconds
Importing 45MP photos
Import 500x Images - 45MP .NEF done in 8.32 seconds
Sort by Filename
Setting 3 star rating for 50 images
Creating Smart Catalogs
Starting Library Scroll Benchmark
Library scroll test 5Dmkiii
Library Module Loupe Scroll - 22MP .CR2 done in 1.01 seconds
Library scroll test Sonya7Riii
Library Module Loupe Scroll - 42MP .ARW done in 0.98 seconds
Library scroll test NikonD850
Library Module Loupe Scroll - 45MP .NEF done in 0.99 seconds
Starting Develop Scroll Benchmark
Develop scroll test 5Dmkiii
Develop Module Loupe Scroll - 22MP .CR2 done in 1.4 seconds
Develop scroll test Sonya7Riii
Develop Module Loupe Scroll - 42MP .ARW done in 2.16 seconds
Develop scroll test NikonD850
Develop Module Loupe Scroll - 45MP .NEF done in 1.44 seconds
Starting Library to Develop Switch Benchmark
Library to Develop Switch 22MP

Posted on 2019-12-23 16:22:09
Fred

Yes, mine was at the import. Sorry, I misunderstood. Thanks for clarifying.

Posted on 2019-12-26 11:15:09
OuiOuiPhoto

So finally i have succeed in having the Bechmark work on a W10 French workstation

1) Install english as a W10 language
2) Modify the LUA and replace

LrTasks.execute (commandLine)

by

writestatus("Command Line Is "..commandLine)
LrTasks.execute (commandLine)

Reload the plugin in LR

Don't ask me why. I have make this modification to follow the steps and find the command that was causing problem and i got the surprise to have at the end the result. Lets say it is the magic of Christmas ;) merry Christmas to all

https://uploads.disquscdn.c...

Posted on 2019-12-24 18:07:36
Fred

If anyone's interested how the i7-4790K (stock) is doing, then here's my result.

Matt, many thanks for the benchmark! Now it's clear that an upgrade is due.

https://uploads.disquscdn.c...

Posted on 2019-12-26 11:21:50
Simon

Hey Thanks Fred - I've got an aging 4460 and it's looking like I can justify an upgrade based on your result

Posted on 2019-12-31 02:40:55

Bravo on the benchmark Matt! This is pretty awesome and finally gives a way to really test different components and settings! Pretty rad. Wanted to throw this out - ran the test with my RTX2080 at stock speeds and then ran it again with no other changes but a pretty stout overclock on the GPU... Pretty big difference on brushes and tools. My CPU is overclocked pretty heavy too but haven't ran the test with it at stock clocks. Hope this helps you guys :) GPU clocks make a big difference https://uploads.disquscdn.c...

Posted on 2019-12-26 21:07:36
FX-8350

Very interesting! I ask me, where is the limit, what is the best result in combination with moderate overclocked CPU, moderate overclocked GPU and moderate overclocked RAM for 24/7 business. The best CPU (3950x), the best GPU (2080ti) and standard RAM was tested and the result was "only" :-) 99,3 in active tasks. Is 140 possible? This would be an extremely huge step forward.

Posted on 2019-12-27 08:46:54

Yes! Well as you can see the main change came from the GPU so not sure what that would look like with a 3950x. Clock speeds are super important too for active tasks... Huge core counts really improve passive tasks but LR isn't coded well enough to take advantage of all those cores on active. I have a 2080 super overclocked so the ti probably wouldn't be a huge step forward. Maybe a Titan RTX or maybe getting into a crazy RTX Quadro card. I also have my 7820x clocked at 4.6 all cores so that makes a big diff.

Posted on 2019-12-27 15:19:15
vfeast

Hmm, makes me want to see what my 1268/121/132 score would go to if I overclocked the 2080 any... right now its at stock.

Posted on 2019-12-27 16:15:58

Do it! My 2080 is at +120 power limit, +30 core (maybe could go higher but it's already an Asus ROG OC card), +1056 memory clock. Afterburner is the easiest way to OC your card. Just open Kombustor, raise your clock til it crashes and then dial it back a bit. You may see a big change. Also in Nvidia control panel you can set your card to adaptive or Maximum performance by app.

Posted on 2019-12-27 16:29:46
FX-8350

Do it for you and for us! :-)

@PugetSystems - This Benchmark is best Lightroom innovation since beginning of Lightroom. It helps every advanced Lightroom user on entire world. A really big, important and great job.

Many THX!!!

Posted on 2019-12-28 08:51:15
PhotoProf

Hi, I'm trying to run the benchmark on my Late 2015 macbook pro, I have the correct settings for the spot removal tool, but I'm getting an " Error - unable to read command line output " message as the program starts to import the test images. Any ideas?

Posted on 2019-12-27 17:56:26

This benchmark does not currently support MacOS.

Posted on 2019-12-27 17:58:39
PhotoProf

Okay! Thanks for your quick response, I'll try it on my Windows machine!

