Puget Systems print logo

https://www.pugetsystems.com

Read this article at https://www.pugetsystems.com/guides/649
Article Thumbnail

Adobe Lightroom CC/6 Multi Core Performance

Written on May 12, 2015 by Matt Bach
Share:

Introduction

When designing a computer there are literally thousands of different hardware components to choose from. Each one will affect the performance of your system in different ways and depending on the software you will be using some will be more important than others. In the case of Adobe Lightroom CC and Lightroom 6, one of the most critical hardware components is the CPU. Even then, there are still hundreds of different CPU models to choose from and each have their own unique specifications.

So the question is: how do you know which CPU will give you the best performance in Lightroom?

Before even attempting to answer this question, it is important to understand the two most basic CPU specifications:

  1. The frequency is how many operations a single CPU core can complete in a second (how fast it is).
  2. The number of cores is how many physical cores there are within a CPU (how many operations it can run simultaneously).

This doesn't take into account the differences between CPU architectures, but in an ideal world a CPU that has the same frequency but twice the number of cores would be exactly twice as fast. Unfortunately, making software utilize multiple cores (and do so effectively) is difficult in most situations and almost impossible in others. Add in the fact that higher core count CPUs tend to have lower operating frequencies and it becomes even more difficult to ensure that you are choosing the best possible CPU for your application.

In this article, we want to find out how well Lightroom can utilize multiple cores - known as multi-threading - to help determine what type of CPU (either one with a high frequency or a high core count) will give you the best possible performance. If you want to skip over our testing procedure and individual benchmark results and simply view our conclusions, feel free to jump ahead to the conclusion section.

Test Setup

To help us determine how many cores Lightroom CC and Lightroom 6 is able to use in a variety of situations, we performed a number of actions and timed how long each took with various numbers of CPU cores. These actions were chosen based on input from a number of our customers who use Lightroom professionally. In order to have a large amount of CPU cores available to help us determine exactly how well Lightoom is able to utilize multiple cores, we used the following hardware:

Since we want to determine how many CPU cores Lightroom can effectively utilize, we are going to alter how many cores are available by setting the affinity for Lightroom in Task Manager. This way we can accurately benchmark Lightroom with anywhere from a single core to the full twenty cores available with this setup. To help with consistency - and since the benchmarks we performed ran for several days - we programmed a custom script using AutoIt to start Lightroom, set the CPU affinity, import the relevant images, perform the benchmark action three times, then clear the catalog and restart Lightroom to get ready for the next round of testing.

To analyze the data, we will be presenting our results in terms of how long it took each action to complete with the various number of CPU cores made available to Lightroom. From the results, we will also be using Amdahl's Law to estimate the parallel efficiency for the action (or how good at utilizing multiple CPU cores the software is). For more information on Amdahl's Law and how it works, check out our Estimating CPU Performance using Amdahl's Law article.

Importing Images from USB

Since importing images from a SD card, USB drive, or other external drive is often the very first thing you would do in Lightroom, we decided to start our testing there. 

Lightroom Import Benchmark

As you can see from the graph, the time it took to import our 80 RAW test images into Lightroom did not change at all based on the number of CPU cores available. However, this is the one action we tested where we do not believe the CPU was the main bottleneck. Instead, it appears that even with the fast Kingston USB 3.0 drive we used (which has a sequential read speed of about 120MB/s) the USB drive is what primarily limited performance. We are confident that importing is only single-threaded (we can determine that by watching the load % on each CPU core while importing the images), but using a CPU with a higher frequency is unlikely to improve the time it takes to import images unless you are comparing a very low-frequency CPU to a high-frequency CPU.

Exporting Images to Disk

Just like importing images, exporting them is something that any Lightroom user is going to do at some point. If you are exporting directly to a CD/DVD, the speed of the export is very likely to be limited by the speed of your optical drive. However, if you are exporting to a hard drive (either to upload to the internet, to email, or for storage) your CPU is likely going to be what limits the speed of the export.

Lightroom Export Benchmark

Unlike importing, exporting is able to use a large number of CPU cores. Also unlike importing, we found that the drive you are exporting to actually has very little impact on how long it took to complete our test export. We tested a number of drives that had sequential write speeds ranging from just 80MB/s up to 1200MB/s and the difference between the slowest drive and the fastest was at most about 2%.

Because of this, the speed of the CPU is the main limiting factor when exporting images from Lightroom. Based on Amdahl's Law we were able to estimate the parallel efficiency of exporting images to be about .97 (97%) which means that it highly benefits from having a high core count CPU. However, one thing we will point out that isn't immediately obvious in the chart above is that once we started to utilize the second CPU (so from cores 11-20) the parallel efficiency dropped from .97 (97%) to about .5 (50%) which is not very good in terms of efficiency.

In other words, exporting images in Lightroom greatly benefits from higher core count CPUs, but only when you can get that high core count from a single CPU. Once you need a second CPU the benefits drop dramatically. The reason for this is not entirely clear, but our best guess is that is has something to do with the fact that once a second CPU is involved some data needs to go across the QPI link (which is how the two CPUs "talk" to each other) which introduces an additional bottleneck. 

Convert from RAW to DNG

After you have imported images into Lightroom, you may want to covert them from RAW to DNG which - depending on the number of images you have - can take quite a while:

Lightroom Convert to DNG Benchmark

Converting to DNG definitely sees a benefit from having multiple CPU cores, but the efficiency is only about .69 (69%). Just like exporting images, the benefit from having more cores is greatly reduced once you start utilizing a second physical CPU. In this case, however, instead of just reducing the parallel efficiency we saw absolutely no benefit to having a second CPU.

Due to the lower efficiency, the best CPU for converting images to DNG is actually going to be one with between four and eight CPU cores with a relatively high (3.5GHz+) frequency.

Generate 1:1 Previews

1:1 previews are extremely useful to have if you often zoom into images while in the Library module to see if the image is in focus or not. Without 1:1 previews, when you zoom into an image it can take Lightroom 3-5 seconds or even longer to load the full-sized RAW or DNG image. With the previews already generated, however, it is much faster - often taking less than a second to load the image.

Lightroom 1:1 Previews benchmark

While the curve of this graph looks very similar to the one from converting images to DNG, the efficiency is actually a bit higher at approximately .77 (77%). The efficiency actually drops a bit once you hit about 5 CPU cores, but not by enough to really impact your choice of CPU. This parallel efficiency of .77 is one of the highest we found in Lightroom (with only exporting images being higher). It is high enough that an eight or ten core CPU would be useful, although with any fewer cores you will still want to prioritize a CPU with a high frequency.

Generate Smart Previews

While Smart previews are not 100% necessary to use in Lightroom, they are great if you are using a laptop (or even a desktop) that is often disconnected from the external storage that has your images.

Lightroom Smart Preview Benchmark

Unlike generating 1:1 previews, generating smart previews has a fairly low parallel efficiency of approximately .51 (51%). Again, we see no benefit from having a second CPU, although with this low of efficiency you are likely to hit the point of there being no measurable benefit to having that many CPU cores before you even think about needing a second CPU.

Create HDR Image

While easy to do, creating HDR photos is actually a multi-step process in Lightroom. After you have selected the photos you want to merge, you first need to wait for the preview to generate. After that, you can adjust any settings you want changed then actually create the HDR photo.

Lightroom Merge HDR Photo Benchmark

In the chart above, you can see that generating the preview (the blue line) actually takes much longer than actually creating the HDR photo (the red line). Generating the photo has a parallel efficiency of about .56 (56%) while creating the HDR photo has an efficiency of about .75 (75%). Combining these two steps (the green line) results in an overall efficiency of approximately .6 (60%) while performing the entire HDR photo creation process.

