Microsoft responds to Vista network performance issue

Microsoft responds to Vista network performance issue

Summary: I have received a response to the Vista network performance issue from Microsoft.

SHARE:

For background to this issue read previous posts: Post 1 | Post 2 | Post 3.

I have received a response to the Vista network performance issue from Microsoft.  Here are some points of interest:

  • "We have been looking into this problem and are working on a doc that will go into the technical details of what we have found."
  • "Please note that some of what we are seeing is expected behavior, and some of it is not.  In certain circumstances Windows Vista will trade off network performance in order to improve multimedia playback.  This is by design."
  • "The connection between media playback and networking is not immediately obvious.  But as you know, the drivers involved in both activities run at extremely high priority.  As a result, the network driver can cause media playback to degrade.  This shows up to the user as things like popping and crackling during audio playback.  Users generally hate this, hence the trade off."
  • "In most cases the user does not notice the impact of this as the decrease in network performance is slight.  Of course some users, especially ones on Gigabit based networks, are seeing a much greater decrease than is expected and that is clearly a problem that we need to address."
  • "Two other things to note.  First, we have not seen any cases where a users internet performance would be degraded, in our tests this issue only shows up with local network operations." 
  • "Second, this trade-off scheme only kicks in on the receive side.  Transmit is not affected."

I've been doing some more research into this and I'm coming to the conclusion that the issue is related somehow to Multimedia Class Scheduler service (MMCSS).  This is a service that makes sure that multimedia applications have prioritized access to CPU resources.  I can't prove my theory because killing MMCSS also disables Windows audio.

As more information is made available, I'll keep you in the loop.

Topics: Windows, Microsoft, Networking

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.

Talkback

65 comments
Log in or register to join the discussion
  • bogus

    i can deminstate the issue effecting transmit and receive using iperf this response is a load of crap
    fix it NOW
    GillianSeed
    • Bafflespeak

      [B]Two other things to note. First, we have not seen any cases where a users internet performance would be degraded, in our tests this issue only shows up with local network operations.[/B]

      This is techospeak psuedo plausible bafflegab. Can anyone tell me what this means in either plain English or correct technical terms?

      They acknowledge a problem exists, but can't find the problem and this only affects local network operations. There is a concise statement if I ever read one.

      TripleII
      TripleII-21189418044173169409978279405827
    • Double bogus

      If Audio was such an expensive CPU hog and needed massive CPU power then it would affect things other than networking, n'est pas?

      It doesn't, so...
      jinko
    • Problem looking for a solution

      Usually it's the other way around. XP is the best of the worst for now. LOL.
      Tin Man
  • I can't believe we have to put up with this in the era of dual cores

    "?The connection between media playback and networking is not immediately obvious. But as you know, the drivers involved in both activities run at extremely high priority. As a result, the network driver can cause media playback to degrade."

    I can't believe we have to put up with this in the era of dual core and quad core computers. Slap the network driver on one CPU core and put the audio playback on another core and problem solved. But even single core CPUs are so fast that this shouldn't ever be a problem even if audio playback gets priority over network-related CPU usage. It's not like network-related CPU consumption uses more than 50% CPU on a modern dual-core processor even when throughput hits 500 mbps. There?s just no excuse for this.
    georgeou
    • See George criticizes microsoft!

      I agree w/ you George. I do not understand the need for a tradeoff. Besides, I thought Vista's networking stack was completely rewritten to be more efficient.
      otaddy
      • Of course he does

        But it appears to be his position that there's nothing wrong enough with MS' latest and greatest to justify not buying or using it.

        George: Please correct me if I'm wrong.
        John L. Ries
        • Be reasonable John

          "But it appears to be his position that there's nothing wrong enough with MS' latest and greatest to justify not buying or using it."

          In the context of this bug, I'd say any rational person would take that position. Vista's been out for 6 months and this is the first we're hearing about this bug? Obviously not a showstopper.

