Clawfinger said that good IRs sound great at 96khz and more....

@Clarky:
More or less, and basically, yes. When *done porperly* :D

Really hoping AlbertA sees this and provides his take on it :D
 
barhrecords said:
Could you clarify what you mean?

The internal sample rate of the convolution engine? The input and output sampling rate of the device? A DAW plugin?

Richard
Click to expand...
The recording rate.
I just tried a Mytek at different sample rates and i was surprised by the results.
I had RW and God's Cab IRs loaded and while the God's cab begun to sound "grayny" the RW sounded less mushy with more depth and realistic. It makes me though to what clawfinger wrote here, when he said that under 96khz it sounded digital.
It's a very anal thing (and you need a good monitoring system) but i think there is something here.
 
Last edited:
jimfist said:
mba has been MIA for a while, and he needs some love...how many pages you need this time, mba?
Click to expand...
I think my posts are more useful than your owns but you should ask to clawfinger if he needs love too.
I'm amazed how a lot of people here are involved in such posts.
 
shasha said:
Anyway the thing that bugs the living crap out of me about this whole thing in terms of sample rate is that I have heard on more than one occasion the same exact signal recorded at 44.1kHz and at 96kHz with the same gear and can hear the difference.
Click to expand...
Yes ...and i should say of course. You are not dreaming.
Some people said that you experiment a difference with cheap audio interfaces.
Mytek is not a cheap audio interface and there is a difference.

If some people has some PrismSound stuff here...?
 
Last edited:
It really boils down to the implementation.

Nyquist is a theory. Then you have various manufacters implementing hardware and software.

Bit depth and sampling rate have theoretical limits but each manufacturer is free to mess that up :)

So, in theory, the 48K sampling rate and IR length in the AxeFx are sufficient to convolve the impulse response of a guitar speaker.

Richard
 
mba said:
The recording rate.
I just tried a Mytek at different sample rates and i was surprised by the results.
I had RW and God's Cab IRs loaded and while the God's cab begun to sound "grayny" the RW sounded less mushy with more depth and realistic. It makes me though to what clawfinger wrote here, when he said that under 96khz it sounded digital.
It's a very anal thing (and you need a good monitoring system) but i think there is something here.
Click to expand...
This has nothing to do with the math/science of IR use (convolution). This is an observation that provides for a discussion of ADC/DAC quality. Though a Mytek is higher quality, it still *suffers* from the simple, real-world, applied side of conversion. If you monitor the applied results (convolution) of an IR captured @ 48k and then at 96k via the same converter @96k (for both IRs), what you have done is provided a reason for scrutiny of the ADC, and/or the convolution method in use.

Please keep in mind that this is in reference to this case. As soon as one decides to consider any other case, new variables, inherently, are being introduced, and thus such case would require new consideration(s).
 
mba said:
I could add that SONY uses 80-100khz for their SA-CD technology. Nyquist does not say all.
Click to expand...
Nyquist/Shannon is not in doubt. Unless you are able to provide credible, scientifically confirmed citation, please do not spread (more) malicious misinformation. This is similar to a pedophile using candy as bait IMO. Please think and research before simply regurgitating.

SA-CD technology does not "use 80-100k," it is DSD that the SA-CD spec allows for. The DSD spec is a 1-bit, ~2.8MHz (go find the exact value after the decimal; I am not in the mood to search at this moment). This particular format, while technically a PCM variant, utilizes a different approach for audio reproduction (and capture, processing). Studies have been performed over one year+ durations comparing *traditional* PCM (24/96k+) with DSD, and no user discernable difference was discovered. This is counter to user reports of "audiophiles" claiming to hear a definite difference. That, however, is not pertinent to the discussion here... but, I will add that, as a DAW(quencer) format, DSD likely remains a far too expensive (processing overhead) format to implement as a replacement for wav, with (likely) very little benefit in the end.

As far as the original comment is concerned..
The claimed 80-100k is a matter of debate, and has been shown to have a practical, reproducible top of ~50k. It is not the "extended range" of frequency content that provides for a user perceived "better" of DSD over PCM wav; rather, it is the method itself that is the point of debate, as PCM wav content of comparable frequency content was utilized during studies. Please read that again to understand why Nyquist/Shannon is irrelevant: Both studies allowed for the same frequency range of content to be reproduced (upper 170k for PCM wav, IIRC), thus reducing the variables to focus upon method.

