The X-Fi Audio Ring: Powerful and Flexible

Putting aside marketing buzz words and the slightly tacky Xtreme Fidelity moniker, the new generation of sound cards from Creative Labs is quite impressive. The technological advancements come from the inclusion of four key new features: the ring based architecture, a high quality sample rate converter, a powerful DSP, and on-board RAM. First on the chopping block is the Audio Ring.

Current sound card architectures are based on a linear flow of data. For straight up audio listening or recording, this is not a problem - there isn't any need to deviate from the norm in these cases. And that's why we haven't seen a fundamental change in audio architecture until now. It is when we want to intensively multiplex these components and perform operations on large numbers of audio streams that we run into problems. The backbone of X-Fi is its ability to process any audio stream on any component in the audio chain in any order, and any number of times without having to leave and come back in.

The ring supports up to 4096 internal audio channels that can all be taking different paths through the hardware. Don't confuse these internal channels with the number of voices that the card can handle. The X-Fi is maximally capable of playing up to 127 simultaneous voices. These 4096 channels include channels necessary for advanced filtering and effects processing. The bus is time division multiplexed rather than interrupt driven or otherwise mastered. Time division multiplexing involves handing out the entire bandwidth of the bus to a single channel on a set time interval. The result is that each channel has a deterministic bandwidth and latency - aspects important to effective audio processing. For example, in a household with two teenage daughters, we could time division phone bandwidth by letting one daughter talk during even hours of the day and the other during odd hours. This way, the entire bandwidth of the phone (or audio ring) is fully allocated and you would always know which daughter (or channel) is using the bandwidth at a given time.

Another reason why Creative may have chosen to support so many internal channels is so that it can handle algorithms that call for splitting or duplicating an audio stream for multiple different types of processing and analysis. This could range from using CMSS-3D (Creative's new surround upmixing feature) to efficient implementations of complicated high order N-tap and feedback filters. Creative doesn't talk much about the uses of these internal channels, but many of their new features include detecting aspects of the audio being played in order to enhance the sound. It seems likely to us that the algorithms that implement their new features would rely on the high number of channels and the flexibility of the ring architecture to get as much done at one time as possible in order to keep from introducing an unacceptable amount of latency.

Now that we know how the ring works, here's what each node on the ring does:

Transport: This handles moving audio streams in and out of local RAM as well as over the PCI bus to system RAM. The Transport engine manages over 1000 DMA channels to both the PCI bus and local SDRAM. Being able to manage so many DMA requests at a time is important for dealing with the latency of PCI and effectively managing the volume of data with which the new solution is capable of working.

Tank: Its name is based on the tanks that held mercury used to create high quality analog delay lines. The tank engine is capable of fractional and modulated delay lines (delays that don't line up with the sample pattern and delays that change over time). The tank engine supports up to 1024 accesses per sample. This means that the tank engine can assist in many types of effects including: chorus, reverb, reflections, and interaural time delays (for positional audio effects). These effects are employed quite a bit in many of the audio processing features of the X-Fi, so avoiding the use of a DSP for the creation of a delay line is very helpful in spreading out processing power.

SRC: The Sample Rate Converter seamlessly transforms any audio stream or channel to any other supported sample rate with very low ripple and THD+N. We will cover this node in detail as it is one of the key features of the new architecture, allowing Creative to convert the sample rate of all audio multiple times independently of any other audio with little to no discernable loss in quality.

Filter: The filter engine implements 512 floating point 2nd order IIR filters. This is the fundamental building block of sound synthesis, 3D spatialization, equalizers, speaker calibration, and a host of other features. There are 13 filter types implemented in hardware from direct and parametric 5 parameter EQs to notch and peak filters. Complex (higher order) filtering and synthesis can be achieved by looping through the filter engine multiple times until the desire result is achieved.

Mixer: Consisting of 256 audio summers, 1024 parameter combiners, and 4096 single segment parameter rampers, the mixer is where internal audio channels come together. Parameter mixing and ramping are used to control effects combining for things like 3D audio effects in game. Instead of single segment, the parameter rampers can also support either 1536 multi segment rampers or 4 multi segment shapes (possible shapes are linear, pseudo exponential, and pseudo logarithmic).

DSP: The X-Fi Quartet DSP is so named because it supports 4 hardware threads. Each thread has access to two SIMD (single instruction, multiple data) units for easy stereo and complex math processing. NVIDIA has called a multiple SIMD processing unit MIMD (multiple instruction multiple data), but Creative has dubbed the overall architecture TIMD (Thread Interleaved Multiple Data) due to the hardware threading support alongside traditional methods. The Quartet DSP is central to many of the X-Fi features and we will cover this hardware at length as with the SRC engine.

