Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Music Clip where lowpass 16 is obvious? (Read 12538 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Music Clip where lowpass 16 is obvious?

I recently did a sweep test and I can easily hear past 20k (I don't have to turn up the volume or anything).  But I had a hard time telling ff123's mustange clips Original from 15kHz cutoff.

I then encoded a some music at --lowpass 16 and could distinguish no loss of frequency.

Does anyone have a clip where the difference is blindly obvious to them. Lowpassing at 16 saves a lot of bits and if I can't tell the difference, why waste them? Perhaps my material does not make it obvious enough.

Peter


Music Clip where lowpass 16 is obvious?

Reply #2
Quote
Originally posted by JohnV
Try castanets.
http://lame.sourceforge.net/download/samples/


Actually once I settled on my settings, I DL'd and encoded everything in that directory. I gave them all a casual listen with no real issues. I could tell on Fatboy and couple of others, but nothing objectionable. I will give Castanets a more critical listen after work.

Peter

Music Clip where lowpass 16 is obvious?

Reply #3
I tried the castanets on the ABX training site and can not reliably pick out the 15KHz versus 22KHz sample. I consider this issue settled for me. lowpass 16 should be fine.

Peter

Music Clip where lowpass 16 is obvious?

Reply #4
Quote
Originally posted by guidryp
I recently did a sweep test and I can easily hear past 20k (I don't have to turn up the volume or anything).
I seriously doubt it. Your equipment is probably not all that good and you're just hearing harmonics around 18kHz when the tone is any higher because the equipment can't play any higher frequencies.

Taking my equipment into account (stopping assuming that I can hear something when the tone doesn't seem to be increasing in pitch anymore) I've determined my own upper hearing limit to be about 19.5 kHz.

Are you sure about yours? Sorry if you are.

Music Clip where lowpass 16 is obvious?

Reply #5
Quote
Originally posted by NeoRenegade
I seriously doubt it. Your equipment is probably not all that good and you're just hearing harmonics around 18kHz when the tone is any higher because the equipment can't play any higher frequencies.

Are you sure about yours? Sorry if you are.



A longer slower sweep might help. I am running digital out right into my Denon 1802 reciever. Its not junk and should easily reproduce clean 20KHz when not loaded.

I get the same result with the Denon through my Paradigm Monitor 3's as I do with my Koss R/80 Phones.

When I ran a sweep at work with the same Phones coming out of  my H/K computer speakers. It was silent from 16KHz or so onward.

Is there a longer sweep out there?

Music Clip where lowpass 16 is obvious?

Reply #6
Since I'm at the bottom of the results of the "how high can you hear" poll, I can't hear lowpass, but I've gathered three samples from my CDs, that have very much treble up to 22 kHz. They're electronic music.

Cosmic Baby - Heaven's tears - Funny how time flies remix.



Air Liquide - Sun Progress



Transwave - Vision quest.



I tried to upload them, but Brinkster crashed, Geocities don't want uploads >5MB, and I don't want to use the bandwidth of my homepage.
I can mail them to someone for hosting them (4, 1 and 2 MB), if someone is interested in.

Again, I can't hear the 20 kHz tones, but as far as I can tell, the mixes are perfectly clean, and I doubt they're noise or artifacts. They must be part of the music.

Music Clip where lowpass 16 is obvious?

Reply #7
Thanks for the thread. I was able to find the neat little tone generator in them:

http://www.marchandelec.com/fg.htm


At work on a DELL with built in sound and running through HK speakers to my phones.  The cutoff is around 15.5KHz.

I did notice something odd in the ff123 sweep. Kind of like another superimposed sweep.

Just looking at the Winamp spectrum analyzer it looked like the sweep was always covering a range of freqs.

Hopefully this tone gen will be less likely to produce harmonics or alaising error with the single pure tone.

Peter

Music Clip where lowpass 16 is obvious?