Posted on 2019-12-27 18:03:40
ptakeuchi

Any chance you will make a version for MacOS? I'd love to test it on my Hack Pro.

Posted on 2019-12-29 23:04:12

Eventually, we definitely want a MacOS version. Many of our other benchmarks work on MacOS, but for Lightroom Classic we have to use too many external scripts for a Mac version to be feasible right now. We are hoping to work with the LR dev team in the future to add additional functionality to the built-in API that will let us easily create an maintain both Mac and Windows versions.

Posted on 2019-12-30 00:36:15
ptakeuchi

Thanks for the quick reply, Matt. I appreciate all the hard work and research that you do.

Posted on 2019-12-30 03:02:43
Artyom999

Here is what I'm getting. I'll try again one more time another day to make sure there is no user error because I have probably made a mistake somewhere (some scores are high, some are lower than people using older hardware). https://uploads.disquscdn.c...

Posted on 2019-12-28 12:37:06
Nicolas M

Hello

I have a Intel Core i7E-3930 with 32Go Ram, a 2080ti and a Samsung sata SSD

I have perform only a part of the bench (I'm still working on that) but my result look WAY different than yours.
For exemple : Import 500x Images - 45MP .NEF done in 281.65 seconds

...and when I look at some of your results : a Ryzen9 3900x could do that in less than 7seconds ??? Am In correct?

I was expecting something like x3 or x4 but it's something like x40 for this task.

Posted on 2019-12-28 21:39:12

The i7-3930 is more than 8 years old now, so those results could be 100% accurate. Many applications like Lightroom Classic are not just about the raw core count and frequency of a CPU, but other features like AVX2/AVX512 that are present in modern processors. I can't say for 100% that this is what is going on with your results since we have not tested a CPU that old in Lightroom. It could also be a problem with the ingest settings. You could open the benchmark catalog and verify that when you import an image that "generate smart previews" is not checked. It shouldn't be since that settings is saved in the catalog file, but if it is that may be what is causing the long import times.

Posted on 2020-01-02 18:55:20
Nicolas M

I have checked and no smart preview. I have bought today a Ryzen 3900 and Motherboard to replace my CPU and Motherboard and now it's 7.1s to import the 500Raw !

The active task are also now far more better. In fact I have a Nikon Z7, and now my Lightroom reply with no noticable delay :)

SO MANY thanks for your tool.
I am in France, but if I was in america, I would have been, one of your customer for sure! :)

Posted on 2020-01-02 20:30:56
computerhersteller

I wonder how consistent the results are. I get rather big differences between runs, and I get better results with GPU support off...

My test system: Core i5-7600, 16GB, GTX 1070, clean Windows 10 (full up to date, with only Lightroom and Photoshop, no anti virus or other background applications, latest GPU driver)
OS and applications are on one SSD, the benchmark folder on another SSD.

The first run with auto GPU settings:
- Overall: 536.5
- Active tasks: 61.9
- Passive: 45.4

The second run with GPU support disabled:
- Overall: 623.5
- Active: 73.2
- Passive: 51.5

Third run with auto GPU:
- Overall: 587.5
- Active: 70.3
- Passive: 47.2

LR is restarted between all runs. On the first run there was the extra step of copying the 500 images, I don't know if this is taken in account for the results.
Does GPU support have an influence on the test results? Or can't this be measured?

I've retried this on the same system, but with my normal Windows installation and have the same differences.

https://uploads.disquscdn.c... https://uploads.disquscdn.c... https://uploads.disquscdn.c...

Posted on 2019-12-29 12:32:10
Juan Guerrero

After several trials, I got the results. CPU temperature went quite high, more than with the Photoshop Bench run. First result is with a conservative motherboard configuration

https://uploads.disquscdn.c...

and second result after pushing memory to the limit (3200Mhz)

https://uploads.disquscdn.c...

Posted on 2019-12-29 15:40:35
Gabe Chapman

Hello! I'm really just getting into editing photos on my machine, and I was happy to find your Lightroom benchmark tool. My results are obviously not that good and short of buying a new machine, I'm wondering if there are any adjustments/upgrades that can be made to make the Lightroom experience better... Also, I'm not very advanced when it comes to the ins/outs of the inner workings of a PC, but I did already upgraded my SSD to the Sabrent 1TB Rocket Nvme PCIe 4.0 M.2 2280. I 'm assuming my mother board is limiting my performance as far as accessing the full potential of the M.2 SSD?

Well in any case, I look forward to your feedback and again, thanks for the tool. Hopefully I can make some system improvements until I have the funds to get a new machine.

https://uploads.disquscdn.c...

Posted on 2020-01-01 18:04:41

I think you are simply limited by your CPU to be honest. Storage speed can help a little bit, but in most cases even things like importing/exporting is going to be limited by the CPU. Your motherboard does support up to a Core i7-7700K if you wanted to improve performance, but my guess is that you would see around a 15% performance gain by doing so. If it was me, I would start saving up for a platform upgrade to a mid-range AMD Ryzen CPU if you want better performance in Lightroom Classic. You should only need a new motherboard and CPU - although you would also see a benefit from DDR4-2666+ RAM if you do go to Ryzen.