Audio I/O: This node handles moving the final processed audio streams to output either digitally or through a DAC, and can also acquire input from each of the source options on the hardware. Audio sources in memory do not enter the path through this node, and likewise for audio written to memory (or a file). This block manages all the physical I/O ports on the X-Fi card.

Having easy access to any of these structures at any time during audio processing will greatly simplify the process and increase the complexity of operations possible on X-Fi hardware. In order to further optimize the architecture, Creative has introduced three distinct modes in X-Fi. These modes outline the basic type of path that audio streams will take through the ring. For instance, professional audio programs require very low latency audio from the sound card when recording. In order to accommodate this, the Creation mode limits the types of processing done to less complex (and faster) algorithms. Creation mode also allows synchronized audio streams for proper integration into a studio setting. These features aren't necessary when in Gaming or Entertainment modes, as the focus of these modes is on generating or heavily processing audio before input or output.

Architecting the card like this isn't useful unless the nodes on the ring are powerful enough to exploit the potential. Creative has spent quite a bit of time in making sure that this is the case. Their components are high quality and introduce as little distortion and noise as possible in order to make heavy processing of audio on a consumer sound card a reality. (Whether this is really a good thing or not is still up to the end user.) Over the next couple of sections, we will cover the most important components of the audio ring, as well as explore what is possible when all of this hardware sings in unison.

Index X-Fi Processing Elements: Exploring the SRC
Comments Locked

110 Comments

View All Comments

  • yacoub - Wednesday, August 31, 2005 - link

    As does "it's just bloatware and marketing".
  • Eskimooo - Wednesday, August 31, 2005 - link

    PCIe is designed for graphics and high data transfer, but audio sends very small packets and the overhead can be big. Performance of PCIe is bad for audio, therefore no-one was tempted to come up with a product yet...Correct me if I am wrong.
    Creative says they are working on that, I have just came across this interview:
    http://www.hardwarezone.com/articles/view.php?cid=...">http://www.hardwarezone.com/articles/view.php?cid=...
  • Araemo - Wednesday, August 31, 2005 - link

    From that interview:
    "So what we have to do is go back to the drawing board and work on the transport part of the chip and re-design it to add more silicon to overcome some of the problems we had with PCIe. So for us to come up with a PCIe solution is going to take a while because we have to overcome the problems we're facing with that bus."
    It sounds like they thought it would be a simple transition, and had their product mostly designed(around PCI 2.x) before they tried to adapt it to PCI-Express.

    In order for sound cards to make efficient use of an overcrowded PCI bus(common in older chipsets, before the nforce era or so), they had to pull a LOT of tricks, including using a very low(relatively speaking) IRQ. W/ the PCI-express bus, they have to re-engineer their chips to send data earlier, since it is sent serially instead of parallel, meaning they can't buffer one big clump and expect it to get to the destination all at once, they have to send the first part when it's ready, and then the second part.. sequentially.. I just honestly thought that the delay in the X-Fi WAS to get pci-express working.. It's not like they haven't had plenty of time (What, 3? 4? years since the first working silicon was available for engineers to test on?)

    I think if creative had a serious competitor, they may have been more eager to have pci-express support, but since noone else is even making a dent in their market, they only need to improve over their past product... and that isn't hard to do.
  • Spacecomber - Wednesday, August 31, 2005 - link

    I was a little surprised to see that the X-Fi has a higher level of CPU usage and results in a sligthly lower FPS average when playing Battlefield 2 than what you can get with an Audigy card. I really would have hoped that all that onboard processing power would have taken a significant load off of the CPU, but perhaps this will improve with newer drivers down the road, as you seemed to suggest.

    I wondered if this might not deserve a bit of a closer look. An average of FPS, as you know, doesn't really do a very good job of capturing how well a game runs on particular system. This seems especially to be true with Battlefield 2. For example, when I run a timedemo of this game and then exam the cvs file, which captures the time for each frame rendered, I see that while my system gets an average of 50 fps, the range is from 2-106 fps. If you look at a graph of all these framerates, it looks prety ugly with all those momentary dips into single digits and low teens.

    My wishful thinking is that a sound card that is capable of taking more of the load off of the CPU might help with this. Even if the average frame rate was slightly lower, if it cleaned up some of those framerate dips, it would still be a good thing.

    By the way, how did you disable sound in Battlefield 2 for you tests? Did you simply lower all the sound settings to 0, or is there a command line switch or console command that will do this? I've been meaning to run a timedemo with no sound to compare to a timedemo with sound just to see if the audio contributes significantly to the framerate dips mentioned previously.

