Sound Blaster Audigy 2's equalizer measurements

[Zemus]

In a thread a while back Tom said that he preferred to record while both the bass and the treble controllers in the Audigy's mixer were set to 0%. Now, I've been wondering how much of a difference the equalizer really makes and if it's true that its neutral position is at 0%.
I used RightMark for this test (available here) for this test. Since I own the platinum pro version, I can use the sound card for both output and input since the ground on the external box's line-in 3 is separated from that on the sound card so there's no feedback loop being created.
Here are my findings:



As you can see, when both bass and treble are set to 0%, the sound card attenuates the signals between 0-200Hz and 4000-20000Hz. Between those there's a slight amplification to about 6dB. Both set to 100% gives an almost inverse graph, but with more distortions as the sound levels became too high for the input.
50% on the other hand, is almost completely flat, except for the expected parts at less than 20Hz and above 15000Hz where it decreases to about -1.9dB at 2000Hz. This should also show that the Audigy 2 is a great card for recording. The only limitation I know of is the limitation where the ADCs operates at 48kHz even if you record at 44.1kHz. Recording at 48kHz easily avoids this problem, and you can always convert to 44.1kHz later if it's necessary.
The other tests I did (stereo crosstalk, dynamic range and so on) were also positive. I can post the rest of the graphs if anyone's interested.

[Tom]

Record at 48kHz, eh?  I should start doing that, then.  I usually don't even think of it; I'm so used to clicking the record button without even thinking...

[Alistair]

Remember Tom, that's with a SB AUdigy 2 ZS Platinum Pro- a far cry from the humble 'Audigy'. Maybe the convertors aren't 48 on that?

I better find out if my Audiophile is 48 even at 44.1!!

- Alistair

[Zemus]

Audigy operates at 48kHz when you record too. Audigy 2 (the ZS being the 7.1 version with a slightly upgraded external box, I think) is also capable of recording at 24-bit/96kHz which bypasses the 48kHz "engine", otherwise it records everything at 48kHz and converts in hardware if your software is set to 44.1kHz. I tried to compare 44.1kHz and 48kHz modes and 48kHz did have a bit less distortion and gave generally better readings. 48kHz at 24-bit gave even better except for the frequency response (the graph I posted above) where it varied between 0.2dB and -0.2dB on the higher frequencies instead of being flat. Not something that's noticeable, just a bit weird... 3dB is said to be the smallest difference in volume the human ear can hear.

It'd be interesting to see the readings on other cards, like a terratec or an audiophile to see if there's any big differences.

[Glottis]

It'd be interesting to see the readings on other cards, like a terratec or an audiophile to see if there's any big differences.

Some readings I got with RMAA 5.5 for my card.

Analog I/O loopback tests, unbalanced connection.  

Results might be a bit better when using balanced output and input for testing, but I don't have the cables to do that.

44,1kHz, 16-bit
96kHz, 24-bit
192kHz, 32-bit int

Digital I/O loopback tests

44,1kHz, 16-bit
96kHz, 24-bit

[Zemus]

I created reports with RMAA too and uploaded them:

48kHz, 16-bit
48kHz, 24-bit
96kHz, 16-bit
96kHz, 24-bit
96kHz, 32-bit

As you can see, it does something funny in the frequency response when you try 48kHz, 24-bit. I have no idea why... It doesn't do it when I try 24-bit at 44.1kHz.
Seems like it scores best at 96kHz, 24-bit so I'll be using that from now on.

[Glottis]

As you can see, it does something funny in the frequency response when you try 48kHz, 24-bit. I have no idea why... It doesn't do it when I try 24-bit at 44.1kHz.

Here's a blast from the past when I still had my SB Live! and the Hoontech DB3 digital i/o addon board for it. Can't help but to notice similarities in the 48kHz figure.

RMAA 5.1 digital loopback tests

44,1kHz, 16-bit
48kHz, 16-bit

I do remember reading from somewhere that the SB Live! still resamples to 48kHz internally even if it's set to record 48kHz via software. No idea if this applies to the Audigy series.

[Zemus]

Wow, the frequency response at 48kHz is way off!
Were those really done with the digital input? They should be completely flat.
What I read was that the ADCs in the cards were running at 48kHz no matter which frequency you recorded at, so if you used any frequency OTHER than 48kHz, it'd be resampled before getting to the software.
Resampling from 48kHz to 48kHz should result in a 1:1 conversion.
I tried looping SPDIF-out to SPDIF-in, but the signal's way too low to do any measuring. I can't get it louder than -64.1dB.

