More on HD, DRM, and CPU usage

More on HD, DRM, and CPU usage

Summary: I've been reporting my experiences with Windows Vista and playback of HD DVD and Blu-ray discs. Several commenters have expressed skepticism over my contention that Windows Vista's DRM didn't come into play at all, and I've also seen some raised eyebrows over my test results involving CPU usage. To put those questions to rest, I dug up an older, slower system without a single HDCP-compatible part. All it takes is a single two-buck part to produce perfect HD playback - at least for now.


This is a very quick follow-up to my previous post documenting my experience with HD DVD and Blu-ray playback on Windows Vista. (If you haven't been following this story, you'll want to read that post and its predecessor, Blu-ray, HD DVD, and Vista to get the proper background.) DRM doom-and-gloomers have tried their best to scare you into thinking that you'll need to scrap your older monitors, video cards, and even HDTVs to play back HD content. They're wrong, as I was able to demonstrate with a two-buck VGA cable.

In the Talkback section, several commenters expressed skepticism over my contention that Windows Vista's DRM didn't come into play at all. Here's one typical comment:

You said that Vista's DRM was not used. If PowerDVD supplied the complete end to end protected pipe, then why did MS add it into the OS? I think you mistake the PowerDVD app displaying the HDCP non compliant warning that Vista supplied to the application as not Vista DRM? (i.e. Vista's monitoring reported to PowerDVD the problem, and PowerDVD displayed the information)

Another commenter thinks my measurements of CPU usage (Blu-ray disks required only 9% CPU on average) were out of line:

The statement regarding CPU usage is complete fancy. HD playback beats the crap out of your CPU. On an AMD Opteron 180/8600GT on an Abit mobo it pegs both cores at 90% on Vista using the XBox 360 drive and on a AMD 6600+/8800GT on an Asus board it hit's 50% across both cores.

Well, there's a very easy way to put both assertions to the test. I pulled the HD DVD/Blu-ray drive out of the system I had been using and plugged it into an older, slower system running Windows XP Media Center Edition 2005. I installed the same copy of PowerDVD Ultra. Neither the monitor nor the video card were HDCP-compatible.

When I tried to play either of the HD discs using a digital (DVI) connection, I was greeted with the exact same HDCP error message I showed in the previous post. The older operating system reported HDCP information to the player software, which in turn decided whether to allow playback. That proves to my satisfaction that Windows Vista isn't involved at all in this playback restriction.

Ah, but that error message says I should try plugging in an analog connection. So I powered down the system and connected the same monitor using a VGA (D-Sub) connector instead. When I started the system back up and tried to play the same HD disc, everything worked just fine. As promised, PowerDVD Ultra pays no attention to HDCP over analog connections.

Now, the monitor I used for these tests is an old 18-inch LCD with a native resolution of 1280 x 1024. As a result, it displayed the HD content in letterbox format, at 1280 x 720 (720p) resolution. Obviously, the results couldn't compare with the output of a 50-inch living room display, but the picture was rich and detailed and it looked great from a reasonable viewing distance. If I had connected it to a larger LCD monitor with a 1920 x 1280 resolution, there's no reason why I shouldn't have gotten full 1080p output.

To measure CPU usage, I ran Performance Monitor as a background task while I played a Blu-ray and HD DVD disc in the foreground. For a video adapter, I used a spare Nvidia 7600GS board I had lying around (similar adapters sell for $80 or so new). That's nowhere near as capable as the 8600 GT I used earlier. The CPU in this system is an AMD Athlon 64 X2 3800+ (2.0 GHz). It's considerably less powerful (and less expensive) than the Intel Core 2 Duo E6600 (2.4 GHz) on the XPS 410 I used for the earlier tests. These benchmarks at Anandtech peg the difference at 30-40%, and that feels about right to me. So how did this lesser system do?

  • On the Blu-ray disc, CPU usage was consistently in the 35-36% range. That's considerably more than the 9% I measured using the other, more powerful PC, but it still leaves plenty of room to do other tasks in the background without overheating.
  • On the HD-DVD disc, CPU usage was in the 50-52% range, compared with approximately 24% for the same disc on the more muscular Core 2 Duo-based system. That still isn't even close to overtaxing the system, though. (And I certainly wouldn't recommend this older system as the centerpiece of a high-definition Media Center.)