Just like the previous tests, adding a second CPU did absolutely nothing to improve performance. In fact, in the case of HDR photo creation adding a second CPU actually increased the time it took to create an HDR photo by an average of about 5%.

Create Panorama Image

Creating a panorama photo is something Adobe just added to Lightroom CC and Lightroom 6 so we were very interested to see how it performed. Just like creating an HDR photo it has two distinct steps - generating the preview and creating the new panorama image. 

Lightroom Merge Panorama Photo Benchmark

In this case, generating the preview (the blue line) is much faster than creating the panorama photo (the red line) which is exactly the opposite of what we saw with creating a HDR photo. Also, both these steps have a lower parallel efficiency than anything else we tested with the exception of importing images. Generating the preview had a parallel efficiency of about .47 (47%) while creating the photo was only about .41 (41%). Combining these two together results in an overall efficiency of only about .44 (44%) which means a CPU with a high operating frequency (as long as it has 2-3 cores) is going to give you the best performance when creating a panorama photo .

Facial Recognition

In addition to native support for creating panorama photos, the latest version of Lightroom (CC and 6) also added support for facial recognition. To test this, we timed how long it took Lightroom to detect a total of 116 faces from a collection of 32 images.

Lightroom Facial Recognition Benchmark

While there was a benefit to having two CPU cores, the benefit completely disappears after that point. In fact, we calculated the parallel efficiency to be just .2 (20%) which is pretty much nothing. In other words, the number of CPU cores you have is going to play almost no role in how long it takes Lightroom to detect faces in photos.

Conclusion

Based on our testing, there are two major conclusions we can come to regarding the multi-threading capabilities of Lightoom CC and Lightroom 6:

  1. Lightoom does not work well with multiple physical CPUs. When exporting photos, you do see a benefit from having a second physical CPU, but the efficiency drops like a rock. In most other cases you simply do not get a benefit from having multiple physical CPUs and for some actions like creating HDR and panorama photos you may actually see a small (~5%) drop in performance. This isn't as bad as we have seen in other applications like Photoshop where saw a 30-50% drop in performance with multiple CPUs, but we can safely say that multiple-CPU configurations are not ideal for Lightroom.
  2. Lightroom has a mix of single, light, and heavily threaded components. The parallel efficiency of the actions we tested ranged from nonexistent, to moderate, to fairly good. This makes it very difficult to determine a single CPU that will give you the best overall performance in Lightroom.
Action Parallel Efficiency
(higher is better - 1 is perfect)
Importing images from USB 0
Exporting images to disk .97
Convert from RAW to DNG .69
Generate 1:1 Previews .77
Generate Smart Previews .51
Create HDR image .6
Create Panorama image .44
Facial Recognition .2

Most of the actions we tested have a fairly low parallel efficiency (below .7 or 70%). They still benefit from having more cores, but in most situations a CPU with a lower core count but higher operating frequency will give you the best performance. However, exporting images - which pretty much any Lightroom user is going to do - has a very high parallel efficiency which means a CPU with a high core count will allow you to export images faster. This means that the best CPU for you is going to depend on exactly what actions in Lightroom you regularly find yourself waiting on.

If you find that the only thing you are ever waiting on is exporting images to a disk, then a CPU with a high core count will likely give you the best performance. However, the more time you spend waiting on almost anything else in Lightroom makes a lower core count, high frequency CPU better and better in terms of performance. If your workflow just consists of importing images, converting them the DNG, creating 1:1 Previews, then exporting half of them to a disk, the overall parallel efficiency (based on our testing) works out to be about .75 or 75%. Based on this, the Intel Xeon E5-1680 V3 3.2GHz Eight Core 20MB 140W CPU would be a great choice being about 6% faster than the Intel Core i7-4790K. 

Once you start adding in other tasks like creating panorama or HDR photos, the balance shifts away from a high core count CPUs towards higher frequency CPUs. If you were to import your photos, convert them to DNG, generate Smart previews, create 8 HDR and 2 panoramas photos, then export only a handful of photos (say 10% of the total you imported) the overall parallel efficiency drops to about .57 (57%). At that point, the CPU that would give you the best overall performance would actually be the Intel Core i7 4790K 4.0GHz Quad Core 8MB 88W. This CPU only costs about $400 so in addition to being the fastest CPU for that use-case will also save you quite a bit of money which you could then spend on other aspects of the computer like faster storage or more RAM.

Lightroom CPU performanceBased on our testing, there are four CPUs that we would recommend for Lightroom:

Which CPU is best for you is going to depend on a number of factors including what you do in Lightroom, what other programs you use, which chipset you want to use and how much system RAM you need. The Intel Core i7 4790K is by far the most affordable of the four CPUs and while it may be slower for exporting photos than the other CPUs for almost anything else in Lightroom it will actually be the fastest. However, the chipset it uses can only have at most 32GB of system RAM. This is more than enough for Lightroom, but if you use other programs that need more RAM it may not be the best choice for you. 

If you need more RAM, one of the other CPUs may be a better option. The Xeon E5-1680 V3 is a good choice if you hate waiting for exports to finish while the Xeon E5-1650 V3 or Core i7-5930K (which have identical performance) are both very balanced options. 

Tags: Lightroom, Hardware, Multi-threading
Adrar

thank you very much for this valuable test! i was just about replacing my 2012 i7-3770 imac with a custom 12-core 2009 Mac Pro. Mostly doing timelapses using lightroom cc and lrtimelapse 4. My workflow in lightroom includes everything from importing, previews, edits and exporting to about equal amounts. Guess i will keep my machine for a little longer :)

Posted on 2015-05-20 05:14:34
Kotlos Kotlos

Excellent analysis!

Could you do another one with the GPU?

Posted on 2015-06-11 15:12:29

We've done a bunch of articles about GPU acceleration in the past for Photoshop, Premiere Pro, AutoCAD, and a few others (https://www.pugetsystems.co..., and I wanted to do one for Lightroom as well. Unfortunately, I couldn't figure out a good way to reliably benchmark video card performance in Lightroom. The only thing that uses the GPU right now is performing image editing actions in the Develop module. The problem is that when you do something like change the contrast of an image with a mask, all that happens with a better GPU is that it is a bit smoother. It isn't like applying a large effect in Photoshop where you can actually time how long it takes the action to complete.

I might be able to figure out something at some point, but another factor is that right now I've seen a ton of reports of performance actually being worse sometimes with GPU acceleration on than with it off. So it sounds like this first implementation of GPU acceleration in Lightroom is a bit buggy. I'm sure Adobe will resolve the reported issues, but right now I'm not 100% sure that any benchmarks we performed for the GPU would be completely accurate.

If I (or someone else on the internet) figure out a reliable way to benchmark the GPU though, I'm sure we will do a Lightroom GPU Acceleration article.

Posted on 2015-06-11 17:54:14
Kotlos Kotlos

Yes people report all kinds of things, that is why a proper test like yours would make sense. Now I know that is hard to benchmark the smoothness of a slider, but even a simple write-up of your subjective experience with different CPUs/GPUs/Resolutions would be of great value!

Posted on 2015-06-12 03:57:11
Stephen Partington

I know Right now for me GPU acceleration is a negative impact, I have a Nvidia 750 GTX and i7-4790K and it overal is much better with the GPU acceleration off. I need to go back and turn of the Intel GPU and disconnect that screen and see if there is a chance with just the one Display device.

Posted on 2016-02-25 14:49:36

For what it is worth, I've heard from at least some Lightroom users that the latest versions (the ones which include GPU acceleration) actually perform slower for them than previous versions. You lose out on some other features as well, of course, but if they aren't things you need then rolling back to an older version of Lightroom may help if you are having performance issues. We found similar problems with other Adobe software as well, specifically After Effects.

