Testing Creative X-Fi cards: a new audio revolution. Ultrasound


new generation of processors and sound cards Xtreme Fidelity

What is Xtreme Fidelity?

On August 2, 2005, Creative held a conference in Germany where the new generation of sound cards - SoundBlaster X-Fi - were demonstrated to the European press.

Earlier, on May 13, 2005, Creative announced the Xtreme Fidelity technology, abbreviated as X-Fi, and the sound processor of the same name.


If it’s more or less clear with the new processor, then what is “Xtreme Fidelity technology”? According to the manufacturer, its essence is to provide 24-bit quality, sound clarity with at least 110 dB signal/noise and new CMSS (Creative Multi Speaker Surround) technology for 3D sound in headphones and multi-channel speakers. As stated by Sim Wong Hoo, founder and chief executive of Creative “Xtreme Fidelity is the result of our vision for a new audio standard that can bring dramatic improvements to MP3 playback, PC gameplay and digital movies, thereby laying the foundation for a whole new digital home entertainment platform.”.

Speaking about Xtreme Fidelity, the manufacturer places special emphasis on two new technologies: 24-bit Crystalizer and CMSS-3D. According to Creative's vision, the technologies are designed to remaster, enhance and remix in real time existing conventional and compressed 16-bit stereo material into 24-bit quality, with multi-channel sound in speakers of any configuration, as well as headphones.

To provide the capabilities of the new standard, Creative is creating the X-Fi processor under the motto "Xtreme Audio Needs An Xtreme Fidelity Processor!"(“exceptional sound requires the highest quality processor”).

Retrospective

Previous sound cards, starting with Live! model 1998, were based on modifications of the EMU10K processor. This processor was designed primarily for use in E-MU synthesizers and modules as a hardware MIDI synthesizer with a built-in effects processor. At a time when the main requirement for a sound card was the presence of MIDI hardware with the largest memory bank, wide capabilities and high-quality effects, the EMU10K was a progressive chip. However, at present, as everyone understands, it is morally outdated. Despite 32-bit audio processing, programmable high-quality professional effects, support for 8 digital I2S or SPDIF interfaces, shortcomings (the main one being the rigid architecture) reduced everything to nothing.

Since the development of the EMU10K, the interests of sound card users have changed significantly. In fact, the audience was divided into three camps: 1) office and other users who are satisfied even with integrated sound; 2) gamers who are not particularly interested in quality and who do not mind good 3D sound in games as long as it does not compromise performance; 3) music lovers and audiophiles who use a computer for a convenient music library; they need a card exclusively for high-quality music playback.

However, lovers of high-quality sound expressed distrust in Creative products, even despite the high-quality I2S Cirrus Logic DACs installed in the latest models. The main "horror story" destroying the image of Live/Audigy sound cards for many years was the presence non-switchable mediocre hardware resampling of all frequencies to a reference frequency of 48 kHz. To obtain high-quality playback of the 44.1 kHz format, in which most recordings exist, card users began installing DirectSound or ASIO plug-ins for software sample rate conversion (SSRC software sample rate conversion) for popular mp3 players Winamp and Foobar.


SRC Audigy vs. SSRC Winamp

Starting with Audigy2, the developers introduced a P16V block into the chip, allowing cards to directly play 96 and 192 kHz frequencies for DVD-Audio playback. But this mode turned out to be available only for these frequencies and only through the DirectSound interface. The effects processor running at 48 kHz had to be disabled in this case. Otherwise, oversampling comes into play again.

In the low-end below, Creative begins to catastrophically lose the sound card market due to the expansion of AC"97 and HDA codecs forcedly installed on motherboards. This greatly reduces the incentive for users to spend money on a separate sound card. In addition, the onset of the Internet and file-sharing networks makes the computer a means accumulation of a collection of files in compressed formats.So, for high-quality listening to music, many prefer semi-professional cards on Envy24, with a minimum of hardware capabilities, but with a choice of a fixed sampling frequency, an honest 44.1 kHz in digital and a minimum of unnecessary bells and whistles.

Obviously, Creative needed a new processor that would not have the disadvantages of the previous one. X-Fi has been in development for the last 5 years. Since the release of the first Audigy model, work has begun on a new generation processor. Development location Creative Advanced Technology Center, Silicon Valley, California.

X-Fi architecture

According to the manufacturer, the new Xtreme Fidelity processor has more than 51 million transistors and performance of more than 10,000 MIPS (million instructions per second). X-Fi is also 24 times more powerful than its predecessor.

The heart of the processor is the "Quartet" DSP, as there are 4 subprocessors operating in parallel. The architecture is called TIMD (Thread Interleaved Multiple Data). To achieve parallelism, a processor architecture with very long instruction words is used ( Very Large Instruction Word (VLIW).

Such a command combines several ordinary commands that are executed simultaneously (in parallel) by different functional blocks of the processor to increase its performance. The performance increase is achieved due to the fact that the processor does not need to spend time organizing parallelism at the instruction level. Grouping of concurrently executed operations is performed by the compiler.)

Each subprocessor has a 2xSIMD architecture. 2x comes from the fact that the subprocessor usually operates with two data streams - stereo data, complex numbers. By the way, both floating-point and fixed-point data types are used. The instruction set consists of 235 operation codes, and about 60 specialized instructions often found in frequency audio processing algorithms, for example, complex data multiplication from two streams, decimal logarithm, exponent.

The internal processor frequency is 400 MHz. External synchronization is supported for S/PDIF and I2S. The jitter value of the processor's PLL (PLL) is given. Sound processing algorithms (reverb, equalizer, 3D sound propagation modeling effects) use 512 IIR filters ( IIR, infinite impulse response = IIR, infinite impulse response) second order. This will not surprise anyone in modern audio processing tools, but it should be taken into account that everything here works in real time for dozens of threads simultaneously. The number of hardware filters includes 4 and 5 band parametric equalizers, bandpass, notch and other specific filters.

There are also rich mixing capabilities - up to 4096 signals with hardware volume control and combining two sources. This can be used both to build a mixer that is flexibly configured in terms of signal routing, and is also used in games, since it requires adjusting the level of each applied effect.

To connect to digital and analog interfaces, the processor has 4 I2S input/output buses, providing 8 input and 8 output channels. In addition, there is a proprietary protocol for transmitting 8 channels on one wire, thus resulting in 32 x 48 kHz, 16 x 96 kHz, 8 x 192 kHz channels. For internal needs and routing there are 4096 audio and 4096 parametric channels.

SRC

Sample Rate Converters (SRCs) are used when the sample frequency does not match the current mode's reference frequency (otherwise the sample will play at the wrong speed and change the tone). 256 interpolators are used, with quality higher than a typical 100th order polyphase FIR filter (FIR, finite impulse response = FIR, finite impulse response). The tone shift range is from 0 to 8.


SRC X-Fi operation diagram

The picture illustrates the work of SRC X-Fi to convert 44.1 kHz to 48 kHz. First, the signal samples are doubled to 88.2 kHz. The polyphase FIR filter converts signals up to 192 kHz at twice the 48/44.1 ratio. In the final stage, the frequency is reduced to 48 kHz. This scheme is more efficient in terms of computational costs and gives a better result, since at the last stage aliasing does not form due to non-multiple frequencies.

When converting a standard 997 Hz tone from 44.1 kHz to 48 kHz, the standard parameter responsible for noise and distortion power is THD+N equal to -136 dB, frequency jitter in the passband ±0.00025 dB. Creative recommends checking these parameters, at a minimum, using an AudioPrecision measuring station.

Considering the headroom of 6 dB and 32 stages of the processing cycle, THD+N can deteriorate to –124 dB, and frequency response unevenness increases to ±0.01 dB. But the parameters of even the best DACs today are inferior to such indicators in terms of noise and distortion.


Performance quality of SRC X-Fi when converting 44.1->48 kHz


It should only be taken into account that quality parameters in the frequency domain do not indicate the operation of filters in the time domain. But, given the positioning of the cards primarily for games and films, such conversion accuracy can be considered excellent. In previous products, intermodulation distortion from SRC reached 0.1% in the high-frequency region, while only a few users complained about a drop in quality.

There is a very important point that is not much advertised by the manufacturer - unlike previous generation sound processors, In X-Fi you can turn off oversampling along with other DSP effects! In this case, of course, the unique opportunity to simultaneously listen to several signals of different sampling frequencies, with the subsequent application of all sorts of effects and enhancers, is lost. However, fans of the highest quality sound need this least of all.

