The cleaner the system, the more likely you'll be to hear it. However, you are certainly getting into diminishing returns, there.
As you might imagine, it tends to benefit the systems with more extreme levels of horsepower and with multi-driver systems. But there is no reason the right levels of compensation might benefit your situation. I'm sure you've got the parts around to get confused (as to what exactly it is you will be hearing-or not) by!
As you might imagine, it tends to benefit the systems with more extreme levels of horsepower and with multi-driver systems. But there is no reason the right levels of compensation might benefit your situation. I'm sure you've got the parts around to get confused (as to what exactly it is you will be hearing-or not) by!
eh, how I am supposed to make quotes here? Is this forum engine supposed to support Firefox? 
"Talking already on the level of system comparison, there is a relatively new approach that is promising though not fully mature.
PC based XO's are the ultimate to merge advanced XO functionality and feed forward compensation of imperfections of circuits, channel matching , drivers and room behaviour carried out by convolution – kind of almost unlimited EQ ." -quoted from migeQ
I definitively would not recommend that, unless you happen to be software engineer by occupation. That is: not yet. That's still really die- hard.
AS promising as it might look,
1) There's really nothing you could not do with a good active XO you can do with a sound card. At least, in XO -and EQ, if you feel like- duties. Most of XO's have BTW some kind of a digi input today, I am feeding SPDIF to XO. And, I remember some better BSS's having an ADAT option already.
2) With a sound card and PC you get a _whole_ lot of a new problems, like how to best deal with the deeds of mr. Gates? Simplifying it a little:You got to keep your DSP below the OS, and, that's not a laymans hobby. To me looks like the only feasible way to deal with it is to run something like ProTools, and that's really a little bit an overkill.
Separate XO unit does just what it is supposed to do, and it does it pretty well. It does not makes any unexpected sample rate conversions, or dynamic compression, or latency problems,as might happen with PC, unless you dont know really precisely what are you doing. And, if you have read my mail this far, you probabaly dont, as I neither.
Guess how many phase linear automatically latency compensated EQ plug ins supporting multichannel output there is in the market today?
It's tricky enough to get decent 2-channel output from PC.
Are you happy with Foobar/ASIO/EAC? Try to keep it that way, and do not look what's behind the curtain...
And, cheap sound cards are just pure c***. With a price of a really good sound card, and associated PC you can buy the best available XO and still have 1000$/500€ to modifications, if you feel like. Or to something else.
-quoting again:
"PC based XO's are the ultimate to merge advanced XO functionality and feed forward compensation of imperfections of circuits, channel matching , drivers and room behaviour carried out by convolution – kind of almost unlimited EQ "
PC or an active digi XO is _not_ a magician's hat which is supposed to compensate imperfections elsewhere in the system.
It does not make a bad system good, usually it does it worst.
It might greatly help with some problems otherwise pretty hard to tackle, but basically, it is just an crossover, that's all.
"Talking already on the level of system comparison, there is a relatively new approach that is promising though not fully mature.
PC based XO's are the ultimate to merge advanced XO functionality and feed forward compensation of imperfections of circuits, channel matching , drivers and room behaviour carried out by convolution – kind of almost unlimited EQ ." -quoted from migeQ
I definitively would not recommend that, unless you happen to be software engineer by occupation. That is: not yet. That's still really die- hard.
AS promising as it might look,
1) There's really nothing you could not do with a good active XO you can do with a sound card. At least, in XO -and EQ, if you feel like- duties. Most of XO's have BTW some kind of a digi input today, I am feeding SPDIF to XO. And, I remember some better BSS's having an ADAT option already.
2) With a sound card and PC you get a _whole_ lot of a new problems, like how to best deal with the deeds of mr. Gates? Simplifying it a little:You got to keep your DSP below the OS, and, that's not a laymans hobby. To me looks like the only feasible way to deal with it is to run something like ProTools, and that's really a little bit an overkill.
