Creative's X-Fi Audio Chip Reviewed 336
theraindog writes "The Tech Report has posted an in-depth review of Creative's new X-Fi audio processor. The 51-million transistor chip employs a unique audio ring architecture that pushes an apparent 10,000 MIPS, supports up to 128 hardware-accelerated voices for 3D audio, and can upsample and upmix stereo 16-bit/44.1kHz audio to multichannel 24-bit/96kHz. Creative says that the X-Fi's upsampling and upmixing capabilities can make MP3s sound better than the original CD, and although that claim isn't validated by listening tests, the X-Fi does sound better than other consumer-level audio cards. It also performs better in games, in part because precious few sound cards feature hardware acceleration for 3D audio."
Creative Left Out (Score:5, Insightful)
My point is merely that sound cards provide great sound, but if your not in the Music industry, all the cards sound pretty much the same.
Why? (Score:4, Insightful)
Yes you can get the live drive, but on a media PC that's designed to be on show, it makes sense to have the digital outputs out the back, where they can be easily concealed.
HiteC [ihda.co.kr] do one, as do turtle beach [turtlebeach.com], why not Creative?
Better than a CD? (Score:5, Insightful)
In other news, Creative have created a new image compression standard that makes compressed images "look better than the original uncompressed version". A Creative spokesperson has announced that this compression standard uses the same technology as X-Fi to create information out of thin air.
Seriously, there is no way to make a recording that is compressed by a lossy algorithm such as MP3 sound as good as the original without creating information out of thin air. Of course, X-Fi can't do this, so it must be "guessing" what the original information was. This would of course mean that what you are listening to is just a moderately close approximation of the original recording that has had information added to it to sound "better" (by some Creative engineer's definition of "better").
upmix and upsample? (Score:1, Insightful)
Umm, how exactly does this work? Once you've sampled and quantized a signal you've already thrown out information--you can't get it back. Has there been some new development since I took digital signal processing, or is this simply marketing?
Re:Better than a CD? (Score:3, Insightful)
It's very possible to take even a lossy MP3 and via processing make it "sound better" to the average listener than the more accurate reproduction given by the original CD.
Just like smoothing can make an image look better even though it loses more information.
Of course there's no reason why the same processing couldn't also be applied to the CD output, so claiming it makes MP3s sound better than the original CD is a little silly, but otherwise I don't see a problem with the claim.
Marketing BS on the sample rate (Score:5, Insightful)
There are reasons for 24bit/96kHz, but upsampling just to play it out of a speaker isn't one of them. That's kind of like printing out something at 2400dpi only to scan it back in again at 300. At best, you're going to wind up with exactly the same thing, while at worst you're going to have a bunch of aliasing artifacts from the upsampling.
Upsampling for playback is worthless even if your source material is perfect CD audio. Taking something even worse than that (MP3) and upsampling it is just turd polishing.
Want better sound? Buy better speakers. And a sound card that has high-quality analog components. The digitial part is not the weak part of computer sound playback. Hard to market that, though: "Now with 10db more S/N! And better capacitors!"
24bit/96kHz is good for doing high quality recordings, then manipulating the sound and mixing it. Once that's done there's no point in distributing it in anything better than 16/44.1, if all that's ever done with it after that is playback. If you want your listeners to be able to do their own remixes, that may be another story, but then you have to distribute separate mixer tracks anyway...
Re:Why are there no other contenders? (Score:3, Insightful)
people also used to complain about many issues with it, including hissing, crackling and popping (no not those 3).
what we need is for nvidia or someone else to invest in a high end gaming sound card to compete with the x-fi.... i just don't see that happening. all we have now are semi-pro music cards with eax1/2 (if even that) and relatively bad performance in games.
creative owns way too many patents though... maybe that's why there aren't any competitors. they even owned a patent related to shadows in games (doom3 debacle)
Define Better - The masses don't look for accuracy (Score:5, Insightful)
I was going to make much the same observation and then something occured to me:
Quality of sound is subjective.
It's why every crappy CD player and walkman comes with a Bass Boost. Boosting the bass doesn't make the sound more authentic than the original but, for the average listener with no idea what clear music should sound like, more bass is appealing and a selling gimmick.
Similarly, you upsample, apply smoothing algorithms, apply fractal algorithms, whatever, you may be able to give a perception of clarity, of spacial separation, etc. far in excess of what the original CD had. That doesn't mean it's what the artist and engineers intended, it doesn't mean it's more accurate to the original performance, but you'll still get the average 13 year old telling you that Britney's latest masterpiece sounds even better now.
So, you can make a track sound "better" to an average sampling of listeners without it being more accurate to the environment of the original recording. It's all about their definition of better.