SRC converters also operate in DMA mode, allowing conversions directly into the computer's RAM without the participation of the CPU. To avoid collisions when the PCI bus is busy, there is a configurable cache.

X-Fi capabilities

Data processing for human head and ear transfer functions (HRTFs) and headphone audio calculations is based on several technologies. These are developments by UCDavis, Aureal and Sensaura. Patents for MacroFX and binaural 3D panning are used. The implementation uses 48-tap FIR (FIR) filters and processes up to 128 3D audio sources.

CMSS (Creative Multi Speaker Surround) technology has been upgraded to CMSS-3D. The manufacturer claims that the sound field quality is higher than that of Dolby Prologic II/IIx, and in headphones the spatiality and naturalness of HF tones is superior to Dolby Headphone.

Unlike previous generations of processors, effects are now supported for 24/96 and 24/192 modes. To process signals of higher resolution, for example, 192 kHz, Quadrature Mirror Filter technology is used - dividing the signal into 4 frequency bands of 48 kHz, which are processed separately, and the results are added together. Thus, when processing 192 kHz, the calculation speed increases compared to conventional methods.

The chip's architecture allows switching between modes without rebooting, so sound cards can make the most of their resources.

Supports ASIO 2.0 with latency For WDM and OpenAL, drivers and hardware have been optimized to obtain the lowest latency and higher sound quality.

Comparison

The table shows approximate performance ratios between different generations of Creative cards (according to the manufacturer).

Prod. calculated
MIPs		Total production
MIPs		Int. channelsMIPS vs Live!Number of effectsNumber
transistors
Sound Blaster Pro≈1 3+ - 0.0001x- 100K
AWE 32 (EMU8000)67 200+ - 0.2x- 500K
Live! (10k1)335 1000+ 16
(effects)		1x1 2M
Audigy (10k2)424 1250+
64
(effects)		4x4
4.6M
X-Fi10340 30000+ 4096
(Total)		67x
8 51.1M

X-Fi has a completely different architecture from previous generations of Creative processors, as well as all other processors, including graphics. The concept is a programmable ring-based architecture. This gives flexibility in signal routing, allowing any of the 4096 audio channels to redirect the signal to any processor element - Tank, SRC, Mixer, DSP.



Thus, the new X-Fi architecture consists of the following main innovations:

  • "Ring" architecture
  • New DSP "Quartet"
  • Ultra-quality SRC
  • Hardware acceleration Mixer, Tank Engine (effects), Filter Engine, DMA.
  • Simultaneous development and interaction of hardware and software parts

A huge amount of information about the X-Fi processor can be found on a special website.

Sound Blaster X-Fi sound cards

On August 8, 2005, Creative announced a new line of Sound Blaster X-Fi sound cards based on the processor of the same name. 4 card models will be produced: Sound Blaster X-Fi Elite Pro, Sound Blaster X-Fi Fatal1ty FPS, Sound Blaster X-Fi Platinum, Sound Blaster X-Fi XtremeMusic.

The Sound Blaster X-Fi Elite Pro features a professional-quality 116 dB S/N DAC, plus an external interface module with 24-bit Crystalizer, X-Fi CMSS-3D, 3DMIDI and EAX control buttons. The card is equipped with 64 MB of on-board X-RAM sound memory, as well as microphone and guitar preamps. Estimated cost US$399.99.

The Sound Blaster X-Fi Fatal1ty FPS is designed to satisfy the needs of renowned gamer Jonathan "Fatal1ty" Wendel. The card has 109 dB signal/noise, 64 MB of onboard X-RAM. The kit includes a switching block for a 5" bay and an IR remote control. A PCI card with lower quality DACs. Cost US$279.99.


Sound Blaster X-Fi Fatal1ty FPS

Sound Blaster X-Fi Platinum. Same as Fatal1ty, but PCI card without onboard memory. US$199.99

Sound Blaster X-Fi XtremeMusic. The simplest board option. Platinum without switching block. US$129.99
By the way, even for this card there is a mention of the possibility of ordering an IR receiver on the USB bus and a remote control.

In this article we will look at the most interesting model - Sound Blaster X-Fi Elite Pro. A distinctive feature of the card is maximum quality and performance, plus a huge external switching unit with a remote control.

Sound Blaster X-Fi Elite Pro


Sound Blaster X-Fi Elite Pro


The card itself, labeled SB0550, has gold-plated connectors and a mysterious rectangular piece in the top corner of the board. In fact, it turned out that this is an LED with the inscription X-Fi, which lights up blue when turned on and will be clearly visible in translucent modder cases.


Sound Blaster X-Fi PCI Card (SB0550) from the Elite Pro Kit


Indeed, the phonogram is remastered using a “widely known in narrow circles” mastering plug-in called “multi-band compressor.” To be 100% sure I was right, I compared the Crystalizer with three mastering multiband compressors: Waves LinMB, iZotope Ozone, Steinberg MultiBand Compressor. When you turn on these plugins, the character of the sound changes in a similar way to turning on Crystalizer.

To be 200% sure, I tested Crystalizer with our RMAA program and compared it with the results Waves LinMB Linear Phase MultiBand compressor(with standard preset Low-Level Enhancer- I inform you for those who want to repeat the measurements). Tests were conducted with the Crystalizer's default setting, in the middle position.

Pay attention to the equalization of the frequency response and large intermodulation distortions - according to measurements, the nature of the impact of the plug-ins is similar. However, the situation is much better by ear. This is due to the fact that the nature of the music signal differs from the test signal. According to psychoacoustics, an overload of less than 6 ms is not detected by a person, and a high-quality compressor does not allow long-term overload, quickly reducing the signal level, which is regulated by the attack time.

You can find an analogy for Crystalizer technology in the world of digital photography. Suppose a certain plugin appears that converts a JPEG obtained by compression from a certain high-quality TIF into a 16-bit color format, then tightens the levels (roughly speaking, increases the brightness/contrast) and sharpness by 10-20%, depending on the source. Ready! We get a new ingenious “48-bit Picturelizer” that makes JPEGs better than the original TIFs. Attention, question: in which case will the quality be better? Correct, only if JPEG has a quality reserve for such abuse, and the display device has problems transmitting the lightest and darkest halftones of the original image.

In sound, the situation with MP3 and WAV is approximately the same, but there are also differences. Mastering studios use very expensive equipment that analyzes the source file not in real time, but with some anticipation or even more than one pass, using calculation accuracy not of 24 bits, but of 64 bits with floating point. Next, mastering compression parameters must be painstakingly adjusted individually for each recording on high-quality mid- and far-field reference monitors. Entrusting a similar function to an automatic machine in real time and talking about superiority over the studio result is presumptuous. At the final stage of mastering, the stage of forming a psychoacoustic distribution of dithering noise (noise shaping) takes place, which audibly turns a 16-bit recording into a 20-bit one, so all that remains is to send the recording bit by bit to the input interface of the DACs. Any interference in this process has a detrimental effect on the quality, and there is no way to surpass the original phonogram from an MP3 recording, and multi-band compression can distort the tonal balance beyond recognition and lead to a redistribution of signal energy, which will cause overload at high or low frequencies.

It’s another matter if we’re talking about mediocre household acoustics with a limited dynamic range and defective frequency response, so that even the most primitive tone control subjectively significantly increases the comfort of perception. In this case, additional band compression of high and low frequencies makes it easier to perceive the phonogram through these specific speakers, especially in combination with a moderately compressed original phonogram, with less aggressiveness (ratio of peak power to average value). In this case, the resulting distortions from processing will not exceed the distortions of the reproducing path.

This is how things worked in practice. On inexpensive active stereo speakers, some recordings subjectively sounded more intelligible, although with more aggression. On high-quality acoustics (studio monitors and mid-class Hi-Fi speakers), I easily found several MP3s, as well as original CDs, the quality of which was significantly degraded by Crystalizer, so that overload was audible, or the sound was made too aggressive, which led to rapid fatigue from listening.

During the measurements, it turned out that in addition to multiband compression of the signal, the level is raised by about 3 dB. So any quiet recordings will seem subjectively better even without a compressor.

Thus, the 24-bit Crystalizer will benefit owners of inexpensive acoustics or budget headphones, which almost automatically means a lack of lows and highs, as well as problems with midrange detail. For owners of high-quality acoustics, the good news is that this technology can be easily turned off.