Separate XO unit does just what it is supposed to do, and it does it pretty well. It does not makes any unexpected sample rate conversions, or dynamic compression, or latency problems,as might happen with PC, unless you dont know really precisely what are you doing. And, if you have read my mail this far, you probabaly dont
, as I neither.Guess how many phase linear automatically latency compensated EQ plug ins supporting multichannel output there is in the market today?
It's tricky enough to get decent 2-channel output from PC.
Are you happy with Foobar/ASIO/EAC? Try to keep it that way, and do not look what's behind the curtain...
And, cheap sound cards are just pure c***. With a price of a really good sound card, and associated PC you can buy the best available XO and still have 1000$/500€ to modifications, if you feel like. Or to something else.
-quoting again:
"PC based XO's are the ultimate to merge advanced XO functionality and feed forward compensation of imperfections of circuits, channel matching , drivers and room behaviour carried out by convolution – kind of almost unlimited EQ "
PC or an active digi XO is _not_ a magician's hat which is supposed to compensate imperfections elsewhere in the system.
It does not make a bad system good, usually it does it worst.
It might greatly help with some problems otherwise pretty hard to tackle, but basically, it is just an crossover, that's all.
AJinFLA said:
I've used a few that might qualify. Those lovely all balanced Studer boards at Radio France.
Or the Crest "Gamblin Series" live mixing board of the mid to late 90s. Very sweet. Actually promoted as having high quality caps and low phase error. You could hear it right away.
But mostly they are pretty awful, for sure. It's easy to tell the difference from mixer to mixer.
" Better-Sounding Active Crossovers " !!!
Hi
Miksin, convolution technique IS close to " a magician's hat " – and nothing you can do with any ordinary active XO .
To all your other reservations I agree almost completely.
Use "QUOTE" and "/QUOTE" in squared brackets at the beginning and at the end of your quote manually if its not working otherwise.
Greetings
Michael
Hi
miksin said:
Miksin, convolution technique IS close to " a magician's hat " – and nothing you can do with any ordinary active XO .
To all your other reservations I agree almost completely.
miksin said:eh, how I am supposed to make quotes here? Is this forum engine supposed to support Firefox?
Use "QUOTE" and "/QUOTE" in squared brackets at the beginning and at the end of your quote manually if its not working otherwise.
Greetings
Michael
Re: " Better-Sounding Active Crossovers " !!!
Now it works. _IT DID IT NOT BEFORE_!
I really dont understand. Perhaps it's just plain stupidity.
On topic, perhaps I've just been unlucky, but my experieces so far have only confirmed that any system -or room- needeing excessive electronic corrections is just a bad system, or a bad room, and the end result with corrections is never as good as could be achievable with a good system or in a good room.
We'll see whether that will change with time, perhaps I have just been badly informed.
Good that you pointed out rewards in analogue parts of the chain with DSP. Perhaps i've got to reveal my secret perversion:
I liked to try something like this:
DL103 to LL9206, to 6F12P, to AD, DA transformer coupled to 6N1P-EV"Aikido" to 6C45Pi.
I did not found any good phono eq software -that was a year or so ago. It's the only reason I'm not using a Rosetta 800 as a XO now.
Getting rid of phono amp...
One man's SOTA is always anothers unbearable compromise, ain't it?
mige0 said:
Now it works. _IT DID IT NOT BEFORE_!
I really dont understand. Perhaps it's just plain stupidity.
On topic, perhaps I've just been unlucky, but my experieces so far have only confirmed that any system -or room- needeing excessive electronic corrections is just a bad system, or a bad room, and the end result with corrections is never as good as could be achievable with a good system or in a good room.
We'll see whether that will change with time, perhaps I have just been badly informed.
Good that you pointed out rewards in analogue parts of the chain with DSP. Perhaps i've got to reveal my secret perversion:
I liked to try something like this:
DL103 to LL9206, to 6F12P, to AD, DA transformer coupled to 6N1P-EV"Aikido" to 6C45Pi.
I did not found any good phono eq software -that was a year or so ago. It's the only reason I'm not using a Rosetta 800 as a XO now.
Getting rid of phono amp...