Posted on 2016-02-25 16:45:35
Stephen Partington

Right now turning off GPu acceleration and Letting my 4790 do the work is acceptable.

Posted on 2016-02-25 16:48:22

Question: What if you split up these workloads into multiple clients running concurrently each performing the same operation (such as generating 1:1 previews) on their own set of images?

Let's say you've 2,000 images you need to generate 1:1 previews for and a 2x12 cpu/core configuration, and the practical limit appears to be around 3-4 cores for such an operation; 4 cores per client (either automatically managed by the OS or manually via core affinity configuration) and each does its own batch of ~333 images. Would this not be completed faster than having a single client do all 2k images' previews alone?

Or what about setting up concurrent 1:1 preview, standard preview and smart preview processes in one client?

Posted on 2015-06-19 20:37:00
Pesto DaCat

Why doesn't the software do that for you AUTOMATICALLY!?!? Why should I have to manually do this...

Posted on 2015-11-06 14:40:54
mclaren777

This article is fantastic and I look forward to reading more Lightroom tests, especially once GPU functionality improves.

Posted on 2015-06-20 04:00:41

Thank you very much for this testing! I was wondering whether to go for a 6 core CPU next upgrade cycle and you've given me a lot of great data to ponder.

Posted on 2015-06-30 23:47:29

Not a problem, glad we could help!

Posted on 2015-06-30 23:50:28
Jeff

Just wanted to say THANK YOU for the great write-up and a very helpful Google Doc! I've been searching for a
data-driven analysis of Lightroom performance relative to CPU clock
speeds and core counts since Lightroom 4 was released (due to the more
computationally intensive Process Version 2012) and was almost ready to attempt to try doing an analysis on my own in pursuit of better LR performance. This article has not
only properly analyzed Lightroom's threaded performance, you even
calculated the parallel efficiency which will help extrapolate the data
beyond what we have today in terms of hardware. This was the final piece
to help me decide to invest in a 5820K (assuming OC @ 4.3-4.4GHz)
instead of a 4790K (est. OC @ 4.7GHz) which should provide similar
performance at those speeds now and (hopefully) provide increasing performance as the software
is optimized for more cores in the future across all functions.

Just wanted to point out, as a friendly observation, that I
thought it was a bit odd to bottleneck the import test with a USB 3.0
drive rather than the same internal drive used in the export test (sounds like you have a nice and fast PCIe or M.2 SSD). Also, since export is a CPU intensive task, and the exported files (~23MB RAW converted to ~8MB JPEG) should either fit in the HD's cache or be written fairly quickly/easily to the drive itself, testing various drives during import would probably yield more varied results and insight.

Curious, was Hyperthreading enabled when running the tests? In the past, I had read that HT did not benefit LR (v.4 at the time) at all, a big part of why I built a 2500K system over a 2600K system. Curious how that helps performance, though all of the processors recommended support HT.

While, LR 6/CC now seems to utilize more cores for a single export (supposedly it kicks off 3 exports behind-the-scenes), Theo's comment about trying to run multiple processes (exports, 1:1 preview renders, etc.) would be interesting to try, especially since it seems that LR doesn't benefit from a 2nd CPU at all. With independent data sets that don't require cross-CPU communication, perhaps the additional cores or 2nd CPU would show higher gains.

Posted on 2015-07-04 07:31:15

I'm glad this was helpful! There is certainly more we could test and what you pointed out would make a great "part 2" to this article. For the importing, we limited to USB drives simply because one of the questions we get asked a lot is whether a faster CPU will improve the time it takes to import photos from their flash cards. We have done a bit more testing which you can find in the "hardware recommendations" section for our Lightroom Recommended Systems ( https://www.pugetsystems.co... ), although that was more to determine how much the speed of the USB drive affected import times.

I'll make sure I keep in mind hyperthreading comparison, importing from a fast internal HD, and running simultaneous actions when we have the time to expand on this article. Thanks for the feedback, it really helps us tailor future articles!

Posted on 2015-07-06 18:44:00
Wim Pollet

I'm also very curious about turning Hyperthreading off and also about starting 2 export sessions simultaneously.

Posted on 2015-07-06 21:22:06
Jeff

Just an update now that I've built my 5820K PC OC'd to 4.4GHz. I've found that 1:1 Previews do not use more than 25% of my available logical cores - this is still true if I kick off 2 1:1 Preview renders, each rendering 1/2 the gallery. It simply does not use more threads. This is frustrating considering LR's insistence on re-generating the previews every time an image is edited.

Wim: For LR6 (and I'm fairly sure LR5, though I didn't reinstall it on the new PC), the export process is much more efficient at using all available CPU cores. As long as you're not running a dual-CPU setup (which I have no information or experience with), kicking off one export will use near 100% of your CPU. Kicking off two exports actually seemed to slow down the process, though I didn't time it. LR6 also uses up a ton of RAM during export, and since it tries to export 4 images at once, I've seen LR6 use up to 5-6GB of RAM when exporting.

Posted on 2015-08-20 21:57:52
Mark

I would add: exporting 2000+ images from Raw(Raw being mixed files Canon 5DMKIII, 1DX, 5DmkII) to jpg (quality 74 @ full file size)....
My current PC will utilize 7-8 gig out of 16gig on board RAM.

Posted on 2015-10-29 00:03:41
John Murphy

Does exporting one very large file benefit from more than 4 cores the way exporting multiple files does?

Posted on 2015-08-07 14:20:24

I don't have a really high core count machine available to test that right now, but I did do a quick test with a 4 core i7 6700K. With just four cores, it looks like the parallel efficiency is about .4 or 40%. That is really, really low so I suspect for single image exports you will want a high frequency CPU with ~ 4 cores.

I tested this just by resizing a single 18MP RAW image to 500MP (500 PPI) JPG during the export though, so it may be different if you are starting out with a very large image rather than resizing.

Posted on 2015-08-07 18:00:07
John Murphy

I'm just now trying to figure out whether to get a 6700k or a 5820k... I see advantages in both. Most of my slowdowns in Lightroom are with previews, and exporting. Most of my slowdowns with photoshop are load and saves and applying plugins within smart objects. My images are typically 1G or higher with all of the layers and I use NIK and other plug ins frequently. Since you've used a 6700k and 5820k, do you have any guidance for picking one vs the other for PS/Lightroom for large image photography workflows? I see advantages with the 5820k in more cores and being able to put the storage directly on the PCI lanes connected to the CPU instead of routing through the pch.. the 6700k has more ghz. choices choices

Posted on 2015-08-07 19:23:08

Between those two CPUs, I would recommend the 6700K. We just finished testing the 6700K the other day in Lightroom and Photoshop and it has a pretty decent performance bump compared to the 4790K ( https://www.pugetsystems.co... ). That combined with the higher frequency should make the 6700K noticeably faster than the 5820K for most tasks - even for tasks like exporting multiple images in Lightroom that are well threaded.

I don't think the difference in PCI-E storage speed is going to be very significant. We run PCI-E storage over the PCH quite often and have never seen a real measurable drop in performance. Especially since the Z170 chipset now has PCI-E 3.0 (Z97 was just PCI-E 2.0) I wouldn't worry about CPU vs PCH PCI-E lanes.

The main advantage the 5820K has over the 6700K is the amount of RAM the platform can use. With the 5820K you can get up to 256GB of RAM if you really need it, but the 6700K can only do 32GB today (once 16GB DDR4 sticks come out it will be able to do 64GB though). Photoshop usually needs more RAM than Lightroom so that is going to be what determines how much RAM you need. From what you described, 32GB of RAM should just about right ( https://www.pugetsystems.co... ) so I don't really see a need for the higher RAM capacity of the 5820K. The nice thing is that if you find you need more RAM in a year or two, you can upgrade to 64GB just by swapping out the RAM with 16GB sticks.