The 24-bit Crystalizer technology has a right to life, but the presentation itself, with wishful thinking, is not at all pleasing. In reality, Crystalizer does not expand, but narrows the dynamic range, and 24 bits are indeed used, but only so that rounding errors do not accumulate (this is normal practice; no modern DSP operates at the same resolution as the original data).

СMSS-3D

CMSS-3D technology is designed to solve the main problem of the younger generation of users of multi-channel sound cards: “I have a 5.1 set of speakers. Everything is great in movies, but only 2 front speakers play MP3! And I want everything!!!”

As mentioned above, CMSS (Creative Multi Speaker Surround) technology has been upgraded to CMSS-3D. Developments and patents of IRCAM, UC Davis, Aureal, Sensaura and Creative's Advanced Technology Center are used.

CMSS-3D virtualizes a sound source of any channel to any device, thus providing:

  • CMSS-3DHeadphone: surround sound in headphones;
  • CMSS-3DVirtual: surround sound in stereo speakers;
  • CMSS-3DSurround: converting stereo to multi-channel audio;
  • CMSS-3DInteractive: 3D audio from multiple sources.

Testing was carried out using Sennheiser HD600 headphones. The problem of double overlay of virtualization effects was identified, with the deterioration of the volumetric effect. The fact is that all musical recordings of recent years already contain, to one degree or another, an expansion of the stereo panorama, so the result may be unexpected.

At the same time, the demo file "The Rising of the Sun" from the "Essential Anuna" CD, containing almost raw stereo, showed good results, and even outperformed DolbyHeadphones technology.

In addition to music, CMSS-3D technology can also be used in games. In this case, the game calculates three-dimensional sound on 5.1 speakers, and CMSS-3DHeadphone converts multi-channel audio into surround sound in headphones.

X-RAM

Unlike cards from previous decades, X-Fi's onboard memory is not used for MIDI samples, but for storing or caching sounds while gaming applications are running. There is a problem with a shortage of RAM, so developers either reduce the bit depth of samples to 8 bits 11 kHz, which degrades the quality by an order of magnitude, or compress them into lossy formats, MP3 or OGG, which leads to CPU load when unpacking at the stage of playing the sample.

Thus, the introduction of on-board memory on a sound card allows you to: save main memory, speed up memory access, eliminate the need for real-time decompression, eliminate loss of sample quality, increase graphics performance and FPS.

Today there is only one game with full support for X-RAM. This is the UT2004 X-Fi Edition, which is not yet available to a wide audience. We invite you to review the results provided by Creative. As you can see, the introduction of X-RAM can increase performance by up to 20%, which is equivalent to upgrading the graphics accelerator, and the sound quality will also increase! Definitely a useful innovation.


UT2004 X-Fi Edition

Tests in RMAA 5.5

We can't resist publicly quoting Creative's manual for journalists testing X-Fi.

"As stated, Creative designed the Sound Blaster X-Fi family with sound quality in mind. To achieve this goal, we took testing very seriously. However, we recognize that not all journalists have access to Audio Precision", the industry standard for testing professional audio equipment. We see many people using "RightMark Audio Analyzer" for audio quality tests due to its accessibility and attractive interface. However, due to the difference in methodologies between the two platforms, their results will vary. In the past, we've received some questions about these differences, so we want to show the prepared results and test procedures for both Audio Precision and RMAA. We hope you find something useful in them."

Instructions from Creative for installing X-Fi cards for RMAA measurements can be downloaded from the official website.

Audio Precision Test Results

Sound Blaster X-Fi
XtremeMusic/
Platinum/
Fatal1ty FPS		Remark
Digital PlaybackFrequency ResponseLo –1dB Hi –1dB ≈ 45kHz
N.B. "Digital Playback" refers to the fact that only the quality of the line-out is tested using a digital source file. It is not a test of the digital outputs
Cross-talkL-R: -105 dB
R-L: -105dB
Signal to Noise Ratio (SNR)109 dB
0.004 %
Line-IN / Line-OUT
Record & Playback		Frequency ResponseLo –1dB ≈ ~17 Hz
Hi –1dB ≈ 42 KHz
Output: 2Vrms
Cross-talkL-R: -86 dB
R-L: -86 dB
Signal to Noise Ratio (SNR)98 dB
Total Harmonic Distortion + Noise (THD+N)0.004%

Sound Blaster X-Fi Elite ProRemark
Digital PlaybackFrequency ResponseLo –1dB Hi –1dB ≈ 46kHzPCM: 24-bit/96kHz, 997Hz Output: 2Vrms
N.B. "Digital Playback" refers to the fact that only the quality of the line-out is tested using a digital source file. It is not a test of the digital outputs.
Cross-talkL-R: -112 dB
R-L: -112dB
Signal to Noise Ratio (SNR)116 dB
Total Harmonic Distortion + Noise (THD+N)0.0008 %
Line-IN / Line-OUT
Record & Playback		Frequency ResponseLo –1dB Hi –1dB ≈ 45 kHzPCM: 24-bit/96kHz, 997 Hz Input: 2Vrms
Output: 2Vrms
Cross-talkL-R: -106 dB
R-L: -106 dB
Signal to Noise Ratio (SNR)112 dB
Total Harmonic Distortion + Noise (THD+N)0.001%

RMAA 5.5 test results

Sound Blaster X-Fi Elite Pro

SB0550SB0550SB0550SB0550
+0.01, -0.07 +0.01, -0.07 +0.01, -0.07 +0.01, -0.07
Noise level, dB (A):-94.8 -95.1 -113.0 -113.3
Dynamic range, dB (A):94.7 95.1 112.8 112.5
THD, %:0.0009 0.0009 0.0007 0.0007
IMD + Noise, %:0.0051 0.0049 0.0010 0.0010
Stereo crosstalk, dB:-95.1 -95.5 -102.7 -102.8

Sound Blaster X-Fi XtremeMusic
Sound Blaster X-Fi Platinum
Sound Blaster X-Fi Fatal1ty FPS

TestSB0460SB0460SB0460SB0460
Frequency response (from 40 Hz to 15 kHz), dB: +0.02, -0.08 +0.01, -0.09 +0.01, -0.09 +0.02, -0.17
Noise level, dB (A): -94.1 -94.5 -102.1 -102.2
Dynamic range, dB (A): 94.0 94.3 101.6 102.0
THD, %: 0.0009 0.0009 0.0008 0.0008
IMD + Noise, %: 0.0057 0.0054 0.0026 0.0025
Stereo crosstalk, dB: -97.0 -94.5 -101.5 -98.6

Comparing the test results of RMAA loopback vs. AP loopback, I would like to note that the discrepancy is a few percent, which is due to the difference in calculation algorithms. In the latest version 5.5, maximum compliance with existing international standards has been maintained (while RMAA uses more accurate calculations than required by the standard). Considering the kilo-buck price of Audio Precision and the free nature of RMAA, we can fully recommend a program for checking the passport data declared by the manufacturer. The THD+N value should be taken from the detailed RMAA report.

Of the tests, only the distortion graphs are of interest.


THD cards SB0550


IMD (DIN) cards SB0550


IMD (CCIF) from frequency SB0550


IMD (CCIF) from frequency SB0460

Judging by the measurements, the problem with a lot of distortion is a thing of the past. The 44.1 kHz mode in X-Fi cards is no different from 48 kHz.

Hardware SRC quality test

It is interesting to compare the quality of the new hardware SRC X-Fi against the common real-time SSRC MP3 player plugin WinAmp, which is of fairly high quality and puts a decent load on the CPU.


IMD (CCIF) vs. frequency


IMD (DIN)

Reference 16/44 - analysis of the source data file generated by the test; shows the maximum achievable quality for a 16-bit format.
X-Fi HW SRC OFF - disabling hardware SRC for X-Fi is achieved by setting the reference mode to 44.1 kHz and the Enable Bit-Matched Playback option.
X-Fi HW SRC ON - enabling hardware SRC in X-Fi is achieved by playing a 44.1 kHz file with the reference mode set to 48 kHz and disabling Enable Bit-Matched Playback.

WinAmp SSRC plug-in - playing a 44.1 kHz file in WinAmp with the SSRC plug-in installed, configured to convert to 48 kHz, while the card has a 48 kHz reference mode and the Enable Bit-Matched Playback option.

As we can see, the SRC of the card is superior to the SSRC of the plugin, and there are no distortions visible in the diagram from its inclusion.

RightMark 3DSound


The diagram is deliberately built on a scale of 100% to show that the difference, even several times, with a small number of buffers is insignificant, and enabling EAX in 3D mode is practically free.

