Review and Measurements of SoundBlaster X-FI HD

[raif71]

Apologies for reviving an old thread. I own the Sound Blaster X-Fi HD USB, and found it quite useful for very many years. The connections and features are simply ideal for my environment. I appreciate the simplicity of a single USB connection for both data and power. I use the 1/4" headphone jack almost exclusively (ATH-M50x). I find the volume knob on the front of the box convenient, despite it being somewhat small and stiff. For me, I will never try to plug it into anything other than my desktop PC. It just sits in the corner, and does its job...

...until my teenage son found it and took it for his own use. Now I'm looking to replace it with a similar device. I was ready to just go buy another SB1240, but recognizing how old it is I thought that surely there must be something improved. Yet, I cannot seem to find a simple USB connected box with 1/4" headphone jack and a volume knob. And at this price point, perhaps I am willing to overlook the performance figures. After all, the rest of my equipment is hardly audiophile grade either.

I would appreciate any suggestions.

I'm kind of interested in the optical out of this unit and I'm wondering whether the unit itself acts as an ADC. Does it convert line in audio into digital via optical out? I'm asking this as I'm looking for such a device. I have bought some cheap ADC, something like this https://www.amazon.com/avedio-links..._1_6?keywords=audio+adc&qid=1571193211&sr=8-6 but the digital out into the receiver sounds bland/weak coming out of the speakers. Also, I don't plan to use the soundblaster with a PC/laptop ie hopefully no need to control which outputs to be activated or not with software. Any help will be appreciated. Thanks

 

[pma]

I have this sound card and it works well for me under Win 8.1. It is good to download and install the newest driver. No problems with ADC mentioned by Amir. I get about 0.0005% THD at about -9dBFS. The spectrum is very clean with only 2nd and 3rd harmonics visible. This is measured with X-Fi HD USB.

 

[Hifihedgehog]

Back in the day I dropped $400 on the then top of the line X-Fi Elite Pro. 4 DAC chips. Sounded great to me. Wish I still had it so I could send it in to be measured.

It is going for just $9.99 here on eBay with 15 hours left.

@Amir, if I were to win this in auction, could I have it shipped direct to you from the seller to put in your measurement queue? Note it is PCI-based and though it is well over a decade old, there are Windows 10 drivers. I would be willing to donate it. Afterwards, I personally wouldn’t have any need for it. So you could ship it to someone who wants it in the forums after the testing is completed.

 

[Svperstar]

It is going for just $9.99 here on eBay with 15 hours left.

@Amir, if I were to win these in auction, could I have them shipped direct to you from the seller to put in your measurement queue? Note it is PCI-based and though it is old, there are Windows 10 drivers. I would be willing to donate it. I personally wouldn’t have any need for it. So you could ship it to someone who wants it in the forums after the testing is completed.

Wow great find. I got rid of mine years ago. I would bid but I have no way to use it as I have been laptop only for years now. Uses the same DAC chip as my D30, very very clean sounding soundcard and the games that took advantage of the hardware features sounded amazing.

 

[Jimster480]

Wow great find. I got rid of mine years ago. I would bid but I have no way to use it as I have been laptop only for years now. Uses the same DAC chip as my D30, very very clean sounding soundcard and the games that took advantage of the hardware features sounded amazing.

Not sure if its worth at all today though. Just having the same DAC chip doesn't mean it will be good though.

 

[bennetng]

Not about DAC/ADC chips but just to point out that none of the USB products with the "X-Fi" brand name has the DSP capabilities on the bona fide PCI/PCIe X-Fi cards which use the CA20Kx processors. They are very different products.
https://en.wikipedia.org/wiki/E-mu_20K

With Audio Precision measurements:
http://ixbtlabs.com/articles2/multimedia/creative-x-fi.html

 

[AnalogSteph]

Not sure if its worth at all today though. Just having the same DAC chip doesn't mean it will be good though.

The X-Fi Elite Pro must be one of very few consumer-level devices to ever use an AK5394A ADC, a TOTL part (their EMU lineup already used these). Probably not nearly making full use of it, but anyway. With how good the digital filters on that chip are it ought to be taking excellent recordings even at 44.1 kHz. Probably a great vinyl ripper. Looks like they shipped with a heatsink on the EMU20K by default which is probably a good thing, X-Fis without them are known for failing.

Cards using a CS4398 DAC are almost a dime a dozen but usually a CS5381 ADC is about the fanciest you'll ever see, and most of the time it'll be a CS5361, as in this X-Fi HD. Which is a pretty good little chip in its own right (and apparently easier to get low distortion out of than its bigger brother), not to mention the standalone equivalent of the ADC in the common CS4272 codec found in numerous audio interfaces, but it's not quite top flight.

 

[bennetng]

The X-Fi Elite Pro must be one of very few consumer-level devices to ever use an AK5394A ADC, a TOTL part (their EMU lineup already used these).

