Announcement

Collapse
No announcement yet.

nVidia likely to remain accelerated video king?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #21
    Originally posted by MU_Engineer View Post
    The reason your 2.8 GHz P4 is able to play back two 1080p videos with 3-5% CPU utilization is because you are using GPU hardware decode assist. There is absolutely no way you are not using it with getting those CPU usage figures on that CPU. A 2.8 GHz P4 will be running at a very high load playing back one 1080p video encoded in something easy to decode like MPEG-2 and will fail spectacularly at something in a tougher-to-decode codec like H.264.
    He's not testing this, he's just spewing BS. There are a few things that prohibit this:

    -The lack of CPU horsepower required to decode h264
    -The lack of bus bandwidth to accommodate for the decoded video stream
    Also, any P4 running with an AGP card (AGP was far more prevalent in the early-mid P4 era) automatically has zero video decoding capabilities through blobs, as NVIDIA never added VDPAU support for older generation purevideo (7xxx series and lower were the latest cards to run on AGP)

    The only way he could do it would be to use a newer PCI card or PCI Express card on a P4 motherboard that supported it.
    -PCI however doesn't have enough bandwidth to fit DVD resolutions comfortably let alone HD so he's either using GL output (in which case his CPU usage claim is bogus) due to its low bandwidth requirements or VDPAU for the same reason.
    -PCIe is more or less the same story. GL or Xv would produce the same results, with VDPAU being the only solution to grant low usage to the CPU.

    I'd nominate for the "he imagined everything and really has no idea how computationally or bandwith intense video decoding can get" option.

    Originally posted by MU_Engineer View Post
    My far-from-the-latest GTS250 pushes a minimum of ~60 fps in ETQW at maximum detail and AA settings at 1920x1080, god only knows what a high-end card can do.
    I could run Bioshock and Fallout 3 in WINE at 1080p_60 while decoding a 1080p h264 stream on one of my GTX260's. I haven't even been able to properly load my GTX470, but I don't do alot of gaming beyond Linux benchmarking and testing the waters.

    Comment


    • #22
      Originally posted by kazetsukai View Post
      He's not testing this, he's just spewing BS. There are a few things that prohibit this:
      I've done some testing of video decode performance on my system, which is at least as good as any 2.8 GHz P4 system:

      CPU: Socket 939 A64 X2 4200+ (dual-core, faster than a 2.8 P4)
      Chipset: NVIDIA NF4 SLi (CrushK8-04), PCIe 1.0a x16 slots configured as x16/x0
      GPU: NVIDIA GTS250 512 MB
      Test files: several transcoded 720p 15 Mbps H.264 files
      Video player: Mplayer SVN-r29796-4.4.3 (amd64, VDPAU supported)

      I've done some testing and concluded his claims are 100% absolutely impossible since my system is grossly unable to do what he has claimed, and it's at least moderately faster and potentially much faster than his setup.

      -The lack of CPU horsepower required to decode h264
      Absolutely. My X2 4200+ cannot decode the 15 Mbps H.264 files in software (using Xv), and it's a faster CPU operating in a 64-bit environment and also has a second core. No 2.8 GHz P4 is going to be able to do this if the X2 4200+ cannot.

      -The lack of bus bandwidth to accommodate for the decoded video stream
      I'd agree with this as well. I can play one stream with my setup fine, but adding a second leads to a bottleneck and a ton of lag. CPU usage is in the 20% range, so I guess it's probably PCIe bus bandwidth. My setup is equivalent to the best P4 setup as far as bandwidth is concerned. At best, his 2.8 GHz P4 is a P4 620 HT on an i975 platform. That's an x16/x0 or x8/x8 PCIe 1.0a board just like my NF4 SLi is, so bandwidth is at best equal. If he's on AGP, he's running at best half of the bus bandwidth as me (AGP 8x = 2133 MB/sec, PCIe 1.0a x16 = 4000 MB/sec). If he's running an older Northwood-A or Northwood-B on an i845/855 system, he's getting AGP 4x at 1067 MB/sec and going to be even worse off.

      Also, any P4 running with an AGP card (AGP was far more prevalent in the early-mid P4 era) automatically has zero video decoding capabilities through blobs, as NVIDIA never added VDPAU support for older generation purevideo (7xxx series and lower were the latest cards to run on AGP)
      They can use AMD's XvBA video decode assist on HD 3xxx or HD 4xxx-class AGP cards, which has roughly the same capabilities as VDPAU. However, the bus bandwidth still is low compared to my system which is apparently still bottlenecked.