One man's SOTA is always anothers unbearable compromise, ain't it?
Re: Re: " Better-Sounding Active Crossovers " !!!
Hi
Generally spoken if you " need it " – I agree .
BUT consider the issue of matching for a while.
I once had a system I throughout ( signal path AND power supply AND each supply bias ) have matched to << 1% left / right with more or less each resistor / capacitor / transistor from the phono pre to the power amp and its stability and extension of soundstage in depth and width and also the imaging was extraordinary .
This is what matching can do for us.
OR what convolution can do for you INCLUDING to even up the series variations of speakers - and of course you can do that for bass / mid / tweeter separately .
This is at least one really big benefit that makes XO's with convolution processing in the PC so appealing for me .
An other one is the ability of advanced EQing - be it for drivers or room .
Greetings
Michael
Hi
miksin said:
Generally spoken if you " need it " – I agree .
BUT consider the issue of matching for a while.
I once had a system I throughout ( signal path AND power supply AND each supply bias ) have matched to << 1% left / right with more or less each resistor / capacitor / transistor from the phono pre to the power amp and its stability and extension of soundstage in depth and width and also the imaging was extraordinary .
This is what matching can do for us.
OR what convolution can do for you INCLUDING to even up the series variations of speakers - and of course you can do that for bass / mid / tweeter separately .
This is at least one really big benefit that makes XO's with convolution processing in the PC so appealing for me .
An other one is the ability of advanced EQing - be it for drivers or room .
Greetings
Michael
I recently started to study 3rd and 4th order active filters built with just one opamp. I would appreciate any advise on these topologies. I thought this topic fit Lynn's search for better sounding active crossovers, since expensive discrete JFET and tube amp designs have been discussed. A one amp filter could sound better than a two amp filter.
Attachments
What about ditching the riaa filter section of the "Pacific Riaa Pre" and replacing with a pllxo setup? Should sound pretty good. Basically a jfet buffered pllxo.
"Pacific Riaa Pre" http://www.diyaudio.com/forums/showthread.php?threadid=28223
pllxo
http://www.t-linespeakers.org/tech/filters/passiveHLxo.html
regards,
psz.
"Pacific Riaa Pre" http://www.diyaudio.com/forums/showthread.php?threadid=28223
pllxo
http://www.t-linespeakers.org/tech/filters/passiveHLxo.html
regards,
psz.
Contrary to popular belief, you CAN easily obtain perfect dynamic phase compensation (crossover shift issues) within the use of analog line-level and passive crossovers.
Much of the drive signal simply has to be dumped into the compensation network.
For example, dump a 5watt or more class A buffer into a passive compensation and crossover network. Out the other end comes about maybe 5% of the original signal.
But it IS done fully passively. With one single stage. This seems totally counter-intuitive, but there is no rule against it, whatsoever.
Much of the drive signal simply has to be dumped into the compensation network.
For example, dump a 5watt or more class A buffer into a passive compensation and crossover network. Out the other end comes about maybe 5% of the original signal.
But it IS done fully passively. With one single stage. This seems totally counter-intuitive, but there is no rule against it, whatsoever.
What are you using for a preamp, or is there a link to the schematic if it is your own design?
It helps to know what the crossover will have as a driving source.
I messed with this a few years ago when upgrading some JBL crossovers used in the 4343/4345.
Essentially I found biased Jfet discrete opamps provided the best sonic proeprties and lowest noise. Simple is best and even a white follower can work nicely.
Erno Borbely is the best source of knowledge on the subject if hi resolution discreteopamps
I trimmed out the DC offset to avoid coupling capacitors- see passlabs diy projects
The better digital crossovers hold a lot promise.
Ian
Essentially I found biased Jfet discrete opamps provided the best sonic proeprties and lowest noise. Simple is best and even a white follower can work nicely.
Erno Borbely is the best source of knowledge on the subject if hi resolution discreteopamps
I trimmed out the DC offset to avoid coupling capacitors- see passlabs diy projects
The better digital crossovers hold a lot promise.
Ian
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.