The OpenAL API is ahead of DirectSound due to more direct access to hardware and optimization of the complete library CT_OAL.DLL (Creative OpenAL Driver, 5.12.1.1141) exclusively for X-Fi. The exception is 2D mode. But, in this mode, OpenAL most likely will not be used, and the speed there is unimportant. Most likely, OpenAL's 2D is 3D with fixed coordinates. The download is the same:

Games

As mentioned, Creative had the opportunity to test X-Fi on a special version of UT2004 with support for 128 voices. I can’t help but quote the results:

  • Realtek HD Audio: 7.1 Audio, EAX 2.0, 32 Voices
  • Sound Blaster Live!: 5.1 Audio, EAX 2.0, 32 Voices
  • Sound Blaster Audigy: 7.1 Audio. EAX 3.0, 32 & 64 Voices, CMSS-3D
  • Sound Blaster X-Fi: 7.1 Audio, EAX 3.0, 32, 64 &128 Voices, X-Fi CMSS-3D, X-Fi 24-bit Crystalizer

So, in UT2004 the difference between HDAudio and X-Fi at 32 voices reaches 17%, with better sound quality and CMSS-3D and 24-bit Crystalizer enabled. The computer configuration is quite average: P4-3.4, 1 GB, GeForce 6600.

Among real games, direct support for X-Fi is currently only available in Battlefield2 (even in the demo version!), implemented via OpenAL. If you have a card, in the sound options you can select a separate item Creative X-Fi, EAX and Ultra High sound settings. In this case, on-board memory is used to cache samples and speed up sound processing. It must be said that the samples in this game are mostly of average quality, many are 22 kHz, plus they are compressed in OGG with a variable bitrate of 100 kbps. So with the X-Fi option selected and EAX activated, the sound deserves a "Very Good" rating, but not "Excellent" due to the terrible quality of the samples. The sound in Doom3 with patch 1.3, which includes EAX4, is much more interesting.

We carried out our measurements in Battlefield2 using a hacker mod. The system used was P4 3.4 GHz, 1 GB DDR400, ATI X800, graphics settings 800x600, Medium. Creative X-Fi + EAX ON + Ultra High mode gave 52 FPS, compared to Hardware + EAX OFF + Medium mode, which showed 55 FPS. Thus, the difference in performance between modes with maximum and medium settings does not exceed 5%, for which we must thank the card’s fast and powerful processor. And this is with such video settings, when the performance is not yet limited by the video card. In gaming resolutions, you can safely set the sound to maximum.

Based on the results of testing a real product for a new processor
Creative X-Fi award is given website Original Design

To be continued…


24.09.2014 17:05

Creative Sound Blaster X-Fi HD is no longer a new product on the market, but this is one of the brightest attempts of this company to please the already completely inexperienced audiophile public.

An external sound card is definitely a specific product. Which means automatically audiophile. Because the average user is unlikely to bother with the search for the right device to play any kind of music, and even more so for scoring acoustic scenes, for example, in computer games.

A buyer who decides to organize a mini-studio at his workplace approaches the choice of a device for reproducing incoming and outgoing signals purely from subjective preferences. The technical characteristics of sound recording/playback, external ports for connecting acoustics and auxiliary devices, for example, musical instruments or a microphone, are also important here. Thus, there are quite a lot of subtleties in a device called an external sound card.

Using factory drivers and without the use of bundled or third-party software to enhance and change the sound, the Creative Sound Blaster X-Fi HD sounds really good.

Creative Sound Blaster X-Fi HD is no longer a new product on the market, but this is one of the brightest attempts of this company to please the already completely inexperienced audiophile public. By and large, this device can implement two main functions - reproduce very high-quality sound, and also record it using one input connector ¼ jack format located on the front panel black box.

The main characteristics of the Creative Sound Blaster X-Fi HD stereo card are as follows: signal-to-noise ratio - 114 dB, the X-Fi processor already known to many (from the gaming series) is installed inside, the playback and recording quality is 24 bit/96 kHz, which is quite good for home needs. Headphone Gain - 330 Ohm.

In addition to the sounded instrument jack located on the front panel of the Creative Sound Blaster X-Fi HD, there is another headphone output with exactly the same form factor. On the back side of the device there are analogue tulips(two pairs), as well as TOSLINK optical input and output. The sound card is connected via the connector USB 2.0. Creative didn’t forget about grounding the device, which is very nice.

The true power of Creative Sound Blaster X-Fi HD hidden in the control panel, which is installed from the disk or the manufacturer’s website.

During operation, a blue diode notifies the user that the device is operational. On the front panel of the Creative Sound Blaster X-Fi HD there is also a volume control in the style of classic receivers. Many users criticize it, but in vain. Perhaps it is too miniature, but the device itself is quite modest in size. In any case, it’s convenient to control the volume; the smooth and moderately tight movement of the plastic handle is created just for this.

Using factory drivers and without the use of bundled or third-party software to enhance and change the sound, the Creative Sound Blaster X-Fi HD sounds really good. To be honest, after using the external Behringer U-Phoria UM2 card, no difference was noticed (passive Dual speakers from Germany were used as an acoustic pair, connected to a Telefunken RA200 amplifier). The volume reserve is more than enough, the frequencies of all ranges are excellent.

But the true capabilities of the Creative Sound Blaster X-Fi HD hidden in the control panel, which is installed from the disk or the manufacturer’s website. It contains various EAX effects, equalizer, THX TruStudio Pro functions. The panel is not overloaded with options, which means any user can adjust the sound at their discretion. It's really worth it. The familiar sound of your favorite compositions is literally updated with every turn of the slider of one or another parameter.

Creative Sound Blaster X-Fi HD does not support ASIO, which means recording, as well as subsequent playback of the track, occurs with a significant delay.

Another significant one plus Creative Sound Blaster X-Fi HD consists of a soldered instrument input, with which you can record sound from musical instruments, such as a guitar, as well as from a full-fledged microphone. The sound card copes with this task very well and the output is audible The quality of the track is not studio quality, of course, but sufficient for personal experiments. But there is one drawback - the Creative Sound Blaster X-Fi HD does not support ASIO, which means recording, as well as subsequent playback of the track, occurs with a significant delay. Thus, when overlaying one track on another, especially when playing the first one in parallel, difficulties may arise in the workflow. This is the disadvantage of any non-professional solution; after all, the Creative Sound Blaster X-Fi HD is an amateur device.

Creative Sound Blaster X-Fi HD can be purchased for 3700 rubles and higher. By and large, this is a worthy offer for the capabilities that the device under review offers the user, if we are talking about non-professional external cards for the home.

For beginning musicians who want to constantly evaluate the quality of their work, as they say on film, this device is also suitable, but with the reservations described above, which are associated with a large recording delay. In any case, when recording exclusively one track in mono or stereo mode, there will essentially be no problems.

And the sound quality reproduced through the analog inputs is truly excellent, without any exaggeration. You just need to get the appropriate speakers, preferably more powerful ones.

Two USB sound cards with high-quality DACs

In today's material we compare two interesting sound cards: Creative Sound Blaster X-Fi HD vs. Asus Xonar U7. These are two products that are close in price and capabilities, in which the emphasis was placed on high-quality converters. Moreover, both cards at the time of writing were on sale at a very low price, a little over 3 thousand rubles.

On the forum, the Asus Xonar U7 was called the “optimal choice” in absentia, based on a review of photographs and specifications. Many people completely in vain forget about the need to directly compare devices by sound. Most are also unable to correctly interpret measurement results.

Name

  • Creative Sound Blaster X-Fi HD
  • Asus Xonar U7

The Creative card has a confusing name, as it is not related to the X-Fi cards and is not related to the Titanium HD card. As a result, the card's release went unnoticed. This product is not known at all to anyone except some computer audio enthusiasts.

Asus Xonar U7 also has an unclear name. Quite old cards U1 and U3 are budget products with integrated sound quality. U7 has no relation to previous products in this series.

Price

  • Creative: 3700 rub.
  • Asus: 3100 rub.

The cards cost almost the same, the difference in absolute values ​​in price is not serious. More importantly, such an affordable price makes them interesting to a wide range of buyers. They have virtually no competitors among USB cards!

At the same time, the Asus card appeared quite recently. The manufacturer also reports that the Asus Xonar U7 Echelon Edition model will soon appear with an army camouflage paint job and more functional drivers.