    Thanks for this review. As an avid gamer, despite my frustration with Creative's cards, they still seem to be the undisputed gamer's card, simply because no one else fully supports all the 3D effects (such as the latest EAX features) in hardware that Creative does. And certainly, no other game card manufacturer is in the position (that Aureal was at one point) to challenge Creative with a competing set of 3D audio effects and features. Much as I'd like to jump to someone else's solution (I loved my Aureal card while it was supported, and I also was happy with my Santa Cruz when Turtle Beach kept their drivers up to date), I'm skeptical that any one really has the clout, now, to establish themselves as a solid card for game players or that they could survive under Creative's shadow.

    Spacecomber
  • jr9k - Wednesday, August 31, 2005 - link

    [quote]We therefore see slightly lower performance from the X-Fi card. With the X-Fi being brand new, driver improvements could also change the performance picture over time. [\quote]

    Given creative's past record, don't count on them.

    Seems a nice card though, once it is available at 50$
  • ceefka - Wednesday, August 31, 2005 - link

    So if you do run a comprehensive audio test, please also differentiate between Intel and AMD machines and subsequently Intel, nF3 and nF4 chipsets. Also vary on graphics cards. It is said that a higher grade PCI-E graphicscard on nF4 increases latency. Maybe you can shed your own light on that. Please use excessive amounts of audio and processing and push the systems real hard.

    While the X-Fi looks to have some nice tech to it, I am afraid that a lot of its tech will sit unused for years to come while missing a few attractive things like Dolby Digital Live and balanced inputs.
  • Eskimooo - Wednesday, August 31, 2005 - link

    I have read DDL may introduce delays like the DICE did with first XBOX. This whole DDL does not mean cinema quality and does not make a card. DDL is not as goog as full dolby. And it always means compression. So it is just a feature which I can live without. I have good analog set anyway. Balanced inputs are important for those who need them. I presume EMU will come up with something but I do not think I want to wait. X-Fi is the best option for me, thanks for all your comments that helped me a lot. Does anybody know where I can order it in Europe?
  • stmok - Wednesday, August 31, 2005 - link

    I've been using dual CPU (and now dual-core) systems for quite a while, and Creative have dissappointed me time and again with their Windows drivers...Constant "pops and clicks" is always present with the Live and Audigy series. However, this doesn't occur in Linux...For some reason its fine. (I guess its a different driver team working on it).

    So my question is, does this new Creative solution suffer from the same issues as its older ones? (Where you are forced to either try third-party drivers OR disable Hardware Acceleration just to get rid of the "pops and clicks" in games/videos/music)



  • The Blue Moose - Tuesday, August 30, 2005 - link

    Let's assume you actually need a semi-pro sound card. Is Creative really going to be your first choice? At the $400-$500 price point you have a few options. Given Creative's less than stellar track record with drivers and software support, only those who insist on the absolute cleanest sound above all else will jump right into this one. This will be those who can (or at least think they can) hear the difference between Audigy sound and X-Fi sound.

    So, now let's look at the more mainstream version. $130 for a card that uses lesser components (therefore will show even less difference from current Audigy cards), won't have the on board RAM, and doesn't do Dolby encoding. I can get the same sound quality, WITH Dolby encoding for $100 or less right now. Plus the same driver/support issues apply here as well.

    Now, most hardcore gamers aren't gonna pop for a semi-pro card they don't need, nor will they skimp by grabbing the low-end board. I'm assuming the $280 version will have middle of the road components, but it will definitely have the RAM. They won't buy it for cleaner sound necessarily, but more for the potential performance boost of the RAM.

    But, how much of a boost will there be? The (admittedly limited) benchmarks would seem to show there's a good bit of driver overhead. Also, with dual core chips filtering down to the masses, will the RAM actually get you that much more? Many game designers have said how very hard it will be to make games that can effectively use a dual core rig. So, if they can't send graphics related work to the 2nd core, what are they gonna do with it? The most likely candidates are sound, physics, and AI. If you've already got a decent sound card, I think you'd get more milage out of your $280 by dropping it on a 2nd core than a new "faster" sound card.

    While the tech behind this card, certainly seems impressive. The card just seems to be a lot more about hype and clever marketing than any real leap in sound cards. The article mentions plenty of analog and digital I/O, but the picture used doesn't even show a digital output (but it does have a outdated game port, strange). There appears to be a connector for a daughter card, but I don't recall a mention of it. I'm not saying the thing is a piece of crap, but at the prices they're asking, you're paying more for the SB name than the card itself.
  • DerekWilson - Wednesday, August 31, 2005 - link

    oh, and there is a clear audible difference between X-Fi and Audigy cards.

    The coloration from the poor frequency response and IMD sweep at 16-bit 44.1kHz really deadens cd auido and mp3s. Its not about a slightly higher noise floor or a little less dynamic range. It's about poor sound reproduction and bad sample rate conversion.

    I agree with you on the point of your post though -- it is hard to justify the price of this card.

Log in

Don't have an account? Sign up now