All that to say that unless you need more than 32GB of RAM today (not in 2-3 months, but TODAY) I would recommend the 6700K over the 5820K. It should be faster for most of the tasks you are performing, a hair less expensive, and be quieter than the 5820K. Basically, this is the configuration I would recommend for you: https://www.pugetsystems.co... . I just made a guess on the storage options, but for the core system I think a 6700K, 32GB of RAM, GTX 970 (for the 4GB of VRAM which will be nice in Photoshop for large images), and an Intel 750 NVMe drive would be simply the fastest possible configuration you could buy today for what you are doing.

Posted on 2015-08-07 19:49:28
John Murphy

Interesting points. Assuming I overclock the 5820k, it seems I might get the best of both worlds? (5820k at 4.3-4.5 vs. 6700k at 4.6-4.7). Seems like the gap might narrow at that point? Do you have 6700k's available now?

Posted on 2015-08-07 20:35:10

Overclocking might narrow the gap, but I don't know if it would be enough. The last 5820K we overclocked we got to 4.3GHz while maintaining stability, the 6700K we get to about 4.6. .3MHz may not seem like much, but with the newer architecture I still think it is more important than having two more cores for Photoshop/Lightroom.

We do have plenty of 6700K CPUs and Z170 motherboards in stock at the moment. Right now we list them on our Serenity https://www.pugetsystems.co... , Serenity Pro https://www.pugetsystems.co... , Deluge https://www.pugetsystems.co... , and custom systems.

Edit: Slight correction - we have the Asus Z170-A (used in Serenity Pro and Deluge), but the Maximus VIII GENE we use in Serenity is still in shortage. So if you want the system quickly you will want to go with either a Serenity Pro or Deluge.

Posted on 2015-08-07 20:49:07
John Murphy

I'd buy the processor from you if you'd sell it, but I have a asus z170 hero unopened sitting at home, and I'm a build it myself kinda guy:) was going to return that MB and get a x99, but you may have talked me out of it!

Posted on 2015-08-07 20:59:58

Unfortunately, supply is tight enough right now that we are holding all our Skylake-S CPUs for full system builds. We should have just enough to last us until supply improves, but that means that we are not in a position to sell them as individual parts. Sorry! Hopefully supply will improve soon and you can get your hands on a CPU.

Posted on 2015-08-07 21:22:45
Jeff

Matt, love how you're keeping the discussion alive after all this time, I'm sure we all appreciate it.

Are you observing a higher IPC improvement than Anandtech's review suggests (~6% from Haswell to Skylake)? It just feels like even an avg. 10% IPC improvement (lower for most things based on your Haswell vs Skylake article) + the ~7% clock speed improvement over the 5820K @4.3GHz would still lose out w/ the 50% increase in core count on the 5820K (assuming the software used is fully optimized to use all available cores/threads, which LR's Export is, though the rest of LR isn't). For most single-threaded applications, you're absolutely right, Skylake is the clear winner. It's also the clear winner for anyone who doesn't want to dabble in overclocking, at a huge 30%+ improvement.

When comparing my 4.5GHz 2500K to my 4.4GHz 5820K (on air!) I've observed a 2x increase in Export speed, but 1:1 Preview rendering was only 28% faster on a single photo, and only 15-16% faster on a batch of 182 RAW CR2 files (approx. the IPC improvement from Sandy to Haswell). From my observations, Lightroom is just really poorly optimized software. It's fine to say certain processes are limited to 1-2 cores with diminishing returns but batch processes should be able to kick off jobs for multiple images at once to utilize all resources.

In the end, my decision to go w/ the 5820K was due to the fact that I got mine brand new for over $100 off retail, making it significantly cheaper than a 4790K/6700K, and the hopes that having a 6-core will last longer in the long run. Not to undermine anything you said since it's quite valid and much easier to achieve guaranteed performance but I think the 6700K should give you the upper hand for rendering 1:1 Previews (which I usually walk away from) but exporting would greatly benefit from more cores as long as clock speeds are w/in 0.2-0.3GHz. Overall time savings would then be w/ the 5820K.

Question: Did you observe 1:1 Preview rendering to use all 4 cores / 8 threads for either the 4790K or 6700K? The ~2% improvement is well within margin of error and even if it isn't, doesn't seem like a very nice improvement.

Keep up the great work!

Posted on 2015-10-29 01:49:30
Toh Gouttenoire

Thanks alot I was looking exactly for that info! I am now waiting for BH to receive their i7 6700...

Posted on 2015-09-17 14:49:51
Lori

Here you see 50 cpu test in lightroom
http://www.pcgameshardware....

Posted on 2016-06-09 07:42:08
AnonymousBillionaire

Hey Matt. I see most of the charts even out after about 7 "cpu cores" listed on the chart. Is there a difference here between cores and threads? I currently have a 6600k with 2 cores and 4 threads. If I get a 6700k with 4 cores and 7 threads will I be hitting that "7 cpu cores" on the chart where things even out?

Posted on 2016-09-22 22:58:05

The charts in this article are all for actual cores, not individual threads. Hyperthreading (where each core has two threads) is great for multitasking, but when you have a CPU decently loaded with a single application, spliiting up one core into two threads tends to not make much of a performance difference. So while the CPU we used for testing does support Hyperthreading, we always report results in terms of how many cores are being used.

Also, your i5-6600K should have four cores and four threads (since it does not support Hyperthreading). Upgrading to a i7-6700K would give the same number of cores, but technically 8 threads since it does support Hyperthreading. I don't think Hyperthreading itself will give you much of a performance increase in Lightroom, but the i7-6700K is about .3-.5 GHz faster than your i7-6600K which should make it faster. Exactly how much is really hard to saw without actually benchmarking both CPUs in Lightroom side-by-side, but I would estimate you would see about a 10% improvement in general Lightroom performance by upgrading.

Posted on 2016-09-22 23:16:09
elektronaut

eventually a very nice article and with the sers' comments and your support is staying it up-to-date!!
quick opinion needed. after i read your previous comment.i m going for a new system and i am between the i5 6600k ( i may not overclock it ) and the i7 6700nonK.
do you think the price difference of 70-100 euros worth the gain in performance for lightroom use and bit of photoshop? i am an amateur photogropher ( just a hobby of mine ) bringing in approx. ~ 50GB of photos annually.
and would you consider the new kaby lake cpu's ( i5 7500 or i5 7600 ) ?

thanx a lot for your time !!

Posted on 2017-01-10 22:57:19
hugemon

Just upgraded from 3930k (6 core) to 5960x(8 core). While speed of each core would be about the same (oc'd to 4.8GHz Sandy bridge architecture vs. 4.5GHz Haswell architecture) Lightroom feels much faster. Of course time to complete each task not that much different, (would be only about 33% faster even if we assume 100% efficiency) usability while doing something in the background is hugely improved. (ie: browsing thru photos while they are still being converted to dng...)

In real life usage users often use lightroom while it is still doing something else. I think (feel) more cores helps much more than the synthetic(?) tests in this article show.

Posted on 2015-08-30 10:51:50
Stefano

Interesting test, thanks.

For me the biggest bugbear in LR was always the laggy performance when using the brush and spot removal tools in the develop module. It seems that LR doesn't very effectively use more than 4 cores anyway and I've found that upgrading to a significantly overclocked (5.1ghz) i5 has given it the headroom to process these changes in real time with very little lag. Of course, video rendering etc. will benefit from more cores, but I don't do that so for me higher clock speed and less cores seems to be the sweet spot.

