Announcement

Collapse
No announcement yet.

TMPGEnc Video Mastering Works 5 and cuda

Collapse
X
 
  • Time
  • Show
Clear All
new posts

  • TMPGEnc Video Mastering Works 5 and cuda

    I am trying the trial and so far I can not get the software to use the cuda processing. Is there something I am missing?

    Doesn't the softwre use the cuda capability of my Nvidia graphics card?
    Edius 8 Workgroup, Intel 3770K, Asus P8Z77-V Motherboard, 32GB DDR3 1600 ram, SSD for C, external Raid box with WE RE4 1TB drives Raid 0 for video assets. Overclocked 4.3ghz, Asus GTX 660 ti, water cooled system. Windows 10 Professional 64bit

  • #2
    Yes, but you have to enable it. Click on preferences, and then under CPU/GPU click Nvidia Cuda, and then click enable. Understand that Video Masterworks 5 does not use CUDA for converting all codecs. It only uses CUDA on certain codecs. Experiment with a few different ones.
    EDIUS Workgroups 8 with NX Hardware plus HD expansion Card,
    EDITHD Workstation: Intel Core i7 [email protected], Nvidia GTX-950 Video Card, Windows 7 Premium, 64bit, 6GB Ram, Vistitle 2.6.0.6, Robuskey, ChromaKeyer Pro,

    Comment


    • #3
      Will you tell me which codec will utilize the cuda on the graphics card? I tried quite a few but always get 100%cpu 0%cuda.

      Or exactly how to set up the project to get the cuda processing. I would like to compare the speed difference.
      Edius 8 Workgroup, Intel 3770K, Asus P8Z77-V Motherboard, 32GB DDR3 1600 ram, SSD for C, external Raid box with WE RE4 1TB drives Raid 0 for video assets. Overclocked 4.3ghz, Asus GTX 660 ti, water cooled system. Windows 10 Professional 64bit

      Comment


      • #4
        try an mpeg source clip or mpeg transport stream source clip and it should kick in

        I personally have it disabled because my CPU outperforms CUDA

        I think CUDA has a benefit on slower computers
        Anton Strauss
        Antons Video Productions - Sydney

        EDIUS X WG with BM Mini Monitor 4k and BM Mini Recorder, Gigabyte X299 UD4 Pro, Intel Core i9 9960X 16 Core, 32 Threads @ 4.3Ghz, Corsair Water Cooling, Gigabyte RTX-2070 Super 3X 8GB Video Card, Samsung 860 Pro 512GB SSD for System, 8TB Samsung Raid0 SSD for Video, 2 Pioneer BDR-209 Blu-ray/DVD burners, Hotswap Bay for 3.5" Sata and 2.5" SSD, Phanteks Enthoo Pro XL Tower, Corsair 32GB DDR4 Ram, Win10 Pro

        Comment


        • #5
          Originally posted by antonsvideo

          I think CUDA has a benefit on slower computers
          Then it must be very slow systems, my old Xeon 5050 setup doesn't really speed up with cuda used or without.
          Maybe it only works with a Monster of a video card, but, then the CPU will always win the price / performance battle.
          I also have it set to OFF as testing on several systems with varying CPU's and GPU's showed me little to none speed improvement in use with T4 or T5.

          Maybe it works better with Premiere & Mercury playback engine ................
          Tony D.

          EDIUS WG 9.54 / MATROX MXO2 LE / WIN10 Pro WS / LENOVO P71 Workstation laptop / XEON E3-1535M v6 / nVidia QUADRO P3000 6GB / Toshiba M.2 NVMe 512GB / 2x Samsung 860 EVO 4TB / 32GB ECC RAM.
          EDIUS WG 8.53 / HDRX-E1 + HDBX-1000H / WIN10 Pro / DUAL XEON X5470 / SUPERMICRO X7DWA-N / SUPERMICRO SUPERCHASIS SC745TQ-920B / INTEL 520 SSD 240GB / WD CAVIAR BLACK 4TB (many) / 32GB ECC RAM / GTX 770 4GB / 2x MOTU 896HD.
          wavearts / neatvideo / tmpgenc / hitfilm / imaginate / affinity

          Comment


          • #6
            I think CUDA is only used for filters . Never used on any of my TMPGenc encodes either so I too have turned off. Always showed CPU 100% CUDA 0% after optimization. I even tried some of the filters that I do not use in TMPGenc thinking that it was used just for filters but that had no difference either. Maybe there are some filters it is used for that I don't use or it has decided that the i7-2600K is faster anyway.

            Ron Evans
            Ron Evans

            Threadripper 1920 stock clock 3.7, Gigabyte Designare X399 MB, 32G G.Skill 3200CL14, 500G M.2 NVME OS, 500G EVO 850 temp. 1T EVO 850 render, 16T Source, 2 x 1T NVME, MSI 1080Ti 11G , EVGA 850 G2, LG BLuray Burner, BM IP4K, WIN10 Pro, Shuttle Pro2

            ASUS PB328 monitor, BenQ BL2711U 4K preview monitor, EDIUS X, 9.5 WG, Vegas 18, Resolve Studio 18


            Cameras: GH5S, GH6, FDR-AX100, FDR-AX53, DJI OSMO Pocket, Atomos Ninja V x 2

            Comment


            • #7
              On my older Xeon system I can get 20 to 30 percent cuda usage on some encodes, when measured the total duration there is actually no difference.
              This was tested with 1 hour timeline (quite a while back) on short clips it doesn't show anything at all.

