Audio ASUS Xonar Essence STX Sound Card with 124dB SNR Now Official

Ah, that's quite a catch.

The 1212 is not going on sale anytime at all. I spent a lot of money and time upgrading and maxing it out, so it's well above the level of the off-the-shelf product. It may be relegated to a lesser rig, but the card is a stellar performer now and I'm very happy with the software setup and ease of use.

Nor is this card in the picture for me. I'm looking at a Firewire unit that can be cracked open to provide I2S feed, and a Buffalo 32S (the new version), but this is only by the end of the year. I'm broke now. The Buffalo that I use now will then be up for grabs, along with the Delta 66 which is currently in the backup rig - that card is even more heavily modded, with Fine Gold and Black Gate caps, new opamps and regulators, EMI suppressors, that sort of thing.
 
Update to an error on my part. The pop comes only when switching between PCM and Dolby Digital Live and when switching between speaker and headphone. It doesn't appear when enabling/disabling spdif output.
 
sangram said:
Ah, that's quite a catch.

The 1212 is not going on sale anytime at all. I spent a lot of money and time upgrading and maxing it out, so it's well above the level of the off-the-shelf product. It may be relegated to a lesser rig, but the card is a stellar performer now and I'm very happy with the software setup and ease of use.

Nor is this card in the picture for me. I'm looking at a Firewire unit that can be cracked open to provide I2S feed, and a Buffalo 32S (the new version), but this is only by the end of the year. I'm broke now. The Buffalo that I use now will then be up for grabs, along with the Delta 66 which is currently in the backup rig - that card is even more heavily modded, with Fine Gold and Black Gate caps, new opamps and regulators, EMI suppressors, that sort of thing.
Click to expand...
good for me, i guess. I m assimiliating moolah for a complete lyrita setup and this might just be the source i could get started with. The savings could be done only by this year end. (useless awards in IT field FTW)

Keep me posted.:p
 
I had a listen to the Xonar at Chaos's place and i have to say even I(im partially deaf wrt high end audio :p) could make out the difference! Great card :D
 
The experiment of swapping opamps didn't go too well. Tried 2x LM4562 and 2x OPA2134 for I/V and neither sounded as good as the cheap stock JRC2114D. I'm quite surprised by this.
 
I'm not. You obviously mean OPA2134.

Those sockets destroy any potential benefits of better opamps.

Also, LM4562 is a little sensitive to the layout and decoupling.

The card operates off a very low voltage, 12V at best. No -12V rail on the PCI-E slot. I posted about this on the other forum. No opamp likes that kind of voltage, and any voltage doubler makes the whole thing pointless.

I'm actually surprised you were so delighted by the performance of the analog section. I'm not about to bash equipment I haven't heard, but a PCI-E card is tough to get good analog performance out of, given that the max an opamp can swing is off a 12V line.

I was offered a card for review IF i wrote favorably and IF I joined another forum and posted it there. Of course, I refused :p or I would have contributed a bit more to this topic.
 
^^Err yea... corrected :).

The quality of the sockets ain't great to be honest. So what you say could be true. The card doesn't feed off the slot and as you said PCI-E doesn't have a -12V in any case. So yes there is no physical -12V but I'm sure internally the card has some sort of inversion to get a -12V line. An inverter is not too hard to build.

Either way, I'm not disappointed. Considering the limitations of what exists in this price range, this card is a whiff of fresh air. It sounds great and measures well too.

I'm really surprised about the review offer. Really strange!
 
There is no inverter in the card iirc. The opamp is probably biased at the mid point of the supply voltage, making it unnecessary to have the -12V supply (its equivalent to running it on a +6V,-6V supply).

@Chaos, Good that you were able to compare the JRC with the BB. Im surprised too that the OPA2134 sounds worse.
 
sangram said:
I was offered a card for review IF i wrote favorably and IF I joined another forum and posted it there. Of course, I refused :p or I would have contributed a bit more to this topic.
Click to expand...