The target audience

  • Creative: mass, audiophiles
  • Asus: mass, home cinema

The Creative model is a card with full-size TRS headphone and microphone connectors. Interestingly, there is a Phono input for digitizing vinyl, with grounding. It also works in normal linear input mode. THX and other proprietary processing are software only. For movies, there is an optical digital output for connecting a 5.1 receiver with a decoder.

You can connect 7.1 analog speakers to the Asus card using minijack connectors. The emphasis is on the included Dolby software. Digital output - combined, optical/coaxial. Both cards have a line input, but in the case of Asus the line input is combined with a microphone input, which is quite difficult to guess.

It remains a mystery why it was necessary to load both cards with so many functions and software in addition to a high-quality output. This inevitably blurs the audience and distracts from the main advantage of the cards, namely high-quality DAC. All the other dregs are available in abundance in almost any sound card. In the description of sound cards and in the design of the boxes, as they say, “you can’t see the forest for the trees.” It is not clear who will organize a 7.1 cinema around a computer or laptop - this is inconvenient and in practice unclaimed. Fortunately, the user does not overpay much for all these additional functions, since the price of the cards is very low.

Volume knob

  • Creative: Miniature metal volume knob with original location. There is a Mute function when you press the knob. Using the handle is not very convenient, since it is located very close to the surface of the table: the fingers are located in a very unusual position and slip over the surface.
  • Asus: The smooth handle is recessed into the body. Pressing the knob turns off the speakers in favor of the headphones and vice versa. It is inconvenient to use: considerable effort is required, otherwise the finger will slip. The recess for the finger is made for beauty - it is much easier to adjust by pressing your finger where there is no recess.

In both cases, a non-stop digital regulator is used. The ergonomics of the regulators are very conditional. Due to the combination of the functions of a regulator and a button, the Asus regulator plays quite noticeably. Many users would prefer to remove the button, but so that there is no backlash. Also, Asus's recessed dial makes it easy to carry the audio interface. In all other cases, this decision looks controversial.

Illuminations

  • Creative: blue LED on top panel
  • Asus: 3 signal LEDs on the top panel

The activity indicator on the Creative card lights up brighter. In principle, for most this will not be a problem. But we had a strong desire to seal it with duct tape so that it would not be an eyesore and would not be constantly visible in peripheral vision.

Headphone output location

  • Creative: front
  • Asus: front

Both cards have convenient connectors.

Filling

  • Creative: DSP Creative CA0189-2AG, 2-channel DAC AKM AK4396 (123 dB), 2-channel ADC CS5361 (114 dB).
  • Asus: DSP CM6632A, 2-channel DAC Cirrus Logic CS4398 (120 dB), 6-channel DAC Cirrus Logic CS4362A (114 dB), 2-channel ADC CS5361 (114 dB).

The contents of both cards are very interesting. And this is the main feature of the cards, against which everything else no longer interests anyone. The Cirrus Logic CS4398 DAC is considered a higher class than the AKM AK4396.

All Creative USB sound cards feature a proprietary CA0189-2AG RISC processor with a frequency of 100 MHz. Unlike the usual Creative DSPs, this chip does not have hardware audio processing. Its purpose is a fast USB 2.0 controller and a multi-channel interface to full-fledged I²S/I²C converter chips, such as those used by Cirrus Logic and AKM. There are two crystal oscillators soldered on the board.

Asus uses an interesting CM6632A chip, which supports both HDA codecs and I²S/I²C converters. Number of channels - 8+2 (analog + digital) output and 2+2 input. There is support for 192 kHz frequency. The chip can be clocked from three generators at once, 12, 24.5 and 22.5 MHz, which is what the creators of the card used. The decision to throw pasta from one end of the board with the DAC to the other with connectors remains not very clear.

All processing in both cards is carried out exclusively at the software level. The maximum data rate of 96 kHz isn't much of an issue since 192 kHz content is virtually non-existent. Hi-Res did not give anything to ordinary users and did not reach the masses, and aesthetes and Hi-End lovers are unlikely to be interested in sound cards for $100.

Auto switch to headphones

  • Creative: There is
  • Asus: There is

When plugged into the headphone jack, the speakers are turned off. This feature was first implemented in integrated audio, as part of the AC'97 and HDA standard. First, the detection of connected headphones was carried out by impedance, and then by a mechanical sensor. External cards with good converters are not inferior to integrated sound in this regard. However, in practice this can cause problems.

Bundled software

  • Creative: there is a set of software on the disk, a driver update on the manufacturer’s website
  • Asus: the same installation kit on CD and on the manufacturer’s website

The software disc is the same as all other Creative products.

The appearance and contents of the Entertainment Console of the X-Fi card have already been described more than once in our materials. The Audio Control Panel comes included right away and allows you to control absolutely all settings without the back and forth of a graphical menu, with the exception of the equalizer. Creative Volume Panel - it is not clear why the program is needed, it is supplied for assortment. An important difference between drivers for XP: they provide a choice of frequency and bit depth directly from the Entertainment Console. The frequency is limited to a choice of 48 or 96 kHz, regardless of the OS. This choice is typical for all Creative USB cards.



The Asus map panel is designed very strangely. Parameters can only be selected through the context menu with the right mouse button!

There is an interesting option for setting the headphone output volume limit.

In the jungle of settings, Asus has a separate option - Dolby Home Theater v4, which seems to have surpassed the number of “improvers” of Creative cards.

44 kHz support

  • Creative: No
  • Asus: There is

The Creative card contains two crystal oscillators and also has 44 kHz support at the DSP level. However, drivers for neither Windows XP nor Windows 7/8 do not support operation in 44 kHz mode: in the device settings you can only set 48 kHz or 96 kHz. In a card for audiophiles, this indicates either inadequacy or complete incompetence of the developers. To listen to music, 99.9% of cases use a frequency of 44 kHz. We are looking forward to comments from the map developers themselves.

The Asus card supports all frequencies: 44, 48, 88, 96, 192 kHz.

Availability of ASIO driver

  • Creative: No
  • Asus: There is

The lack of an ASIO driver for Creative cards is beyond our understanding. The best listening option for a Creative card is the WASAPI interface and software resampling in the player. Conversion from 44 kHz to 48 or 96 will always be forced by the Windows sound engine, or you need to select the conversion on the fly in the media player. For example, in the foobar 2000 player:

Sound from speakers

Listening was carried out on active monitors Adam S2.5A and Adam ARTist 5. For comparison, the very well-known sound cards E-MU 1616M (CS4398) and E-MU 0204 USB (AK4396) were used.

First surprise: all the cards play completely differently! In first place in terms of sound quality is the E-MU 1616M card, in second place is the E-MU 0204 USB, in third place is the Creative Sound Blaster X-Fi HD, and in fourth place is the Asus Xonar U7.

The sound of the Asus card is least similar to the E-MU 1616M (despite the installed identical DACs). The problem with distortion at Asus is to blame, since we know for sure that the 1616M card has an unrivaled analog output implementation. This includes a measured 120 dB signal/noise, a measured 0.0003% distortion, and a wonderful clear sound experience. The 1616M is not much inferior in sound to the best implementations of rack-mounted studio devices, comparable to Pro Tools HD and Lynx Aurora, so we confidently choose the 1616M sound as a reference for devices in the category under $150. E-MU 0204 USB is inferior quite a bit. X-Fi HD sounds brighter and more accentuated on the high frequencies, therefore, despite all its effectiveness, it is a little further from the original sound. The Xonar U7 really disappointed us with its poor highs and rather sluggish sound. Perhaps this problem can be solved by flashing the firmware or drivers; We will definitely clarify this issue with the manufacturer. In any case, there should not be 0.01% distortion. Both the digital and analog parts of the card are capable of more.

Let's also criticize Creative for the lack of a 44 kHz mode. This is simply inexplicable, nonsense! How can you make an audiophile card, install two crystal oscillators and prohibit it from operating at 44 kHz? Don't provide the user with an ASIO driver that bypasses Windows? This is simply the wildest incompetence of the developers. The E-MU 0204 USB - a device for the same money - has no trace of such problems! There is a feeling that one division of the Creative company is mercilessly competing with another: persistently developing a bicycle with square wheels, when in the next room a racing sports car has long been developed and tested.

Sound in headphones

We used high-impedance headphones Sennheiser HD600 (300 Ohm) and Beyerdynamic DT990 Pro (250 Ohm).

