Creative Communities of the World Forums

The peer to peer support community for media production professionals.

Activity Forums Adobe Premiere Pro Premiere’s PC Desktop performance is dismal Any ideas on what I can do?

  • Premiere’s PC Desktop performance is dismal Any ideas on what I can do?

    Posted by Adrian Lyons on August 19, 2014 at 11:48 pm

    Maybe it’s just me but I find Premiere CS6’s performance on Windows dismal. I’ve used other software that blows it out of the water when it comes to rendering times.

    I am in the process of editing a multicam sequence and am rendering out an H264. A 30 min sequence is going to take just over 7 hours to render. I have two GoPro 1080/30p RAW ProTune streams and a Nikon 720/60p. There’s a separate WAV audio track as well.

    I am confused as to whey PP does not utilize all of the available resources. Here is a screen shot of the perfmon in Windows showing 37% CPU and only 176kb disk throughput. My source clips are on an SSD and I am writing to a Sata3 HDD. My swaps are on a separate drive as well.

    Here’s my system stats

    OS Name Microsoft Windows 7 Professional
    Version 6.1.7601 Service Pack 1 Build 7601
    Other OS Description Not Available
    OS Manufacturer Microsoft Corporation
    System Name MONSTER
    System Manufacturer System manufacturer
    System Model System Product Name
    System Type x64-based PC
    Processor Intel(R) Core(TM) i7-2600K CPU @ 3.40GHz, 3401 Mhz, 4 Core(s), 8 Logical Processor(s)
    BIOS Version/Date American Megatrends Inc. 4101, 5/9/2013
    SMBIOS Version 2.6
    Windows Directory C:\Windows
    System Directory C:\Windows\system32
    Boot Device \Device\HarddiskVolume2
    Locale United States
    Hardware Abstraction Layer Version = “6.1.7601.17514”
    User Name MONSTER\adrian
    Time Zone Eastern Daylight Time
    Installed Physical Memory (RAM) 32.0 GB
    Total Physical Memory 32.0 GB
    Available Physical Memory 28.0 GB
    Total Virtual Memory 63.9 GB
    Available Virtual Memory 59.3 GB
    Page File Space 32.0 GB
    Page File D:\pagefile.sys

    NVIDIA System Information report created on: 08/19/2014 18:47:32
    System name: MONSTER

    [Display]
    Operating System: Windows 7 Professional, 64-bit (Service Pack 1)
    DirectX version: 11.0
    GPU processor: GeForce GTX 660 Ti
    Driver version: 340.52
    Direct3D API version: 11
    Direct3D feature level: 11_0
    CUDA Cores: 1344
    Core clock: 1006 MHz
    Memory data rate: 6200 MHz
    Memory interface: 192-bit
    Memory bandwidth: 148.80 GB/s
    Total available graphics memory: 19183 MB
    Dedicated video memory: 3072 MB GDDR5
    System video memory: 0 MB
    Shared system memory: 16111 MB
    Video BIOS version: 80.04.4B.00.0F
    IRQ: 16
    Bus: PCI Express x16 Gen2
    Device Id: 10DE 1183 844E1043
    Part Number: 2004 0001

    [Components]

