It's software-based X-Fi card, like Creative X-Fi XtremeAudio PCIe (same CA0110 chip).

Daniel_k's X-Fi Support Pack is designed for real X-Fi cards with CA20k1/20k2 chips, so there is no error here.

As mentioned earlier, this sound card doesn't have true X-Fi.
It's actually Live! 24-bit on the PCI-e bus (HBA - software Host Based Adapterin which the payload is processed not in the controller itself, but on the CPU).
The Audiotrack was initially created using the VIA ENVI24 controller, which had a great feature: two oscillators for different sampling rates at 44 and 48 kHz.
That’s why they were so highly regarded.
These cards were highly regarded by audiophiles, music lovers and semi-professional musicians for home studios.

I know that the Audiotrak does not feature a hardware dsp, but the real question is (when a seemingly high quality soundcard such as this is concerned) does it really matter? For ante mid 2000s period correct builds, offloading the sound effects processing from the cpu it would, but for any decent semi-modern system or an overkill retro one -featuring anything equal or better than a good core2duo-, i think it wouldn't. You could compare that with the X ram feature. When the first x ram equipped cards hit the market it maybe was a useful, resource saving feature (highly debatable though even then), but for any half decent system build in the last 20 years with contemporary components, it is irrelevant. So the real question is: is there a discernible audio quality difference when a software dsp based, but (unlike the relatively low quality Xtreme Audio for example) high quality X fi card, such as this, is used in comparison to the the "real" X fi cards such as the Platinun or the Xtreme Music? That's what i would really like to find out and the reason why i am asking if someone here had a first hand experience with the Audiotrak, cause this one, along with the Auzentech Bravura, despite lacking a "true" X fi chip, seem to be really good quality cards and if there is no perceptible sound quality difference in Eax capable games between them and the other "real" X fi cards, then there is no reason someone would not consider one for his system if he had the chance to acquire one.

An interesting read here: https://audiophilesoft.com/publ/3d/software_x_fi/7-1-0-73

*Use Google translate or something similar, the consensus of the article is that there is definitely a perceived audio difference between hardware and software produced gaming effects and that stretches beyong Eax.

The main problem with games from those days, when audio streams were offloaded to the sound card rather than processed by the game engine, was that different samples had different sample rates.
In other words, having 22 kHz, 44 kHz and 48 kHz in a single scene was perfectly normal, because 22 kHz came from the first game in the series, 44 kHz from the sound bank, and 48 kHz had just been recorded.
A decent DirectSound SRC (sample rate converter) for Windows only appeared in Windows 7 SPx KBxxxxxx (and, as far as I recall, this update isn’t mandatory)
In this situation, dual-frequency cards based on the VIA ENVY24HT (2 crystals) or C-Media (PLL) perform poorly. Even Live! with its hardware SRC performs poorly (because E-MU designed it as a dual-quartz solution for music-oriented sound cards)
Audigi performs well, and X-Fi performs excellently.
On the X-Fi, the SRC blocks take up almost half the entire chip.
With HBA, it all depends on the driver, and unfortunately, they can’t work miracles.

Play this file using a media player that supports bit-perfect playback – such as ASIO or Kernel Streaming.

shevalier wrote on 2026-03-24, 06:40:

The main problem with games from those days, when audio streams were offloaded to the sound card rather than processed by the ga […]
Show full quote

The main problem with games from those days, when audio streams were offloaded to the sound card rather than processed by the game engine, was that different samples had different sample rates.
In other words, having 22 kHz, 44 kHz and 48 kHz in a single scene was perfectly normal, because 22 kHz came from the first game in the series, 44 kHz from the sound bank, and 48 kHz had just been recorded.
A decent DirectSound SRC (sample rate converter) for Windows only appeared in Windows 7 SPx KBxxxxxx (and, as far as I recall, this update isn’t mandatory)
In this situation, dual-frequency cards based on the VIA ENVY24HT (2 crystals) or C-Media (PLL) perform poorly. Even Live! with its hardware SRC performs poorly (because E-MU designed it as a dual-quartz solution for music-oriented sound cards)
Audigi performs well, and X-Fi performs excellently.
On the X-Fi, the SRC blocks take up almost half the entire chip.
With HBA, it all depends on the driver, and unfortunately, they can’t work miracles.

Play this file using a media player that supports bit-perfect playback – such as ASIO or Kernel Streaming.

Unless I am mistaken, that file is a renamed udial.wav or similar. Anyone playing that file should be very careful about keeping volume levels low enough to avoid clipping (which could damage speakers, tweeters in particular).

darry wrote on 2026-03-24, 07:33:

Anyone playing that file should be very careful about keeping volume levels low enough to avoid clipping (which could damage speakers, tweeters in particular).

Pushing the amplifier or tweeters to the point of clipping isn’t generally a very good idea.

shevalier wrote on 2026-03-24, 07:44:

darry wrote on 2026-03-24, 07:33:

Anyone playing that file should be very careful about keeping volume levels low enough to avoid clipping (which could damage speakers, tweeters in particular).

Pushing the amplifier or tweeters to the point of clipping isn’t generally a very good idea.

True, but it appears especially easy to cause accidental damage with this file.

https://www.diyaudio.com/community/threads/ud … d-take-2.67608/

The Windows 7 hotfix you are referring to i think is this: KB2653312. If i am not mistaken Windows Xp didn't suffer from this sample conversion problem as it was introduced from Vista onwards and was not remedied until this fix came out for Win 7.

I know about the audio file you linked to. Indeed it could potentially damage tweeters if played really loud, especially on non full range speakers. I think it is used to detect poorly done oversampling when accomplished by a soundcard on the hardware level. But i think the problem you described stems from the fact that most soundcards come equipped with only one crystal at 24.576 mhz and are therefore capable of doing adequate hardware resampling at 48000 or its multiples so in that case hardware resampling produces worst problems than software resampling. If those cards were equipped with a second crystal at 22.5792 mhz that problem would be solved.

Dimos wrote on 2026-03-24, 11:48:

from the fact that most soundcards come equipped with only one crystal at 24.576 mhz

C-Media (ASUS's preferred chipset) has a single oscillator but adjusts frequencies using a phase-locked loop.
48k&44k both are native.
However, this does not apply to mixed content, since the board switches the entire clock grid.

Dimos wrote on 2026-03-24, 11:48:

The Windows 7 hotfix you are referring to i think is this: KB2653312. If i am not mistaken Windows Xp didn't suffer from this sample conversion problem

As far as I understand, this wasn’t a problem, as it used to be a function of the sound card driver&hardware.