[Laust]

SB Live! cards were notorious for resampling everything, and using a buggy resampler which effectively decreased the resolution due to rounding errors. There's more info on the kX Project website, I think.

[Glottis]

Were those really done with the digital input? They should be completely flat.

Yes I tested the card myself, and ideally they should be flat. Here's a quote from http://www.linuxdj.com/audio/quality/ regarding SB Live!

For example, at http://www.pcavtech.com/soundcards/ct4620/index.htm it is stated that

"Tests of the digital input of this card have been pretty disturbing... the digital input still seems to get unneeded DSP processing and corrupts sound quality by adding fairly broad 1 dB high peaks around 4 kHz, 10-12 kHz, and 16-20 Khz, with 44 kHz sampling. The digital output measures 1 dB down at 13.5 KHz, and -3dB at 17 khz, and -15 dB at 20 kHz. This is atypically bad performance for sound card digital I/O...."
   
And at http://www.maz-sound.com/sblive.html we read that "...the DSP runs fixed at 48 kHz, means every sound runs through the internal 8 point interpolation with 48 kHz ... digital 1:1 copies are impossible (the 48 kHz SPDIF frequency can't be changed to anything else either). [This] means: the SB Live! is no replacement for a digital-only card or an EWS64 L/XL for instance..."

Here's another one from http://kxproject.lugosoft.com/tech.php?language=en

It is widely known that 10k1 and 10k2-based audio cards perform audio resampling even when the incoming audio signal is 16/48. This happens due to not-so-perfect implementation of the SRC algorithms in hardware. For Audigy and Audigy2 cards (and, probably, for 10k1-based cards with chip revision >= 7 as well) the 'modified' 16/48 audio stream can be restored by using 'b2b' or 'FXBusX' plugins in the kX DSP.

The nature of the SRC bug causes all the audio data to contain partially-wrong 16th bit, thus giving you '15.5-bit playback' (and not '16bit'). The known solutions (the DSP plugins mentioned above) restore the 16th bit of the audio data, but only for the original 16/48 content. That is, if the incoming signal is not 16/48, but is, say, performed at 44100, 22050 or any other frequency, the 'b2b' and 'FXBusX' plugins, obviously, won't "correct" it, but will change the signal a little bit. That's why this correction is not turned on by the kX Audio Driver by default.

What is the difference between using b2b and FXBusX?(advanced topic)
The main difference is that FXBusX not only restores the 16th bit, but also performs sound truncation to 24 bits (optionally obtaining the least 8 bits from FXBus2 sources, if available), while the 'b2b' plugin restores the 16th bit only. From the user point of view this causes the following problem: if the truncated audio data generated by the FXBusX plugin is passed thru the Routing and Epilog plugins, it is automatically restored to '15.5bit' state (due to mathematical conversion), while the 24-bit audio data is not. So, when using FXBusX for bit-to-bit 16/48 playback, it is preferred to avoid adding any volume controls and route the output directly to the epiloglt (lite version!). The 'B2B' doesn't truncate the audio data and can be easily inserted between, say, FXBus 0,1 and the Routing.

The output signal (for instance, of the SPDIF outputs) might get truncated / rounded by hardware. This option is card-dependent (certain cards perform that, while the others don't). That's why it is recommended to check bit-to-bit playback and the particular B2B/FXBusX chain before using it (for instance, by trying the 'Direct SPDIF Recording' method and a SPDIF loopback cable).

[Zemus]

Interesting... I wish I was able to test the digital inputs and outputs on my card. In Device Control, there's an option called "Bit Accurate Mode" that says it turns off any modification on the digital input, but it's hard to tell without testing it.

[jbltecnicspro]

If I may put my two cents in... I find it very interesting that everyone is getting different results with SB LIVE, Audigy, Audigy 2, etc.  All of those soundcards are built on the same, exact audio engine processor, so should it not matter what sound card we use?  Now I am aware that one card may have a DAC more than the other and vice versa, but since they are all of the same EMU logic core, I wonder why they all function differently?  Eh, just another signature of Sound Blaster, I guess. :?

[Zemus]

Uh, not exactly. Live! and Audigy used different processors and Audigy 2 also had an update, chips-wise fom the Audigy 1. The ZS added DTS-ES decoding and 7.1 compatibility so that one should be identical to the regular Audigy 2 (if they didn't switch around any other chips that is).
The Live! and Audigy 2 were also released years apart and there had certainly been an improvement in both ADCs in that price range and circuit development with regards to interference from the rest of the computer.

[jbltecnicspro]

Ok, just checking.  I did read somewhere that the everything Live and up (except for X-fi) had the same logic cores to them.  Only difference is better ADCs and DACs, plus more decoding options.