The volume is sufficient in both cases, but the Asus card has practically no headroom. Creative was able to surprise us once again. In terms of sound quality, the places were distributed as follows: in first place is E-MU 1616M, in second place is Sound Blaster X-Fi HD, in third place is E-MU 0204 USB, and in fourth place is Asus Xonar U7. The professional interface E-MU 0204 USB, in principle, performed well, but X-Fi HD looked like a clear favorite against its background and had a significantly larger volume headroom. The Asus Xonar U7 card is the quietest, but its volume increases sharply when processing is turned on. Perhaps this was done intentionally so that there was a so-called headroom for Dolby technologies and other auto-equalizers. Thus, the Asus card is the least suitable for simply listening to music; its emphasis is placed precisely on the effectiveness of processing. The squeals and hoots of the card fans remain completely incomprehensible to us.

Measurements in RMAA




Tests at the RMAA gave an "Excellent" rating. This means that the devices have no obvious flaws. It doesn't mean anything anymore. (But as soon as artificial intelligence makes human measurement analysis unnecessary, we'll be the first to let you know!)

Both cards were tested through their own path, as they have the same high-quality CS5361 ADC.

Asus is slightly better in noise, and Creative is slightly better in distortion. We were concerned about the appearance of the Asus graph in the floating tone intermodulation test: usually anomalies in this test indicate frequency conversion or some other digital processing. We rechecked all the settings several times. The results do not change. We conducted additional tests on Windows XP and Windows 7 that were completely installed. We used the latest drivers and the latest firmware. There is only one result: the Asus card has about 10 times more distortion at high frequencies.

Creative X-Fi Xtreme Audio Notebook is the only sound card with an ExpressCard interface. Let's see how good it is compared to integrated sound.

It's no secret that integrated sound cards are usually not of outstanding quality. Unfortunately, this is true not only for desktop computers, but also for laptops. In particular, in my personal laptop Fujitsu Siemens ESPRIMO M9400 the headphone output is very noisy. In low-impedance and sensitive headphones (for example, such as the Creative EP630 and Philips SHE9500 we have already reviewed), interference caused by the operation of the hard drive and graphics adapter can be easily heard.

At the same time, I quite often use my laptop to listen to music through headphones, and this situation naturally did not suit me. Unlike desktops, where you can install any sound card from dozens of existing models, laptops have limited expansion options. Therefore, when choosing a sound card for use with a laptop computer, you have to focus primarily on external solutions with a USB interface. However, they are usually quite large, and many also require external power, which makes it impossible to carry them with you all the time.

Some professional audio manufacturers - such as Echo or EMU - make PCMCIA sound cards, but my laptop (like most modern laptops) only has an ExpressCard slot. This is precisely the reason for purchasing the Creative X-Fi Xtreme Audio Notebook sound card, which is the only sound card in the world designed for installation in the ExpressCard slot. So, meet the Creative X-Fi Xtreme Audio Notebook.

Delivery set, appearance, interfaces

The card comes in a small box with a traditional Creative design.

In addition to the sound card itself, the box contains a traditional set of waste paper in five languages, a terrible quality wired headset, a driver and a very nice carrying case.

The Creative X-Fi Xtreme Audio Notebook is almost completely hidden in the ExpessCard slot; only the “head” remains outside, on which there are three connectors: output, input and a connector for connecting the so-called Speaker Docking Module. This very module is necessary for analog connection of a sound card to 7.1 or 5.1 format acoustics. It is not included in the package, and the possibility of purchasing it in the CIS countries is questionable.

The card's stereo output is combined with a digital optical output, and the input (to which you can connect both a microphone and linear sources) is combined with a digital optical input. Thus, the owner of a laptop X-Fi receives a complete set of digital audio interfaces. In my opinion, when watching movies with a multi-channel soundtrack in Dolby Digital and DTS formats, it makes sense to output the sound to an external receiver via the optical output. The Creative X-Fi Xtreme Audio Notebook copes with this task brilliantly.

Functions

Despite the name, the X-Fi Xtreme Audio Notebook does not include the Creative X-Fi chip (EMU20K1). First of all, this means that the card is poorly suited for recreating a three-dimensional sound environment in computer games, but this does not matter for a laptop, since mobile computers (even the most powerful) are not intended for games. Much more important is that X-Fi Xtreme Audio has higher quality digital-to-analog converters (DACs) than the built-in sound card. Creative doesn't mention the components used anywhere, but as far as I was able to understand while scouring the Internet, the XFXA uses a CS4382A DAC manufactured by the highly respected company Crystal Semiconductor and a WM8775 ADC manufactured by Wolfson Microelectronics. If the DAC level is quite comparable to household Hi-Fi equipment, then the ADC is much inferior to it in quality (this theory will subsequently be confirmed by measurements). The sound card supports two proprietary Creative technologies - CMSS-3D and X-Fi Crystalizer. The first of them makes the sound in headphones pseudo-three-dimensional and is absolutely useless in real life. The second technology is supposed to significantly improve the quality of compressed audio (such as MP3), we will discuss it a little later.

Drivers and software

The sound card was tested on a Fujitsu Siemens ESPRIMO M9400 laptop (Core 2 Duo T5250 processor, 3 GB of RAM, Intel GM965 chipset) running the Windows Vista Ultimate operating system. I would like to point out right away that the drivers supplied on the included disk are simply terrible. Firstly, they are very outdated, and secondly, their installation procedure takes no less than 1 hour and 15 minutes. I recommend not even unpacking this disk, but immediately downloading the new version of drivers from the Creative website. The only caveat is that the version of PowerDVD recorded on the included disc cannot be installed separately. If you need it, you will have to first install all the bundled software, and then update the driver to the latest version. In general, Creative is not very willing to write drivers for its devices; new versions are usually released every six months or less.

Addition from 01/16/2009: After installing Windows Vista SP1, problems began to arise with the sound card - for example, periodic “stuck” sound. As of January 2009, Creative has not yet released a driver version that is fully compatible with Vista SP1. In addition to the drivers themselves, you will need the Creative Console Launcher utility, since some sound card settings (for example, the analog input operating mode) can only be changed with its help. Unlike other X-Fi sound cards, the Xtreme Audio Notebook is limited to Entertainment Mode only. There are no Game Mode or Audio Creation Mode. The card does not support the ASIO interface and therefore cannot be used for professional audio work.

It is worth noting that X-Fi Xtreme Audio works well with the sound card integrated into the laptop, which allows you to perform various interesting tricks, for example, two people listening to different music on one laptop.

Measurements

It so happened that just while I was writing this material, I briefly came across the professional external audio interface EMU 0202. I took this opportunity to take some measurements of the analog path of the Creative X-Fi Xtreme Audio Notebook sound card using the well-known RightMark Audio Analyzer package.

Operating mode: 16 bit 44 kHz

Noise level, dB (A)
Dynamic range, dB (A)
THD, %
THD + Noise, dB (A)
IMD + Noise, %
Stereo crosstalk, dB
IMD at 10 kHz, %
General performance

Excellent

As you can see, the X-Fi Xtreme Audio Notebook shows excellent playback quality even in 16-bit 44 kHz mode, traditionally the most difficult for non-professional Creative sound cards. Let's check whether the sound card suffers from forced oversampling at 48 kHz:

Creative X-Fi Xtreme Audio Notebook Line-Out

Operating mode: 16 bit 48 kHz

Frequency response (from 40 Hz to 15 kHz), dB
Noise level, dB (A)
Dynamic range, dB (A)
THD, %
THD + Noise, dB (A)
IMD + Noise, %
Stereo crosstalk, dB
IMD at 10 kHz, %
General performance

Excellent

Judging by the fact that the values ​​of intermodulation distortion have almost halved, forced oversampling is taking place. It is of quite high quality, but in any case, when listening to music from a computer, it is better to use the SSRC plugin for your favorite player. Now let's check the quality in external loop mode by connecting the line output of the X-Fi Xtreme Audio Notebook to its line input (16 bit 48 kHz):

Frequency response (from 40 Hz to 15 kHz), dB
Noise level, dB (A)
Dynamic range, dB (A)
THD, %
THD + Noise, dB (A)
IMD + Noise, %
Stereo crosstalk, dB
IMD at 10 kHz, %
General performance

Very good

As I expected, the input of X-Fi Xtreme Audio is significantly inferior in quality to the output. And now the deadly number - measuring the quality of the Realtek HD Audio audio codec integrated into the laptop. Since I had the EMU 0202 at my disposal for literally 20 minutes, I had to test the integrated sound using the X-Fi line input. Despite this, Realtek fully lived up to the expectations placed on it. :)

Realtek HD Audio Line-Out

Operating mode: 16 bit 48 kHz

Frequency response (from 40 Hz to 15 kHz), dB
Noise level, dB (A)
Dynamic range, dB (A)
THD, %
THD + Noise, dB (A)
IMD + Noise, %
Stereo crosstalk, dB
IMD at 10 kHz, %
General performance

The advantage of X-Fi over integrated sound is obvious. At the same time, I would like to note the undoubted progress of built-in audio cards: Realtek HD Audio is at least as good in quality as Creative SB Live sound cards! ten years ago.

