Creative Soundblaster ZxR Asio Internal PCIe Sound card - review

Want to talk about music hardware or software that doesn't include Reason?
Post Reply
sleeper0013
Posts: 18
Joined: 20 Feb 2015

20 Feb 2015

Just bought one of these recently and I am blown away with this card. I have gotten a 30% increase in DSP on an 8core 4.7 9590FX CPU. Along with this it has Optical in/out can hook optically into a focusrite 18i20, or what ever you choose, as well as out to an Beranger Ultracurve for near 0 latency, . it has its own spunky A/D D/A it self. This card is has a 3D sound processor meant for PC gamers with real time audio environment Shaders, that means room position relative reverb while in motion processing. In other words when you move in game reverb from the room you hear changes. Think of it as virtual reality processor for audio. It has RCA in/out, has a Headphone Mic dongle for easy access, if your a gamer who does voice over IP this is nice. Best part of all the card is 200$ from amazon.

drivers are stable, Only had reason crash once on me while i was screwing around with my latency buffer too much while audio was playing.

User avatar
orthodox
RE Developer
Posts: 2286
Joined: 22 Jan 2015
Location: 55°09'24.5"N 37°27'41.4"E

20 Feb 2015

I quit hunting for computer audio.
Because optical in/out are the same no matter what card.
Even on my $1 realtek alc 8xx.

sleeper0013
Posts: 18
Joined: 20 Feb 2015

20 Feb 2015

the boost in DSP is why i purchased this card Considering I mix into my master channel, rather than EQ a bounce. I'm happy with it. The optical was just an added bonus.

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

21 Feb 2015

orthodox wrote:I quit hunting for computer audio.
Because optical in/out are the same no matter what card.
Even on my $1 realtek alc 8xx.
No they're not. If your digital input doesn't recreate the clock but uses the one it gets from the PLL over the input connection you get quite a lot of jitter.

And @OP: No you don't have almost zero latency, try using the VMG measuring feature on an anlog loopback from the output back to an input of your card :)

And as said in the other thread, the DSP use decrease is most probably from different buffer settings or problems with the previous audio setup.

User avatar
orthodox
RE Developer
Posts: 2286
Joined: 22 Jan 2015
Location: 55°09'24.5"N 37°27'41.4"E

21 Feb 2015

normen wrote:If your digital input doesn't recreate the clock but uses the one it gets from the PLL over the input connection you get quite a lot of jitter.
Are there any that use PLL over the signal clock at all? Realtek datasheets imply that the clock is read and data is buffered. With jitter tolerance of ~3%.

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

21 Feb 2015

orthodox wrote: Are there any that use PLL over the signal clock at all? Realtek datasheets imply that the clock is read and data is buffered. With jitter tolerance of ~3%.
A PLL is the normal way to retrieve the clock from the cable.. What other way could be used..? And still, good interfaces then create their own clock and sync the signal back to that. 3% is actually pretty abysmal.

User avatar
orthodox
RE Developer
Posts: 2286
Joined: 22 Jan 2015
Location: 55°09'24.5"N 37°27'41.4"E

21 Feb 2015

normen wrote:A PLL is the normal way to retrieve the clock from the cable.. What other way could be used..? And still, good interfaces then create their own clock and sync the signal back to that. 3% is actually pretty abysmal.
Well, honestly I don't know if that's enough.
It seems to me that synchronizing data streams with progressive drift is more common problem.

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

21 Feb 2015

orthodox wrote: Well, honestly I don't know if that's enough.
It seems to me that synchronizing data streams with progressive drift is more common problem.
Not really, with the clocks nowadays you get very few drift and dropping or inserting a sample here or there is certainly less of an issue than continuous jitter. But it seems you agree now that digital inputs are not all the same quality :)

User avatar
orthodox
RE Developer
Posts: 2286
Joined: 22 Jan 2015
Location: 55°09'24.5"N 37°27'41.4"E

21 Feb 2015

normen wrote:But it seems you agree now that digital inputs are not all the same quality :)
Not quite. I'm still not sure if that quality is excessive and makes no sense.

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

21 Feb 2015

orthodox wrote:Not quite. I'm still not sure if that quality is excessive and makes no sense.
You get 3% when you record the audio into the system, then you get 3% when you send out the audio again (because its still synced to the input clock). Then you might get another 3% if you go into your low-cost digital input of your amplifier.. If you think that doesn't amount to considerable degradation thats okay but I rather try to avoid this and have a proper audio interface. But for home use I guess you're right, unbalanced audio connections and non-treated rooms are common there as well so you won't get accurate reproduction or low error margins anyway.

Yet you went from "the same" to "excessive quality", thats the point I made. There IS differences. Syncing digital audio connections are a different thing than other digital connections where you transfer data packages (e.g. network connections, USB etc.).

User avatar
orthodox
RE Developer
Posts: 2286
Joined: 22 Jan 2015
Location: 55°09'24.5"N 37°27'41.4"E

21 Feb 2015

normen wrote:, then you get 3% when you send out the audio again (because its still synced to the input clock).
I don't use pass-thru or monitor mode so the clock is new.
normen wrote:Yet you went from "the same" to "excessive quality"
It's the same, just like audiophile's stuff.