Posted on 2020-01-02 18:11:05
Paul Birkbeck

https://uploads.disquscdn.c... This is my result on my Dell XPS 15 laptop! Core i9 9980HK + 32gb ram + nvidia gtx 1650

Posted on 2020-01-01 23:53:15
Paul Birkbeck

Can anyone beat my laptop score? Dell Precision 7540 w/ i9 + nvidia quadro rtx 3000 https://uploads.disquscdn.c...

Posted on 2020-01-02 05:08:34
Rob

Yes, I can:

https://uploads.disquscdn.c...

Fast memory with AMD Ryzen is what you need!

Posted on 2020-01-03 22:23:34
dasman

I hope someone can assist. I DL'd the benchmark today and am trying to run it on LR Classic 9.1 with Camera Raw 12.1 (both updated today). The benchmark stops shortly after starting... at the import/copy screen for 5Dmkiii files. It shows and selects the images, but then just sits there... presumably waiting for the script to hit the "Import" button. I left it run overnight, just in case it was doing something in the background, but no change. I hit the cancel button on the import screen, closed LR and tried again, with identical results. Because I can't seem to figure out how to upload/post a screencap, here is a link to one: https://www.mediafire.com/view/pw9l24ut8v0gdwl/Untitled-1.jpg/file

Any ideas?

FWIW, machine is an AMD Ryzen 7 1700 w/ 32GB ram w/ (2) 500GB SSD's. OS is on the first SSD and the catalog/benchmark is on the second. Both drives have plenty of space and this was the only thing running.

Posted on 2020-01-04 01:38:08

Thanks for reporting the issue. The only thing I can think of is that it is set to "Copy" instead of "Add" like it should be. That should be a setting saved in the catalog file, but it is possible there is some preference that is overriding it. You could try backing up and resetting Lightroom Classic's preferences to default and see if it works after that.

Posted on 2020-01-04 01:53:40
dasman

Thanks... I ended up reading all the comments and saw Fred suggest the same thing to someone else. I imported a single file in another catalog to change the settings and they carried over to the benchmark catalog, allowing the benchmark to run.

https://uploads.disquscdn.c...

Going to turn on graphics card acceleration (old Nvidia GTX 970) just to see if anything improves, but otherwise I guess it's time to start thinking about a CPU upgrade. Great benchmark and it's nice to be able to make an educated decision benchmarking the software I use everyday.

Posted on 2020-01-04 03:18:55

Hmm, I guess some settings are catalog-based and others are application-wide. That definitely makes things tougher to deal with since the API doesn't have a way to change those settings.

Thanks for the heads up about the issue and the fix for it!

Posted on 2020-01-04 03:21:11
douglas crann

So...leave the mouse/keyboard alone? Running the benchmark, got to the "pano" section, 6 images were selected and progress stopped. Notice in the directions I printed that the keyboard commands were given, safe to assume that I need to intervene here?

Posted on 2020-01-08 18:14:21
douglas crann

I am having some odd issues with LR and figured the bench mark test may be a good place to start. When scrolling thru the library grid things are VERY slow. Assuming I stop, and it catches up, the images are very slow to become sharp. Next part of the problem...say I edit an image. When I go back to look at it in LR when I put it in loupe view I see all sorts of weird, splotchy artifacts. Does not matter what part of the image I look at, shadows, brights, darks...just see these odd splotches. Give it a few seconds and they clear out. Have a friend who is far more computer literate than I, he told me that he feels it is my GPU. The resolution on my monitor is 3840x2160, my friend is telling me that video card is just not fast/powerful enough for this resolution. About ready to order a new GPU, just clueless as to which one. This was the result of my benchmark test. The only settings I changed was I used the Radeon software to "overclock" the GPU. In the past when I left the Radeon Software overclocked I get several "blackouts" an hour, so I am guessing that the GPU is having troubles.... https://uploads.disquscdn.c...

Posted on 2020-01-08 18:42:46

Yea, that sure sounds like a bad GPU to me as well. You might want to try disabling GPU acceleration in Lightroom to see if that helps, but even if it does that is probably more of a bandaid than a real solution. Definitely don't overclock your GPU if it is already giving you problems, that is only going to make it worse until you get the GPU replaced (assuming that truly is the cause of your problems)

Posted on 2020-01-08 18:45:14
douglas crann

I did disable GPU acceleration in LR, did not seem liked it helped. Just so hesitant to buy another GPU, largely because I just have no clue as to what I really need. Have had a few folks tell me that because I have an AMD CPU I should stick with an AMD GPU to avoid any compatibility issues. Have had a few folks tell me that is hogwash. Hmmm....

