Perceived quality of TX audio when PS engaged

JJ4SDR
Posts: 436
Joined: Fri Jul 30, 2021 10:09 pm
Location: TEXAS, USA

Perceived quality of TX audio when PS engaged

Postby JJ4SDR » Sat Sep 24, 2022 8:51 pm

(Perhaps this belongs in another sub-forum?)..........I have been racking my brain for a long time over the perceived quality of audio when PS is engaged (this, I did much before I started using PS myself which is very recent). What I am writing is clearly an over-simplification (or, alternatively, making this much more complex or convoluted than it needs to be), but I would value comments from others.

The ANAN platform brings a plethora of professional audio sculpting tools and that is only one of the reasons for my wanting to acquire an ANAN (a little over 1 year ago) and, being able to use PS is a huge "bonus." :) In my case, I do process audio inside either a DAW or a VST Host, before the resulting audio is fed into the radio. I often receive unsolicited reports on the quality of my audio.

Yes, to most of you, I may be down-right off-base or nuts and getting into the weeds that just isn't worth spending time on. :roll: Also, there are so many different ways to mold audio that attempting to decipher what I (and some others, see below) may or may not be causing what I am hearing just doesn't make sense. Quality of headphones and/or reference monitor speakers and the listener's ability to perceive of frequencies and the separation among sub-bands of lows, low-mid/mid/high-mids/highs (I have striven to "train" my ears in various ways too) play a big role. I don't claim to have "golden ears" either (VERY few people actually possess them), but I do come from a musical family and have been exposed to different sorts of music and vocals starting at a very early age. I even had the audacity and gall to tell a professional pianist visiting our home (I was in my late teens at the time), playing a piece of classical music, that he went off-key at which point he stopped, cracking in a smile, and said that I was right on the money.

I would like to better understand why it could be that I detect a very slight deterioration in the quality of the transmitted audio (a few other hams using ANANs and PS have echoed what I am writing here with some being professional audio engineers). Some also don't avail themselves of the very significant decrease of IMD products that using PS brings, so they don't always switch it on. On the other hand, many do not hear any difference so I am clearly in the minority which is fine by me.

It is hard to put my finger on this, but I have conducted several A/B comparisons with PS ON or OFF with a few different stations that put out very strong high quality SSB signals, with nothing else changing in their audio chains. With PS on, I am hearing a slight "smearing" or "feathering" effect on the very low-end (80-120Hz) of an SSB signal, in a way (my interpretation), reducing the amount of "separation" between the very lows and the frequencies above. This effect makes the perceived audio sound just a smidge less transparent or "clean" in my ears (well, ears don't process the audio but our brains do). Or, perhaps, it is the highs that are being somehow de-emphasized, rendering the overall transmitted signal just VERY little less articulate. Some operators roll off at 0 Hz when their TX bandwidths are 4K or wider too. Perhaps, the frequency at which the low-end roll-off is set plays into my perception of their signals when they switch PS ON and OFF? From a theoretical perspective, there really isn't much information at 0Hz ;) ! Even professional music studios that mix/master music with scores of "stems" (or tracks) don't work at 0Hz so why should we, as ham radio operators? What would be the benefit?

OK, I got it out of my system....you're welcome to shoot me down. It wouldn't be the first time :lol: !

Juha
NI2M
PC: 8 Core i7-10700 CPU @ 2.90GHz, NVMe SK Hynix 512 GB SSD, 32GB RAM
Windows 10 Home, Version 22H2
Thetis v2.10.4.3 x64
Protocol 2 v2.1.18
User avatar
w-u-2-o
Posts: 5539
Joined: Fri Mar 10, 2017 1:47 pm

Re: Perceived quality of TX audio when PS engaged

Postby w-u-2-o » Sat Sep 24, 2022 10:10 pm

By definition turning on PureSignal will result in more accurate reproduction of audio. Period. Dot. Less IMD is less IMD.

You are suffering from a psychoacoustic phenomenon whereby you merely prefer what you hear with PureSignal turned off. This is quite common, particularly in audiophiles. Technical accuracy does not always equate to experiential preference. It is sensation vs. perception.

This is immediately drifting away from PureSignal effects, so I'm spanking myself for thread drift ;) , but since you are branding yourself an audiophile...

If your test cases involve an Apache/Thetis setup at both ends, try changing both ends to use linear phase filtering instead of low latency, and reduce the DSP filter size to 1024. This will soften the filter skirts and reduce any ringing that you might be sensitive to. As an audiophile that may make a much larger difference to you than the state of PureSignal.

I gave up my Stereo Review subscription way before the Carver Challenge proved that engineering can trump artistry, not that the artists or any of their customers care!
JJ4SDR
Posts: 436
Joined: Fri Jul 30, 2021 10:09 pm
Location: TEXAS, USA

Re: Perceived quality of TX audio when PS engaged

Postby JJ4SDR » Sun Sep 25, 2022 3:42 pm

Scott,

I don't "suffer" from a "psychoacoustic phenomenon" but rather I relish and savor every second of it and I am grateful for and happy over the unique ANAN platform that I have at my disposal ;).

To me, audio is very much about the "art", having to do with (at times, visceral, depending on signal strength) feelings, sensations, associations being evoked by a tone or a frequency. Yes, the "artiste" in me trumps my electrical engineering penchants and wants (I am an electrical engineer), having played classical music on the piano for years. To me, this conversation reminds me of the constant and never(?)-ending debates in pro audio circles that pit "analog warmth" (classic hardware consoles and other gear) against all-digital audio processing (mixing/mastering "in a Box") where it is becoming evident that technological progress in digital is clearly improving the art with incredible results. :)

Waxing philosophic, I don't think art in audio and engineering are mutually exclusive so the question of which one emerges as the "winner" can only be determined by the fellow down in the trenches ;).

Thanks for the reminder on linear phase vs. low latency settings. I have experimented with minimal phase, linear phase and "natural" phase (VST plug-ins) a lot and they do generate different sound signatures.

Dug into the Carve Challenge article too. Very interesting and illuminating although the scope was limited to Audio amplification only.

Talk about "thread drift!" ;)

Best!
Juha
NI2M
PC: 8 Core i7-10700 CPU @ 2.90GHz, NVMe SK Hynix 512 GB SSD, 32GB RAM
Windows 10 Home, Version 22H2
Thetis v2.10.4.3 x64
Protocol 2 v2.1.18

Return to “PureSignal Operations (PowerSDR & Thetis)”