              To me it sounds more like a marketing name game gimmick than actual useful working time saving feature ........ similar to the much hyped about ivy bridge with its marginal improvements over sandy bridge besides power savings.
              Tony D.

              EDIUS WG 9.54 / MATROX MXO2 LE / WIN10 Pro WS / LENOVO P71 Workstation laptop / XEON E3-1535M v6 / nVidia QUADRO P3000 6GB / Toshiba M.2 NVMe 512GB / 2x Samsung 860 EVO 4TB / 32GB ECC RAM.
              EDIUS WG 8.53 / HDRX-E1 + HDBX-1000H / WIN10 Pro / DUAL XEON X5470 / SUPERMICRO X7DWA-N / SUPERMICRO SUPERCHASIS SC745TQ-920B / INTEL 520 SSD 240GB / WD CAVIAR BLACK 4TB (many) / 32GB ECC RAM / GTX 770 4GB / 2x MOTU 896HD.
              wavearts / neatvideo / tmpgenc / hitfilm / imaginate / affinity

              Comment


              • #8
                Originally posted by Ron Evans
                I think CUDA is only used for filters . Never used on any of my TMPGenc encodes either so I too have turned off. Always showed CPU 100% CUDA 0% after optimization.
                I never got the cuda thing to work either with my nvidia gtx260, no matter if I rendered to a mpeg2 or mpeg4 bluray, it stayed at 0%. I do get much more benefit from spursengine and my leadtek winfast card with mastering works and I"m very happy that they included this as well in authoring works. Yesterday I rendered a 1080p HQAVI movie to a mp4 24mbs bluray, on my I7 950 processor only this took 2,5 hour to render and with spursengine enabled only 30 minutes.

                Comment


                • #9
                  In Video Mastering Works 5, you can use CUDA for encoding H.264 video by selecting it as the encoder in your MPEG-4 AVC/h.264 output format settings. By default, the encoder is set as x264.

                  Otherwise, it will only be used for filtering and/or decoding if you have those options checked in the preferences.
                  Last edited by TMPG_Garett; 04-26-2012, 12:59 AM.

                  Comment


                  • #10
                    I have the trial version of this software. I would like to buy it but some things for me are a little unclear. Such as this cuda thing. Why don't they publish a list of the codecs that work with the cuda. Instead of us having to go through the whole list and test what works. Not a big thing but what does what and when would be a big start for me to actually buy the program.
                    Edius 8 Workgroup, Intel 3770K, Asus P8Z77-V Motherboard, 32GB DDR3 1600 ram, SSD for C, external Raid box with WE RE4 1TB drives Raid 0 for video assets. Overclocked 4.3ghz, Asus GTX 660 ti, water cooled system. Windows 10 Professional 64bit

                    Comment


                    • #11
                      Send a question to Support team. You have a link: https://ssl.pegasys-inc.com/inquiry/...&category=ffen
                      Last edited by MPPais; 04-26-2012, 06:59 PM.
                      MPPais

                      Comment


                      • #12
                        Originally posted by burnandreturn
                        Why don't they publish a list of the codecs that work with the cuda. Instead of us having to go through the whole list and test what works. Not a big thing but what does what and when would be a big start for me to actually buy the program.
                        First of all, do you want to use CUDA for filtering, or for encoding?

                        Encoding with CUDA is pretty straightforward; select any MPEG-4 AVC output format and change the video encoder to CUDA instead of x264. You might still get some CPU usage with this, but it should be mostly CUDA.

                        For filtering, it's more about the strain on the hardware and not what codec is being used. The program simply looks at what will be more efficient between the CPU or CUDA and chooses one over the other. CUDA filtering is not limited to certain codecs, so there is no codec list for CUDA filtering.

                        Decoding and encoding IS limited to certain formats, so that can also contribute to the usage %. Decoding is limited to MPEG-1, MPEG-2 and MPEG-4 AVC.

                        Since everyone has different hardware and usage habits, it's hard to guarantee when CUDA will kick in. I typically encode in MPEG-4 AVC (x264) 1080p and usually get over 80% CUDA usage for decoding/filtering. The CPU is a i7-2600 and the GPU is a GeForce GTX 460; the only filters activated were Video Fade-in/out and Audio Fade-in/out.

                        My personal advice to you is not to change your habits just to get CUDA usage. Encode in the format you want/need and if it kicks in, great.

                        A note on optimization: you should run the optimization with your typical work environment. That is, if you are usually browsing the internet or working in Photoshop while you encode, then you should have those programs open when you are running the optimizer.

                        Also note that when Video Mastering Works says "10% CPU usage" it doesn't mean that only 10% of your CPU is being used; it could very well be using 100% of your CPU. It means that the CPU is contributing to 10% of the encoding process. Same with CUDA usage.

                        Comment

                        Working...
                        X
                        😀
                        🥰
                        🤢
                        😎
                        😡
                        👍
                        👎