I wouldn't dream of trying to do HD playback with an underpowered video card. The latest generation of GPUs from ATI and Nvidia (even those found in relatively inexpensive cards) do an excellent job of offloading decompression from the CPU.

Analog playback has its own set of complications. If you use composite or S-video connectors, you get only SD output, regardless of the source media. A composite component connection works just fine up to 1080i (sorry, no 1080p), but very few video cards offer composite component connections, and adapters cost as much as a new video card. A VGA connection like the one I used here is your best bet. Just about every LCD monitor has this type of connection, although they're not as common on HDTV equipment. And, of course, the entertainment industry has the option to disable or constrain analog output anytime, although it's unlikely to happen for at least another three years, and maybe considerably longer. In hardware terms, that's a long, long time.

Coming up next: Is it worth it?

Topics: Hardware, Microsoft, Processors, Windows

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


Log in or register to join the discussion
  • So answer the question, I was one of the quoted

    [B]f PowerDVD supplied the complete end to end protected pipe, then why did MS add it into the OS? [/B]

    You will also note that I put a ? mark at the end of the statement. If PowerDVD supplies all the DRM, and Vista doesn't actually do anything, doesn't that make all the Vista DRM useless? That is a serious question. If, using PowerDVD eliminates the need for the complete end to end secure pipe, etc, doesn't that make gaining access to the content quite easy?

    As I noted following that...

    [B]See, if Vista is doing nothing, and PowerDVD is doing it all, that makes all the DRM in Vista absolutely useless, bypass the app's protections and then do what you want.[/B]

    Isn't the purpose of the DRM to protect the content? If it truly is as flimsy as the above, and quoting myself again,

    [B]On a side note, can you really play full 1080p on XP? Do you need an HDCP monitor or not? If you can, then I can only chuckle at all the vista DRM effort.[/B]

    If my last statement is true, then I completely stand by the new assertion that the DRM is NOT to protect any content whatsoever, if anyone with XP and an HD drive can just pipe the 1080p to whatever they want, all the protection is beyond useless. Does such a hole really exist? It makes DRM about nothing more than hardware churn. Why is a powerful enough video card allowed to play unprotected 1080p over VGA suddenly obsolete in Vista.

    Again, I will ask, and honestly looking for clarification, what is the DRM in Vista doing when an app that supplies their "own" DRM causes Vista to do nothing to protect the content? I do hope that the DRM is as flimsy as all the above, and what you proved by playing 1080p in XP (OK, 720p, but I agree, no reason to think 1080 wouldn't work) because a Linux player will be trivial to craft to beat the AACS DRM that really doesn't exist.


    P.S. I hope you can forgive me for assuming that the new DRM subsytem to ensure protected content in Vista might actually be used in tandem with the player, if I am wrong, I apologize, and as I said, it will make for fair use workarounds quite easily.
    • Two words

      Cable TV.

      Seriously, I didn't design this subsystem and I don't defend it. But I can tell you that it will be the only way to get CableCARD content or IPTV via a PC.
      Ed Bott
    • In addition...

      I've said all along that Vista's DRM is an option, and I stand by that. Developers can take advantage of the protected medis infrastructure if they want to, or they can build their own support for the restrictions on the media. And we are free to use or not use any of those players.
      Ed Bott
      • That kind of doesn't answer the question...

        If the DRM in vista is optional, then what's the point?

        You say that cablecard will be the one exception. Why so, exactly?
        • It's a platform

          Third parties can write their own apps with their own support for content restriction features. Vista adds support in the OS so that third parties can use that if they want. I'll do a follow-up post to answer this question, which seems to confuse the heck out of everyone.
          Ed Bott
  • Do you have access to a full non protected HD movie?