Reply #8
Quote
Originally posted by NeoRenegade
I seriously doubt it. Your equipment is probably not all that good and you're just hearing harmonics around 18kHz when the tone is any higher because the equipment can't play any higher frequencies.

Are you sure about yours? Sorry if you are.


I stand corrected. I was playing with the tone generator, and it seemed beyond 15KHz the tone didn't necessarily sound like it was increasing, but it was still outputing all the way, but seemed to be varying.

Crap now I guess I will have to burn some test tracks on a CD and try that. I guess trying to ABX anything over 15KHz is pointless. I don't know if it my computer outputs correctly at that point. I trust the Reciever. Its the Nforce MB that I question. I didn't expect it would do anything to the digital stream. But apparently so.

I get the same from the analog outs on the card as the Denon gives me after a digital out.

I would like it better if it went silent after 15KHz, all those Harmonics pretty much means spdif out is not worth crap on my setup.

Hey maybe I can use lowpass 15. :-)

Peter

Music Clip where lowpass 16 is obvious?

Reply #9
I have the same situation.  My on-motherboard "sound" chip gives bad aliasing distortion.  When I had a different computer with a Sound Blaster AWE32 there was no such thing.  Alas, my hearing petered out at somthing like 17k if I remember right.

Unfortunately, I can't decide whether to order a Turtle Beach Santa Cruz or ignore the problem and shop for better speakers for my mini-stereo...
A fearless audio explorer in search of Truth, Beauty, and the elusive MPC-busting sample!

Music Clip where lowpass 16 is obvious?

Reply #10
Now I burned a CD with with most of the sample directory:

http://lame.sourceforge.net/download/samples/

original wav, followed by APS --lowpass 16 -V3 -b90 then back to wav.

Plus the sweep (but it was limited to some lower frequency on disk).
The only spot where I thought I noticed a difference was on Castanets. And it might not be the low pass that was doing it, but just my solution not using enough bits for the attacks.

The sweep sounded the same lowpassed as original, once I got away from the nasty aliasing distortion of my computer. I think I can peg my cutoff around 15-16KHz.

What kind of sound cards is everyone using that doesnt give them this crappy alaising???

Peter

Music Clip where lowpass 16 is obvious?

Reply #11
If you're looking for a new sound card I would suggest looking at either M-Audio Delta Audiophile 2496 or Delta 410 2496. Depends if you also want 5.1 analog output from DVD's.

Both have great sq, check here

http://www.midiman.net/products/m-audio/delta.php

You can buy them from Digital Connection's store: https://www.metro-one.com/dc_store/system/P...asp?MFG=M-AUDIO


To make a proper decision I would also look here
http://www.avsforum.com/avs-vb/forumdispla...p?s=&forumid=26

Oh, and btw: encode a few cds with a 16khz cut-off then buy one of these and a decent set of headphones and then kick yourself in the nuts, twice, but real hard

Music Clip where lowpass 16 is obvious?

Reply #12
Have a look at terratec products too, the EWX and the 6fire are worth. http://www.terratec.com
I use a Marian Marc 2 (not Terratec)

Music Clip where lowpass 16 is obvious?

Reply #13
Sorry for doubting your hearing, guidryp.

Music Clip where lowpass 16 is obvious?

Reply #14
Quote
Originally posted by NeoRenegade
Sorry for doubting your hearing, guidryp.


What are you sorry about, you were absolutely correct. My computer does really nasty alaising. After burning to CD, I estimate my cut-off around 15.5 - 16k.

Peter (in Canada too, eh)

Music Clip where lowpass 16 is obvious?

Reply #15
Quote
Originally posted by guidryp
Thanks for the thread. I was able to find the neat little tone generator in them:

http://www.marchandelec.com/fg.htm


At work on a DELL with built in sound and running through HK speakers to my phones.  The cutoff is around 15.5KHz.

I did notice something odd in the ff123 sweep. Kind of like another superimposed sweep. 