Posted on 2020-01-08 19:32:57

There is absolutely no issue using an NVIDIA GPU with an AMD CPU, or an AMD GPU with an Intel CPU. We almost exclusively use NVIDIA even on our AMD systems because NVIDIA is much more stable (especially in terms of drivers) and in most content creation apps, NVIDIA simply gives better performance for your dollar. The only thing I would recommend doing when you install an NVIDIA GPU is to run "DDU Driver Cleaner" https://www.guru3d.com/file... to fully remove the AMD GPU driver.

Posted on 2020-01-08 19:39:20

I can't speak to the LR benchmark results specifically, but in general if you think your system (or part of it, like the video card) is struggling at a high resolution you can turn the resolution down in Windows temporarily to see if that alleviates the symptoms. Here is a guide to how to do that easily:

https://www.laptopmag.com/a...

I would suggest trying 1920x1080, as that is 1/4 the total number of pixels - and if it is indeed the video card, that should lessen its workload substantially.

Posted on 2020-01-08 18:46:46
douglas crann

I tried this the other day. Kind of odd looking images...made it a bit of a challenge to edit lol....I was on my way out the door so I only edited a few pictures. Will give it a try here.

Posted on 2020-01-08 19:34:33
douglas crann

Well....did not take long to convince me. I set the resolution to 1920-1080. Opened LR and just started scrolling. WAY more responsive.....hmmm.....

Posted on 2020-01-08 19:40:10
douglas crann

I just ran the benchmark test again, with the monitor set to the resolution that you suggested... https://uploads.disquscdn.c...

Posted on 2020-01-08 20:33:57

Looks like there was some improvement, but more than the scores I would ask if it feels any better at the lower resolution? If so, then that would indicate to me that something (very possibly the video card) is holding you back when at higher resolutions.

Posted on 2020-01-08 20:58:48
douglas crann

LR was noticeably smoother scrolling.

Posted on 2020-01-09 00:06:51

That is great to hear! Hopefully a more powerful GPU gives you a similar experience even at the higher resolution your monitor supports. Just make sure to order from somewhere with a good return policy, in case it doesn't help (if the bottleneck was somewhere else). :)

Posted on 2020-01-09 00:17:28

I can not run the test. After he asks what is the folder where the images are and import them, he simply stops and stays there, nothing else happens.

Posted on 2020-01-08 21:40:37

Razer Blade 15" i7-9750 2070 16GB with Upgraded 970 EVO Plus 1TB. Plugged in with max performance. To be honest, very disappointed in the performance. Expected a lot more. Thinking of jetting this laptop and going for the MacBook Pro 16" 5500. Sorry for the lack of a proper screenshot. Overall score: 653. Thanks Puget Systems for the fantastic benchmarking program. Hopefully there is a comparable one for the Macbook Pro 16" soon for Lightroom CC. https://uploads.disquscdn.c...

Posted on 2020-01-09 19:10:36
Travis Haughton

Any word on MacOS support for the benchmark tool?

Posted on 2020-01-12 07:17:19

No time soon, unfortunately. We have to do too much through an external app and while we could make that work on MacOS, it simply isn't our focus so we can't dedicate the time to it that it would need. We are hoping to work directly with the LR devs to add additional API hooks so that we don't need to use an external app for anything - at which point Mac support is easy to add.

Posted on 2020-01-13 18:15:26

any chance that you can perhaps have some numbers in general for the latest MacBook Pros 16" specifically i9 2.4GHz /64GB RAM /2TB /5500 8GB GPU ^_^ TIA

Posted on 2020-01-26 08:22:21

Not anytime soon unfortunately. We do want to have a Mac version at some point, but that is going to require some help on the part of the Lightroom dev team to add in some API hooks that don't currently exist.

Posted on 2020-01-27 19:23:05
Klaus

Ran the benchmark, couple of times. Score seems okay for my computer (3950X/32GB@3200 16-18-18-36 CR1/2070 Super). However, it seems to lag behind on the two active tasks 'Develop Module Loupe Scroll' and 'Library to Develop Switch'. Seen similar or slower cpus scoring around a 100, that's a 15% difference. About 20% is the threshold people tend to notice a difference, so there is actual room for improvement I would think. But how? I've tried a different graphics driver (latest Game and Studio drivers) and also ran the benchmark not full screen, set 'Prefer maximum performance' in nvidia's Control Panel and made sure LR is using the gpu, but the score is about the same.

I noticed that the gpu stays at 1-3% usage tops during the bench. The 8GB graphics memory however, is being used for almost 90% by LR once the Development module is triggered.

https://uploads.disquscdn.c...

Posted on 2020-01-12 13:48:52
FX-8350

Hi Klaus,

seems like pushing Memory to the Limits is your answer. RAM OC or for example G.Skill Trident Z Neo DDR4-3600, CL14-15-15-35

+ Maybe a new Monitor with smaller resolution for example Full HD (1920 × 1080 Pixel).

It would be also interesting to see the performance Differens between GPU activation on/off and/or on/off in editing mode.

