[ibm_jennifer]

A plain PCM wave file with NO compression would take ~400 MB for a 6 minute song.

You are perfectly right, its for a stereo and I am speaking about 5.1 channel.

[navin]

Originally Posted by ibm_jennifer

Leave about ARC. Do you know whats RTA and what it does? have you ever tried Alphine 605 - Do you know where from they get their algorithm?

I agree with you and I am not using with any EQ or DSP for SQL.

Its True. But can you show me a Car speaker which reproduce the frequencies perfectly in a car environment?

Noise and rattling issue are trademarks of TATA and you need some brain and time to solve it yourself. if you don't have any of that, I advise you stay away from a Safari.

And in my world, a software is not a true RTA, if it wont support 1/24th octave resolution FR and 1/24th octave Phase Response with Magnitude & Coherenc and inbuilt professional mic and sound card calibration.

Originally Posted by ibm_jennifer

There is no big difference bw this High End and my poor JBL 607 if I am playing Lossless 24 bit 192KHz samples.

The secret is where the Distortion happens, it happens at your source, for me its at the destination.

But its impossible now because there is no High End Speakers available to reproduce the input with a Flat FR Curve.

There is no Cross-over available here without introducing Group Delay,

There is no Amplification/EQ available here with out Phase delay.

Originally Posted by ibm_jennifer

There are no more advantages in DTS-MA/TrueHD/24-192 Flac when comparing to DVD/CD other than High Dynamic range, Less Quantization Errors, Pure Transients, Lower Noise Floor, More Headroom, Less Jitter, Zero DC Offset, less Aliasing effects and finally more frequency spectrum.

If you cant feel any difference, You are really very OLD or you need to train your ears.

Originally Posted by ibm_jennifer

There is no use of explaining this mastering techniques to an end user unless your ear cant find any difference. The main Aim of this 24/192 is to simulate the old day Analog Harmonics in the Digital Music with pure reproduction..

Originally Posted by DerAlte

* High Dynamic range: cannot exceed that of the recorded stream

Using an Dynamic Range expander (in the 80s DBX made analog versions - I used the 3BX-DS)

* Less Quantization Errors: but no one will be able to figure out LSB errors in playback, right?

On a 44/16 signal the errors in lower 2 bits are very diffcult to tell - 3rd LSB on it is audible. On a 192/24 bit signal LSB error should be more tolerant.

* Less Jitter: OK, what is the practical impact?

easily resolves using a buffer.

Hello boss, please do not assume that others have an inability to discriminate.

These concepts have been around since ages..What you are seeing today as path breaking discoveries are actually another step in the process of evolution of signal reproduction techniques.

DA,
I have put my comments to your post in bold.

IBM_J.
1. Please do NOT underestimate the knowledge of many on this forum. I do not consider myself a guru - there are many here whose depth of knowledge far exceeeds mine - but even I would not like to be underestimated. And BTW I am OLD, old enough to have first used RTAs in 1976-7 even if it was only a simple 1/3 octave 33 band RTA.

2. I have used an Alpine 650. In understand it's limitaitons. I might not use it again unless the application demanded the use of such a product. My wife's car has a 650.

3. You say you are not using any EQ or DWSP for SQL but on antoehr thread did you recommend the same to B&T? If so why? Forgive me if I am mistaken.

4. There is no perfect speaker (car or home) not even a speaker that has a flat frequency response (car or home), power response (car or home). All SUBJECTIVE listening sessions (not tests) have indicated that such a speaker is not nesscarily desireable.

5. You mention an RTA with "1/24th octave resolution FR and 1/24th octave Phase Response...". Really why do you need 1/24th octave RTA? That would mean 240+ bands of display! Do you view music or listen to it?

6. ofcourse no XO has 0 group delay - as long as the group delay is constant it is fine as the transiest response will be good. However even better people that you or me have had no problem with speakers even if the XO has a NON-constant group delay. Sigfried Linkwitz for example.

BTW it is mathematically possible to simulate a XO that has 0 group delay. all one needs to do is add an all-pass network to the XO that has the inverse group delay of the XO in question however in practice this would mean that you'd need the negative group delay meaning that the correction All pass network responds before it receives its triggering signal, which is obviously not possible!

The way to create a 0 group delay would be to saple the signal and use DSP to "reverse" the signal in the time domain (say using LIFO buffers) etc.. I dot know if such a system is practically implementable but the theory allows for it to be.

7. Similar DSP techniques can be used to sole phase delay.

we can go one but you get the picture.