Posted on 2015-09-29 11:29:51
Diko Jelev

Now that is what interests me most. Brush and Spot tools are the biggest bottleneck in my workflow. It is highly interesting that CPU Speed is more important than number of cores. Even worse why there's still no true GPU utilization of this sort of computation...

Posted on 2016-06-02 20:39:50
Borconi Szedressy Emil

Hi.

Thank you for the nice and detailed test.

I'm a funny Lightroom user. I'm not doing any of the RAW to JPEG or panorama or HDR stuff but I do export huge amount of JPEGS. My work process looks at this: import 20k+ images (jpg), apply preset, export them. Currently I have 2 computers a laptop with i7-4712QM and a desktop AMD FX-8350. I have the feeling that my laptop exports faster than the pc. I'm thinking to get 1 or 2 more PC's, and based on your review and the spreadsheet you provided and my work style it looks to me that the using an intel Atom with 8 cores (C2750) should more or less have the same performance as an i7-4790K. Am I being naive? once again I stress out, I do not generate 1:1 preview, do not do any fancy stuff with the images, only import, apply preset, export. I always use SSD for the catalog location drive, to make sure import speed is as good as possible, the killer is the export time. Thanks for the advice.

Posted on 2015-10-11 12:07:26

I can pretty much guarantee you that an i7-4790K will be much, much faster than an Intel Atom. The Atoms are weird because they look great on paper (8 core, 2.4GHz looks amazing!), but they are only 20W CPUs. In order to get that low of wattage, they have to cut out a ton of features that have a huge impact on performance. I don't know what those features are off the top of my head, but I did find a small set of benchmarks for the Atom C2758 (which is similar to the C2750): http://www.silentpcreview.c... . Notice that they are comparing it to an i3-3217U and A8-5600K and those processors are overall much faster than the Atom even though in terms of core count and frequency they are much lower.

One last recommendation I would make is to look at a i7-6700K over a i7-4790K if you can. The 6700K just launched a month or so ago and should be somewhere around 10% faster for the same price. You will have to spend a bit more on the RAM (since DDR4 is a bit more expensive than DDR3) but I think it would be well worth it.

Posted on 2015-10-12 17:01:53
Borconi Szedressy Emil

Hi Matt.

Thank you for the input, somehow I had the feeling this will be the case. I will have a look at the recommended CPU. Today I did a small test, just to compare Intel VS AMD (i7-4712QM 8GB DDR3 VS FX-8350 4GB DDR3). Set the same files and started the export and waited for exactly 4 minutes. the AMD managed to export 104 files, from which 2 where fails (just a white image, happens quite a few times) and the Intel managed to export 120 files, with none fails. I guess the fails might have to do something with the RAM. Thank you for your inputs.

Posted on 2015-10-12 20:02:38
bbourizk

Hi
Thanks for the great article.
Im looking at updating my editing pc.
Currently running intel 2790 i7 ssd drive in raid as well as 16gb ram and a recent addition nvidia gtx970

When skylake came out I was set on the 6970k.
Then someone recommended the haswell 5820.

I mainly use light room to import and adjust images and then send to photoshop for my main work.

Im finding that when applying actions and plugins the wait is too long. Some actions take a couple of minutes converting to 8bit mode speeds it up a bit but a process id rather not do.

With the new build I will be using the newly released Samsung 950 pro pcie m.2 ssd and using my current ssd as scratch disk for photoshop. 32gb ram minimum.

Since PS likes a lot orf ram im leaning towards the 5820k but not sure now.

Any advice would be appreciated.

Posted on 2015-10-20 15:26:14
Jeff

Curious if there's a consensus on whether the faster SSD should be used for PS scratch disk rather than as an OS drive. Once everything is loaded, including the PSD file, that should all reside in RAM and the scratch disk would be used for PS operations right?

Posted on 2015-10-29 01:55:42
bbourizk

The plan is to buy a 512gb M.2 Drive for OS. With all programs installed it should be no more than 60gb of space used. it will have a Windows 10, Lightroom, capture one pro, photoshop and a couple of other programs for panorama and photo stacking.

On importing photos I will put them on that drive to work with also backed up to three other locations, two internal and one external. I want to access the photos as fast as possible and I'm thinking having them on the fastest drive would help. The scratch disk will be for PS to store temp files etc.

Some of the motherboards I've been looking at have two m.2 slots, so another option is to get a 256gb drive for OS and a 512gb for photo imports to work on the files. These drives support up to 32gb transfer speeds so theoretically accessing data should be faster. That's my thinking but I could be all wrong on this.

Posted on 2015-10-30 03:02:15

Practically speaking, current M.2 drives max out around 2.5GB/s for reads, and around half that for writes (with many drives being slower than that). We've also seen that most M.2 drives get really hot during extended file transfers, which can limit performance once the drives hit a high enough temperature to start throttling. Just something to be aware of :)

Posted on 2015-10-30 04:34:01
bbourizk

Hey mate.
Thanks for the reply.
Those speeds are still better than my current ssd. So I can live with that.

The samsung drives I'm looking at supposedly fixed the overheating issues that the sm951 I was originally looking at. Although I haven't seen hard proof of that. Basically I just want to access the files as quick as possible but maybe there's a better option in not aware of?
Regards
Bud

Posted on 2015-10-30 05:37:34

Are you planning on the Samsung 950 Pro series M.2 drives? If so, they don't get as hot as the SM951... but they still get hot enough to throttle performance: https://www.pugetsystems.co...

Posted on 2015-10-30 15:35:51
bbourizk

Thanks for the link William
Even with the heating issues I think that's the drive for me. I've all but made up my mind to go with the haswell E 5820. Unless broadwell E launches in the next few weeks.

Posted on 2015-10-31 05:35:37
Mark

Really appreciate the article. Love the spreadsheet too.
Question .... any thoughts to RAM amount? 8gig vs 16gig vs 32gig when exporting?

Variable:
1.500 image to export vs 3000+.
(I edit indian weddings... kind of on the large amount)
2. DDR3 vs DDR4

Posted on 2015-10-29 00:05:41
Jeff

DDR3 vs DDR4 shouldn't show much improvement (or none at all) for LR Exports. As long as you observe having unused RAM during your export, you should be fine. I don't think I've seen LR use up over 6-7 GB of RAM when exporting a full wedding (RAW to JPG). Having more RAM would mainly be useful if you have other programs open at the same time (PS uses a ton for a PSD with tons of layers for example, or in my case, I have a billion tabs open in Firefox & Chrome) so your computer doesn't start kicking things out of RAM which would slow down your system significantly.

Posted on 2015-10-29 01:13:05
Pesto DaCat

As figured.. LR is a dog no matter how many cpu's you have.... sad!

Posted on 2015-11-06 14:40:05
Patrick van Baarle

I am planning on building a 5820K rig ( and over clocking it) as we speak this is my current set up http://de.pcpartpicker.com/... (first build ever so I am a newbie!)

The main concern I have as a photographer is how to manage all this data (4tb per year) So I want to upgrade my systems so that I can stick to the "3-2-1" rule back-up rule
http://windowsitpro.com/blo.... For me every "extra" second Lightroom takes to load a picture ends up to quite a lot of time wasted.

Therefore thinking of using a super fast SSD Samsung 950 but will it make a "real" difference? I read somewhere that a couple of SSD
in raid 0 will be similar and also that more latency is expected in real world use as software
(unlike benchmarks) like light-room are not being written with NVMe in mind. Is it worth upgrading to a NVMe card for lightroom? Or is it better to wait a bit build it without the M2
storage and wait until is a bit more mature? So I want to stick to the "3-2-1" rule back-up rule
http://windowsitpro.com/blo....