+ A GeForce RTX 2080 Ti is my last advice.

The Lightroom performance is so confusing and hardly predictable, therefor is this Benchmark so helpful!

Have a nice Day

Posted on 2020-01-14 08:41:57
Klaus

I couldn't believe it was hardware related. Increased the Lightroom cache (5GB -> 25GB) and decreased the preview size from 3440 ->2048. Results are dramatic: https://uploads.disquscdn.c...

Posted on 2020-01-15 17:32:26

Just remember that changing from the default settings for things like preview size makes the results invalid as a benchmark. It absolutely speeds up Lightroom, but that score is no longer accurate when comparing your system to others.

Posted on 2020-01-15 17:37:13
Klaus

But everybody has its own preview size, depending on the monitor or resolution used (mine was 3440, as wide as my screen). Or does the bench sets it own preview size? In that case I haven't changed anything - I set it prior to starting the benchmark script. Anyway, if what you say is correct, the monitor and resolution used are a variable invalidating the benchmark as well.

Posted on 2020-01-15 19:00:00

Yep, you are exactly right - there are a ton of variables that are really difficult (or impossible) to control when benchmarking a real-world application like Lightroom Classic. Many we can at least detect and prompt people to change (like the spot heal brush settings), others we can temporarily change, and others (like the import and preview settings) we simply have no way to detect through the current API. I'm hoping to work with the devs to add additional API hooks for those things before this benchmark hits release (version 1.0), but that is one of the things that is keeping it as a beta for now.

Other things like display resolution is going to be impossible to control. I don't think it will actually affect things too much, but one thing I do want to do is a set of testing to determine exactly how much the display resolution and number of displays affects performance. That way, we at least know that someone with a HD monitor will see 5% (or whatever) higher performance than someone with a 4K monitor. This is just the joys of this kind of real-world benchmark. It is never going to be 100% perfect and repeatable for everyone, but even if it does end up having a 5-10% margin of error after we work all the settings stuff out, that is still 100% better than what we have right now, which is... nothing.

Posted on 2020-01-15 19:11:42
Klaus

I fully understand. But I'm glad you took the effort creating this bench: it is not exact science due to all the variables, but it is a highly useful tool to tune the Lightroom performance, and a good indicator because you can compare (as long as you compare with averages, not with the single extremes). I'll try/tune a bit more once I have some time because I'm rather curious why my active task numbers are that high of a sudden (compared to most other scores), perhaps an extra preview setting I did set or something like that. Lightroom can be rather complicated.

Posted on 2020-01-15 20:01:50
Klaus

Did some testing with a 4K and HD monitor, using version 0.85:
HD (1920):
https://uploads.disquscdn.c...

4K (3840):
https://uploads.disquscdn.c...

Did also a run with a UWA (3440) screen, score: 1221. As to be expected. Just a few percent difference, in practice unnoticeable :)
My earlier score of 1385 with the 0.80 version seems to be a fluke.

Posted on 2020-02-08 17:36:16
Myga

Hey guys, just curious if you're working on a Mac version of this bench? Mostly to see if there any differences between different operating systems.
So many times I've heard people saying LR works better on OSX - would be awesome to see the the numbers behind it if that's true. Thanks for the awesome work as always!

Posted on 2020-01-14 16:18:11

Eventually, we definitely want a MacOS version. Many of our other benchmarks work on MacOS, but for Lightroom Classic we have to use too many external scripts for a Mac version to be feasible right now. We are hoping to work with the LR dev team in the future to add additional functionality to the built-in API that will let us easily create an maintain both Mac and Windows versions.

Posted on 2020-01-14 16:20:46
Juan Guerrero

Fine tuning the system, and adding a new 25$ cooler to the CPU, improved a bit the results. https://uploads.disquscdn.c...

Posted on 2020-01-15 07:35:57
Samuel Pettersson

Just upgraded my system since Lightroom Classic was very slow on my old system(The benchmark hadn't finished after 8 hours so had to abort it). Switched to a Ryzen 5 3600 and 32gb 3200mhz ram which gave me a good active score but my passive score seems quite low compared to the results you get in your benchmarks with the Ryzen 5 3600. Any idea what could be causing this? https://uploads.disquscdn.c...

Posted on 2020-01-15 18:07:35

The import and smart preview scores are way off. My guess is that the import settings are not at the defaults. That is supposed to be set in the catalog file, but you might try resetting LR preferences (backing up if you want) and see if that corrects that.

Posted on 2020-01-15 18:10:13
Robert Taylor

Can't do anything with the catalog. There are no images and all Develop settings are greyed out. The plug-in is not listed, either.

Posted on 2020-01-19 16:46:45

You can't do anything in the develop module unless at least one photo is imported. As for the plugin, did you install it?

Posted on 2020-01-19 16:48:54
Robert Taylor

So, I should import one of my photos? The written instructions and the video don't even match, much less give full instructions. Seems like the instructions should list all the steps and list them in order.