    NvGFTrayPluginr.dll 15.3.33.0 NVIDIA GeForce Experience
    NvGFTrayPlugin.dll 15.3.33.0 NVIDIA GeForce Experience
    nvui.dll 8.17.13.4052 NVIDIA User Experience Driver Component
    nvxdsync.exe 8.17.13.4052 NVIDIA User Experience Driver Component
    nvxdplcy.dll 8.17.13.4052 NVIDIA User Experience Driver Component
    nvxdbat.dll 8.17.13.4052 NVIDIA User Experience Driver Component
    nvxdapix.dll 8.17.13.4052 NVIDIA User Experience Driver Component
    NVCPL.DLL 8.17.13.4052 NVIDIA User Experience Driver Component
    nvCplUI.exe 7.8.760.0 NVIDIA Control Panel
    nvWSS.dll 6.14.13.4052 NVIDIA Workstation Server
    nvViTvS.dll 6.14.13.4052 NVIDIA Video Server
    NVSTVIEW.EXE 7.17.13.4052 NVIDIA 3D Vision Photo Viewer
    NVSTTEST.EXE 7.17.13.4052 NVIDIA 3D Vision Test Application
    NVSTRES.DLL 7.17.13.4052 NVIDIA 3D Vision Module
    NVMCTRAY.DLL 8.17.13.4052 NVIDIA Media Center Library
    nvDispS.dll 6.14.13.4052 NVIDIA Display Server
    PhysX 09.13.1220 NVIDIA PhysX
    NVCUDA.DLL 8.17.13.4052 NVIDIA CUDA 6.5.12 driver
    nvGameS.dll 6.14.13.4052 NVIDIA 3D Settings Server

    c:\windows\system32\cfhd.dll CineForm Inc. CineForm HD VFW Codec OK C:\Windows\system32\CFHD.DLL 8.6.3.670 1.38 MB (1,443,328 bytes) 1/29/2014 1:55 PM
    c:\windows\system32\iyuv_32.dll Microsoft Corporation OK C:\Windows\system32\IYUV_32.DLL 6.1.7601.17514 53.00 KB (54,272 bytes) 1/2/2014 10:02 AM
    c:\windows\system32\lvcod64.dll Logitech Inc. OK C:\Windows\system32\LVCOD64.DLL 13.40.840.0 172.28 KB (176,416 bytes) 12/15/2011 5:15 AM
    c:\windows\system32\msrle32.dll Microsoft Corporation OK C:\Windows\system32\MSRLE32.DLL 6.1.7601.17514 16.00 KB (16,384 bytes) 1/2/2014 10:02 AM
    c:\windows\system32\msvidc32.dll Microsoft Corporation OK C:\Windows\system32\MSVIDC32.DLL 6.1.7601.17514 38.00 KB (38,912 bytes) 1/2/2014 10:02 AM
    c:\windows\system32\msyuv.dll Microsoft Corporation OK C:\Windows\system32\MSYUV.DLL 6.1.7601.17514 25.00 KB (25,600 bytes) 1/2/2014 10:02 AM
    c:\windows\system32\tsbyuv.dll Microsoft Corporation OK C:\Windows\system32\TSBYUV.DLL 6.1.7601.17514 14.50 KB (14,848 bytes) 1/2/2014 10:02 AM

    RAMMon v1.0 Build: 1013 built with SysInfo v1.0 Build: 1063
    PassMark (R) Software
    https://www.passmark.com

    Memmory summary for MONSTER:
    Number of Memory Devices: 4 Total Physical Memory: 32735 MB (32768 MB)
    Total Available Physical Memory: 28422 MB
    Memory Load: 13%

