This shows the difficulty of dynamic measurements - wouldn't the dynamic circumstances that are causing the DAC's internal noise generation be at it's lowest during processing of such low level signals as reverb tails? But I wouldn't just focus on low level sounds as the possible area where perceptual effects might occur.
At the other end of the scale, I have a sense that the generated noise may effect the perception of the start of & rise time of the initial transient of all sounds. This, IMO, would effect the solidity of the soundstage - the start of each sound is now less distinct & therefore the perceived timing effected (we exclusively use ITD for locating sounds below ~700Hz & it becomes of less importance as frequency rises)
The generated noise could also be effecting the attack portion of the sound - a portion of the sound that we are particularly sensitive to?
The FFT gives an accumulated reading of this noise but does it show the dynamic range of this noise or the noise at any point in time? It would be of interest to know the range of this fluctuation.
I'm asking myself a question. How long take our ears to analyse the harmonic profile of a sound ( or an instrument) ?
I used often in studio to clip on purpose an instrument to add some attack on it.
It is tricky, but, if carefully done, it works and the effect is the contrary from what some could expect: a feeling of added dynamic. Of course, it have to be short and moderate.
Our ears are very trained to analyse the evolution of the harmonic contents of an instrument. At the attack of an acoustic guitar chord, with a nail or a pick, you have a lot of impair and high orders harmonics that stop very fast, while during the resonance of the note, the harmonic content decrease slowly from lower but high orders (5, 3, 2) to the fondamental.
Yes, I agree with this - our auditory perception has evolved with these two distinct levels of focus - the initial attack is quickly analyzed to categorize the sound & the evolving spectral & timing nature of the sound envelope can occur over a number of seconds. Both these factors (along with many others) are in play when we listen to music.
Can you define "timbre"? Not being confrontational - I just don't think it's very clear what it is & what parts of sound determine timbre?
Certain effects on auditory perception occur with age such as the ability to follow speech in a background of noise i.e a roomful of people speaking - this has been studied fairly well
I'm not sure the effect on perception of the attack portion of sounds has been studied but remember ITD is the perception of the interaural difference in arrival time of the same sound - doesn't need HF sensitivity & in fact is mostly used below ~700Hz & increasingly so up the frequency range where ILD becomes the dominant factor for locating the source of sound.
Again, the risetime of the attack portion of the sound i snot about HF sensitivity either - a bass note can have a sharp risetime
I am speaking in terms of audibility.
Not instrumentation.
Try an experiment yourself.
This reminds me of "It's already been through so many I.C.'s"
Bee. Ess.
Put it through 100 and you can still hear the 101'rst
Simple experiment:
Put 2 fuzztones in series.
Adjust either/and/or both .......
It sort of defies intuition until you experience it .
And here lies the problem (forcing fractions of my half brain to work)
Please notice the x-over mod at post #23 (cutting off the woofer peaking at 8-10kHz). Maybe Tournesol can try it and tell us what he thinks of it.
George
More stuff from the days of uA741's and uA709's, nothing here needs a new name.
Not at all, you're listening to nothing but the distortion, it's the message. Take two amplifiers maybe a Benchmark and a Pass Labs F5 now put back to back diodes on the inputs like some popular fuzz pedals and tell the difference. Some fuzz tones use one germanium and one silicon diode, I knew one person that swore by base emitter junctions on transistors removed from old IBM 1400 boards.
Probably so. The reverb used in recordings often consists of layered reverbs and delays. The artful layering of two different time effects is what makes the effect sound pleasing. The reverb tails end up being complex sounds low in level and diffuse in time and frequency. In some cases the tails are modulated in some way. Any thoughts on how best to proceed? Always complicated when we have A/Ds and D/As and are trying to pin something down on only one of them (except for with fixed tones
).
Last edited:
I wish I could help but the modulator designers are the right folks for this and they have obvious financial interests. BTW did your friend get the AD1955 board working, the software should all be a free download?
I'm sorry 38 yr. ago we've moved on slew rate, etc. are no longer relevant and the asymmetric waveform stuff was always fuzzy thinking.
Not my friend exactly, but a forum member with interest in trying that particular dac chip. He said had a computer that could run the software, and that he would ask for help if he had any trouble getting it working. That was the last conversation.
Say, you just gave me an idea about modulators. HQplayer has several 1-bit modulators that range from kind of experimental to optimized for certain DSD sample rates. Might be interesting to compare reverb tails between modulators and ask the author about any differences. Also, not sure if matlab has some modulator synthesis capability, sort of think it might.
Several open source platforms should be able to run algorithms off line in not real time to create files for listening tests.
My version is called "wireless": active, wifi, blue-tooth, USB, optical with a DSP filter ;-(
At the price of this speakers, isn't supposed to be nice, out of the box ?
Mr. Cook was listening to his speakers. Obviously, his successors do not !
It was the first time in my life I bought a speaker without to had listened to it before: not the best idea.
They are good for the bay.
- Status
- Not open for further replies.