Skip to main content

⚠️ Please note that this topic or post has been archived. The information contained here may no longer be accurate or up-to-date. ⚠️

Exporting benchmarks v2

Igor Dmitriev
  • Edited

There is a thread with attempt to collect CO export benchmark statistics on different machines

So everyone could use this information to build/buy a good computer for CO workflow.

I suggest to improve this bench with fixed set of RAW files, so the collected bench data would be more accurate: it will depend mostly on computer specification (CPU and GPU first), but not on RAW files type (Canon, Nikon, Sony and all other RAW types affect export speed).

So here is the benchmark algorithm:

  1. Download the following zip with RAW files and unpack them to some folder
    50 photos made with Canon 5D mk3, size about 1.3GB. Theese photos were made by myself so no property rights are violated 😊

  2. Start CO and ensure that GPU acceleration is enabled: menu Edit - Preferences - General tab - Hardware acceleration - Processing set to Auto (if it was set to "Never" before, you need to restart CO so changes are applied)

  3. Open RAW files that were downloaded on step 1 and wait until they are fully imported (previews are built by CO).

  4. Set up the following export parameters:
    Format: JPEG
    Quality: 100%
    ICC Profile: sRGB IEC61966-2.1
    Resolution: 300 px/in
    Scale: Fixed 100%
    Open with: None

    ❗️ It is very important to set up export with theese parameters! Otherwise the bench results won't be reliable!
  5. Make the first export (GPU accelerated) noting the duration of this process.
    At least one export process is needed. However it would be great if you'll be able to make it 2 or 3 times, because results may vary depending on HDD caching, computer background processes and others. Take the shortest time duration - it is the first bench result.

  6. Disable GPU acceleration: menu Edit - Preferences - General tab - Hardware acceleration - Processing set to Never. Restart the CO.

  7. Make the second export (no GPU acceleration) in a similar way as in step 5.
    That will be the second bench result.

  8. Make a post at this thread with the following format:
    - Computer type (PC/Mac) and model (if any) and OS version
    - CPU+GPU - benchmark time 1
    - CPU only - benchmark time 2
    - CPU model
    - GPU model
    - CO version

You can also make the second benchmark for TIFF export format - it would be good addition to previous JPEG benchmark. Current version of CO (11.0) is limited in perfomance when exporting to JPEG format due to internal algorithms. So TIFF export would show better results in some circumstances, consuming more hardware power. Theese TIFF export parameters are needed to be set up:

. . .
Here is a myself-written Windows utility that will help to easily calculate benchmark results: it analyzes all files in selected folder and automatically calculates time span between opening of first file for writing and last file modifying.
. . .
Suggestions for benchmark improvements are very welcomed!
Hope my post doesn't look to rigorous 😊
The only goal of this thread is to help each other with selecting hardware components for best CO workflow experience.

0

Comments

135 comments

Date Votes
  • Thomas D.
    Threadripper 3960X Stock, 64 GB 2666 ECC.

    Tiff 16 Bit, 2x W8100 GPU: 18 sec. 90° both

    Tiff 16 Bit, 1x RTX 5000 GPU: 24 sec. 65°

    Tiff 16 Bit, CPU: 87 Sec.

    From a Perfomance point CO is a Strange piece of Software.
    0
  • Cary Hatakeda

    Can you post the link for the original zip file to download to do these tests.

    0
  • Igor Dmitriev

    Updated links to new forum format

    0
  • Andrea wee

    with the new phundervolt, I am now no longer able to undervolt my zbook. 


    CPU+GPU+JPG=90s
    CPU+GPU+TIF=22s
    CPU+JPG=134s
    CPU+TIF=115s

    2286m 128Gb ram with RTX5000

    CPU seems to idle at @26% when using GPU. 