    Item | Slot #1 | Slot #2 | Slot #3 | Slot #4 |
    ——————————————————————————-|———————–|———————–|———————–|———————–|-
    Ram Type | DDR3 | DDR3 | DDR3 | DDR3 |
    Maximum Clock Speed (MHz) | 800.00 (XMP) | 800.00 (XMP) | 800.00 (XMP) | 800.00 (XMP) |
    Maximum Transfer Speed (MHz) | DDR3-1600 | DDR3-1600 | DDR3-1600 | DDR3-1600 |
    Maximum Bandwidth (MB/s) | PC3-12800 | PC3-12800 | PC3-12800 | PC3-12800 |
    Memory Capacity (MB) | 8192 | 8192 | 8192 | 8192 |
    Jedec Manufacture Name | G Skill Intl | G Skill Intl | G Skill Intl | G Skill Intl |
    Search Amazon.com | Search! | Search! | Search! | Search! |
    SPD Revision | 1.1 | 1.1 | 1.1 | 1.1 |
    Registered | No | No | No | No |
    ECC | No | No | No | No |
    DIMM Slot # | 1 | 2 | 3 | 4 |
    Manufactured | | | | |
    Module Part # | F3-1600C7-8GTX | F3-1600C7-8GTX | F3-1600C7-8GTX | F3-1600C7-8GTX |
    Module Revision | 0x0 | 0x0 | 0x0 | 0x0 |
    Module Serial # | 0x0 | 0x0 | 0x0 | 0x0 |
    Module Manufacturing Location | 0 | 0 | 0 | 0 |
    # of Row Addressing Bits | 16 | 16 | 16 | 16 |
    # of Column Addressing Bits | 10 | 10 | 10 | 10 |
    # of Banks | 8 | 8 | 8 | 8 |
    # of Ranks | 2 | 2 | 2 | 2 |
    Device Width in Bits | 8 | 8 | 8 | 8 |
    Bus Width in Bits | 64 | 64 | 64 | 64 |
    Module Voltage | 1.5V | 1.5V | 1.5V | 1.5V |
    CAS Latencies Supported | 5 6 7 8 9 10 | 5 6 7 8 9 10 | 5 6 7 8 9 10 | 5 6 7 8 9 10 |
    Timings @ Max Frequency (JEDEC) | 9-9-9-24 | 9-9-9-24 | 9-9-9-24 | 9-9-9-24 |
    Maximum frequency (MHz) | 666.67 | 666.67 | 666.67 | 666.67 |
    Maximum Transfer Speed (MHz) | DDR3-1333 | DDR3-1333 | DDR3-1333 | DDR3-1333 |
    Maximum Bandwidth (MB/s) | PC3-10600 | PC3-10600 | PC3-10600 | PC3-10600 |
    Minimum Clock Cycle Time, tCK (ns) | 1.500 | 1.500 | 1.500 | 1.500 |
    Minimum CAS Latency Time, tAA (ns) | 13.125 | 13.125 | 13.125 | 13.125 |
    Minimum RAS to CAS Delay, tRCD (ns) | 13.125 | 13.125 | 13.125 | 13.125 |
    Minimum Row Precharge Time, tRP (ns) | 13.125 | 13.125 | 13.125 | 13.125 |
    Minimum Active to Precharge Time, tRAS (ns) | 36.000 | 36.000 | 36.000 | 36.000 |
    Minimum Row Active to Row Active Delay, tRRD (ns) | 6.000 | 6.000 | 6.000 | 6.000 |
    Minimum Auto-Refresh to Active/Auto-Refresh Time, tRC (ns) | 49.125 | 49.125 | 49.125 | 49.125 |
    Minimum Auto-Refresh to Active/Auto-Refresh Command Period, tRFC (ns) | 260.000 | 260.000 | 260.000 | 260.000 |
    | | | | |
    DDR3 Specific SPD Attributes | | | | |
    Write Recovery Time, tWR (ns) | 15.000 | 15.000 | 15.000 | 15.000 |
    Internal Write to Read Command Delay, tWTR (ns) | 7.500 | 7.500 | 7.500 | 7.500 |
