I'm in the process of testing the Aureon Sky (loop-back) but I have encountered a few problems:

1. Adjusting the levels: I get different results, depending on how I choose the output-volume (master/front) and the line-in-recording-volume, although I keep the combined level at approximately the same niveau (e.g. -0.2, -6.2). Basically the noise level is increased (5+ dB) with higher rec-volume. Is this behaviour normal?

2. In the "adjust level" dialog the value of -400 dBFS is displayed, when I select 24-bit (3-byte packed), although the 24-bit mode is said to be supported and can be tested in RMAA 3.4 (currently im using version 4.3). 32-bit mode (PCM) works fine though.

3. Crosstalk: I get weird results in this category: one time the value is about -88dB, the next time about -74dB (with a logarithmic growth tending to higher frequencies in the graph).

4. Frequency response: Seems to be okay, but in 44.1kHz it is better around 30Hz than with 96kHz. Any reason for this?

5. Noise level: I'm not really impressed by the numbers (~95dB) ... Is it likely, that reomving a TV-card and a modem card will improve this value? Is shielding an option?

Here are the results:
Aureon Wave, 32-44
Aureon Wave, 32-96
Aureon Wave, 32-96 CT problems

Thanks for your time!

1. This behaviour is normal because the line-in tends to be (in most cases) of lower quality than the line-out. You should set the line-in volume as low as possible, this gives the most accurate test results.

2. I've seen similar problems with the Terratec 24/96 dmx 6 fire (look in the forum), and it seems this card doesn't support 24-bit, 3-byte packed mode, most likely this is the same problem with the Terratec Areon Sky. Developers of RMAA (thumbs up ) said they will support other 24-bit testing mode(s) in future release(s). So i think we have to wait.

5. I suggest you try by removing all other pci-card(s) and putting the souncard as far as possible from the graphics card to avoid (most) electromagnetic interference (EMI). If this helps, put the other pci-cards back one by one and see how they affect the test results. This should isolate the cause. If not than i'm afraid it's the card limitation.
By the way it is important that you use a high quality cable for line-in, line-out testing. But with these test results I assume you do.

I hope this helps you out

Owhh and one question can you also make test(s) in 16-bit mode (samplerate at 44.1, 48 & 96). I would be very gratefull.

Greetz Sikalot

1. You're right.

2. We are not planning any more 24- or 32-bit modes. Seems that we've done all we can here. Use 32-bit modes. Bad support of 24 bits is only a driver problem.

3. Maybe it's a problem of RMAA. We'll revise it in future releases.

4. No ideas. Is it stable from test to test?

5. Creative cards never showed super low noise levels (as far as I remember). Try your removing some hardware away from the sound card.

quote:

---

Sikalot:
By the way it is important that you use a high quality cable for line-in, line-out testing. But with these test results I assume you do.

---

quote:

---

Owhh and one question can you also make test(s) in 16-bit mode (samplerate at 44.1, 48 & 96). I would be very gratefull.

---

Aureon Wave, 16-44.html
Aureon Wave, 16-48.html
Aureon Wave, 16-96.html

------------

quote:

---

Alexey Lukin:
2. We are not planning any more 24- or 32-bit modes. Seems that we've done all we can here. Use 32-bit modes. Bad support of 24 bits is only a driver problem.

---

quote:

---

5. Creative cards never showed super low noise levels (as far as I remember). Try your removing some hardware away from the sound card.

---

I'll try removing cards later, I have not much time right now.

Thanks for your comments!

In RMAA 3.4 version we wrote "24-bit" but tried to open the card in 32 bits (that doesn't matter really).

Sorry, that's Terratec!

Thnx for the test results . I will buy this card (or the 7.1 Sky Version) soon. I'll make test(s) of my own and send them in. See if I get the same 'crosstalk'-problem.

quote:

---

gLiNa:
Could you please check the components used on this Aureon card. Who is the manufacturer of DACs and Opamps ? Is it still AKM and JRC ?

---

Envy 24 HT (the DSP, you probably knew that)

Takamisawa NA5W-K (could be the headphone amp?)

TL0726
28T
DDVO

XILINX
XC9536XL

WM8770IFF
2AAAEX1

SIGMATEL
STAC97447

Addition of 01-21-2003 10:11 PM:

quote:

---

Try your removing some hardware away from the sound card.

---

I did that. The effect was neglectible small though (like 1 dB).

So, how do these results compare to those of "similar" cards? (Older Terratec, M-Audio cards, ...) I haven't seen any of them on the results page.

What's this Creative story? I thought they were (only) attacked for their resampling 44.1 -> 48 kHz issue.

Still, I'm sure that Terratec did a good job with the Aureon and I'm looking forward to hearing one. I'd also love to see some details about M-audio revolution 7.1 and maybe some RMAA results.

Big thanks go to Hylas as i wouldn't have done that reasearch without the data he supplied. I'm still curious about the opamps used. Hylas : Could you please have another look at your card. There should be 4 or 5 identical chips, 2.5 or 5 milimeters wide with 8 "legs" each. Every one of them close to the line-out inputs. These are the opamps. What are they ?

Addition of 01-22-2003 10:42 PM:

Oh .. wait. Aureon Wave is a 5.1 card - 3 or 4 opamps are more probable ;)

4580
206
JRC

4 of those can be found near the output jacks.

So, how do these results compare to those of "similar" cards? (Older Terratec, M-Audio cards, ...) I haven't seen any of them on the results page.

There is a very good review of the Terratec DMX 6 Fire compared with other cards made by Maxim Liadov (better as Tom did ).
Here's a link : http://www.digit-life.com/articles2/terratecdmx6fire/index.html
It's seems Terratec DMX 6 Fire crosstalk performance is limited around -90db, so assuming Terratec's Aureon Sky has the same circuity as the 6 Fire (exept for the Envy24HT chip which enables 192 kHz samplerate)..the performance should be the same.

Btw does anyone know if the Aureon Space version also uses the same circuity as the 6 Fire?

On Terratec's site they describe the 'Space' as having an "exemplary frequency response and a signal-to-noise ratio of better than –100 db". The 'Sky' is described as having an "excellent frequency response...you can hear the difference...For 2, 4 or 6 loudspeakers and with a signal-to-noise ratio of more than -100db".
Is this just a marketing technic?

Addition of 01-25-2003 12:51 AM:

To add something, if you read it carefullly does more than -100db means less? example -90db. I hope this is the right conclusion, it would mean the 'Space' version schould give better test results .

Thanks for the link! But I would be more interested in loop-back results, since this is only test I can perform.

Of course, testing only either the output or the input of a card (using a better card, say Lynx Two) gives more detailed and representative results.

But as I cannot perform these tests (no Lynx, ), a loop-back result gives me a comparision point for my own measures.

I've had excellent results with testing the DMX in lineout-linein loop. Please have a look : http://free.of.pl/g/glina/DMX.htm . Accompanying hardware, motherboard and power supply seem to have a bigger impact on the results then the type of cards used for anylyzing. In my case removing or replacing the componens has no effect. I guess Tomshardware weren't that lucky.