    0
  • Whitesnake

    Ryzen 3900X + Radeon VII + 32 GB RAM (3800 CL16) + very very fast SSD + C1 v12:

    CPU+GPU+JPG:31
    CPU+GPU+TIFF: 10

    Before I sold my Mac Mini:

    Mac Mini i7 (Hexa-Core), 32 GB RAM + Vega 56 + C1 v12:

    CPU+GPU+JPG: 35
    CPU+GPU+TIFF: 21

    I think the Radeon VII is the fastest GPU for C1.

    0
  • Ladislav Starý

    iMac 27" late 2015, i5 3.3 GHz, AMD Radeon R9 M395, Capture One 20

     

    CPU+GPU: 

    JPEG: 76 s

    TIFF: 65 s

     

    only CPU:

    JPEG: 140 s

    TIFF: 132 s

     

    0
  • Ladislav Starý

    iMac 27" 2019, i5 3,7 GHz, AMD Radeon Pro 580X 8GB, 40 GB RAM DDR4, Capture One 20

    CPU+GPU:

    JPEG: 50 s

    TIFF: 39 s

     

    only CPU:

    JPEG: 77 s

    TIFF: 78 s

     

    0
  • Robert Whetton
    • Edited

    Been a couple of years. Thought I'd throw in some numbers with V20
    Spec is still the same
    Ryzen 1700
    R9 Fury
    32GB RAM

    JPEG
    V11
    GPU: 0:34


    V12
    GPU: 0:31


    V20
    GPU: 0:31

    Upgraded GPU to a Vega 64
    GPU: 0:27

    0
  • Sebastian Cohen

    C1 v20.04 / WIN10
    3900XT / 32GB DDR4-3200 / GTX1080 / EVO970 Pro

    JPEG CPU+GPU = 0:43
    JPEG CPU only = 1:33

    TIFF CPU+GPU = 0:30
    TIFF CPU only = 1:22

     

    0
  • Igor Dmitriev

    Windows PC (Win10 b2004)
    CaptureOne 21 (14.2)
    CPU – AMD Ryzen-3600X (6-core, 12-threads)
    GPU – NVidia GTX 1080

    JPEG
    CPU+GPU – 39s
    CPU only – 93s
    The GPU speed up factor is about 2.38x.

    TIFF
    CPU+GPU – 26s
    CPU only – 79s
    The GPU speed up factor is about 3.04x.

    0
  • Oscar Lai
    • Edited

    Lurking around a while, set up my new desktop back in 2020 Dec and finally have time to test. Contribute my parts to the community, as I decided to buy 2nd hand Radeon card back in the fall of 2020 after referencing to this thread (decide between Nvidia vs AMD card, and between the new RX 6xxx and Vega series).

    Windows 10 Professional

    OS build 19042.1348

    Version 20H2

    MB - Asus Crosshair VIII Hero WIFI

    AMD 3900XT no OC

    64gb DDR 3600 small OC to 3733

    MSI Radeon RX Vega 64 8gb Liquid Cooled edition

    Samsung 970 pro 512gb m c drive

    Samsung 860 pro 1tb d drive

    Fresh Windows 10 pro install, with Office 2016 and Capture One 21 v.14.4.1.6

    JPEG

    CPU+GPU

    1st run - 37.98s

    2nd run - 37.13s

    3rd run - 37.20s

    CPU only - 73.22s; only have 1 run

    Did not try TIFF output as I know it will be faster from all others' result above

    Vega 64 LC edition is a few seconds slower than Vii or Frontier Edition, but the cost is much lower. Hope this info will be helpful to choose between the new RX 6xxx services and the previous generation from AMD.

    0
  • Brian Jones
    • Edited

    I've updated my benchmarks for the 4 Macs that are my production computers. The 2013 Mac Pro and 2018 MPB have been replaced by the M1 MBPs.