    Internal Read to Precharge Command Delay, tRTP (ns) | 7.500 | 7.500 | 7.500 | 7.500 |
    Minimum Four Activate Window Delay, tFAW (ns) | 30.000 | 30.000 | 30.000 | 30.000 |
    RZQ / 6 Supported | Yes | Yes | Yes | Yes |
    RZQ / 7 Supported | Yes | Yes | Yes | Yes |
    DLL-Off Mode Supported | Yes | Yes | Yes | Yes |
    Maximum Operating Temperature Range (C) | 0-95 | 0-95 | 0-95 | 0-95 |
    Refresh Rate at Extended Operating Temperature Range | 2X | 2X | 2X | 2X |
    Auto-self Refresh Supported | Yes | Yes | Yes | Yes |
    On-die Thermal Sensor Readout Supported | No | No | No | No |
    Partial Array Self Refresh Supported | No | No | No | No |
    Thermal Sensor Present | No | No | No | No |
    Non-standard SDRAM Type | Standard Monolithic | Standard Monolithic | Standard Monolithic | Standard Monolithic |
    Module Type | UDIMM | UDIMM | UDIMM | UDIMM |
    Module Height (mm) | 30 | 30 | 30 | 30 |
    Module Thickness (front), (mm) | 2 | 2 | 2 | 2 |
    Module Thickness (back), (mm) | 2 | 2 | 2 | 2 |
    Module Width (mm) | 133.5 | 133.5 | 133.5 | 133.5 |
    Reference Raw Card Used | Raw Card B Rev. 1 | Raw Card B Rev. 1 | Raw Card B Rev. 1 | Raw Card B Rev. 1 |
    DRAM Manufacture ID | 1229 | 1229 | 1229 | 1229 |
    # of DRAM Rows | 0 | 0 | 0 | 0 |
    # of Registers | 0 | 0 | 0 | 0 |
    Register Manufacturer | | | | |
    Register Type | | | | |
    Register Revision | 0 | 0 | 0 | 0 |
    | | | | |
    XMP Attributes | | | | |
    XMP Revision | 1.2 | 1.2 | 1.2 | 1.2 |
    Enthusiast / Certified Profile | | | | |
    Module voltage | 1.50V | 1.50V | 1.50V | 1.50V |
    Clock speed (MHz) | 800.00 | 800.00 | 800.00 | 800.00 |
    Transfer Speed (MHz) | DDR3-1600 | DDR3-1600 | DDR3-1600 | DDR3-1600 |
    Bandwidth (MB/s) | PC3-12800 | PC3-12800 | PC3-12800 | PC3-12800 |
    Minimum clock cycle time, tCK (ns) | 1.250 | 1.250 | 1.250 | 1.250 |
    Supported CAS latencies | 7 | 7 | 7 | 7 |
    Minimum CAS latency time, tAA (ns) | 8.375 | 8.375 | 8.375 | 8.375 |
    Minimum RAS to CAS delay time, tRCD (ns) | 9.625 | 9.625 | 9.625 | 9.625 |
    Minimum row precharge time, tRP (ns) | 9.625 | 9.625 | 9.625 | 9.625 |
    Minimum active to precharge time, tRAS (ns) | 29.625 | 29.625 | 29.625 | 29.625 |
    Supported timing at highest clock speed | 7-8-8-24 | 7-8-8-24 | 7-8-8-24 | 7-8-8-24 |
    Minimum Row Active to Row Active Delay, tRRD (ns) | 6.000 | 6.000 | 6.000 | 6.000 |
    Minimum Active to Auto-Refresh Delay, tRC (ns) | 39.625 | 39.625 | 39.625 | 39.625 |
    Minimum Recovery Delay, tRFC (ns) | 260.000 | 260.000 | 260.000 | 260.000 |
    Minimum Write Recovery time, tWR (ns) | 15.000 | 15.000 | 15.000 | 15.000 |