What would be the best set up for Primary storage when I was done working on photo's on the SSD? Put them on a NAS? I am worried about bottlenecks between the NAS and the PC if I wanted to retouch some pictures. Is there something like a thunderbolt 3 NAS because a standard WIFI conection will not cut it for me. I was thinking of installing 2 Hitachi Deskstar 4TB 7200RPM perhaps in Raid 6 directly on my PC. My Final solution would be to use blue ray the rest of the data to off-site. But perhaps its best to have a couple of TB of SSD in my pc and a big NAS?

Regarding
the 950 will the MSI x99A gaming 7 motherboard support it well? I went for the X99A because of USB 3.1. Would a 5930 with 40 PCI lanes be better for M2 based internal high speed storage, by adding PCIe 3.0 x4 storage on the 3 remaining PCIe slots be better future proofing? .

TL:DR, I think LR is about I/O, Where are my bottlenecks?

Posted on 2015-11-07 21:21:24

When actually working in Lightroom, we found that your storage drive makes almost no difference to performance. We have some information on storage performance here ( https://www.pugetsystems.co... ) where we tested export and import times. When exporting, you can get ~2% faster exporting with an SSD, but there is no need to go up to a NVMe drive or RAID. For importing the speed of your memory card makes a big difference, but the internal drive doesn't. We didn't test other LR tasks, but those should be the most storage intensive so if they don't show a difference nothing else should either.Basically, with current hardware your bottleneck in Lightroom is almost always going to be the CPU - not the internal storage drive.

Now, moving around the files before/after working with them in Lightroom will definitely depend on the speed of your storage since that is a straight file copy. If you need as fast of performance as possible, I would recommend looking at a Thunderbolt external storage or possibly one that uses USB 3.1. I'm not too familiar with external storage myself, but either of those will definitely be faster than a NAS (although they may not be quite as convenient). If you do go with a NAS, definitely use a gigabit wired connection since you will see much, much faster transfer speeds than using WiFi.

Posted on 2015-11-08 17:26:47
Pim

Great article guys! It seems to be the place to be with great expertise on these subjects.

I currently use a i7 2600k@4,4Ghz, 32GB ram combined with a new Samsung 950 pro m2 as OS disk and 4x 4TB raid 0 as my work discs. Do you guys think there is a noticeable difference in upgrading to a 6700i, 5960x or skylake Extreme? Or will it be only noticeable in benchmarks? I work with 50MP (Canon 5DsR) files now and really need some extra speed and 64GB's of ram (32GB ram is pretty quickly filled when opening a few images from lightroom to PS).

Posted on 2015-11-30 20:13:57

Upgrading to a 6700K would definitely give you a performance boost and is what I would recommend. It is hard to know exactly how much faster, but just rough guessing I would expect anywhere from a 15-25% performance improvement (possibly more) over your current 2600K. Also, just recently 16GB DDR4 sticks came out so you can now pair 64GB of RAM with the 6700K. If you think you need even more RAM, I would recommend a Core i7 5930K (Or Xeon E5-1650 V3 which is essentially the same thing) although that will be slower for most Lightroom and Photoshop tasks than a Core i7 6700K.

Posted on 2015-11-30 20:28:48
Stoney

Thank you for this extremly useful post!

I'm building a lightroom system, but photography is "just" a hobby so my budget is very limited.
What would you recommend if I had to choose between a Intel Core i3-6320 or a Intel Core i5-6500?

Posted on 2015-12-02 11:52:09
Stoney

FYI: Mostly I do edits and exporting, sometimes I use presets

Posted on 2015-12-02 11:53:46

Definitely the Core i5! The i3 series are dual-core chips (with Hyperthreading) but the i5 are quad-core, and for several parts of Lightroom those extra cores will be worthwhile. If you can swing it, though, one of the slightly faster i5 models (like the 6600K) would be even better, without costing a lot more.

Posted on 2015-12-02 16:25:38
LewisL

Great articles on LR and PS builds.
The only issue I have is the unpleasant lag when browsing through small to medium size thumbnails in LR.
Using an older machine with an I5 but with Samsung SSD for programs and system.
If I build a new PC fixing that lag would certainly be the one issue I'd want to fix.
Any hints?

TIA,

L

Posted on 2015-12-25 02:52:25

Unfortunately, that isn't something we tested (although we do have it on our list of things to look at when we re-do this testing in the future). I can, however, make some educated guesses based on everything we have tested.

If I understand exactly what you are asking, the only way I know of to completely get rid of that lag is it generate 1:1 previews ahead of time for all the images. This way, Lightroom doesn't have to generate the previews on the fly which really cuts down on lag. Generating those previews ahead of time can take a while, however, so it is really a matter of waiting ahead of time, or waiting while you are browsing the images.

Most likely, the only thing that will help with the lag on the fly is a higher frequency CPU that is the latest architecture. I don't know what i5 you have, but if you are two or more generations old, a new PC with a Core i7 6700K should be ~15-20% faster for browsing through thumbnails than what you have right now. Maybe a bit faster if you have a lower-end i5 or if it is more than a generation or two old, maybe a bit slower if the CPU is a higher-end i5 and very new.

Posted on 2015-12-28 20:21:51
LewisL

thanks, I'll try generating 1:1 previews on a selected folder and seeing how that effects browsing.

Posted on 2015-12-30 02:30:21
Adel Mandani

Thank you

Posted on 2016-02-21 02:21:39
Adel Mandani

Thank you

Posted on 2016-02-25 09:54:07
Stephen Partington

So this was a great read, However It seems there is GPU acceleration in Lightroom now as well. do you know how that would affect some of this testing - https://helpx.adobe.com/lig...

Posted on 2016-02-25 14:47:33

Edit: I just realized you already commented on that thread so I didn't really need to link you to it! I'm leaving this anyways since it is a question we get somewhat often and your comment is currently at the top of the comment list so it will improve visibility.

Yes - this article is solely about CPU performance. We got a similar comment to yours a few months ago, but it is pretty buried so rather than repeating everything, here is the link to the thread: https://www.pugetsystems.co...

Basically, GPU acceleration is only in the develop module and we haven't figured out a great, consistent way to benchmark that module yet. In addition, there have a been a ton of reports about lower performance with GPU acceleration enabled than with it disabled so it appears to still be pretty rough. The good news is that everything in this article is not in that module so it does not support GPU acceleration - which means that the GPU wouldn't impact any of these tests.

Posted on 2016-02-25 17:41:12
Ryan B

I have an i5 4690k with 970 and it runs fast in LR. I had to turn off gpu acceleration because it was buggy. I also tried my friends computer and he has a 980ti which is also buggy too. Now at work I have an Imac 5k that is an i7 and that works flawless and very responsive with gpu acceleration turned on.

My question is, are the problems within the Nvidia cards? The imac running only 2gb AMD video card is smooth and fast. Why?

Posted on 2016-03-07 22:08:26

I know of folks with AMD video cards on PCs who are also having poor GPU acceleration performance in Lightroom. I would suspect that what you are seeing may come down to differences between the Mac and PC versions, and how Adobe has coded them / implemented GPU acceleration. Only someone at Adobe could answer that completely, though.

Posted on 2016-03-07 22:12:02
Ryan B

Thanks for quick reply!

I spoke to someone at adobe and they say it is my i5 4690k. My friend has an i7 4790k with similar performance in LR. So I know it's not only being an i5 chip.

I was going to upgrade to an 6700k chip, but I am sure it will not change things except for faster exporting times.

It just bothers me that they state gpu acceleration is faster and smoother, but in reality it is not.

My LR runs quick now with acceleration off, but I want it to be just as blazing fast with gpu acceleration like my imac 5k at work. I don't understand how the Macs can be that much better using an AMD card when in fact adobe recommends nvidia?