    All times are on C1 14.4.1 Interestingly the times are slower than the last test on 12.2

    2021 MBP M1 Pro 16core video 16gb

    Tiff with GPU 31.08

    Tiff Without 54.52

    JEPG with GPU 39.05

    Jpeg without 1:25.35

     

    2021 MBP M1 Max 32 video core 32gb

    Tiff with gpu 18.65

    Tiff without 46.08

    Jpeg with 25.63

    Jpeg without 1:02.40

     

    2018 MPB i9

    Tiff with epgu 29.85

    Tiff egpu preferred 34.33

    Tiff gpu 57.0   

    Tiff no gpu 1:20.31

    Jpeg egpu 51.0

    Jpeg no gpu 1:14.05 

    Jpeg epgu preferred 48.95

    Jpeg gpu 1:13.72

     

    2013 Mac Pro 3.5ghz 6 core D500

    Tiff GPU 34.71

    Tiff no gpu 1:44.31

    Jpeg gpu 55.40

    Jpeg no gpu 2:08.08

     

    0
  • Matthew Varney
    • Edited

    Results from version 15.2 and the new Mac Studio.

    2022 Mac Studio
    macOS 12.3.1
    M1 Max (10 core CPU, 32 core GPU)
    32GB RAM
    Internal SSD
    C1 v15.2.0

    JPEG CPU+GPU: 26s (0:26)
    JPEG CPU Only:  61s (1:01)

    TIFF CPU+GPU: 19s (0:19)
    TIFF CPU Only:  54s (0:54)

    Mac Pro 5,1
    macOS 10.14.6
    X5675, 12 cores, 3.06 GHz
    48GB RAM
    Vega 64 8GB
    SATA III RAID 0 SSDs
    C1 v15.2.0

    JPEG CPU+GPU: 51s (0:51)
    JPEG CPU Only:  118s (1:58)

    TIFF CPU+GPU: 27s (0:27)
    TIFF CPU Only:  101s (1:41)

    Asus Prime X470-Pro
    Windows 10 Pro 21H2
    Ryzen 7 3800X
    32GB DDR4 3600 CL16
    Vega 64 8GB (Sapphire Nitro+)
    NVMe SSDs
    C1 v15.2.0

    JPEG CPU+GPU: 34s (0:34)
    JPEG CPU Only:  80s (1:20)

    TIFF CPU+GPU: 17s (0:17)
    TIFF CPU Only:  73s (1:13)

    Best of three runs. Restarted Capture One between runs (this makes a difference).

    0
  • Fernando Javier Giménez Cepero

    Capture one 16.1
    I7 920
    Gpu r9 280x 3Gb ram Gpu
    18gb ram ddr3
    Ssd sata ii

    CPU +Gpu
    jpeg 85 secs
    Tiff 72 secs

    CPU only jpg 285 secs

    I tried using 2 r9 280x without crossfire but had several crashes
    I manage to get 72 secs in cpu + 2 Gpu in jpegs

    I observed that the first 50 first secs the cpu load goes to 80-90% and the falls to 20% while the Gpu usage in windows only goes 2-3% in one Gpu. And nothing on the other. Only the ram is used in the 2 gpus

    Does anyone observed this behavior in using two gpus in windows?
    it doesn’t seem to be very efficient

    Thanks

    0
  • BeO
    • Edited
    Top Commenter
    Notebook DELL XPS13
      (model 7390 2-in-1 from year 2020)
     
    NVMe disk drive
    RAM 16 GB
     
    CPU: Intel Core i7-1065G7 CPU @ 1.30GHz, 1498 Mhz, 4 Cores, 8 Logical Processors
     
    GPU: Intel(R) Iris(R) Plus Graphics
      OpenCL Compute Units : 64
      OpenCL CL_DEVICE_GLOBAL_MEM_SIZE : 6461 mb
      C1 ImageCore Benchmark: 0.268607
     
    Windows 10 Home 22H2 
    C1 version 15.2.1.9
    ____________________________________________________
    JPEG CPU+GPU: 02:50 (170s)
    JPEG CPU Only:  02:44 (164s)
     
    (best of two runs each. C1 not retarted in-between)
    Export Adjustments: "no output sharpening"
     
    0

Please sign in to leave a comment.

Didn't find what you were looking for?

New post