If one considers the scientific studies that have shown indiscernibility between DSD and PCM wav (24/96k+) by (audiophile) end users, this might open a discussion that could consider the (Lavry associated) contention that a ~64kHz conversion (end) rate might be the best compromise. However, and to reemphasize!!! This does NOT counter Nyquist/Shannon in any way, PERIOD. What this simply provides is for the conversation to be refined, so as to focus upon: methods of capture/process/reproduce (PCM variants? Something else? Push into ultra-sonics ala DSD? Does that push introduce (possible) undesirable bleed/aliasing/artifacting/etc? Is there a better method?); Should we be focusing more upon the converters (ADC/DAC)? Is the human listening experience truly constrained at ~20k? And more.

Once again, Nyquist/Shannon is finite. Unless one has citable information derived of scientific method (and not correlative, lazy hypothesis), please look elsewhere for your reasonings, and above all, please do not spread (regurgitate) misinformation. Please. And this goes doubly for horrid analogies, especially using pixels as a comparative (screams and beats the walls).
 
barhrecords said:
It really boils down to the implementation.

Nyquist is a theory. Then you have various manufacters implementing hardware and software.

Bit depth and sampling rate have theoretical limits but each manufacturer is free to mess that up :)

So, in theory, the 48K sampling rate and IR length in the AxeFx are sufficient to convolve the impulse response of a guitar speaker.

Richard
Click to expand...
I feel bad correcting this...
Nyquist hypothesized.. tada! Theory.
Shannon proved.. tada! Theorem.

There is no try. Do, or do not.
In any case of application of said theorem, the task falls to the applier to provide the proper method(s). Nyquist/Shannon remains absolute. It is not a variable.

(I do understand the intent behind your statement though! I think I do.. :D)
 
nikki-k said:
This has nothing to do with the math/science of IR use (convolution). This is an observation that provides for a discussion of ADC/DAC quality. Though a Mytek is higher quality, it still *suffers* from the simple, real-world, applied side of conversion.
Click to expand...
I agree that it maybe comes from the conversion and even the best ADC can provide differences with different sample rates, but i don't know for sure where it comes from, that's why i said..."there is something here".
By the way the SA-CD technology take care of the sampling rate and allows a higher bandwith than the CD which set me in trouble.(but perhaps Sony engineers try to fool people? That's what you mean?)
I sent a mail to a french guy who owns PrismSound stuff and he told me that he experiment differences between sample rates too. (very slight between 96 and 192khz according to him)

By the way "This has nothing to do with the math/science of IR use (convolution)." is an assertion or you have to expose your scientific argumentation.
 
nikki-k said:
Please think and research before simply regurgitating.

SA-CD technology does not "use 80-100k," it is DSD that the SA-CD spec allows for. The DSD spec is a 1-bit, ~2.8MHz (go find the exact value after the decimal; I am not in the mood to search at this moment).
Click to expand...

Wow...
It's here :
Super Audio CD - Wikipédia

"Le format DSD :
Le Super Audio CD utilise une technologie de numérisation DSD ((en)Direct Stream Digital)[SUP]4[/SUP] à très haute fréquence d’échantillonnage de 64 fs (1 fs = échantillonnage CD = 44100 Hz) quantifié sur 1 bit qui autorise une bande passante allant jusqu’à 80 kHz voire 100 kHz et une dynamique de 120 dB ; elle est donc bien supérieure à celle recommandée par le théorème d’échantillonnage de Nyquist-Shannon. "

Translation :
"The DSD format :
The Super Audio CD uses a DSD sampling technology at very high sampling frequency of 64fs (1 fs equal the CD sampling frequency of 44100hz) quantified on 1 bit which allows a bandwith of going to 80-100khz and a dynamic of 120dB. It is much higher than the one recommended by the Nyquist-Shannon theorem."