      The only way he could do it would be to use a newer PCI card or PCI Express card on a P4 motherboard that supported it.
      -PCI however doesn't have enough bandwidth to fit DVD resolutions comfortably let alone HD so he's either using GL output (in which case his CPU usage claim is bogus) due to its low bandwidth requirements or VDPAU for the same reason.
      My HTPC runs a PCI GeForce 6200 and I can tell you that the only stuff you're going to be playing smoothly is low-def (480i) video using Xv or OpenGL for video output. There isn't enough bandwidth to run XvMC on the PCI bus for decode assist on low-def video. Playing 720p video is impossible on a PCI-fed card even with Xv output. I'd guess you could probably run one HD stream on an AGP 8x system though.

      -PCIe is more or less the same story. GL or Xv would produce the same results, with VDPAU being the only solution to grant low usage to the CPU.
      I can't vouch for PCIe 2.0 not having enough bandwidth as no system I have has such support and I can't test that possibility. But there aren't any PCIe 2.0-supporting chipsets that I know of that support P4s since Intel removed P4/Pentium D support from the 3-series and 4-series chipsets that do have PCIe 2.0 support.

      I'd nominate for the "he imagined everything and really has no idea how computationally or bandwith intense video decoding can get" option.
      I concur!

      Comment


      • #23
        Originally posted by MU_Engineer View Post
        I can't vouch for PCIe 2.0 not having enough bandwidth as no system I have has such support and I can't test that possibility.
        PCIe (including 2.0) has plenty bandwidth for HD video, I suppose I phrased it incorrectly.

        GL however in my experience seems to have a much smaller bandwidth footprint than Xv.

        Comment


        • #24
          Originally posted by kazetsukai View Post
          -PCI however doesn't have enough bandwidth to fit DVD resolutions comfortably let alone HD so he's either using GL output (in which case his CPU usage claim is bogus) due to its low bandwidth requirements or VDPAU for the same reason.
          Well this is false. PCI does have plenty of bandwidth available for HD content. There are many users out there using vdpau on such a setup on 8400 pci cards or 1x PCI-e cards. Even at the highest bitrate for Bluray (40 mbit or 5 MB) the PCI bus can handle it just fine. Bluray players themselves typically operate on a much slower bus then PCI.

          Comment


          • #25
            Originally posted by deanjo View Post
            Well this is false. PCI does have plenty of bandwidth available for HD content. There are many users out there using vdpau on such a setup on 8400 pci cards or 1x PCI-e cards. Even at the highest bitrate for Bluray (40 mbit or 5 MB) the PCI bus can handle it just fine. Bluray players themselves typically operate on a much slower bus then PCI.
            If they're sending the compressed video stream yes, however with traditional output methods which send a decoded video stream (Xv) this is impossible.

            Comment


            • #26
              Originally posted by kazetsukai View Post
              If they're sending the compressed video stream yes, however with traditional output methods which send a decoded video stream (Xv) this is impossible.
              How's this so?

              PCI bw (bytes/s) / 720p60 bw = 1.26, ie that and some extra.

              (266 * 1 024 * 1 024) / (1 280 * 720 * 4 * 60) = 1.26103704

              Comment


              • #27
                Originally posted by curaga View Post
                How's this so?

                PCI bw (bytes/s) / 720p60 bw = 1.26, ie that and some extra.

                (266 * 1 024 * 1 024) / (1 280 * 720 * 4 * 60) = 1.26103704
                Where did you get the 4 in there? You must be doing some crazy palette work

                Even 24fps 24bit color doesn't fit in PCI.

                1280 * 720 * 24 * 24
                530841600

                Comment


                • #28
                  4 bytes = 32 bits, ie the max color depth usually used.

                  If you use 24 (instead of 3), you'll need to multiply the other side by 8 as well, so the units match up.

                  Comment


                  • #29
                    Well, you're right, the math adds up. However, I've never been able to play 720p video even on a PCIe x1 slot, which has ~250MB/s peak rates.

                    Perhaps in practice the dataset is larger?

                    Comment


                    • #30
                      I think you guys got carried away with the maths and forgot the essential. It's obvious a P4 2.8GHz can't decode 1080p video with only 5% cpu usage :P Unless that particular cpu is overclocked to something like 10GHz. I suppose that would do the trick.

                      Back on topic: Don't the intel drivers support VA-API? How well does that work? It seems to only work with the i965 driver which is unfortunate for netbooks based on 945GM

                      Comment

                      Working...
                      X