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

21 Feb 2015

orthodox wrote:I don't use pass-thru or monitor mode so the clock is new.
I doubt that. Almost all audio interfaces use one clock source, else you get all kinds of issues. And if it actually uses a new (self-generated) clock it could just as well sync back the audio to that, as its done in proper audio interfaces.

Actually, if you would use direct pass-thru you'd only get the degradation once. Exactly *because* you send the same audio out at another clock time you get twice the jitter.
orthodox wrote:It's the same, just like audiophile's stuff. 
No its not. As I said, you are mistaking a digital audio connection that uses direct syncing (like ADAT, S/P DIF, MADI etc.) with a digital connection that works packet-wise with a sync thats independent from the signal (like that 10.000$ ethernet cable which actually makes no sense).

User avatar
orthodox
RE Developer
Posts: 2286
Joined: 22 Jan 2015
Location: 55°09'24.5"N 37°27'41.4"E

21 Feb 2015

normen wrote:Exactly *because* you send the same audio out at another clock time you get twice the jitter.
I don't send the same audio. The digital input is input to a DAW which then outputs something new. The DAW uses two sync clocks, input and output, and deals with sync problems.
normen wrote:you are mistaking ... direct syncing ... with ... packet-wise
What makes you think I am?
I just don't believe that the jitter from a crystal osc source after an optical link may exceed 10ns. I could just make a test and transfer 10GB file through 3 s/pdif cables at 192k*24bit and then compare it bitwise.

User avatar
ScuzzyEye
Moderator
Posts: 1402
Joined: 15 Jan 2015
Contact:

21 Feb 2015

I'm not sure if both of you are on the same page. When Normen says "audio" I'm assuming he means analog audio. That's the only place jitter matters: moving to or from the analog domain. If you're doing digital to digital transfers, and the actual sample rate is known, and no bits are lost, then jitter doesn't matter. But if you're recording an analog source, or trying to achieve high-quality playback, then the bits need to moving in a very consistent manor, or there is going to be distortion.

sleeper0013
Posts: 18
Joined: 20 Feb 2015

21 Feb 2015

Here is my question now, IF i optical out from a Focusrite 18i20 into the computer and out from the PC to a Beringer ultra-curve with word-clock strung between the 18i20 and Ultra-curve. what problems does anyone see with this? If i am correct the DAW should regulate sync on its own, and the A/D D/A should regulate through the word-clock link thus the only problem would be discrepancy between the 18i20 A/D and the ultra-curve D/A converter quality. Is this correct?

User avatar
ScuzzyEye
Moderator
Posts: 1402
Joined: 15 Jan 2015
Contact:

21 Feb 2015

sleeper0013 wrote:Here is my question now, IF i optical out from a Focusrite 18i20 into the computer and out from the PC to a Beringer ultra-curve with word-clock strung between the 18i20 and Ultra-curve. what problems does anyone see with this? If i am correct the DAW should regulate sync on its own, and the A/D D/A should regulate through the word-clock link thus the only problem would be discrepancy between the 18i20 A/D and the ultra-curve D/A converter quality. Is this correct?
Where is the 18i20 getting the signal that's going out it's optical out? Its analog inputs? The 18i20 doesn't get sync from the DAW, the DAW simply tells it at what sampling rate to operate. The 18i20 generates its own clock internally (or from an external clock source if you hook one up). So what ever jitter is present in its clock will imprint distortion during the A/D process. As the bitstream passes through the computer, there will be no jitter, but then the Ultra-Curve will be also generating its own clock, and passing the bits to the DAC based on that clocking. So the distortion caused by that jitter will be added to what was created during the original capture.

The Focusrite probably has a pretty solid clock, I wouldn't bet on the Behringer, but you never know. It doesn't look like the Forcusrite has a world clock input, but it can output the clock that its generating. I'm not sure if it would improve anything if you connected it's clock output to the input on the Ultra-Curve, but you can try.

All that said, distortion caused by jitter is minimal, and probably not audible. So what you're doing isn't likely causing any harm.

User avatar
normen
Posts: 3431
Joined: 16 Jan 2015

22 Feb 2015

orthodox wrote: I don't send the same audio. The digital input is input to a DAW which then outputs something new. The DAW uses two sync clocks, input and output, and deals with sync problems.
Again, theres no input and output clock, just the one clock your interface gets somewhere. In the worst case over a cable.
orthodox wrote: What makes you think I am?
This makes me think you do:
orthodox wrote: could just make a test and transfer 10GB file through 3 s/pdif cables at 192k*24bit and then compare it bitwise.

Ostermilk
Posts: 1535
Joined: 15 Jan 2015

22 Feb 2015

If Creative/E-Mu have managed to make an ASIO driver that actually plays nicely with one of their cards on a 64 bit Windows box that would be something of an advance from them.

This doesn't engender much faith in me that much has changed there.

Probably a great gaming card, with all those 'sound enhancement' technologies going on (which would be a bane for production purposes) but for $200 you'll probably not have too much trouble finding an interface better suited to audio production, balanced I/O, Mic Pre's etc.


Post Reply
  • Information
  • Who is online

    Users browsing this forum: No registered users and 12 guests