I'm confused. 😀

Posted on 2016-03-07 22:38:28
Ryan B

Also when I turn off gpu acceleration on the imac, the performance is similar to my i5 pc build. Once when gpu acceleration is turned on in both systems, the imac is blazing fast and smooth. My 4690k system is buggy. You think it could be the 970?

Posted on 2016-03-07 22:42:32

You aren't alone - I think it is more about Lightroom (on the PC) being buggy, not your system :)

Posted on 2016-03-07 22:55:29
Draycat

Thanks for an excellent article, it's certainly given a lot to think about in terms of my imminent upgrade. I'm waiting to see what the Broadwell-E chips are like and maybe upgrading to a 6 core 9800k when they come out and overclock it - my files are on a NAS with my catalogue on the cloud so I can edit and access files on the go when necessary so I often have to remake previews in order the keep the size of the catalogue down - I'm a semi professional photographer with about 3.5TB of files. I have a question related to the GPU. On your site you recommend the GTX960 4GB and above. Can I ask the rationale for that since you say it's difficult to assess the impact of the GPU. Would something like a GTX750Ti do just as well or is there any significant difference. I'm not a gamer so the GPU is basically for Lightroom and on the odd occasion that I use Photoshop, so something like the GTX960 feels like it will be overkill. Currently I'm using a an old GTS450 with i7-3770.

Posted on 2016-04-27 14:03:08

Unless the 9800K has a very high frequency (3.8GHz+ or so), I doubt it will be faster than a 6700K for most tasks in Lightroom. I could be completely wrong, though, so if you can wait for the launch of Broadwell-E to see how the specs compare that is probably a good idea.

As for the GPU question, the reason we recommend a GTX 960 or higher is largely due to the 4GB of VRAM. I will readily admit that we haven't been able to fully evaluate the GPU performance in Lightroom yet, but typically photo editing tasks can easily require more than the 2GB of VRAM found on lower end cards. Lightroom right now is a bit odd since the GPU accelerated portions are so new, so in many ways our recommendation is in anticipation of future versions of Lightroom. With the current version of LR, you probably would be fine with a GTX 750Ti though, so that isn't a bad card at the moment.

Posted on 2016-04-27 17:46:06
Draycat

Thanks for the clarification Matt, much appreciated. This page is one of the few sources of concrete info I've found related to these issues - I really appreciate you answering questions on such an old thread.

The reason I'm considering the Broadwell-E is because creating previews is probably my biggest slowdown as the catalogue is on the cloud and I discard previews after a week to keep the size of the catalogue down, so often when I looking at something after a week it has to remake all the previews. From what I understand from your article more cores would help to speed up this particular process, although other things may be slower than on a 6700k. This would be the reason for overclocking the Broadwell-E, just to increase the speed in other functions.

In terms of the GPU I figure I have to replace my GTS450 regardless - Lightroom seems to recognise it but I'm not really sure I getting any benefit from it within Lightroom. Because new graphic cards come out constantly I can't decide whether it worth spending more now (on the GTX960) and then keeping the card for the next 3-4 years, or just getting something like the GTX750Ti, and then upgrading later when I can actually see more benefit, and also benefit from new tech introduced in a newer graphics card.

What are your thoughts?

Posted on 2016-04-28 00:18:52

Overclocking a Broadwell-E CPU would certainly help, but keep in mind that if you are going to overclock that you can overclock a 6700K just as easily (if not more easily due to the lower wattage). So any performance gain you could see with a overclocked Broadwell-E CPU would pretty much be completely offset if you were to overclock a 6700K. I think you still have the right idea, but just wanted to point that out for you to take into consideration.

As for the GPU, its always hard to know when you should purchase something for your needs today versus future proofing. One thing I will mention is that NVIDIA has announced the new Pascal architecture recently and although there are no details on release date or specific model information publicly available, it is very possible that they will either be available for purchase or at least more information released by the time Broadwell-E is out.

GPU acceleration in Lightroom is pretty light at the moment, however, so something like a GTX 750 Ti should be fine for now. I think it comes down to spending less money now, but needing to replace the GPU in a year or two versus spending more money up front, but not needing to purchase a second card in the future. Neither way is really wrong or right, its just a matter of how you want to handle your budget.

Posted on 2016-04-28 16:55:03
Draycat

Hi Matt, thanks for that, and especially the heads up on the new GPUs - I'll wait and see what that brings. Yeah, I understand its a trade-off with the CPUs, but because my biggest slowdown is previews and your article seems to suggest that more cores will help to speed up the process I think it's the right way to go.

Posted on 2016-05-22 11:52:54
Art

I have a question about ram usage during the panorama process, was your ram maxed out? I had 8gb single Channel, and it took 13:30minutes to process a 21 image panorama, each file was about 27 mega bytes. I bought another 8gb stick and installed for dual channel, and my time dropped to 4:30 minutes. Both times my ram was maxed out, CPU was about 20-30%. I'm wondering if I should spend some more money on another 16gb of ram, or just save money toward a 4790k. I'm not expecting another 16gb to drop my time a whole lot. My brother ran the test with his 4790k and 16gb of ram, and it took him about 2:30 minutes. Not sure if his ram was maxed out though.

Posted on 2016-05-17 03:59:00
Diko Jelev

Hi.

The ultimate hardware review for a photographer. Bravo! :-)

Now as some already mentioned whenever you make a second round of tests & reviews it might be nice to include heavy 50 megapixel files (if need some, let me know to assist) and some local adjustments spot/brush tool usage.

I think and hope it will be soon since as a proud owner of 32GB,6700k (4 cores +HT), GTX970 rig am wondering can I improve the lack of LR performance by upgrading to i7-6900K (8 cores +HT)? Now that unfortunately will have to include a new motherboard, which I really don't like. But as Anandtech said it: this generation is aimed at other than me people with older systems.

The weirdest things of them all that most of your current graphs don't show any significant improvement beyond 4 cores even on a single CPU. And since the new 6900k would be at slower speeds and as someone here mentioned that Brush/Spot removal are most likely to utilize more speed rather than parallel computing, could you be so kind to test that for us?

Thank you in advance! ;-)

Posted on 2016-06-02 21:20:10
Lori

http://www.pcgameshardware....

Posted on 2016-06-09 07:39:39
spyros

looking at the results, which i agree with, i did some simple tests of my own. i copied a lr catalogue with 100 pictures both to my ssd, local c disk (samsung evo 850,250gb) and to a conventional hard disk 1Tb 7200rpm. then i counted the time it took to export the 100 pictures as jps to a file on the same disk that originally the catalogue resided. it took the same time to create the 100 jpgs. it means that the speed lightrrom cc6 was built towork is not as fast as the conventional hard disk can absorb. the extra speed of the ssd wasnt used here. Now when i went to the develop module and zoomed in the picture, i was a difference in time between the two disks. on the conventiona disk it took 12 seconds to create the 1:1 preview, where the ssd did it in 9 seconds. exporting improves only with a faster processor, where working with the pictures, the ssd improves speed

Posted on 2016-08-11 16:54:41

Thanks for the benchmark! Adobe Lightroom had it's ups and downs regarding multicore performance. I remember back in the LR4 days when I upgraded to 4.3 (I believe), multicore performance went down. I eventually upgraded to LR6. Well, as can easily be seen in your benchmarks, Adobe has still a lot to fix. I can't imagine that panorama stitching or HDR can't profit from additional cores. The same goes for processing the photos. I'm hopeful that incremental releases will address these shortcomings.
About your benchmark: I'm not familiar with using affinity to alter the cores at disposal to LR, but I could think of a number of flaws doing so. For one, performance of LR is dependent on system resources. Throttling Lightroom while having all the CPU power in the world for system tasks (supporting LR functions) is not exactly simulating a low-core CPU. On a system with 20 cores, determining that up to 6 cores for LR would be beneficial does not exactly show what a 6-core CPU could do, as there would be little resources left to the system (for I/O, etc.).
I run LR on Windows 10 in a virtual machine with VGA passthrough using kvm, as described in one of your earlier articles (and thanks for mentioning me!). I currently give Windows 10 VCPUs (out of a total of 12 threads/VCPUs on a 3930K 6-core processor). If I find the time I will run some benchmarks modifying the VCPUs. This should allow me to simulate a regular multicore CPU where some of the (V)CPU resources are allocated to system tasks. It would be interesting to run your benchmark on your hardware, but in a kvm virtual machine.
You mentioned performance as a function of number of cores and processor speed (GHz), but what about memory speed and PCI channels. Aside from supporting 64GB memory out of the box, my old 3930K 6-core CPU uses 4 memory channels with a total bandwitdh of 51.2 GB/s. That is twice as much bandwidth compared to the 4790K you suggested above. The 4790K also supports only 16 PCIe lanes, the 3930K has 40. I have no idea how this impacts LR performance, only benchmarks or experience will tell if these specs pose bottlenecks. I guess you had a reason for recommending the 4790K.