Posted on 2020-01-19 16:59:49

No, you don't need to import a photo unless one of the Lightroom settings we list is different from the defaults we listed. Keep in mind that this benchmark is in beta currently - it is very much still in development and things like documentation is not completed.

Posted on 2020-01-20 01:01:28
Serban Dumitrescu

Thanks Matt for putting this together!! As a heavy Lightroom user, it's been tough to asses the exact impact on Lightroom performance that different hardware configurations or settings have. This is a great tool for doing so!

One problem first: on my side, the benchmark always stops at:
"Starting Panorama Benchmark
Panorama 22MP". Nothing happens after this step (left computer running overnight, morning was same).
However, if I manually Merge to Panorama the next 3 batches of photos it's able to continue and complete.
I have attempted many things, many resets of Lightroom, many installs of the benchmark, different Win10 users with same point of getting stuck.
Does the above make sense or is it related to anything obvious? Any way to "automate" the panorama part?

As far as results, a stock Ryzen 3900x with some mildly aggressive RAM timings looks as below. Will experiment more... https://uploads.disquscdn.c...

Posted on 2020-01-21 02:53:57

The Panorama issue was probably because the Lightroom window lost focus before that test. We have to send keystrokes in order to start the panorama/HDR tests, but I'm hoping we can get the LR devs to add some API hooks for that before we take the benchmark out of beta.

Posted on 2020-01-21 18:39:33
Jim Kiefer

https://uploads.disquscdn.c...

The benchmark was working great, but now getting this error when I set GPU auto or GPU full. Do not get error when GPU off or GPU basic. Resetting LR to defaults did not help. Have 2070 Super using latest gaming driver.

An internal error has occurred: [string "BenchCode.lua"]:1815: attempt to perform arithmetic on local 'activescore' (a string value)

Here are the last few lines of the log when it happens:

Active Error: Develop Module Brush Lag - 22MP .CR2 - Result: 0 - Score: 0
Develop Module Brush Lag - tonumber: 100
Develop Module Brush Lag - Result: 0.15 - Score: 100 - activescore: 1091.7 - activetotal: 13
Active Error: Develop Module Brush Lag - 45MP .NEF - Result: 0 - Score: 0

Thoughts about how to fix? Even with this small challenge, this benchmark is GREAT! Thank you, Matt!

Posted on 2020-01-25 17:05:04

One of the brush lag tests probably broke, and somehow the catches I have in place didn't trigger. To be honest, I am very close to removing that test in the next version. A brush lag test in general I definitely think is useful, but the method we are using right now doesn't seem to be as reliable as I had hoped.

Posted on 2020-01-27 19:31:32
Jim Kiefer

Oh no! The brush lag is my favorite part of the benchmark. Brush and spot tool are two biggies where lag can be really annoying because those can be invoked literally thousands of times in a session of editing. Hope it stays In, but do what you must. Thanks again for creating the benchmark!

Posted on 2020-01-27 22:35:30
Mario S.

Well done PugetSystems! But it's still a long way to go, don't give up!

Is there a chance to see in nearly Feature a GPU Benchmark for Lightroom?

Posted on 2020-02-02 13:11:50

A GPU-oriented test is definitely something we want to do at some point. To start, we just need to throw different GPUs at our current benchmark to see if anything shows a difference. With the RAW processing being GPU accelerated now, our scrolling tests may be useful already.

Going beyond that will be difficult though. The plugin API really doesn't have much in the way of features th at we would need (which given what it was made for is not surprising), so we may need to try to convince the LR team to carve out dev time to add in some API hooks.

Posted on 2020-02-02 17:29:20
Don Andrews

https://uploads.disquscdn.c... I tried running the benchmark several times. It went through the imports and I think some of the HDR stuff and then stopped in Library module displaying a picture of the ocean with rocks (sideways). It stopped at the same point three times. Current version of Lightroom Classic and all settings set as suggested. Screen shot is attached.

Posted on 2020-02-03 06:54:42
Don Andrews

Benchmark failed to complete. Tried three times. Each time, it stopped in the LIbrary Module with the same picture displayed. I have attached a snapshot of the screen. https://uploads.disquscdn.c...

Current version of Lightroom Classic (9.1), all application set to default, and I ensured all settings indicated above where set as requested.

Sincerely, Don

djandrews@gmail.com

Posted on 2020-02-03 14:49:28
Armand Tanase

Same here, I tried it few days ago and it stopped at exactly the same photo, multiple times. I tried changing other settings but no luck.

Posted on 2020-02-03 18:21:51
Don Andrews

Please let me know if you get it working.

Posted on 2020-02-03 18:24:19

Can you post (or upload and link to) the StatusLog.txt file?

Posted on 2020-02-03 18:25:59
Don Andrews

Yes. Please let me know where to find it.

Posted on 2020-02-03 18:26:47

Log files are all inside the "PugetBench.lrplugin" folder. The "LRStatusLog_____.txt" are the most important for troubleshooting this kind of thing.

