"When for instance a sine wave is used of say 5kHz which is stopped at the 0 crossing what would the ringing be?"
Interesting how I've never seen measurements of this... We always see just the worst case scenario with a single maximum amplitude impulse response when as I mentioned previously, this kind of thing does not exist in real music.
So then folks, let us have a look at what a single "pulse" looks like with the different filters at 5kHz and 10kHz. Again, remember that I am measuring the waveform back using the E-MU 0404USB at 24/192 so there will be some "ringing" imposed by the ADC itself when looking at the TEAC UD-501 NOS mode especially.
Note: Ignore the phase inversion between 5kHz and 10kHz - I accidentally caused this inversion and it's not due to the TEAC hardware.
I. SHARP (Linear Phase) Filter:
Reminder of the impulse measurement:
5kHz (16/44):
10kHz (16/44):
Ringing evident at 10kHz but lower amplitude and shorter duration than impulse (ie. worst case scenario). Notice that the frequency of the ringing is the same as the impulse so we're looking at around Nyquist (22kHz).
II. SLOW (Linear Phase) Filter:
Impulse response:
5kHz (16/44):
10kHz (16/44):
III. Digital filter OFF ("NOS-mode"):
Impulse response:
5kHz (16/44):
10kHz (16/44):
IV. Minimal Phase Upsampling to 24/192 with Digital Filter OFF:
Impulse response:
5kHz (16/44):
10kHz (16/44):
V. Summary...
Some comments about the linear phase pre-ringing which some folks obsess over:
1. Lower frequencies like 5kHz essentially is associated with no pre-ringing to worry about in linear phase filters... Since the ear is most sensitive to tones from 1-5kHz, this may be reassuring. As you can see, pre-ringing does show up higher like at 10kHz.
2. Ringing amplitude also correlates with the frequency. The lower the frequency, the less the pre-ringing amplitude. For example, the SHARP filter 10kHz pulse pre-ringing amplitude is ~1/2 of the impulse.
3. The pre-ringing frequency itself remains high at around Nyquist like the impulse response so I remain skeptical that it's even audible (seriously folks, real musical recordings have a noise floor as well as complex harmonics - how anyone can claim that it's low-amplitude, high frequency pre-ringing causing any defect in the sound is a mystery to me!).
Of course, there is an "easy" way to not have to worry about the pre-ringing phenomenon in the audible spectrum... Go download a hi-res copy of the music ;-).
SHARP filter, 10kHz sampled at 88kHz:
Presto! Pre-ringing gonzo!
Note: Actually, there will still be some very high frequency pre-ringing for 20kHz sampled at 88kHz but we're talking 40+kHz ringing at even lower amplitude... I'm pretty sure I can live with this! (Again, please ignore the phase inversion compared to the graph above.)
Thanks for the measurement efforts !
ReplyDeleteThe closer you get to the filter's cut-off frequency the worse the ringing will be.
This is to be expected of course but haven't seen measurements of this.
As squarewaves have higher harmonics these will exite the filter's ringing more.
That's indeed the worst case scenario and why this test is often performed, to show worst case.
Most people, however, think that the ringing is always present and thus also is present in the audible band and therfore detrimental to the sound.
Music consists of plethora of sinewaves (perhaps with the exception of some synth tones) and these tests show the ringing with music signals is not present.
For those who may not know; a squarewave essentially consists of a sinewave fundamental tone (the same frequency as that of the squarewave) plus its odd harmonics (3rd, 5th, e.t.c.) that decrease in amplitude the higher the frequency of those harmonics is.
There is also little evidence of this ringing being present in the audio signal with null tests using actual music signals.
The only eveidence you will find is a small 'peak' at the filter's cut-off frequency in the 'difference file' in this case. Any (small) differences in the audible band will be caused by phase or amplitude (FR) differences or distortion.
Small FR and phase differences are not audible BUT very measurable by the way and caused by the filter or coupling capacitors in the chain.
This leaves us with the question that, besides the obvious FR roll-off, how much of the filters 'action' is indeed audible and with what sonic effect.
For the NOS DAC lovers that do not use an analog brick-wall filter with it... notice how nice a squarewave reproduces (which does not exist in music signals) and how Bad a sinewave is reproduced ?
Music consists of sinewaves.
For a good performing NOS (non Over sampling) DAC a sharp filter is essential that cut's off all signals above 1/2 the sampling frequency.
This is needed to remove the 'steps' one sees in the shown pictures of this fine post.
NOS DACs that must do multiple sample frequencies should therefore have a sharp filter based on the lowest sample frequency that can be present or (hard to make in an analog way) a filter response that can be altered depending on the used sample frequency.
As usual thanks for the response Frans.
DeleteGreat article!
ReplyDeleteIt'll be fun if you could experiment with the DAC in NOS mode and do the processing/filtering in the PC using HQPlayer's different upsamplings, convertion (PCM->DSD and/or DSD->PCM), "advanced" filters, etc. and see if it imporves over the IC's onboard filters (from their website):
Resampling filters:
6 linear phase
5 minimum phase
3 impulse optimal
Dithers and noise-shapers:
4 dithers
4 noise shapers
Delta-Sigma conversion:
3 modulators
6 oversampling filters (64x - 512x)
Direct rate conversions
In case you didn't know about HQPlayer, its site is here: http://www.signalyst.com/consumer.html (I'm not related to it in anyway), they have a fully functional 30 day trial version. IMO it is probablythe most flexible and tweakable player available though a bit of a hair shirt approach functionality wise.
Some details about how to configure and set it up here: http://www.computeraudiophile.com/f8-general-forum/hqplayer-resampling-filter-setup-guide-ordinary-person-13298/
Thanks Javier. I'll look into it in time...
DeleteFor now, working on some tests with Win 8 players and JPLAY like what I did with the Mac OS X players to see whether there are any surprises!
Very interesting but even if you find there are zero differences with other bit perfect players I'm afraid audiof**ls will claim that it still sounds much, much better measurable or not. As Frans put it, they don't trust measures nor graphs only their golden ears: "what do engineers know about sound anyway?".
ReplyDeleteOTOH, HQPlayer is a very "different" kind of player compared to JPlay, AFAIK it doesn't claim to have any "magical" or "esoteric" properties. It offers a huge array of possibilities for the tweaker that can actually make a real difference as it is not meant to be a bit perfect player but a digital audio processor that runs on the modern PC's powerful CPUs. I'm sure there will be measurable differences though whether audible or not is another story. BTW, it is available for Windows, Mac and Ubuntu based Linux distros.
Thanks again for the referral to HQPlayer. Had a look at the site - very interesting! Will definitely try out.
DeleteUpsampling to 32/384 with digital filter off could certainly result in very interesting waveforms for the TEAC to play. Alas, I suspect my ears would be grossly embarrassed as to whether it would be able to hear differences :-). Any specific settings you like or feel have made an improvement?
As for "audiof**ls". Sadly, I think this hobby has become very much associated with the magical thinking expressed by some of those individuals. Doesn't help that the mainstream audiophile press has made no serious attempt at separating fact from fiction. After all, I guess they need the ad revenue from Synergistic Research and their "Tranquility Base", "Quantum Fuses" and "Acoustic ART System" whether any of it makes sense or not (example off the top of my head). Hey, "harmless" fun hobby, right?
I generally don't mind folks who have a strong subjective take on audio but when it totally goes overboard into the bizarre, I'm gonna darn well post a thing or two about that if I can :-).