Nice but (Score:3, Insightful)
As others have said, a good set of speakers is really more meaningful these days than the card. Yes, definitely the card can make a huge difference. But the difference between an Audigy 2 and an X-Fi? Not significant enough to warrant a new purchase unless it is a totally new machine.
Which, by the way, I suspect that is where the majority of Creative's revenue comes from, Dell and others who buy their cards in large quantities for their higher end machines.
"Makes MP3 sound better than the CD" ? (Score:5, Insightful)
That's like saying you can make JPEG look better than the uncompressed image. Yes, you can improve the quality of MP3 by careful interpretation of data and perhaps extrapolating information for higher frequencies (which most often suffer from MP3 compression -- MP3Pro [mp3prozone.com] does something similar), but it will NEVER be as crisp and clear as the original material, let alone better.
Not that you'll be able to hear the difference on your $20 desktop speakers you got at the 'Shack anyway.
Re:Marketing BS on the sample rate (Score:3, Insightful)
You could upsample the audio, by interpolating values between the samples (in which case the higher sampling rate would have some effect on the sound), and you could run some high quality filters and such, maybe split the frequencies better for sending to your subwoofer, tweeter, etc... Maybe with the higher sampling rate you can run a better algorithm for removing certain mp3 artifacts. I have also heard that due to the properties of some speakers, it is possible to get frequencies higher than the nyquist frequency by using the right set of frequencies together (I don't think it is true, but I have heard people who know about sound say it is true, so who knows)... maybe having the higher frequencies helps that!
While a professional sound engineer, or some audiophiles might balk at the idea, there ARE things you can do to make the audio "sound better" to a decent slice of people. Think about FM radio stations: The sound is compressed and EQed in such a way that many audophiles think it sucks... but the larger segment of radio listeners love their music sounding like 1980s television.
It might not be YOUR cup of tea, but don't discount the idea outright. There may be something to it. Only your ears will tell the difference.
Re:Creative Bloat (Score:2, Insightful)
Since then, I have never purchased any Creative product, and I probably never will. I make it a habit to only let someone fuck me ONCE.
Re:Creative Left Out (Score:5, Insightful)
nVidia probably backed off SoundStorm because of either implicit or obvious threats from Creative.
In terms of Companies That Are Evil, I'd say Creative ranks right up there with Microsoft. I don't see why we should give them the time of day.
Very different (Score:5, Insightful)
The Creative cards from the Live on up are all DSPs. They are designed to convolute sound. So in a game if they want it to sound like you are in a parking garage, they give the proper commands to the card and it convolutes the sound to do it's best appromation of a parking garage. This leads to both lower CPU usage and more realistic sound than doing the processing all on the CPU.
So the problem is, because of this consumer focus, sacrafices were made. One was that the Lives and Audigys output (and input) only one sample rate: 48kHz (Audigy 2s have 96kHz, but only in special cases). They'll accept any you like under that, but sample rate convert that. They do an ok job, but not great, distortion is introduced that you can see on a scope and hear on good equipment. So they are right out for good recording. Also, they kinda chepskated on the converters for the cards, so they are noisy, compared to others in their price range.
But, for all that, they are real, no-shit, DSPs. If you get the OSS kX drivers (http://kxproject.lugosoft.com/index.php?skip=1 [lugosoft.com]) you can actually write your own assembly programs for the DSP and control what it does.
Now the X-Fi is extremely exciting as it fixes most of the problems people had. For one it has three different modes it can be set in. In pro mode it dispenses with all teh resampling crap and does accurate 1:1 bit capture at any sample rate up to 96kHz. In other modes where it does resample, it does it with a kickass high-order filter that introduces essentially no distortion.
I am unsure if it has the ability to function as a VST plugin built in, but certinaly nothing precludes it from doing so. It's a powerful DSP and has the capability to route sound in and out of it.
So, really, it's not comparable to pro cards. They are designed to do different things mostly. There are some pro cards that feature DSPs, but very few. These days in pro work, the effect processing is done in software. It's more flexable and real time is non-critical. However in a game, you can't dump 20% of your CPU in to doing a single high-quality reverb, so having a DSP is a real boon.
Personally, I use both. I have an M-Audio Firewire 410 for pro, an Audigy 2 for consumer. I imagine that'll become an X-fi very soon here.
Radio Processing? (Score:3, Insightful)
When it comes down to it, you are (were) doing roughly what they do on the radio- trashing the signal. Bob Katz has a great chapter on the whole process in his main book on audio mastering.
The only thing that matters to me on a sound card is the Clocking, Lowpass filtering, and D/A.