Sound quality (subjective listening)

Having connected headphones to the audio output of the Creative X-Fi Xtreme Audio Notebook sound card, I was pleasantly surprised... by the silence. No hissing, no buzzing, no “sounds of space” or other annoying interference - excellent! Compared to an integrated sound card, X-Fi produces much clearer and more detailed sound. The timbres of instruments are transmitted more accurately. For example, when listening to Astor Piazzola's Five Tango Sensations performed by the Kronos Quartet (Nonesuch Records, recorded 1991), the string instruments began to sound more like themselves. The increased resolution also had a good effect on the sound of complex multi-layered rock compositions: on the album Yes - Close To The Edge, the instruments stopped falling into mush.

Overall, the Creative X-Fi Xtreme Audio Notebook sounds brighter and more detailed than the Philips CD604 desktop CD player (which uses two TDA1543A DACs in parallel). However, the Philips CD604 has a sound that is less dry, more exciting and emotional. It is also worth noting that when using headphones, Creative X-Fi provides a higher volume level than the integrated sound card.

The X-Fi Crystalizer function turned out to be unexpectedly useful. When it is turned on, the level of high and low frequencies rises, plus significant compression of the dynamic range occurs (the audio card artificially increases the volume of quiet sounds). When listening to music, this only spoils the impression, but watching films with a quiet soundtrack becomes much less tiring: you don’t have to listen hard all the time to make out the words of the characters. The effect is especially noticeable when using computer plastic Cheburashka speakers.

Bottom line

The Creative X-Fi Xtreme Audio Notebook sound card has a lot of objective disadvantages, including the lack of a cable for connecting to multi-channel speakers, weak driver support from the manufacturer, unsuitability for professional work with sound, low quality line input and a fairly high cost (450 hryvnia or 90 dollars). In my opinion, almost all of these shortcomings are crossed out by the fact that the X-Fi Xtreme Audio Notebook is the only sound card in the world with an ExpressCard interface. At the same time, its analog output really provides very good sound quality. If you want high-quality sound from your laptop, but can't carry an external USB audio interface with you everywhere, you simply have no other choice. Although, for my part, I have to say that if the headphone output on my laptop was less noisy, I would not even consider the Creative X-Fi Xtreme Audio Notebook as a purchase.

Instructions for setting maximum sound quality on a Creative X-Fi sound card.

2015-08-18T08:43

2015-0408-18T08:43

Audiophile's Software

Note: The information in this article is relevant for the Creative X-Fi XtremeGamer sound card with a PCI interface (as well as similar XtremeMusic, Fatality, etc.) and Windows operating systems from Vista to 10. The settings are focused on sound playback using a sound connected to the output cards. When using interfaces/devices/drivers/programs other than those specified in this article, the settings may not lead to the desired result.

Copyright (C) 2015, Taras Kovrijenko

Full or partial copying of the text is permitted.

1. Introduction

Note: Since I would not like to repeat myself, and also so that you do not miss something important, I highly recommend reading this article from beginning to end.

The first thing I want to say is that the sound card is worth the money. Considering its capabilities (which many probably don’t know about), $100 is a very reasonable price. In addition, other manufacturers simply do not have analogs of this card (in terms of multimedia capabilities).

2. Preparing for installation

This means that we have, as they say, “out of the box”. This is actually the sound card itself, a pack of manuals and other waste paper, and two disks: one for Windows XP, the second for Windows Vista. "What should Windows 7 and 8 users do?" - you ask. “Take these two disks and carefully put them in the trash,” I will answer. Well, at the very least, put it back in the box with the waste paper and throw it away in the closet. Simply because nothing from that disk will be needed. You can, of course, install software from these disks, but the point is that drivers, for example, will have to be installed for XP, and software for Windows Vista. In this case, you will have to resort to a number of tricks (like compatibility modes, etc.). And it’s not a fact that in the end all this will work properly. Well, in addition, a lot of unnecessary garbage is installed from the disk.

Therefore, we will go the other way, downloading in one package the latest versions of the software, already tailored for the new OS - say thank you to Daniel Kawakami.

3. Installing drivers and utilities

So, this is what it should look like initially:

Moreover, in the Start->Programs folder (and in Add/Remove Programs) there should be nothing related to Creative. If this is not the case for you, run Add/Remove Programs and clean it. After removing all software from Creative, unpack the Support Pack, go to the \Audio\Drivers\SBXF subfolder and run the Setup.exe file. Having selected everything as in the screenshot, click OK:

Having gotten rid of the Creative software and having received the device without drivers in the manager, we reboot again (just in case).

Now let's move on to the installation. It is advisable to install the following components from the X-Fi Support Pack:

  • Audio Drivers - latest versions of sound card drivers;
  • Audio Control Panel - provides basic options for setting up a sound card;
  • Console Launcher - control center for sound card capabilities;
  • Speaker Setup Console (THX) - a utility for calibrating 3D effects in accordance with the listener’s position relative to sound sources;
  • Volume Panel is a convenient utility that starts automatically and allows you to adjust the volume directly in the tray and launch other utilities from there;
  • ALchemy is a utility that restores hardware environment effects in Windows Vista+.

We respond positively to the proposal to reboot.

If everything went well, after starting the OS, a silver icon with a volume knob will appear in the tray, and in the start you will see a group of Creative programs:

Well, of course, the system should already make some sounds when starting up (if they are turned on).

4. Setting up the Windows mixer

As I already said, Microsoft, starting with Windows Vista, has significantly complicated the sound subsystem. At the same time, not only the functionality has expanded, but also some pitfalls have appeared.

Well, let's try to get around these stones.

Go to Windows Control Panel->Hardware and Sound->Sound. The following window will appear:

Through the context menu, disable SPDIF (since this device in some cases can cause switching of the reference frequency of the sound card, which is undesirable), in the same way make our device (Creative SB X-Fi Speakers) the default device (if it is not is). Then double click on the device. Now check that the settings on the tabs match the screenshots:

5. Sound card modes

In order to make maximum use of X-Fi DSP resources for a wide variety of tasks, there are three modes:

  • Audio Creation Mode - audio creation, and in our case, playback. Since only in this mode is accurate sound output possible without unnecessary conversions.

  • Entertainment Mode - the mode is mainly designed for playing multi-channel audio (including tracks for video films) with the application of various effects, channel conversion, etc.

  • Game Mode - the name of the mode speaks for itself. Only here you get access to the latest Creative developments in 3D audio for an unrivaled immersive experience in modern games.

It is convenient to switch modes through the Volume Panel context menu:

The settings window (Console Launcher) appears when you double-click on the Volume Panel icon in the tray.

5.1 Audio Creation Mode and Music Playback

So, let's turn off everything unnecessary:

Now a few words about setting up the foobar2000 player. Due to the unreliability of other interfaces (they actually sometimes turn out to be not entirely adequate), it is recommended to output sound via ASIO. Fortunately, this card has a fully functional ASIO 2.0 driver. All you need is to install the foo_out_asio plugin (you can download it on the foobar2000 + plugins page), add an ASIO device in the plugin settings, and select it in the output settings. Pay attention to the channel markings (useful for multi-channel systems).

Now you can enjoy clear sound, not spoiled by any “improvers” and other nonsense.

Another nice touch is that the ASIO driver automatically switches the sound card's reference frequency in accordance with the sampling frequency of the audio being played back. For example, if you first played a track with a frequency of 44.1 kHz, and then you start a track with a frequency of 96 kHz, the reference frequency will switch, accompanied by a characteristic sound. Some people do not like this, however, this function is a significant advantage, because only if the frequency of the material and the reference frequency of the card coincide, bit-by-bit playback is possible. In addition, now you most likely will not need a resampler.