Originally Posted by frankmehta

on a more serious note? Why not a playstation 3?

you funny man! Watching video in a car while driving can be dangerous; you want IBM_J to play GT5!

[NetfreakBombay]

Originally Posted by ibm_jennifer

I am speaking about 5.1 channel.

In commercial media (Audio DVD) 5.1 channel does not exist beyond 96 KHz.

I don't think you can get such music unless you have your own recording studio.

[ibm_jennifer]

Please do NOT underestimate the knowledge of many on this forum. I do not consider myself a guru - there are many here whose depth of knowledge far exceeeds mine - but even I would not like to be underestimated. And BTW I am OLD, old enough to have first used RTAs in 1976-7 even if it was only a simple 1/3 octave 33 band RTA.

No Bhai, no I am not to underestimate anyone else other than the questioner who ASKED me Why RTA? And I respect you because of the way you are asking/correcting my replies - than any one else in the world I have met! You are really a senior and I will be here atleast to learn this behavior from you.

I have used an Alpine 650. In understand it's limitaitons. I might not use it again unless the application demanded the use of such a product. My wife's car has a 650.

But its not a bad product at all and I have no rights to comment that unless I haven't actually tried it myself.

You say you are not using any EQ or DWSP for SQL but on antoehr thread did you recommend the same to B&T? If so why? Forgive me if I am mistaken.

No Sir, I never asked anybody to try VST based EQs. The only Linear Phase EQ I recommended is from McDSp which cost some $1000 and there is no VST version available still now. I asked B&T only to use Digital CrossOver with Phase correction and TA other than this fancy items like EQ, Compression, Dynamics, Limiter, Reverb, Ambisonics, Convolver etc...

Really why do you need 1/24th octave RTA? That would mean 240+ bands of display! Do you view music or listen to it?

Exactly, you dont need them in RTA, if you accept me 512 band of EQ is a marketing gimmick.

ofcourse no XO has 0 group delay - as long as the group delay is constant it is fine as the transiest response will be good. However even better people that you or me have had no problem with speakers even if the XO has a NON-constant group delay. Sigfried Linkwitz for example

No Bhai don't joke me, I am not perfect, I am just a student.

BTW it is mathematically possible to simulate a XO that has 0 group delay. all one needs to do is add an all-pass network to the XO that has the inverse group delay of the XO in question however in practice this would mean that you'd need the negative group delay meaning that the correction All pass network responds before it receives its triggering signal, which is obviously not possible!

Do you referring Analog here? If yes, I agree that its impossible in Analog. But there are Digital cross overs with zero group delay or at least in near zero acceptable time difference.

easily resolves using a buffer.

I agree. But what about the latency introduced by this buffer?

[ibm_jennifer]

In commercial media (Audio DVD) 5.1 channel does not exist beyond 96 KHz.

I don't think you can get such music unless you have your own recording studio.

Its True. And its possible now with TrueHD, but limited because of the throughput of the dics. May be possible in future.

[NetfreakBombay]

Do you have samples at 192 KHz, 24 bit (stereo is fine) that were originally recorded at 192 KHz. I am not talking about "remastered" content.

Any 20 - 30 second clip will do.

All theory is fine, I am trying to see if _my_ ears can spot any difference.

[ibm_jennifer]

Yes, But they are in GBs and only in my car as my M-Audio/X-Fi cant play at 192 - My car can!.

Have you tried the sample at 192Khz Flac in Download our testfiles

All theory is fine, I am trying to see if _my_ ears can spot any difference.

If you cant find any difference bw 48 and 96, then you won't find any difference bw 96 and 192.

192 is nothing but a analog reproduction of a digital cycle using 9 points at 20KHz where 96 is 4 points where the minimum requirement is 2 as per Nyquist theory.

Its like how many points you need to draw a pure circle with hand. Someone like Nyquist from old generation used to draw using 2 points, but I am telling, you need atleast 8 points. But 4 is the correct and adapted for our next generation, thats it!

Its like HD vs SD, if you still love 800*600 for your computer, you don't need 96KHz in Music.

[navin]

Originally Posted by ibm_jennifer

But its not a bad product at all
Exactly, you dont need them in RTA, if you accept me 512 band of EQ is a marketing gimmick.

I agree. But what about the latency introduced by this buffer?

1. The 650 is a limited product. Not bad but limited. and yes 512 is pure marketing! I would have been happy with 30.

2. latency is relative. if the entire signal is time delayed by a few millisecs who will notice.

[ibm_jennifer]

The 650 is a limited product. Not bad but limited. and yes 512 is pure marketing!