          That's not to say that other issues might prevent a migration, but this just isn't it.
          rtk
          • It's definitely annoying and it's probably not a show stopper

            It's definitely annoying and it's probably not a show stopper. Let's wait till Microsoft finishes responding. I don't care for their first response but let's give it a chance.

            I would go as far as saying that if Microsoft absolutely refuses to fix this problem that I'm going to have a problem recommending Vista but it's way to early to say that. This bug is going to feed the DRM conspiracy theorists for sure and Microsoft needs to fix this bug fast.
            georgeou
          • Not the first time

            "Vista's been out for 6 months and this is the first we're hearing about this bug? Obviously not a showstopper."

            It's not the first we're hearing of it (try searching google for "vista networking slow" or something similar) but it's the first time somebody's figured out that it's related to sound playback.
            jinko
          • ignoring the obvious

            There was a patch released quite a while ago that addressed "the network problem" as it was viewed at the time, and it was widely reported by many to fix "the network problem".

            Mixing this bug with the previous bug is hardly fair.
            rtk
          • Given my own experience with Vista

            I would never spend any of my own money on it (even if MS' abusive business practices were to be abandoned tomorrow) or recommend that anyone else use it, but that's just me. Others can reach their own conclusions and mine are subject to revision, based on further evidence.

            George has clearly reached a different conclusion, which is his right.
            John L. Ries
    • Isn't it odd that XP doesn't have this problem?

      There is some logic to the response that Microsoft has offered, but all of those points were just as valid for Windows XP. Vista must need a lot more power to play back audio/video than XP did. Maybe (just maybe) the overhead of the new Vista DRM scheme is the source of the issue.

      George Ou is right, Vista should run the data transfer on one core and run the media player on a second core when possible.
      WiredGuy
      • No it doesn't need more horse power to play audio/video

        No it doesn't need more horse power to play audio/video. In fact, I ran a test last night firing up ten instances of Windows Media Player Classic running ten separate video files. I could see all ten videos running smoothly and the ten audio files mixed perfectly and I could hear conversations from all ten videos smoothly.

        This is definitely a nasty feature turned bug if I?ve ever seen one.
        georgeou
    • amen

      I had never, ever experienced popping or music degredation in XP while transferring files on the lan. I cant believe MS would give such a bulls*** answer.

      I have a Core2 Quad oc'd to 3.2ghz, 4 gigs of ram, an nvidia 8800gtx, and gigabit ethernet. According to task manager, playing an mp3 on this hardware doesn't even register as raising cpu usage at all, it stays at 1% on all four cores! So if my old 800mhz athlon can play music and transfer files at full speed on XP, there is NO EXCUSE modern hardware shouldnt be able to do it with ease in Vista.
      ronjeremy_69@...
      • Don't even need that much power

        The single-core Celeron I use in my home office appears to handle both the company VPN and whatever music CDs or MP3s I want to throw at it just fine, thank you. As noted in a post on a different subject, I dual boot Slackware Linux 12 and Windows 2000 (works equally well either way).

        The official explanation doesn't make me less likely to trust MS, not more, as it doesn't seem to gibe with reality.
        John L. Ries
    • ...or even on hyperthreaded chips

      Is anybody running Vista on less than one of those old hyperthreaded chips...?
      jinko
    • Is this the same George...

      ...who does not understand single-core vs. multi-core?

      http://blogs.zdnet.com/Ou/?p=675&page=2
      Burana
  • I'm still standing by my theory

    that this is DRM related. Ou is right. There is NO reason that multimedia playback and network activity should be sucking so many CPU cycles one has to be throttled at the expense of the other. Unless something more than simple multimedia playback and network transfers are happening (read: DRM checking).
    frgough
    • The network driver stack is all new in Vista.

      The network driver stack is all new in Vista. It's a feature that's gone out of control and it needs to be fixed.

      I ran a test last night firing up ten instances of Windows Media Player Classic running ten separate video files. I could see all ten videos running smoothly and the ten audio files mixed perfectly and I could hear conversations from all ten videos smoothly.
      georgeou