But here we must make a note - sometimes there are tracks with a sampling frequency different from those supported by the ASIO driver (let me remind you: these are 44.1, 48, 88.2, and 96 kHz). For such cases, you can add the SoX Resampler mod to the DSP chain (which can be downloaded on the same foobar2000 + plugins page). It will resample all frequencies not included in the list of supported ones (we specify them in the resampler settings) to a given frequency - to minimize quality losses, we will specify 96 kHz just in case:

5.2 Entertainment mode - watching movies

This mode opens up some interesting possibilities for the X-Fi signal processor.

Do you want to enjoy 5.1 sound of your favorite movie, but don’t have multi-channel acoustics? No problem! After all, a person has only two ears, which means that with the help of two sound emitters you can recreate absolutely any virtual source at any point around the listener, so much so that it will be indistinguishable from the real one!

Using special processing algorithms, the X-Fi chip converts multi-channel audio into 2.0 for acoustics or for headphones (note that the algorithms for creating surround effects in headphones and speakers are very different).

And now let’s talk about how we can take advantage of these opportunities. First of all, let's launch Console Launcher:

Let's look at the main sections one by one.

In this section you need to select the configuration of your speaker system, or headphones. When you press the THX button, the speaker calibration window opens. We are interested in the Speaker Position tab. The fact is that when using a two-channel speaker system (which is what most users have), the position of the listener relative to the sound emitters is not always the same (worse, sometimes it is unstable, but nothing can be done about it). In order for ambience algorithms to work, they need correct data about the listener's position. In this case, you need to indicate the distance to each sound emitter and the angle of rotation relative to the central axis. If it’s not clear, let me explain: this means the angle between two lines:

  • a line connecting the listener's head to a point directly in front of his face.
  • line between his head and the sound emitter

The angle is counted clockwise (when viewed from above), the parameter value can take negative values.

So, if you want a 3D effect, grab a ruler and go ahead and measure it;)

On this tab you can enable the effect of the environment - an amphitheater, concert hall, etc. In case it is selected None, the Enable EAX Effects checkbox does not affect anything, but just in case, let's leave this option enabled.

This option is responsible for enabling X-Fi CMSS 3D - an algorithm designed to improve the ambience effect and positioning accuracy of virtual sources. This function will be useful when playing multi-channel audio on 2.0 speakers or headphones, or vice versa - for the so-called upmix - decomposing sound into several channels (for example, playing stereo recordings on a 5.1 system). If you turn it on when playing, say, stereo material through headphones, it will be of little use - unless it creates some effect of virtual sound sources directly in front of the listener, which, for example, does not sound very pleasant to my ears. But in any case, the choice is yours.

The famous (thanks to the eloquent marketers from Creative) X-Fi Crystalizer - sometimes it can add expressiveness to the sound, but at the same time it inevitably introduces a significant portion of distortion. Of course, it is not recommended for lovers of “honest sound”. Although, it sounds very interesting to the ear when the slider is moved to minimum - the distortion in this case is almost unnoticeable, but the effect is quite good.

Smart Volume Management - to some extent - is the same Advanced Limiter in foobar2000, albeit with a larger pre-buffer. Since the DSP output is inevitably converted to a fixed-point format, measures must be taken to prevent clipping, which is what the Creative programmers did. When SVM detects that the permissible volume in a fragment of the audio stream is exceeded, it lowers its volume. The main disadvantage is that compression occurs (the volume balance between audio fragments changes significantly). Although, dynamic range compression is the second purpose of SVM. If, for example, you want to watch a movie at low volume at night (so as not to disturb anyone), but when you lower the volume, quiet sounds can no longer be heard - this is where a compressor will help. It will increase the volume of quiet areas, bringing them approximately on the same level as the rest - and you can adjust the overall volume to the level you need.

But if there is no need for the above, I would advise when turning on effects (and playing multi-channel sound, among other things) to set the volume (Master Volume) not to the maximum, but to about 50% - usually this is enough to avoid distortion.

Here we see a regular 10-band equalizer with an additional level control. If not needed, turn it off.

And this is the sound card mixer. Here you can enable playback from line input, microphone, etc. We turn off unnecessary sources (because the line input and microphone, for example, can make noise).

With a sound card - that's it. As for the player, there are some nuances here too. First, the player must output the sound "AS IS". That is, all software mixers must be disabled so that in the case of 5.1 tracks, for example, all 6 channels are directly transmitted to the sound card. To watch movies, I can recommend the MPC HomeCinema player and ffdshow audio/video filters (included in the Sam CoDec Pack). I would also like to write in detail about setting them up, but that will come later in another article. Below I will give the main points of setting up sound output (player and ffdshow filter), and then I hope you’ll figure it out:

5.3 Game mode - surround sound in games.

Without a doubt, Creative is the leader in 3D audio for games. It was Creative programmers who created a set of extensions for DirectSound3D, the so-called. EAX. You can read about this technology in the article Questions and Answers on EAX, as well as on Wikipedia.

The latest version of EAX at the moment is 5.0, but our card supports all versions up to and including the fifth.

So, let's start setting up. Some sections of the settings coincide with the Entertainment mode settings, so we will not consider them again.

This tab contains the main settings - speaker configuration (which must be set in accordance with the actual configuration of your speakers), SVM, enabling/adjusting EAX (this refers only to environmental effects - reverberation, etc.). Mic Environment FX is a new feature that allows you to apply effects to the microphone sound to match the environment in a 3D game.

Also, speaker placement (THX) calibration is extremely important in this mode. And I also advise you to pay attention to the Master Volume - in order to avoid distortion, it is advisable to keep this control at a position somewhere around 35-40%.

Of particular interest in this mode are the X-Fi CMSS-3D settings. When you enable this feature (which is highly recommended to significantly improve the 3D effect in games), two new extensions introduced in EAX 5.0 become available:

  • MacroFX- the effect significantly increases the realism of sound sources located in close proximity to the ear.

  • Elevation Filter- the effect significantly improves the positioning of sources along the vertical axis (above/below the listener).

Parameter value On means force the effect to turn on for all sound sources (for which the effect is intended). By clicking on the Test button you can run a small test of environmental effects.

Well, we're done with setting up the sound card. All that remains is to configure the software. As I already wrote, unfortunately in Windows 7 programs do not have direct access to the hardware resources of the sound card through the DirectSound3D API. The only API that has access to hardware 3D sound is OpenAL. So, without hesitation, Creative set about developing the Alchemy utility that converts DirectSound/EAX calls to OpenAL. I’ll tell you about setting it up now.

So, if you followed all the instructions exactly from the beginning of the article, you should already have Creative ALchemy 1.43.06 beta installed. It’s not for nothing that I chose this particular version, since in some incomprehensible way the latest release version does not quite honestly perform its functions - it is obvious that EAX using it is software, plus the MacroFX and Elevation Filter effects do not work.

So, on the left we see a list of installed games that were detected automatically. By moving a game to the right list, we automatically enable Alchemy functionality for it.

You can also add games manually. Let's take advantage of this opportunity. Install the RightMark 3DSound program and add the address of the folder with the executable file of the RightMark 3DSound Positioning Accuracy test utility (RightMark3DSound.exe) to the Creative Alchemy list:

Let's move the added program to the right list. Now the program can be launched.

The screenshot shows how you can recreate a sound source located behind an obstacle, behind, to the right of the listener.

This program is a great opportunity to try out MacroFX and Elevation filter technologies. When turning these effects on and off, try first bringing the sound source in the program closer to the listener's ear, and then moving it up and down along the vertical axis - you will feel the difference!

And don’t be confused by the fact that the program does not include EAX 5.0 - most of the effects of this version are automatically applied to all sound sources, regardless of the EAX version supported by the program/game.

Games can be configured using the same principle - the main thing is to add the address of the folder with the game's executable file to Alchemy. If the game has the appropriate settings, EAX support must be enabled in them. If there is no such option, there is an alternative: enable multi-channel sound in the game. At the same time, it will be converted by the sound card into stereo and you will get excellent positioning of sounds.

If you want to fully experience all the effects of EAX 5.0, I can recommend the game S.T.A.L.K.E.R: Call of Pripyat. This game uses the OpenAL sound engine, which gives it direct access to all EAX effects (no need to use Alchemy) - you just need to enable EAX support in the game settings and select the SB X-Fi Audio device.

6. Epilogue

That's all I wanted to tell you about setting up X-Fi XtremeGamer. I hope the information in this article was useful to you and now you can unleash the full potential of your sound card. Happy listening!


2024 gtavrl.ru.