Diy Question thread

[Tenson]

I don't imagine it used FIR filters? I was under the impression they needed pretty serious processing, perhaps not available in the mid 90's? Still, with a very high order slope the area over which there are phase issues will be minimised anyway so I've no idea how subjectively important that then becomes.

Is it possible to do the same filter with IIR and FIR on the DEQX Shin?

 

[3DSonics]

Hi,

I don't imagine it used FIR filters?

No. I would also argue that "phase linear" crossover and equaliser are not useful, you require a MINIMUM PHASE system, that means analogue or IIR.

In fact, junking traditional FIR Digital Filters from CD replay majorly improves subjective sound quality, to my ears.

Ciao T

 

[ShinOBIWAN]

Hi,

But surely you realise that the main mode of distortion in moving coil drivers relates directly to excursion, the other main sources relate to eddy current distortion which is minimised in my drivers due to a number of reasons inherently.

Your shifting my original argument. Accessing distortion based on excursion for a fullrange driver vs. a midrange driver in a 3-way is pointless since you've completely neglected to mention the dedicated bass driver which is very likely to be more compentant in its operating range because its electrical and mechanical parameters can be optimised in favour of just the bass range without looking toward the upper frequencies too.

Many dedicated mids have SPL and distortion capacity to belittle the fullrange in the same operating range, they don't need the displacement because it isn't needed and actually benefits performance in its optimal range.

If that is what you use the BSS I used all these years ago is DEFINITLY comparable. It cannot do as complex driver equalisation, but the general principles and implementation is very comparable.

And the BSS unit I used was different, build to much higher standards than what is sold currently by BSS (it cost around £ 3,500 at the time IIRC!).

As said, compared to any other hardware crossover the BSS still compares well, nearly 10 years later.

Ciao T

Its not comparable at all, for a start you wouldn't have had access to FIR filters in the mid 90's. It lacks any DRC or correction functions of substantial benefit, PEQ isn't what I use. Its time alignment abilities are basic at best. It lacks the capacity to perform detailed and accurate measurement in a real time manner. In short the BSS is a step up from the DCX2496.

Also £3.5k back then isn't £3.5k now. You only need to look at AV processors and computer hardware to realise that digital technology becomes obsolete sooner rather than later.

 

[ShinOBIWAN]

I don't imagine it used FIR filters? I was under the impression they needed pretty serious processing, perhaps not available in the mid 90's? Still, with a very high order slope the area over which there are phase issues will be minimised anyway so I've no idea how subjectively important that then becomes.

Is it possible to do the same filter with IIR and FIR on the DEQX Shin?

Hi Simon

Yes its possible to use IIR filters as well as FIR.

FIR with phase alignment is noticeably superior in terms of imaging, soundstage and overall detail. For the measured performance, the FIR setup offers many advantages with regards to power response and optimising drivers for their intended pass band. In particular the phase measurements are night and day between IIR and highorder FIR filters in a standalone capacity.

 

[3DSonics]

Hi,

Your shifting my original argument. Accessing distortion based on excursion for a fullrange driver vs. a midrange driver in a 3-way is pointless since you've completely neglected to mention the dedicated bass driver which is very likely to be more compentant in its operating range because its electrical and mechanical parameters can be optimised in favour of just the bass range without looking toward the upper frequencies too.

But my bass driver is bigger than yours.... ;-)

And yes, I have two as well.

Many dedicated mids have SPL and distortion capacity to belittle the fullrange in the same operating range, they don't need the displacement because it isn't needed and actually benefits performance in its optimal range.

Alas, most dedicated midranges do NOT have the SPL capabilities of the Driver I use.

It relates to "normal" Fullrangers about as much as your ATC 3" Midranges relate to "normal" Midranges. Therefore talking "commonplaces" has little benefit in context.

Its not comparable at all, for a start you wouldn't have had access to FIR filters in the mid 90's.

And FIR filters are of extremely limited utility.

It lacks any DRC or correction functions of substantial benefit, PEQ isn't what I use. Its time alignment abilities are basic at best.

PEQ used correctly can come very close in results to DRC algorythms as used by TACT (and I suspect by DEQX). Using excessive resolution in the DRC filtering means you have a perfect response in one spot and there only with the response anywhere else pretty fubared.