Posted on 2020-02-03 18:29:00
Don Andrews

I don't see a way to attach it to this message program. How can I send it to you?

Posted on 2020-02-03 18:32:26

You can copy/paste, upload to your cloud storage of choice (Google Drive, dropbox, etc.), or email it directly to mattbach@pugetsystems.com

Posted on 2020-02-03 18:40:14
Armand Tanase

Sorry, I gave up on the thing after I spent half a day trying to make it work. Unless I have new info on settings or a new software version I don't plan to try again. I was able to get the Photoshop one running for whatever it's worth.

Posted on 2020-02-03 19:32:13
Armand Tanase

Well, I tried again with the same result. I looked into the LRStatus log and there isn't really much too report:

Verifying Images
Getting System Information
Changing to Library view and clearing photos
Removing existing photos
Creating copies of images
Starting Import Benchmark
Importing 22MP photos
Import 500x Images - 22MP .CR2 done in 7.16 seconds
Importing 42MP photos
Import 500x Images - 42MP .ARW done in 10.97 seconds
Importing 45MP photos
Import 500x Images - 45MP .NEF done in 7.2 seconds
Sort by Filename
Setting 3 star rating for 50 images
Creating Smart Catalogs
Starting Library Scroll Benchmark
Library scroll test 5Dmkiii

Posted on 2020-02-05 01:49:02
Joe Redski

Hi there,

I have tried running the plugin a number of times now, but I keep getting the error message: 'Import prep error: Error: Could not import images. Import window wouldn't open.'
Any idea why this is happening please?

Posted on 2020-02-08 16:30:29
BigDave

Same

Posted on 2020-02-14 19:01:28
Tenebs

Just tested the computer I built May 2015. Specs:

Posted on 2020-02-14 03:54:58
Tenebs

Planning on building a new computer so I tested my current old computer I built May 2015.

Specs:

CPU Intel Core i7 5960x @ 3.0 GHz; Video Card Nvidia Quadro K4200 4GB; RAM 32 GB Corsair DDR4 Dominator Platinum 2666 MHz; MB Asus X99 Deluxe; Test files run on OS Drive Kingston Digital HyperX Predator 482 GB PCIe Gen 2x4 M.2; Lightroom installed on Samsung SM951 512 GB M.2 drive.
https://uploads.disquscdn.c...

Posted on 2020-02-14 04:02:52
edispics

Looking at the raw scores for import I have noticed that the times for 500 45MP NEFs are often less than the times for 500 22MP CR2s. Do you have any idea what is going on. I have seen this in Puget benchmark scores and user scores. I tried importing files manually and I found the same thing. 45MP NEFs import at least twice as quickly as 22MP CR2s. Any idea of what is going on or why LR would import one file type so much more quickly than the other?? For example, in your bench for the 9700K, 500 22MP CR2s took 6.58 secs while 500 45 MP NEFs took 6.07 secs.

Posted on 2020-02-18 00:26:50

When dealing with RAW formats, I believe Adobe largely depends on the camera manufacturer to provide the code needed to debayer the image. That level of detail is beyond what I am very familiar with, but I imagine some camera manufacturers are better than others when it comes to how efficient their debayering code is. Even if that isn't the case and Adobe are the ones in charge of that code, I'm sure there are differences in how efficient you can debayer different raw formats.

It definitely leads to some very interesting results where the RAW format is a bigger impact on performance than the resolution of the photo like you pointed out! This seems to be more well-known in the video world where some RAW video codecs like R3D are way easier to process than something like CinemaDNG or ARRIRAW. For some reason, it doesn't seem to be as widely known in photography.

Posted on 2020-02-18 00:56:28
edispics

Thanks Matt, interesting indeed. I spend time on the DPReview forums and I have never seen this come up before - impact of RAW format on performance.!

Posted on 2020-02-18 01:02:26
Alan Haynes

A bug in PugetBench stops the benchmark. I sent an email to support about this, but here are the basics.

Shortly after the start of the "Develop Module Auto WB and Tone Benchmark," it attempts to access Plugin Extras under the File menu. I believe it is trying to start "Subscript - Set Auto" which is part of PugetBench.

Instead, it stops on one of the other entries in my Plugin Extras menu: "Search, replace, append and transfer" which is part of a different plugin of mine. Pugent bench seems to be searching for the first entry in that menu beginning with the letter s. This would work well in the lab where the only plugin installed is PugentBench. But I have nearly a dozen plugins installed and this seems to trip up PugetBench.

- Alan

Posted on 2020-02-19 20:25:46

Yea, there are a lot of little things like that that we have to work out before this benchmark comes out of beta. Unfortunately, the Lightroom Classic API is really not great for this kind of testing (more so than any of the other Adobe apps we test). We're working on it, and trying to get the LR devs involved as well, but little things like that are likely going to be a problem for quite a while.