Lol wtf! :p :rofl:
 
Raghunandan said:
The opamp is probably biased at the mid point of the supply voltage, making it unnecessary to have the -12V supply (its equivalent to running it on a +6V,-6V supply).
Click to expand...

Distortion is much worse when running a card off a low supply voltage. Here is the datasheet page which compares distortion at 5V and 24V, the 12V performance will realistically be in between the two.

 
Looking at those distortion graphs, it seems unlikely that they'd be using a purely 0-12V supply considering the card measures so well with RMAA.
 
Distrotion doesnt scale linearly with voltage. Beyond a certain voltage the change in distortion is negligible. For the 12V I would imagine the distortion to be really close to the 24V curve rather than a linear interpolation.

Also, distortion is of major concern when the output voltage is close to the supply. Notice on the graph, that the distortion is almost the same till about ~0.9V, even for such a big change in supply.

I still dont feel the card uses a -12V supply. Hopefully we can confirm it with a meter, when I make the regular weekend pilgrimage to the Bangalore HQ :p
 
Raghu, download the pdf and zoom into the graph. At 0.9V output, the 24v distortion is .00009%, at 5V it is .001%, or >10 times the distortion.

There is a proven relationship between the distortion products of an opamp and the supply voltage, mostly to do with bias of the output stage. There is a seminal piece by Samuel Groner, which you need to download. You will see that for low-end opamps there is no help offered to distortion by increasing supply voltage, but good audio and instrumentation opamps have substantial improvements in THD+N as the supply voltage is increased to near the limits.

The relationship is not fixed, and is driven by the opamp's architecture, bias and input/output type. I request you to not make blanket statements like

Distrotion doesnt scale linearly with voltage. Beyond a certain voltage the change in distortion is negligible.
Click to expand...

It is a different matter that even .001% is indeed very good, and may not be discernible over .00009%, for most people. But that is another discussion.
 
Please look again. At 0.8V here are the distortion figures for 24V and 5V

24V - 0.0001%
5V - 0.0002%

Its abt 2 times at the worst case (0.8V o/p is worst as the distortion starts increasing for the 5V supply beyond this point). The distortion as we go to lower output voltage only gets closer to each other.

Ofcourse the max output voltage without increasing the distortion is 1.2V for 24V supply, and 0.8V for 5V supply.

I mentioned 0.9V in my previous post as I couldnt make much out of the image, and thought that it might be the minimum distortion point.

I still stand by my earlier statement and maybe I didnt put it clearly then. Heres a better explanation of my stand.

The betterment in distortion due to a delta change in supply at lower supply voltages is much greater than the at a higher supply voltage.
Click to expand...
 
2 times distortion is 6dB, quite a lot. It's the difference between a -114 dB and -120 dB, with the best case 24V distortion being -130dB, and the 34V distortion (in the datasheet, not the pics) is -136dB. Each 6dB drop is half the distortion. And as voltage reduces, max available swign drops off, quite drastically,

Anyway I don't think we'll come to any agreement on this either, so I'm dropping this. No point in continuing the discussion - I don't expect someone of your experience to make comments like this, so here's one more shot from the datasheet. If you so feel inclined, a quick read will tell you why I don't agree with you.



Thread unsubscribed.
 
I just corrected your numbers. Its not 10 times as you made it out to be (you were checking at the incorrect voltage).

And again with the above graph, the IMD is about twice as bad. And you will notice that going from 24V to 34V doesnt give any improvement in IMD, while going from 5V to 24V is a huge change. The improvements saturate beyond a certain supply voltage. Reiterating my first statement.

I will add more points

1) The output impedance of the opamp is likely to be more than 10k rather than 2k, since its driving another opamp.
2) As the output impedance increases the THD+N becomes more independent of the supply voltage .ie. the noise figures will be essentially the same for all supply voltages. This is evident from the graphs for THD+N & IMD for 10k load impedance.

Anyway the best thing is to measure the supply voltage while the card is running. Hopefully someone who has the card can measure and tell us.
 
Back
Top