The time alignment abilities of the old BSS unit was perfectly usable to make sure you got a reasonable squarewave through the system.

It lacks the capacity to perform detailed and accurate measurement in a real time manner.

We used a TEF for measurements.

Also £3.5k back then isn't £3.5k now.

I realise that, however a lot of that money was in the analogue and AD/DA stages, not in the main CPU/DSP.

The point remains, I have used techniques very similar to yours a long time ago. I would use them again if I had a suitable application, however I prefer different methodes nowadays.

DSP and DRC are useful but they are not a replacement for good basic design.

L8er T

 

[ShinOBIWAN]

Hi,

But my bass driver is bigger than yours.... ;-)

And yes, I have two as well.

I've forgotten what we were discussing... I'm pretty sure it wasn't that. Good diversion though.

Alas, most dedicated midranges do NOT have the SPL capabilities of the Driver I use.

It relates to "normal" Fullrangers about as much as your ATC 3" Midranges relate to "normal" Midranges. Therefore talking "commonplaces" has little benefit in context.

Again with the shifting.

You started off with a generalisation and now you want to move away from that to get specific?

And FIR filters are of extremely limited utility.

That's more ignorance from the same back yard as your directivity statements and the fact that you seem to think the system I use is comparible to something you used over a decade ago. I do agree that some FIR windowing functions have limited value regarding loudspeaker filtering but there are ones that are very strong ones too.

The likely reason, and this seems fair given your lack of experience with them, is that you simply haven't heard them properly implemented for loudspeaker filtering purposes.

Let's face it Thorsten, there's your way and then there's the wrong way. How narrow minded of you - which brings me back to my very first contribution in this thread.

PEQ used correctly can come very close in results to DRC algorythms as used by TACT (and I suspect by DEQX). Using excessive resolution in the DRC filtering means you have a perfect response in one spot and there only with the response anywhere else pretty fubared.

Strong filtering that shifts outside the headroom on offer is possibly the worse sounding thing I've ever heard. Completely artificial and an absurdly phasey nature outside of the sweet spot - horrible.

What I use is rather minor amounts since my design is already strong as is. Add on top of that the physical treatments and the digital elements simply further refine the sound of an already good sounding system. This is exactly how it should be, only use it to enhance the sound and the measured performance.

DSP and DRC are useful but they are not a replacement for good basic design.

Of course not, I never said they were.

It does allow me several advantages though and gives me a balanced system with far more strengths than weaknesses.

 

[Johnny]

Size doesn't matter...it's what you do with it that counts.

''DSP and DRC are useful but they are not a replacement for good basic design.''

Neither is FIR or any other drc algorithm a replacement for a correctly designed room.

''Make everything as simple as possible, but not simpler''

~Albert Einstein

 

[ShinOBIWAN]

Size doesn't matter...it's what you do with it that counts.

''DSP and DRC are useful but they are not a replacement for good basic design.''

Neither is FIR or any other drc algorithm a replacement for a correctly designed room.

''Make everything as simple as possible, but not simpler''

~Albert Einstein

I(probably a few others too) don't care what you think Johnny, you've forfeit all right to that with your past performances and extreme lack of offering anything constructive in all area's of discussion.

Most of what you state is plainly obvious and has such brevity and lack of detail that it wastes valuable pixels on my display. Not to mention the seconds that it takes me to read it.

 

[Tenson]

I thought it was just getting hot in my room, but I guess not!

T, I take your point about FIR equalisers when used as corrective tools, but can you expand on the use of FIR for crossovers? It seems a good thing to avoid non-linear phase shifts.

Also, I have often thought the use of FIR equalisers would be great for use in the studio when subjectively altering the sound, not being used as correction in a minimum phase system. That way the signal is maintained even through the mashing about it gets in a mixing session!

Cheers

 

[I-S]

resistivity is the inverse of conductivity

Correct. But that completely fails to answer my question.

whereas I haven't specified the length or csa of the material.

Again correct, you haven't. That's totally irrelevant to resistivity. You would only need to specify those things in order to know resistance

So I ask you again.

It has a higher ratio of direct current to voltage,

Is this different to resistivity and if so, how?