Posted on 2020-02-19 21:32:13
Евгений Краюшкин

Hello there! The benchmark stops after I choose the import folder and all 1500 images are imported. After that just nothing happens and there are no any pop-up info windows.

LogStatus file:

Plugin Path: C:\PugetBench for Lightroom Classic
Catalog Path: C:\PugetBench for Lightroom Classic\Benchmark Catalog\Benchmark Catalog.lrcat
Catalog Name: Benchmark Catalog.lrcat

-----Starting Benchmark Run-----

Verifying Images
Getting System Information
Changing to Library view and clearing photos
Removing existing photos
Creating copies of images
Starting Import Benchmark
Importing 22MP photos

Posted on 2020-02-22 18:22:27

Is Lightroom and your OS both set to English? That is the only thing I can think of that may prevent the benchmark from running. You could also try resetting your Lightroom Classic preferences to default: https://helpx.adobe.com/lig...

Posted on 2020-02-24 16:28:32
Syd Low

Hi Matt and Puget. Just wanted to add my thanks for developing this benchmarking tool.
As I'm looking to update my Lightroom workstation (currently an iMac Pro), I ran a few of the passive elements manually and wanted to share my result.
Spec: iMac Pro 2017. 3.2Ghz 8-core, 32GB RAM, 1TB SSD
500 NEF files...
Import - 9 sec (6.5 for Ryzen 3900)
Build Smart Previews 110 sec (50 for Ryzen 3900)

And, here's a couple of times that's not in your benchmark just for interest:
Build 1:1 Preview 471 sec
Build Standard Preview @5120px: 219 sec
Build Standard Preview @2048: 19 sec

I was surprised at the huge difference with building 1:1 and 5120px previews versus Smart and lower res Standard.
Actually taught me a lot about not leaving the Standard Preview to the Auto setting which is 5120px for this model.

The $/performance looks amazing for a Ryzen 3900 powered Lightroom workstation. Think it's time to jump ship!

Posted on 2020-03-16 07:14:10

Hello,
I tried to benchmarck and I got this message: "could not import images. Incorrect number of images found (excepting 500).

Posted on 2020-03-29 12:01:59

I tried several times and was finally able to go further but at one moment it stops. And each time at the same step:
Verifying Images
Getting System Information
Changing to Library view and clearing photos
Removing existing photos
Creating copies of images
Starting Import Benchmark
Importing 22MP photos
Import 500x Images - 22MP .CR2 done in 12.81 seconds
Importing 42MP photos
Import 500x Images - 42MP .ARW done in 20.75 seconds
Importing 45MP photos
Import 500x Images - 45MP .NEF done in 20.98 seconds
Sort by Filename
Setting 3 star rating for 50 images
Creating Smart Catalogs
Starting Library Scroll Benchmark
Library scroll test 5Dmkiii
Library Module Loupe Scroll - 22MP .CR2 done in 0.93 seconds
Library scroll test Sonya7Riii
Library Module Loupe Scroll - 42MP .ARW done in 0.93 seconds
Library scroll test NikonD850
Library Module Loupe Scroll - 45MP .NEF done in 0.92 seconds
Starting Develop Scroll Benchmark
Develop scroll test 5Dmkiii
Develop Module Loupe Scroll - 22MP .CR2 done in 1.26 seconds
Develop scroll test Sonya7Riii
Develop Module Loupe Scroll - 42MP .ARW done in 2.1 seconds
Develop scroll test NikonD850
Develop Module Loupe Scroll - 45MP .NEF done in 1.53 seconds
Starting Library to Develop Switch Benchmark
Library to Develop Switch 22MP

Posted on 2020-03-29 13:03:21

Exactly the same for me. It stops after "Library to Develop Switch 22MP"

Posted on 2020-04-02 19:20:58

I tried a dozen times and always the same issue.

Posted on 2020-04-03 08:42:46
Justin Ramos

how do i stop the script? i accidentally restarted it (it took 2.75 or so hours) and it won’t stop. is there a process to kill other than Lr itself?

Posted on 2020-04-02 22:43:30

If you close Lightroom, it shouldn't have any way of continuing since it is a plugin. The only thing that might be running is a small "command.exe" utility, but unless something really went wrong, that shouldn't be running.

Posted on 2020-04-02 22:52:15
Justin Ramos

It continued running for about 5 minutes after I closed it, pulling my mouse back to my other monitor to do the cloning brush back and forth motion. It seemed like it stopped as soon as that specific brushstroke was finished, but the brushstroke took a while to run through. It kept going even after I had reopened Lightroom, with the cursor going back and forth in the middle of the screen.

Posted on 2020-04-03 00:43:00
BrianLR

I attempted to run LR Benchmark 0.85 Beta, but it failed near the end with the following error message

'An Internal error occurred ?:0 attempt to index a nil value'

Attached is a screen shot after I had pressed OK on the error message which then disappeared.The bench mark did not proceed after that.

https://uploads.disquscdn.c...

Posted on 2020-04-03 14:32:21