Posted on 2016-08-22 16:11:39

Thank you very much for this great test! This combined with your advised system article has almost set my buying of my upgrade from my i5-2500k. However in this post you mention cores of the processors used, but not hyperthreading (HT). Was HT deactivated during testing or is it not taken into account because all the processors have HT. How would a i5 at simular clock-speed or E5-1650 v2 compair?

Posted on 2016-09-05 20:16:21

We did leave hyperthreading on for our testing (nearly all the CPUs we sell in our workstations are hyperthreading capable), but I'm not sure how much of a difference it would really make for performance. It may be a bit faster with it, but I would guess only by ~5% or less.

I'm not sure how a Xeon E5-1650 V2 specifically would work in Lightroom, but I can say that for a modern i5 and Xeon E5 V4 CPU that have the same frequency and number of cores, the i5 should actually be faster. It won't have hyperthreading, but the 6th generation i3/i5/i7 CPUs actually use a newer architecture than the Xeon E5 or High End Core i7 CPUs. That usually makes them anywhere from 10-20% faster per clock (depending on the application) which should more than make up for the lack of hyperthreading.

Posted on 2016-09-05 21:40:37

Thank you Matt, this gives me much food for thought.

Posted on 2016-09-06 02:48:16
Tom Mathews

Say for Lightroom/Photoscan, what would be the best possible chip to buy today? I'm building a no-holds-barred system, and had ordered an i7-6950X... but now I'm starting to think a single-threaded chip would have been better.

Posted on 2016-09-23 16:11:48

Just to confirm, do you mean Photoscan or Photoshop? They have very different requirements, so I want to make sure that we give you correct feedback :)

Posted on 2016-09-23 16:17:27
Tom Mathews

Thanks for the rapid response! Primarily Agisoft Photoscan, since it takes so darned long and is the main reason I'm putting this build together. Someone was saying I should look at Xenons because of higher clock rate, but even the E5-1680 v4 (8 cores @4Ghz) ends up being slower than the 10@3.5Ghz i7?

Posted on 2016-09-23 16:59:05

We actually have a lot of test data on Photoscan! Check out these articles, about CPU and GPU performance, and then our recommended systems:

https://www.pugetsystems.co...

https://www.pugetsystems.co...

https://www.pugetsystems.co...

Posted on 2016-09-23 17:11:37
Tom Mathews

Yup, I came here from there. :D You don't cover hyperthreading impacts, though... my search on hyperthreading comments lead me to this thread. Sorry to derail a Photoshop conversation a bit, but it was much newer than the Photoscan articles. It does seem the CPU selection was recently updated, as when I looked last week it didn't seem like you had the 1080s.

Posted on 2016-09-23 17:22:48

Ah, yeah - we normally just test with Hyperthreading enabled, since all of the CPUs we are looking at in these articles (Core i7s and the mid to upper-end Xeon E5s) support it. I think Don Kinghorn looked at the effect of turning HT on and off in some scientific applications, but its a topic we just haven't looked at much here.

For Lightroom, the 8- and 10-core CPUs you are considering are way overkill; it would be better off with a lower core count and higher clock speed. However, Photoscan is a *much* more demanding program overall - so I think you'll probably want to aim for the best performance in it, and then just deal with any minor reduction in performance Lightroom might have from not being on perfectly optimized hardware.

So leaning over into the Photoscan area, the choice of CPU depends heavily on your budget and the number of video cards you plan to use. If you are planning on 2-3 GPUs, I would consider the i7 6900K (or Xeon E5-1660 v4, as they are the same specs) to be a strong option. If you are looking at only a single GPU, I would advise you to scale back on the CPU so you can afford at least a second GPU :)

Posted on 2016-09-23 17:29:50
Tom Mathews

Thanks for the feedback, much appreciated! :)

Posted on 2016-09-23 17:46:07
Malcolm Clint

I would like to upgrade my A8 6600K to i7 7700..I think my Lightroom is slow on export. Is it worth an upgrade?

Posted on 2017-01-20 01:41:16
Alessandro Tiberti Bertin

What about doing same benchmarks with "only" 1 CPU? Having 2 CPUs doesn't only double the number of available cores but also the number of memory channels and bandwidth (if all channels are used). Disabling cores doesn't disable memory controllers so testing 2+2 cores (2 active on CPU 1 and 2 active on CPU 2) is not at all like testing 4 cores using only 1 CPU.

Posted on 2017-03-29 11:38:22
João Filipe Terrível

Hello!
But... about Ryzen 5 1600 vs Intel i5 7600?

Posted on 2017-07-19 22:00:35

We don't usually test with mid to low range CPUs, so I can't say exactly what the performance would be. However, based on our latest Lightroom testing https://www.pugetsystems.co... that included the Ryzen 1700X, 1800X, and Intel Core i7 7700K I would expect those two CPUs to perform roughly on par when importing and exporting. For everything else in Lightroom, the i5 7600 will probably have a slight edge but I imagine it would only be by 5% or so.

For Lightroom alone, I don't know if you would really notice a difference without getting out a stopwatch and timing actions so either CPU should be fine. The only thing I would be aware of is that the i5 7600 is on a more established platform so you are less likely to run into bugs and problems. But if you want to support AMD, go for it. I honestly don't think you would be disappointed with either CPU.

Posted on 2017-07-19 23:05:53

Guys, could you take a look at this discussion: http://www.canonrumors.com/...

Posted on 2017-07-20 12:13:20

I didn't go through the entire thread, but starting at the post you linked I thought it was a very good read and likely spot on. I agree with most of their guesses, although only the Adobe employees actually working on Lightroom know for sure what is going on under the hood. I've always thought it would be an easy thing to simply export multiple images at the same time (maybe 1 per every two cores?) but I'm sure that the people working for Adobe have probably already thought of that. Most likely, there is some reason why that isn't as simple of a solution as it seems. Lots of software packages contains tons of old code so it is very possible that doing something like that would require a complete re-write of Lightroom from the ground up.

Again, no way to know for sure, but I certainly agree that it would be nice if Lightroom was better threaded! Thanks for sharing that forum thread, it was an interesting read!

Posted on 2017-07-20 17:29:33
Desertsweeper

What would be interesting for some of the power-hungry users below is to do your setup in this test but using a hypervisor with say 4 or 5 guest systems. This could overcome the dual-socket inefficiency and provide say 4 or 5 cores to each guest, simultaneously. One could rotate between the guests from a low-powered remote desktop and set tasks. Each guest would, in theory, make much more efficient use of those 20 cores than your test scenario.

Posted on 2017-08-13 18:54:34