If people want to translate the other parts of the subject they will know who throws bs here.
Nyquist/Shannon is not in doubt. Unless you are able to provide credible, scientifically confirmed citation, please do not spread (more) malicious misinformation. This is similar to a pedophile using candy as bait IMO. Please think and research before simply regurgitating.

SA-CD technology does not "use 80-100k," it is DSD that the SA-CD spec allows for. The DSD spec is a 1-bit, ~2.8MHz (go find the exact value after the decimal; I am not in the mood to search at this moment). This particular format, while technically a PCM variant, utilizes a different approach for audio reproduction (and capture, processing). Studies have been performed over one year+ durations comparing *traditional* PCM (24/96k+) with DSD, and no user discernable difference was discovered. This is counter to user reports of "audiophiles" claiming to hear a definite difference. That, however, is not pertinent to the discussion here... but, I will add that, as a DAW(quencer) format, DSD likely remains a far too expensive (processing overhead) format to implement as a replacement for wav, with (likely) very little benefit in the end.

As far as the original comment is concerned..
The claimed 80-100k is a matter of debate, and has been shown to have a practical, reproducible top of ~50k. It is not the "extended range" of frequency content that provides for a user perceived "better" of DSD over PCM wav; rather, it is the method itself that is the point of debate, as PCM wav content of comparable frequency content was utilized during studies. Please read that again to understand why Nyquist/Shannon is irrelevant: Both studies allowed for the same frequency range of content to be reproduced (upper 170k for PCM wav, IIRC), thus reducing the variables to focus upon method.

If one considers the scientific studies that have shown indiscernibility between DSD and PCM wav (24/96k+) by (audiophile) end users, this might open a discussion that could consider the (Lavry associated) contention that a ~64kHz conversion (end) rate might be the best compromise. However, and to reemphasize!!! This does NOT counter Nyquist/Shannon in any way, PERIOD. What this simply provides is for the conversation to be refined, so as to focus upon: methods of capture/process/reproduce (PCM variants? Something else? Push into ultra-sonics ala DSD? Does that push introduce (possible) undesirable bleed/aliasing/artifacting/etc? Is there a better method?); Should we be focusing more upon the converters (ADC/DAC)? Is the human listening experience truly constrained at ~20k? And more.

Once again, Nyquist/Shannon is finite. Unless one has citable information derived of scientific method (and not correlative, lazy hypothesis), please look elsewhere for your reasonings, and above all, please do not spread (regurgitate) misinformation. Please. And this goes doubly for horrid analogies, especially using pixels as a comparative (screams and beats the walls).
Click to expand...

By the way what is this comparaison with pedophilia? Are you crazy? It's just unbelievable.
 
Last edited:
By the way i have to correct what Clawfinger said but it doesn't change my own questions.(i can't change the title of the thread)

Clawfinger said this "we've done some extensive testing with DSP's and it felt bad when we tested any frequency under 96KHz."
All apologize Jocke. ;o)
 
Hey, how about that Flux capacitor
burpy.gif
 
mba:
The irrational and scatter-brained nature of your posts defies reasonable, rational discussion.

The pedophile comparison is a very snarky jab. I feel that lazy, willful ignorance.. a preventable condition.. is disgusting. For someone who opts for that way of life to regurgitate information that flies in the face of scientifically proven (theorem, not theory) data is vile.

I had written a lengthy reply, but came back and edited it down to this, as I feel that most of it was unnecessary. One point:

You quote barhrecords stating Nyquist as theory, and follow with:
mba said:
You are right and it seems that a signal spectrum is much more complex than a theory on datas.
A spectrums is not only datas, it has a kind of coherence.
Click to expand...
While what barhrecords states is, in fact, true, it omits the fact that said theory became, in fact, a theorem due to Shannon's proof. Your reply demonstrates not just your willful, lazy ignorance, but a desire to continue the spread of candy.. erm, I mean... falsehoods. Omission, allusion, illusion... delusion?

The absolute inanity you display, and the (apparent) refusal to willingly educate yourself, leads me to bow out of this.. "discussion"... (and stop torturing the other poor forum members with my babble :D ), and to instead engage in a more fitting conversation with my (far) less vapid friend here:
donkey32.jpg
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom