In a sign that M-Audio’s Vista support may be trickling in, DJ Grobe observes that the Fast Track USB driver has been posted to the M-Audio site. Hopefully for Vista users, this means more support fast.

Meanwhile, I’ve quite happily “upgraded” back from Vista to XP, and tweaked my XP system so it’s happy. That equation could change down the road, but given that many of my supposedly compatible drivers either weren’t installing at all (hello, odd USB driver quirk), weren’t fully reliable (hello, chipset-related Blue Screens of Death), or were performing like dogs (hello, NVIDIA graphics), I would strongly advise you test-drive your hardware for Vista before committing to the OS. It may work fine, but a “Vista-compatible” label alone won’t tell you for sure.

  • Michael Una

    I'm on site working on graphics for a big trial going on in the federal courts right now (, and one of our designers showed up with his brand-new, souped up Dell with Vista loaded.

    He spent the first week figuring out how to get it to run powerpoint without hanging on every click. He is very much regretting his switch to Vista so early.

    He said, and I quote, "I didn't know I was going to be a beta tester. I have work to do."

  • Mark IJzerman

    Damn you Bill Gates… it's stuff like this that will be the reason I will be walking around with a mac next year…

  • I recall the same being with most of the NT based OS's, including XP (so many 98 users hated the transition).

    I also remember a similar transition during the move from OS9 to OSX.

    Let's not even talk about having no clue what dependency on a linux or bsd box is causing your binary to crap out, and having to compile with everything statically linked just to get a simple net util or audio tool to work.

  • Pingback: links for 2007-04-16 | ITsVISTA()

  • Yes, but Valis, I don't think this is entirely fair.

    Part of the reason the XP transition was tough for Win98 users was that the consumer Win ME launch was a total disaster. (In fact, for Win2000 users, it was relatively smooth.) And I'd argue the XP launch went more smoothly than Vista.

    The switch from OS 9 to OS X was the move to an entirely new operating system. OS X was indeed launched well before it was ready, which was painful beyond description. But I'm glad Apple chose a driver model from the beginning that covered the majority of what was needed for audio and MIDI. Microsoft has resisted that. So the ongoing reliance on ASIO calls into question whether these difficulties — without a resulting benefit in many cases — are worth it, at least for now.

    And what you're saying about Linux is just not a fair comparison for many modern distros. Compiling without a decent app bundler is a pain. But it's apples to oranges, because the upside is that you're using platform-independent code. And that has really nothing to do with audio drivers; that's a separate issue.

    I'm not bashing Vista; I'm just saying these aren't fair comparisons. Anyway, this is overcomplicating the issue. I don't think this is a reason to move to the Mac, though there might be other reasons for that. It's far, far simpler than that. This is a reason to test-drive a Vista upgrade before committing, because results can vary. And I think there are a number of situations why you might "upgrade" to XP even on a new machine.

    I'm really, really happy with XP at the moment on my PCs. I've tuned the OS for my needs. But to me, the software itself (NI, Cakewalk, FL Studio, Applied Acoustics, Cycling '74, etc., etc., thank you for your work) is what matters.

  • Matt

    Vista works for some of us…

    It's actually faster than XP on the same Alienware sytem. Only one driver issue with Akai MPD 16, but other than that it's smooth sailin'

  • Hi Matt — point well taken. I'm not saying Vista is a lemon or avoid Vista at all costs. I'm just suggesting users exercise caution. Yes, that's true with any OS update. It's especially true in this case because of radical changes to the driver model for graphics and audio. And generally speaking, these changes don't necessarily offer improvements for audio and MIDI. Audio depends on a wide variety of variables, but is generally a step forward more for PCI devices than USB and FireWire, and can be canceled out by new bottlenecks like activating Aero, depending on graphics drivers. MIDI is not an improvement at all. And I find it disturbing, personally, that something as simple as an MPD-16 would stop working; this is a fundamentally very simple device. I've spoken at length on CDM about some of the potential advantages (stability that goes along with the new driver model, better scheduling for apps that support it, and WaveRT for PCI and supported apps). And I agree there are usability improvements. But there are clearly risks to updating that users should consider; that means extensive backups, the ability to roll back, and considering dual-booting, even *after* researching driver compatibility.

    I do wonder about the speed improvements on your Alienware system, however. That could depend greatly on what you mean by "speed" — in what area? So far, the overwhelming majority of tests have shown raw performance to be the same or marginally less than XP, even under ideal circumstances. There's very little in Vista that can improve performance under most circumstances; features like ReadyBoost can improve conditions under low memory circumstances, for instance, and there are multi-core improvements. Is this a multi-core machine (i.e., greater than dual)? I believe it's possible — but then, it's also possible to greatly enhance XP performance in many situations by reinstalling XP. 😉

  • Matt

    Yeah, the upgraded audio (Wave RT) in Vista will be used eventually.

    My Alienware system is a 3.4 hyperthreaded Pentium with 2 GB ram. I upgraded the ram close to the Vista migration, so that must be one reason Vista is faster than XP. There are a lot of little changes in the OS that make production faster and easier. Vista drivers are really started to emerge though.

  • Right, but note that none of those enhancements to audio performance make Vista "faster." You can prioritize audio tasks above others, which can be beneficial in specific situations. And because most users rely primarily on ASIO drivers, some of the WaveRT/MMCSS changes won't be noticeable at all.

    Again, this is not to say you won't want to *eventually* move to Vista. But I have some serious questions about where Microsoft's priorities were placed relative to what's important to the music market. There are some subtle improvements that could be beneficial, but only when driver support is more fleshed out. There are other areas, like MIDI and USB device support, in which widespread requests were blatantly ignored.

    Actually, I would really question how useful that Wikipedia article is in general. It appears to be copy-and-pasted from Microsoft documentation. Very little independently-verified information, real world testing, or third-party developer information is included. Divorced of any actual context or real-world performance scenarios, it borders on the misleading. I love Wikipedia, but I think operating systems and performance are too dynamic to be adequately covered by that kind of article. (I.e., I'm not sure it would be possible to fill in the details there within Wikipedia's style manual. Then again, I wouldn't normally WANT to look to an encyclopedia to understand computer performance.)

  • Here are some specific examples:

    "Also, the new Universal Audio Architecture (UAA) model has been introduced, in favor of WDM audio, which allows compliant audio hardware to automatically work under Windows without needing device drivers from the audio hardware vendor."…

    "Windows Audio Session API – Very low level API for rendering audio, render/capture audio streams, adjust volume etc. This API also provides low latency for audio professionals through WaveRT."

    Okay, tons of loaded statements there. First, it implies that this model is "universal" and can "universally" support WaveRT for low-latency applications — not true. WaveRT only supports PCI drivers at this time, and some of those benefits require support for MMCSS for scheduling. Given that Mac and Linux do have truly universal APIs for audio, it'd be easy enough for a customer to misunderstand what this means.

    Second, what the heck does "audio professional" mean? Low latency is low latency. Do consumers prefer higher latencies? Hell, are most musicians professionals? 😉

    "All the existing audio APIs have been re-plumbed and emulated to use these APIs internally, all audio goes through these three APIs, so that most applications "just work."

    Yes and no. Without specific support for the new low-latency API and scheduler, you're in fact just "going through" the plumbing. If that's an app with a few sound effects, yes, it'll "just work." But for music apps, there's more to the story.

    This is obviously not a problem specific to Windows. You get very technical developer information, which is hard to digest, and it tends to just get repeated in theoretical terms. (I've certainly seen that happen with Core Audio on Mac, for instance, wrapped in Apple marketing speak instead of Microsoft developer white paper speak here.) But I think the real-world situations are equally important here.

    Having a second computer could actually be an ideal way to test Vista. You give it a go, and quite frankly, if it's too frustrating, roll back to XP and try again in a few months, keeping an eye on upgrades and (hopefully) a service pack in the meantime.