    I know it wouldn't be exactly apples to apples, but if you could burn an HD movie, say 15 minutes, to a regular DVD and play that through PowerDVD, you could ballpark and quantify the difference in CPU usage between protected and unprotected HD content. Does PowerDVD allow for unprotected to play over DVI to a non HDCP compliant monitor (i.e. does it always enforce HDCP even when it knows the playback is unprotected).

    This could go a LONG way to putting the CPU question to bed. Are 2-3 year old systems just not powerful enough to play HD content regardless of encryption.

    You could also see if any incorrect restrictions are deployed for unprotected.

    Are these protected or unprotected?

    Maybe some of these samples?

    • I'll try

      I have unprotected HD TV shows which can be burned to Blu-ray disc easily enough (just have to get the media).
      Ed Bott
      • Tres Expensif

        That's why I suggested just using a DVD, I figured that the reading stream would be the same CPU, but then again, if you want to make sure no loopholes. I honestly didn't know that you could ask to bypass the DRM, which to me makes it irrelevant. Any app can he hacked to not ask.

        • Indeed it is

          It would hurt tremendously to burn a Blu-ray coaster when the media runs $15-20 each!
          Ed Bott
          • HD on DVD

            As I understand it you can burn HD to a regular DVD and play it using your HD DVD player. Would that be similar enough circumstances that the comparison would be relevant?
          • No, I don't think so

            At that point, all I'm really doing is playing a WMV or MPEG file from DVD instead of from hard disk. I should be able to accomplish the same thing by playing the original file, and all I'm really doing at that point is measuring the CPU load required to decompress and decode that file format.
            Ed Bott
  • cocclution please!

    I'm just frustrated, what do in need for playing 1080p , is Digital chain , Analog chain ( i understood the only VGA is compatible with 1080p)?
    • Either one

      You need an unbroken chain of hardware and software that can deliver 1080p output. For digital, you can use either HDMI or DVI. For analog, you can only use VGA. If the content is unprotected, you don't need HDCP. If it's protected, you need an HDCP-compatible HDMI or DVI output.
      Ed Bott
  • RE: More on HD, DRM, and CPU usage

    Well 2 things

    1) Which HD Movie were you playing, because the Vista Protection Schemes are invoked based on the content providers(The group that produces the HD-DVD), if the movie you selected uses very few of the content protection features then you will see no load.

    2) As far as playing back HD Movies on non HDCP capable systems this is not an issue, it is the quality of the movie, your article does not speak to the quality of the playback ???
    Also content providers can also control whether you get constricted playback(lower resolution).
    • Please read again

      "your article does not speak to the quality of the playback ???

      Yes it does.

      "As a result, it displayed the HD content in letterbox format, at 1280 x 720 (720p) resolution. Obviously, the results couldn?t compare with the output of a 50-inch living room display, but the picture was rich and detailed and it looked great from a reasonable viewing distance."

      Sometimes I don't know why I bother.
      Ed Bott
      • But did it really ??? ....

        "As a result, it displayed the HD content in letterbox format, at 1280 x 720 (720p) resolution."

        And also how did it look (You just describe the mode not the quality).
        • yes, it really did.

          "the picture was rich and detailed and it looked great from a reasonable viewing distance."
        • I give up

          It's like talking to a wall.
          Ed Bott
  • It is possible

    From this link, 1080p over VGA is allowed, unprotected (for now, I am sure the rules will change). Now, your question, did the player/process actually do 720p natively or upconvert after being made 480i in the player, that is a valid question.

    I know at one point, the rule was, if not encrypted, degrade the signal to regular DVD and upconvert, or limit to 520p, then upconvert to 1080 as best you can, but who can keep the rules straight. I could not find any technical details as to what PowerDVD does at their web site. Allowing 1080p over VGA must be a concession to MS for supporting the DRM in Vista and building an HD into their XBox.

    To be honest, I find all of this amazing. A general consumer is going to be so beyond confused and in many cases think they have true HD when something along the chain limits them to 520p. It would look better than regular DVD so they would be "happy".

    I think what we are seeing is the AACS and the MPAA work their hardest to snatch defeat of their formats from any possibility of success.

    • Meant as reply to mrlinux (NT)