Perfect, But I need atleast 1/48th octave FR scale to test the quality of the 512 band EQ in Alpine 650, Isnt it?

latency is relative. if the entire signal is time delayed by a few millisecs who will notice.

Agreed, as long as its in sync with the Video.

[navin]

Originally Posted by ibm_jennifer

Perfect, But I need atleast 1/48th octave FR scale to test the quality of the 512 band EQ in Alpine 650, Isnt it?

Agreed, as long as its in sync with the Video.

1. Yes and No. Yes if you want to test it's tech specs; No if you want to test it's effect on the audio qualty - for this you only need those 2 appendages we have on either side of our head and the stuff that lies between the above mentioned appendages.

2. Watching video while driving is dangerous. Playing PS3, Wii etc.. is even more dangerous. Ok now I am sounding like an Old man.

[ibm_jennifer]

Watching video while driving is dangerous. Playing PS3, Wii etc.. is even more dangerous. Ok now I am sounding like an Old man.

But not all Old Men are wrong. We don't watch Movies or Play games in the Car while driving. We have technologies to route them to rear passenger and the front monitor is just to control the volume.

[Digital Vampire]

Originally Posted by ibm_jennifer

No, Unless you feed your amplifier with pure Analog signal, you cant feel the transients. The transient response of the analog devices vary based on the quality of the input signal, thats why I ASKED YOU TO TRY FIRST.

Hold on...I missed to understand something here!!

How is the transient response of an analog device variable w.r.t the quality of the input signal?

IIRC, the response to transients is the characteristic of the analog device/system itself to which the signal is fed. For example, even if you have pure transients in the input signal, an amplifier may not respond to them if the amp does not have a fast/good enough transient response characteristics.

I have another doubt in this whole CarPC SQL equation. How are we planning to filter the switching noise and harmonics from the OPUS Power supply. It would run a computer just fine, but we are taking SQL here.

[ibm_jennifer]

How is the transient response of an analog device variable w.r.t the quality of the input signal?

We need atleast good transients in the input signal, isn't it? Transients are high frequency signals which almost aliased in low sampling rated signal. Transient response may not directly based on the quality of the input but the Dynamic range and bandwidth against the noise floor plays a major role for perfect transient reproduction. A pure sinusoidal transient will get good reproduction than a aliased square wave transient. Also in Active crossover the TR is affected by the Phase Delay and Group delay introduced in the input signal.

How are we planning to filter the switching noise and harmonics from the OPUS Power supply.

No switching or Amp thump here and I moved the PSU to the first floor in my cabinet to avoid any induction and cables are shielded from interference and the PSU and Xonar are stared together from ground loop.

[Digital Vampire]

Originally Posted by ibm_jennifer

No switching or Amp thump here

What I meant was the inherent switching noise emanating from a switching power supply (Example: OPUS/M2/M3 or just about any SMPS). This noise and its harmonics gets coupled to the system through the power supply (12V, 5V, 3.3V, GND) lines.

I've heard that the highest quality audio equipments run from traditional power supplies and not from switching power supplies. I even heard of a Nachamichi pre-amp run entirely from re-chargeable batteries so as to have the purest, ripple-free and harmonics-free DC.

[ibm_jennifer]

What I meant was the inherent switching noise emanating from a switching power supply (Example: OPUS/M2/M3 or just about any SMPS). This noise and its harmonics gets coupled to the system through the power supply (12V, 5V, 3.3V, GND) lines.

I've heard that the highest quality audio equipments run from traditional power supplies and not from switching power supplies. I even heard of a Nachamichi pre-amp run entirely from re-chargeable batteries so as to have the purest ripple and harmonics-free DC.

True, But as you are all thinking mine is not actually a SQL setup. I am just using $100 speakers and my Aim is to deliver the highest possible pure sound using that low end speakers.

24bit @ 192Khz doesn't mean that I am using the finest possible digital sound, there is actually a consumer audio which is sampled at 2.8MHz introduced by Sony and already people started to deliver at 32bit for Mastering studios. 24/192 is a pure consumer product just like HD in Video which is nothing when comparing to 4k/70mm. You are all thinking that mine is a SQL because you all are still at 16/44.1, but one day you will find this thread useless and move this thread to off-topics when you are all at 32/384.

But the Acoustic Noise introduced by this SMPS is inaudible to human ear unless they have a fan or are unloaded/malfunctioning as per: Switched-mode power supply - Wikipedia, the free encyclopedia But I am not sure about this as I have zero knowledge here.