Creative marketing at that time actually said the ADC was the same one as the DigiDesign 192 I/O (Pro Tools TDM) as EMU was famous on the synth/sampler reign rather than converters.
https://www.gearslutz.com/board/gee...udio-interfaces-their-ad-da-chips-listed.html

Also, the EMU cards/boxes used the last Gen DSP (EMU10k2) on the Audigy series, native sample rates other than 48kHz and additional studio capabilities are handled by another Xilinx chip. As a result, no hardware effects can be applied above 48kHz. CA20kX on the other hand supports 44.1/48/88.2/96k native sample rates and DSP effects on its own.
https://www.soundonsound.com/techniques/emus-patchmix-dsp-demystified

By the way, from my memory Elite Pro's RMAA loopback results were not really better than Xonar D2, more or less the same. Also, the external box did not use the same TOTL converters, the best ones are all on the card itself.
https://www.dearhoney.idv.tw/6/sound-blaster-x-fi-elite-pro-external-box/

 

[AnalogSteph]

By the way, from my memory Elite Pro's RMAA loopback results were not really better than Xonar D2, more or less the same.

Yeah, not super clean, unlike the Xonars. I'm guessing those just had better boards / board layouts, plus the solid capacitors must have helped. (Still rocking a D1 here; another and a D2 are boxed up. Late-2000s soundcard heroes. )
http://ixbtlabs.com/articles2/multimedia/creative-x-fi.html

Also, the external box did not use the same TOTL converters, the best ones are all on the card itself.
https://www.dearhoney.idv.tw/6/sound-blaster-x-fi-elite-pro-external-box/

PCM1804 + CS4392? Ha, that's exactly like the "old" EMU 1820. Still pretty decent midrange parts. CS4392 is like the CS4272 DAC, and while the PCM1804 is a few dB short of the newer CS5361 and AK5385A in dynamic range and distortion it has some pretty good digital filtering that makes it useful even at 44.1 kHz (stopband starting at 0.547fs, filter ultimate 100 dB, passband ripple +/-0.005 dB, similar to the AKM part except letting up at quad speed a bit). The Audigy 2ZS Platinum LiveDrive used the same ADC apparently as well.

The mid-2000s sure were a wild time for PC soundcards, reflecting the converter quality explosion from a few years prior (see my little history). For around the same price point of around 55€ +/- 15€, you could have had a Terratec DMX XFire 1024 in ~2001 (a CS4294 + CS46xx job, dynamic range = 87 dB, 90 dB with a following wind), a Terratec Aureon 5.1 Sky in 2003-4 (WM8770 + Envy24HT, ~95-100 dB - mind you, you could have spent some more for an M-Audio Audiophile 2496 with its AK4528 codec, even if that's not used to its full potential on there), and an Asus Xonar D1 by 2008-9 (CS4398 / CS5361 + CMI8786, >= 110 dB). (Well yes, all of these cards I've had myself, and two of some. Had two of the Aureons in my old PC for years, one ID hacked as another model so I could use both at the same time. One of them eventually got dodgy, bad caps I think.) You just don't see progress in the consumer space at this rate any day. I mean, PCs did go from Athlon Thunderbirds to Core 2 Duos in the same timeframe, but still.

 

[LKA]

I recommend to change 4 opamps. The ADC input buffers have to be a bipolar version (LME49720), jfets do not work well (I tried OPA2134,OPA1656). The DAC output buffers can be bjt/jfet (LME49720/OPA1656).

ADC(CS5361) FS 1Vrms
DAC(AK4396) FS 2.19Vrms

 

[MCH]

Anyone has tried this as phono input to a raspberry pi? My doubt is how to select phono vs line in with the pi.

 

[MCH]

hello, so i got the X-Fi HD and works fine as audio input for my pi running LMS both as line level input (connected to my Trigon phono stage) and as phono in (connected directly to the turntable).

However, i notice the following, I hope someone can help me:

- the clicks and pops sound somehow louder when i digitalize the signal with the soundcard than when i go all analog with the Trigon. Is it possible i am making something clip or near clip? would it help to decrease the gain in alsamixer?

- the background hum using the X-Fi HD as both phono stage + ADC (connected to the turntable) is much higher than when used as line in ADC (connected to the Trigon line out). Anything i can do to try to diminish the hum of the X-Fi as phono stage?

thanks for your comments.

 

[AnalogSteph]

However, i notice the following, I hope someone can help me:

- the clicks and pops sound somehow louder when i digitalize the signal with the soundcard than when i go all analog with the Trigon. Is it possible i am making something clip or near clip? would it help to decrease the gain in alsamixer?

Which exact scenarios are you comparing -
1. (a) Trigon --> X-Fi HD line-in --> X-Fi HD audio out --> amp vs. (b) Trigon --> amp
or
2. (b) Trigon --> X-Fi HD line-in vs (b) X-Fi phono in?

I don't think the input has a PGA or anything, so alsamixer will only be able to offer you digital gain... best leave at 0 dB.

Can't you see peak levels in recording software?

Overly obvious clicks may indicate ultrasonic IM... other on behalf of a bipolar input stage with slow slew rate or limited overload margin. (It could also be simple FR deviation though... I saw one review that indicated only so-so RIAA accuracy.) One would at least hope that the Trigon phonopre which costs several hundred would be superior to a bonus throw-in.

As a potential further issue, the CS5361 ADC does have pretty decent antialias filters but do make sure that sample rate is consistent everywhere and preferably use 96 kHz to rule out issues in this area.

- the background hum using the X-Fi HD as both phono stage + ADC (connected to the turntable) is much higher than when used as line in ADC (connected to the Trigon line out). Anything i can do to try to diminish the hum of the X-Fi as phono stage?

I suspect it's the power supply you are using for the Pi that isn't exactly short of mains leakage. Try grounding the X-Fi HD via the screw terminal (the tonearm ground is connected there already, right?), and possibly another PSU if that doesn't help.

 

[MCH]

Which exact scenarios are you comparing -
1. (a) Trigon --> X-Fi HD line-in --> X-Fi HD audio out --> amp vs. (b) Trigon --> amp
or
2. (b) Trigon --> X-Fi HD line-in vs (b) X-Fi phono in?

I don't think the input has a PGA or anything, so alsamixer will only be able to offer you digital gain... best leave at 0 dB.

Can't you see peak levels in recording software?

Overly obvious clicks may indicate ultrasonic IM... other on behalf of a bipolar input stage with slow slew rate or limited overload margin. (It could also be simple FR deviation though... I saw one review that indicated only so-so RIAA accuracy.) One would at least hope that the Trigon phonopre which costs several hundred would be superior to a bonus throw-in.

As a potential further issue, the CS5361 ADC does have pretty decent antialias filters but do make sure that sample rate is consistent everywhere and preferably use 96 kHz to rule out issues in this area.

I suspect it's the power supply you are using for the Pi that isn't exactly short of mains leakage. Try grounding the X-Fi HD via the screw terminal (the tonearm ground is connected there already, right?), and possibly another PSU if that doesn't help.

thanks for your suggestions Steph,

to the first question, i am comparing:
trigon -> X-Fi HD line in -> pi server-> wifi -> pi streamer -> dac (D30 pro) -> amp
vs
trigon -> amp

clearly if i could stay with trigon -> amp it would be ideal, but i need to stream to a different room.... i can live with a bit degraded sound that i probably wont notice, but was not expecting the vinyl clicks and pops to become louder than usual (i think). Anyways, its a minimal annoyance, was just curious if there was an easy setting i could try.
Will try to see the peak levels as you mention in a recording software, but that is beyond my current knowledge (i need to study a bit more, i am relatively new to all this...).
Regarding sample rate, i am streaming at 44.1 kHz, but i hear your suggestion, guess i will need to check how to change this in LMS.

testing another power supply i something i can easily do, i know this setup can run on a powerbank, that should be pretty clean, shouldn't it?
Yes, the turntable was properly grounded to the X-Fi ground screw, but what do you exactly mean with "try grounding the X-Fi HD via the screw terminal"?
Yes, maibe i was a bit naive expecting that the phono stage in the soundcard would sound as clean as the Trigon, my dream was to sell the two big (in comparison) boxes that make the trigon phono stage and keep only the small discrete X-Fi, but i see it will come at a price.... If there was only an one box phono stage with good USB out...

 

[datobe]

Today I realized - DAC AMPs matter! I have this device as headphone amp and I have been always thinking It was good. Today I've decided to compare it to my smartphone and surprisingly my smartphone Xiaomi k20 sounds more cleaner with more textured bass than this Creative sound card Hoooow?? I can't wait for my KA13 to be delivered.

maybe this is placebo or something, I wish I could do blind test with someone with good ears ))

 

[Svperstar]

Today I realized - DAC AMPs matter! I have this device as headphone amp and I have been always thinking It was good. Today I've decided to compare it to my smartphone and surprisingly my smartphone Xiaomi k20 sounds more cleaner with more textured bass than this Creative sound card Hoooow?? I can't wait for my KA13 to be delivered.

Might have something to do with settings in Windows

 

[datobe]

Might have something to do with settings in Windows

Apart from that, quite often I hear weird zaps, electric interferences just milliseconds. only when listening to music. I think this has something to do house electric circuit as sound card is not grounded.

 

[Svperstar]

Apart from that, quite often I hear weird zaps, electric interferences just milliseconds. only when listening to music. I think this has something to do house electric circuit as sound card is not grounded.

Try running LatencyMon and check your DPC latency

 

[datobe]

Try running LatencyMon and check your DPC latency

I think everything is ok there. I will try to ground it later, maybe that will fix.

 

[RedCometZero]

The output impedance on the headphone amp is very high. Unless you're using very high output impedance headphones, this DAC/AMP will dramatically impact the frequency response of your headphones.