    Minimum Write to Read CMD Delay, tWTR (ns) | 7.500 | 7.500 | 7.500 | 7.500 |
    Minimum Read to Pre-charge CMD Delay, tRTP (ns) | 7.500 | 7.500 | 7.500 | 7.500 |
    Minimum Four Activate Window Delay, tFAW (ns) | 30.000 | 30.000 | 30.000 | 30.000 |
    Maximum Average Periodic Refresh Interval, tREFI (us) | 7.875 | 7.875 | 7.875 | 7.875 |
    Write to Read CMD Turn-around Time Optimizations | No adjustment | No adjustment | No adjustment | No adjustment |
    Read to Write CMD Turn-around Time Optimizations | No adjustment | No adjustment | No adjustment | No adjustment |
    Back 2 Back CMD Turn-around Time Optimizations | No adjustment | No adjustment | No adjustment | No adjustment |
    System CMD Rate Mode | 2T | 2T | 2T | 2T |
    Extreme Profile | | | | |
    Module voltage | 1.50V | 1.50V | 1.50V | 1.50V |
    Clock speed (MHz) | 800.00 | 800.00 | 800.00 | 800.00 |
    Transfer Speed (MHz) | DDR3-1600 | DDR3-1600 | DDR3-1600 | DDR3-1600 |
    Bandwidth (MB/s) | PC3-12800 | PC3-12800 | PC3-12800 | PC3-12800 |
    Minimum clock cycle time, tCK (ns) | 1.250 | 1.250 | 1.250 | 1.250 |
    Supported CAS latencies | 7 | 7 | 7 | 7 |
    Minimum CAS latency time, tAA (ns) | 8.375 | 8.375 | 8.375 | 8.375 |
    Minimum RAS to CAS delay time, tRCD (ns) | 9.625 | 9.625 | 9.625 | 9.625 |
    Minimum row precharge time, tRP (ns) | 9.625 | 9.625 | 9.625 | 9.625 |
    Minimum active to precharge time, tRAS (ns) | 29.625 | 29.625 | 29.625 | 29.625 |
    Supported timing at highest clock speed | 7-8-8-24 | 7-8-8-24 | 7-8-8-24 | 7-8-8-24 |
    Minimum Row Active to Row Active Delay, tRRD (ns) | 6.000 | 6.000 | 6.000 | 6.000 |
    Minimum Active to Auto-Refresh Delay, tRC (ns) | 39.625 | 39.625 | 39.625 | 39.625 |
    Minimum Recovery Delay, tRFC (ns) | 260.000 | 260.000 | 260.000 | 260.000 |
    Minimum Write Recovery time, tWR (ns) | 15.000 | 15.000 | 15.000 | 15.000 |
    Minimum Write to Read CMD Delay, tWTR (ns) | 7.500 | 7.500 | 7.500 | 7.500 |
    Minimum Read to Pre-charge CMD Delay, tRTP (ns) | 7.500 | 7.500 | 7.500 | 7.500 |
    Minimum Four Activate Window Delay, tFAW (ns) | 30.000 | 30.000 | 30.000 | 30.000 |
    Maximum Average Periodic Refresh Interval, tREFI (us) | 7.875 | 7.875 | 7.875 | 7.875 |
    Write to Read CMD Turn-around Time Optimizations | No adjustment | No adjustment | No adjustment | No adjustment |
    Read to Write CMD Turn-around Time Optimizations | No adjustment | No adjustment | No adjustment | No adjustment |
    Back 2 Back CMD Turn-around Time Optimizations | No adjustment | No adjustment | No adjustment | No adjustment |
    System CMD Rate Mode | 2T | 2T | 2T | 2T |