 

[3DSonics]

Hi,

=T, I take your point about FIR equalisers when used as corrective tools, but can you expand on the use of FIR for crossovers? It seems a good thing to avoid non-linear phase shifts.

I guess it depends on how we view the function of crossover.

If we want a crossover and driver to produce bandpass which can be combined with other bandpass devices and integrates over a wide range of spatial coordinates I cannot see how we can achieve that using a linear phase filter at all.

Ciao T

 

[ShinOBIWAN]

Hi,



I guess it depends on how we view the function of crossover.

If we want a crossover and driver to produce bandpass which can be combined with other bandpass devices and integrates over a wide range of spatial coordinates I cannot see how we can achieve that using a linear phase filter at all.

Ciao T

Your lack of knowledge and experience in this subject is a joke Thorsten. Do I have to explain it all in baby words?

The only thing you seem to be good at shifting arguments surrounding fundamental flaws in products you design, such as the transparence.

 

[3DSonics]

Hi,

Your lack of knowledge and experience in this subject is a joke Thorsten.

As you say so it must be true....

Do I have to explain it all in baby words?

Please do so. Please explain why a zero phase bandpass filter combined with a non zero phase transducer as a bandpass will provide superior integration across a wide range of vertical and horizontal angles when multiple sources with complementary bandpass coverage are arranged non coincidental. Baby language is fine.

The only thing you seem to be good at shifting arguments surrounding fundamental flaws in products you design, such as the transparence.

There are no "fundamental flaws". The design was done to attain certain properties and it does so rather well.

You may disagree with the properties it is designed to exhibit, which is your perogative.

It does not change the fact that conventional speakers such LIKE the ones you use are on fundamental first principles designed to not allow a good acoustic integration into acoustically small spaces that have not been treated to almost anechonic chamber levels.

You may nevertheless enjoy the results of listening to such systems though.

Finally, it is you who shifted the argument from radiation patterns to driver distortion, crossovers used and so on.

If you do not wish to shift the argument into different areas, please shave with Occams razor and do not introduce a multitude of additional strictures and conditions to a single argument, it at the least mudlles the clarity of the argument.

L8er T

 

[zanash]

What useful diy information ?

''No No No ......

you post your ideas here and then we comment or offer advice, not the other way round.''

is what you said to me.

you use silver instead of copper.It has a higher ratio of direct current to voltage, and you use the appropriate insulation with the correct permittivity.

However, there is no such thing as a neutral cable. A cable doesn't sound like anything. It only completes the circuit between say an amp and preamp, so it obviously depends on the two circuits.

So where's the answer to my question ?
I appreciate the responce but I have a full understanding of insulators and their sound effect.

I was talking about metallogy not some indifinable neutrality ...just better sound balance....

 

[lordsummit]

Shinobiwan. Can you please refrain from personal attacks against other members. It is against the AUP.
Thanks

 

[ShinOBIWAN]

L8er T

I'm bored at this point.

You win Thorsten.

BTW: I had an interesting PM from someone that's heard your system. Seems your full of it when it comes to accuracy and uncoloured which was my whole point from the start.

 

[sastusbulbas]

Thorsten's own particular speakers which imbodies his whole philosophy use a full range driver with a horn loaded HF and monopole subs.

My own opinion is that fullrange systems are entirely unsuitable for reproducing bass, mid and partial treble frequencies - if your looking for low distortion both linear and non linear like a truely accurate system would provide.

Do you mean fullrange Single drivers ? as opposed to fullrange systems ?

My own pair of loudspeakers are what I would call a three way fullrange system, quite suitable for reproducing bass, their frequency responce is 20hz - 20khz +/-2db at 2m.

 

[ShinOBIWAN]

Do you mean fullrange Single drivers ?

Yes.

 

[Johnny]

Correct. But that completely fails to answer my question.



Again correct, you haven't. That's totally irrelevant to resistivity. You would only need to specify those things in order to know resistance

So I ask you again.



Is this different to resistivity and if so, how?

Ask Thorsten. He'll tell you.

 

[Johnny]

I'm bored at this point.

You win Thorsten.

BTW: I had an interesting PM from someone that's heard your system. Seems your full of it when it comes to accuracy and uncoloured which was my whole point from the start.

Who heard his system ?

and which system ?