16/44 vs. 24/96 comparison

If you use Audigy cards make sure that all the EAX effects are turned off/disabled or you will always have the sound going through a lot of DSP trash.

IIRC, the Audigy cards share the same basic drivers. If you have a program such as WaveLab you can set it up to work with the Creative ASIO driver.

Hope this helps.

 

I've recently been listening to 96/24 mastered Sergio Mendes & Brasil '66 CDs and comparing their sound with a regular, well- mastered, CD of the same music. The former win hands down.

BTW, is noise shaping involved in downsamples from 192/24 to 96/24?

 

This is what I call the Tao approach. As the "Tao Te Ching" says, "Tao does nothing, and yet, there is nothing it doesn't do."


When at all possible the less is more approach is IMHO the best way to go.

 

I tried late last night on headphones (Sennheiser HD-580s) and could not hear a difference. I thought I heard more ambience in the 24/96 file but it disappeared when I did a blind test. I will give it a shot later with my speakers but my room acoustics aren't that good at the moment so I'm not sure what to expect. BTW, my E-MU card has to be switched between 44 and 96 manually which makes comparison difficult, so I upconverted the 16/44 file to 24/96 in Audition.

 

Hi Liquid Snake,

If you upconverted the 16/44, you can compare the upconversion to the original 24/96. But it is the upconversion you're comparing, not the files.

And that may be fine. Some of the folks I did this for are doing just that and I wanted them to have a means of evaluating how their electronically generated "high res" CDs compare with something that was recorded at high res.

Have fun!

Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

 

IMHO, that is just as bad for getting a real/honest experience with these files as those computers that downsample everything. Do as I did, just play one and load the correct new settings to listen to the second one.

 

that sounds silly. Comparing a native 24/96 song and one that has been re-sampled twice is nuts.

 

Did anybody try the Korg MR-1000 5.6 MHz double DSD recorder? Using this 5.6 MHz recorder really amazes me. It is far the most believable and transparent digital recording gear I have ever owned. For the price it is the deal of the decade.

 

Perhaps. It is pretty annoying and breaks my concentration when listening if I have to go and change the sample rate though.

 

Hi cybermantis,

Much will depend on the SRC algorithm.

I've got a couple I could use on a file multiple times, leaving considerably less sonic signature behind than many algorithms will when used once.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

 

use asio, it will force the card to switch by itself! at least in my experience it does that.

 

i see, it makese sense if it were properly done and time taken to ensure a resampling was done well, but in the scope of this test it would be difficult to ensure it. If the guy doesn't want to switch the card back and forth, I bet he isn't going to ensure that the resampling is done well. w/e, it just sounded like a bad idea.

 

I've always used ASIO but the E-MU cards do not support automatically switching between 44/48 and 88/96 for whatever reason.

 

ah thats a pain in the neck. You could use foobar and switch on the resampler so it is done in real-time. You can also use certain resamplers that may be of higher quality (I think the secret rabbit one is popular for this reason).

 

I know what you are talking about. But, I feel that the only way to do this test justice is to listen to it as Barry posted the files. Have you got a saved 'session' at the different sample rates. This is the simplest way to go about this.

 

I can open the files in Peak and the software automatically switches when I move from one file to the other. There's got to be something like this on the Windows side.

I imagine something like WaveLab or another ASIO-wise program should be able to. Is there anything you Windows users know of that someone who does not want to spend hundreds or more on their software can use?

Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

 

Barry, the thing is that E-Mu cards are quite 'manual' in the sense that you have to chose the target playback or recording sample rate in advance for them to play it. If you do and nothing can be heard it is because you are not getting a true data stream at that sample rate.

There is, however a simple way to have the E-mu soundcards change automatically so that you can listen to music originally at different sample rates: use the 44.1 option. Yet, I believe that this option has the audio data stream going through Windows, which will not deliver the best or most faithful sound quality.

 

True!

I've had to do this only once. I had to clean up an extremly clicky and noise 45 that I wanted to preserve, but could find NO other copy! The song is "That's What I Guessed"/"Shake Your Money Maker" by Clarence "Frogman" Henry on Dial Records. I never knew which side was the A side, nor can I find out what year it was releases, although I suspect it was in the late 60s.

Anyway, I used the "Younglove" method with Cool Edit Pro (the only time the process really worked for me!). It got that sucker clean with no artifacts, but it had no ambiance whatsoever. So, I added some reverb and EQ to bring it back to where it sounds just like the 45 minus the noise. You'd think you were listening to a new pressing!

 

A lost art in this day and age!

 

foobar2000 is free and if one uses the asio component, it is the same- what isn't is that apparently only certain sound cards automatically switch sample rates (eg: as Liquid Snake has said)- I think it is irrelevant if one is on a mac or a pc, it depends on whether the soundcard manufacturer decided to make the switching automatic or not.

 

Thanx for the post, Barry. I did notice a slight improvement in resolution (got 9 in a row right in abx test). However, I too had to upsample the 16/44 because I'm also on an E-MU. The differences were not all that apparent to me, but my system and environment are far from optimal. But I'm afraid what I was hearing was the degradation caused by the upsample.

And for the E-MU users: I'm running foobar through ASIO, 24-bit (internal foo option) and secret rabbit resampling to 96KHz. I play everything like that, because I hate to switch sessions between redbook and hirez files. Is there a better way to go about it, or should I just switch sessions?

Ah one more: Is there a plugin for windows that uses POW-r? So far I only read about Sonar, but I was thinking more in the lines of a plugin that could be used in Audition.

 

Hi cybermantis,

Agreed. I only mentioned PC because, although I use one (and have written software for PC -as well as mainframe and AS400), I do all my audio work with my Mac.

On my machine, it is a question of the software, not the hardware, since my D-A (Metric Halo) will work natively at whatever rate I send to it.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

 

i see, no worries- the two worlds are closer than are often claimed

 

Hi cosmosis,

Any degradation caused by upsampling should increase the apparent differences between the files.

This makes me question how good a job this card, which I assume is internal, is doing with the 24/96 signal. In my experience, internal cards tend to be more jitter prone than external devices. Perhaps this would contribute to the differences being less apparent in some situations than they are in others.

Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com

 

I tend to think it does a great job. It does have two internal PCI cards, but the in/out jacks are on an external rack. Don't know where everything takes place, but the A>D conversion is done outside, sinse the cable connecting it to the PCI is a network cable, not analog. Or at least that's my understanding.