    Adrian Lyons replied 11 years, 11 months ago 2 Members · 3 Replies
  • 3 Replies
  • Tim Kolb

    August 20, 2014 at 1:57 am

    Lots of stuff in there that I don’t understand of course…

    On the GoPro ProTune streams…I’m not that familiar with that process…is that running during all this? I know that some processes in the free version of GoPro Studio do not multi-thread the last time I looked at it.

    have you looked at CPU usage (and I’d like to know what it’s doing per-core, not just overall percentage) when exporting something without GoPro ProTune footage?

    Your processor clock is fast…which is good for H264. The 660Ti has a decent spec…

    How is your sequence set up? Are both GoPro streams being scaled down to 720p AND being frame-duped to 59.94/sec, or are the GoPro streams being scaled and the Nikon footage is dropping every other frame for a 720p29.97 output?

    You have the frame rates all synced at least right? You don’t have 30p and 29.97p on one sequence set one way or the other? I’m not familiar with the kinds of files Nikons record. If you have 29.97 and 30fps and/or 59.94 and 60fps mixed together, you’ll definitely increase your render times…

    TimK,
    Director, Consultant
    Kolb Productions,

    Adobe Certified Instructor

  • Adrian Lyons

    August 20, 2014 at 2:30 pm

    Hey Tim, thanks for taking the time to give your thoughts.

    The CPU load is spread evenly across all cores they just don’t break a sweat at all.

    After further research and testing, I discovered that conforming al the footage to NDxHD would be the best route to go. Sorensen Squeeze 9 does that and in a batch to boot. That was the beginning of another journey through frustrationville!

    Sorensen does a great job, is written well to take advantage of multithreading and hypertheading, it also utilizes the CUDA cores in my GPU so encoding a bunch of files is fast as hell. The problem is, PP does not recognize the MXP files created by sorenson! As a workaround, I can encode the GoPro streams into one file on a timeline within PP and export that to an MXP that PP can read!

    Not sure what Sorenson is doing differently that PP doesn’t like. I can’t even play those MXP files in VLC.

    I’ve checked msinfo32 and looked at installed codecs and funny enough, the Avid NDxHD codecs don’t show in the list. But they are available within PP.

    Anyway, testing a conformed clip on the timeline and PP flies through it without any stuttering, like it should with the hardware I have.

    I’ll reply with the process in total of what I did in order to get this going with some acceptable performance.

  • Adrian Lyons

    September 23, 2014 at 3:15 pm

    Ok, so here we go.

    I’ve completed the project and delivered to the client. Here’s the run around I needed to go through in order to get it done.

    I initially tried editing with the raw files out of camera (2 GoPro’s and a Nikon D7100). That was a lesson in frustration as the native codecs (h264) do not lend themselves to editing with any fluidity or ease in PP.

    I searched and searched and learned that I needed to transcode all the files to an edit friendly codec. DNxHD was the hands down winner. I used Sorenson Squeeze 9 to re-encode all the footage to MOV’s using DNxHD 75 8bit.

    I then discovered after comparing the native files to the recodes that the colour and gamma were massively shifting. No matter what I tried everything looked flat. As it happens, it was an issue with Sorenson Squeeze. I did the same conversions using MPEG Streamclip and the recodes were perfect (shame on sorenson for f’ing with the footage).

    Once all that malarkey was concluded the edit process went fairly ok. I say fairly as I was having fun trying to colour match the GoPro’s and the Nikon together. I did some research and learned of the BlackMagic Decklink and Intensity cards. I picked up an Intesity PCIe which was a little overkill. All I wanted to do was take advantage of the Mercury Transmit out to a TV for colour grading in Speedgrade. Of course it was only after buying and installing did I come to learn that even in SpeedgradeCC, Adobe has yet to support Mercury Transmit for the BlackMagic family of products!. Works fine in PP CC, just not Speedgrade.

    I had to resort to doing all colour grading across the three cameras in PP CC. Time consuming to say the least but I managed to get through it and get it done.

    My takehomes from this project;

    1. Pull all the source clips into Bridge or Prelude and go through all the footage to pick out the shots that you are going to use.

    2. Conform all the footage to the same codec (not necessarily the same framerate in the event you want to pull off some twixtoring {slo-mo})

    3. Keep it simple. I had numerous timelines all nested in other nested timelines. Made hunting around for master timelines fun.

    4. Dynamic linking while fun and allows for tweaking after the fact, seriously slows down render times and adds on another cause of crashing. Export out clips to AE, make your comps and render them back out to conformed clips. The amount of time it takes to render in AE compared to the amount of time PP takes to render them out is a huge difference.

    5. Keep learning and trying. Don’t ever give up. I am really happy with the project and so was the client. So what the client was my brother! lol. It was a 1 week notice job that I feel I pulled off to perfection (as close as perfection as possible).

    Hope that helps other newcomers to the fun of video production.

    A.

We use anonymous cookies to give you the best experience we can.
Our Privacy policy | GDPR Policy