Just looking at the Winamp spectrum analyzer it looked like the sweep was always covering a range of freqs.

Hopefully this tone gen will be less likely to produce harmonics or alaising error with the single pure tone.

Peter


Check the "How high can you hear" thread on this board..

This came up,  and apparently you can get soundcard generated aliasing on very high freqencies,  that's probably what your hearing.    Possibly the sound program ff123 used might have generated some distortion as well..

I used Cool Edit 2000 to make a longer one for more accuracy,  and found similar noises,  but:  1. Bringing the sound level from 0db to -3db reduced them a lot,  2.  Changing from 44.1Khz sample rate to 48Khz made them completely inaudible.  This on a SB Audigy card.

If I remember right,  you can get a demo version of Cool Edit,  crippleware,  even if it dosen't save, you could create and play back a tone sweep.
www.syntrillium.com  If you have a need for a sound editor,  CD 2000 (the non pro version) is pretty good bang for the buck,  and you can buy "addon" packs for more features as you need them.

Music Clip where lowpass 16 is obvious?

Reply #16
Quote
Originally posted by JonPike


Check the "How high can you hear" thread on this board..

This came up,  and apparently you can get soundcard generated aliasing on very high freqencies,  that's probably what your hearing.    Possibly the sound program ff123 used might have generated some distortion as well..

I used Cool Edit 2000 to make a longer one for more accuracy,  and found similar noises,  but:  1. Bringing the sound level from 0db to -3db reduced them a lot,   2.  Changing from 44.1Khz sample rate to 48Khz made them completely inaudible.  This on a SB Audigy card.



Yes, that worked perfectly. Apparently my card handles 48KHz natively as well. The alaising is unreal on the ff123 44.1KHz sweep. Converting the sweep to 48 KHz also worked well. But I was able to get a real good idea of my hearing with a slow sweep and a big frequency plot.

I notice the first drop off right around 15KHz and it is silence by 16KHz.

Too bad about the aliasing since, most stuff seems to be in 44KHz.

Peter

Music Clip where lowpass 16 is obvious?

Reply #17
Quote
Originally posted by tubenut
Unfortunately, I can't decide whether to order a Turtle Beach Santa Cruz or ignore the problem and shop for better speakers for my mini-stereo...

Think twice about the Santa Cruz.  The drivers are the most unstable I've ever seen for a sound card... sound quality is good, but the drivers need work IMHO.  The 'control panel' will crash Windows if you change settings while sound is playing.  I bought the card yesterday, and am probably going to return it tomorrow.

There are some reviews on the Web pointing this out, try searching Google for "Turtle Beach Santa Cruz Review" (no quotes).

Music Clip where lowpass 16 is obvious?

Reply #18
Quote
Originally posted by fewtch

Think twice about the Santa Cruz.  The drivers are the most unstable I've ever seen for a sound card... sound quality is good, but the drivers need work IMHO.  The 'control panel' will crash Windows if you change settings while sound is playing.  I bought the card yesterday, and am probably going to return it tomorrow.

There are some reviews on the Web pointing this out, try searching Google for "Turtle Beach Santa Cruz Review" (no quotes).

Hmm, I change the settings on my Santa Cruz all the time while sound is playing and Windows doesn't crash. I'm using Win2K SP2.

Music Clip where lowpass 16 is obvious?

Reply #19
Quote
Originally posted by fewtch

Think twice about the Santa Cruz.  The drivers are the most unstable I've ever seen for a sound card... sound quality is good, but the drivers need work IMHO.  The 'control panel' will crash Windows if you change settings while sound is playing.

I just installed mine about 30 minutes ago. I'm running Windows XP, using the 4142 build drivers, dated Jan. 10 2002, WHQL certified (from the Turtle Beach website). No crashes yet! (although for some reason, the pressed CD that came with the card would read _very_ slowly on my newly purchased Pioneer DVD-116, making it unusable). I fiddled with all of the control panel settings, and Windows had no problems.

It sounds different from the SB Live I had in there before... Is it better? Hard to say, it could just be louder/different EQ, but I'm enjoying it, and most importantly, resizing windows doesn't cause it to skip!

I'm going to test it myself sometime next week, but maybe someone already knows the answer to this: does the Santa Cruz have the same high-frequency aliasing problems as the SB Live when resampling 44.1khz audio?

Music Clip where lowpass 16 is obvious?

Reply #20
Quote
does the Santa Cruz have the same high-frequency aliasing problems as the SB Live when resampling 44.1khz audio?


Being based on an AC97 codec, which resamples everything to 48 khz, yes, there is some aliasing, but in the case of Santa Cruz the algorithms are much better. Of course, M-Audio Delta is much better than Santa Cruz

Check here:

http://www.pcavtech.com/soundcards/compare/index.htm

Also:

Santa Cruz:
http://www.io.com/~kazushi/audiocard/santacruz/

Audiophile:
http://www.io.com/~kazushi/audiocard/audiophile/

Live!:
http://www.io.com/~kazushi/audiocard/sblive/

Music Clip where lowpass 16 is obvious?

Reply #21
Thanks for the links lucpes. Finally a test of SB Live! using 48kHz and rear output. I'd like to see similar test of Audigy, do you know any?

Music Clip where lowpass 16 is obvious?

Reply #22
Use this cool program to test your soundcard:

http://people.freenet.de/l2001/scan.zip

Sound Card Analyzer
Version 2.0
2001
Author: Alexey Lukin

This program performs various tests of electroacoustical performance of sound
cards and other real-time audio devices. Testing is accomplished by playing the test sig-nals
and recording them after they pass through the testing chain. Because of this, your
sound card must allow full-duplex operation (see “Full-duplex operation”).
In the simplest case the testing chain includes the DAC of your sound card, line
or speakers output, line input, and ADC. In order to test some other devices you must
connect them between the output and the input of your sound card. It is assumed that
you have a high quality sound card, because otherwise poor performance of the sound
card will mask the performance of the external device.
You can use external A/D and D/A converters or digital inputs and outputs of
your sound card.
System requirements: CPU: Pentium or higher, operating system: Windows
95/98/2000.


Things U can test: Frequency response test, Interference and noise level test, Dynamic range test, Total harmonic distortion test, Stereo channels crosstalk test

Music Clip where lowpass 16 is obvious?

Reply #23
for more reviews I would go here:

http://www.digit-life.com/hardware/sound.html

the only hardware-oriented site to do good reviews for sound stuff.

Audigy review: http://www.digit-life.com/articles/audigyp...umex/index.html

edit: http://www.kxproject.com for 3rd party Live & Audigy drivers... try 'em out and see what you think...

AC97 Mixer from http://www.upsystems.com.ua/support/alexmina/ is a nice program for those who like to tweak stuff

Music Clip where lowpass 16 is obvious?

Reply #24
Thanks for the links Lucpes and Deranger.

I downloaded Sound Card Analyzer and carried the tests on my Audigy with kX project drivers and the results are almost identical to those on digit-life's review.
I'm suprised that I got similar results even though I used really long cable between line-out and line-in and didn't have high quality reference card to sample output. Only result that differed more from their results was Stereo crosstalk: digit-life measured -89.7 dB, my result was -96.2 dB.

Update:
I run the test with supposedly bad front out. 44.1 kHz mode was similar to 44.1 kHz using rear outputs, but 48 kHz was way better than in digit-life's test.
Frequency response: -0.56, +0.07 dB (dl: -2.55, +0.01)
Interference and noise: -94.2 dB (dl: -91.1)
Dynamic range: -87.4 dB (dl: 88.3)
THD + noise: 0.009 % (dl: 0.127)
Stereo crosstalk: -97.0 dB (dl: -64.6)

Is the program reliable, or are kX drivers really so superior to Creative's drivers?