I suggest that you go the the FB2K forum and do a search on ASIO. The author of FB2K has stated over and over that there is no advantage to ASIO in Vista or W7 even though he has been forced to write ASIO and WASAPI plug-ins. The only time DS gets involved is when there are two audio streams and this is circumvented by setting your DAC to exclusive output or when your DAC cannot support a particular output. I can guarantee that if your DAC can accept it, DS will output it. I have seen this with my m-Audio DAC.
The whole ASIO frenzy was generated by the very poor audio engine in Win XP. This was fixed in Vista and W7 (and presumably W8). The fact is, you CANNOT hear any difference between ASIO, DS and WASAPI. Of course, with you bias toward ASIO, placebo effect will.
RE: Asus Xonar ASIO. Again search FB2K for this. Asus has notoriously buggy ASIO drivers. The problem is avoided, of course, by using DS.
Do a little research.
Bob
Unfortunately, it is not as clear cut as you say, Bob.
"I can guarantee that if your DAC can accept it, DS will output it."
What you are saying is true, but it is also true that DS will resample to another rate if your DAC cannot accept the rate. When using software such as ARTA and Holm, you have to be very careful in setting the rates, otherwise DS will make everything work but the results will be off. I much prefer getting an error rather than having everything work while getting erroneous results.
I had done plenty of research on this some time ago. But I took your advice and did some more:
Directsound vs WASAPI vs ASIO - Hydrogenaudio Forums
Audio Output Modes - JRiverWiki
The Well-Tempered Computer
One thing is clear, ASIO, WASAPI and DS are all different. Whether we can hear the difference or not is another matter.
For music, in my non-blind listening tests, WASAPI has shown far superior quality. I like to think I'm not biased because it's harder to use WASAPI. I want to use whatever sounds good. Maybe my Windows settings are not optimally set for DS. There may be several reasons why DS doesn't sound good on my computer.
The whole point is DS will make everything work even if you have some settings that are incorrect. The output may be resampled, downsampled or upsampled or a combination. Whereas with ASIO and WASAPI you are guaranteed the correct output straight from the player to the hardware.
"I can guarantee that if your DAC can accept it, DS will output it."
What you are saying is true, but it is also true that DS will resample to another rate if your DAC cannot accept the rate. When using software such as ARTA and Holm, you have to be very careful in setting the rates, otherwise DS will make everything work but the results will be off. I much prefer getting an error rather than having everything work while getting erroneous results.
I had done plenty of research on this some time ago. But I took your advice and did some more:
Directsound vs WASAPI vs ASIO - Hydrogenaudio Forums
Audio Output Modes - JRiverWiki
The Well-Tempered Computer
One thing is clear, ASIO, WASAPI and DS are all different. Whether we can hear the difference or not is another matter.
For music, in my non-blind listening tests, WASAPI has shown far superior quality. I like to think I'm not biased because it's harder to use WASAPI. I want to use whatever sounds good. Maybe my Windows settings are not optimally set for DS. There may be several reasons why DS doesn't sound good on my computer.
The whole point is DS will make everything work even if you have some settings that are incorrect. The output may be resampled, downsampled or upsampled or a combination. Whereas with ASIO and WASAPI you are guaranteed the correct output straight from the player to the hardware.
Well, yea, you have to have a DAC that can support input bit rate. If not DS will resample to something your DAC can support. That's its purpose and are we surprised? If you are running some other software, you need to make sure that the rates match. Are we surprised? Competent software will supply appropriate drivers and all will be well. DS is designed as an audio output engine. Use it for something else and you need dot your I's.
If the input bit rate/depth of the DAC matches the bit rate/depth of the input and exclusive mode is used, DS will output a BIT-PERFECT data stream. WASAPI will aslo. ASIO may depending upon the DAC vendor's drivers. Now, how can one bit-perfect data stream sound better than another?
.
I think that you are missing a point here. In exclusive mode and input matches the DAC, DS does nothing. Period. If you are not in exclusive mode, then DS will find a match between the various inputs and possible DAC capabilities. IIRC that will be 16/48. It incumbent upon you to set the parameters correctly.
Bob
... back to the FAST subject!
If you use the bass drivers up to 120 - 150Hz, they DO need to be phased properly with the mids/FR and if you're using the system for wide music selection (maybe R&R too) the bigger drivers come into their own (IMO, naturally) and there's a BIG difference between the combination of "Room + OB Bass" drivers and "Room + B2 with Link Tranform" (closed box bass with freq response peak added)
There has been inumerable discussions about these 2 different types of bass generation and some of it's really useful too, but FAST is NOT about using subwoofers but is all about good quality accurate clean bass - unfortunately, it's often expensive but with some homework, can get excellent results for reasonable $s, as Dave has mentioned.
I would suggest you avoid the cheap ClassD switcher amps for both bass and mids, altho the Ncore ones from Hypex do a pretty good job for bass, particularly with their traditional type of power supply. Unless you know what you're doing with these classD amps, it's not at all easy to optomise these amps any better than they come "outa-da-box" and the sound is difficult to match up with the higher efficiency FR drivers as they din't seem to dampen the often quite audible cone resonances that are a part of thes FR drivers - maybe okay for the more highly damped cones but no experience to offer there.
I'm still using passive 12dB Xovers and the chokes are quite daunting things - commercially, rather expensive, and at the low cost of good amp kits, it's actually cheaper to go down the active Xover path, as Vix has indicated, and "qusp" is developing/investigating the computer source with Xover into 2 dac chips (HP and LP dac chips) and is an indication of 'things to come'
Sorry Bud, I haven't kept up with the new version of the EnABL but I did get a Fostex 167 from Dave (Planet 10) and the process/driver does work extremely well and Bob has a lot more info on his site.
I haven't heard much about the "Rethym" speakers from India lately, but these have achieved remarkable results with the often difficult Lowther drivers and is definitely worth a study as some of the ideas outlined there can be directly applied to our own speakers with immediate results.
IMO, a well setup/designed FAST system can match the better dynamics of some horn systems for overall sound balance, particularly in a smaller room, but both require some work with room acoustical control methods, so our FAST speaker systems aren't so simple any more but include amp selection, driver selection/matching plus room acoustics to complicate the design.
Good sound reproduction isn't at all easy, but what a challenge, eh!
If you use the bass drivers up to 120 - 150Hz, they DO need to be phased properly with the mids/FR and if you're using the system for wide music selection (maybe R&R too) the bigger drivers come into their own (IMO, naturally) and there's a BIG difference between the combination of "Room + OB Bass" drivers and "Room + B2 with Link Tranform" (closed box bass with freq response peak added)
There has been inumerable discussions about these 2 different types of bass generation and some of it's really useful too, but FAST is NOT about using subwoofers but is all about good quality accurate clean bass - unfortunately, it's often expensive but with some homework, can get excellent results for reasonable $s, as Dave has mentioned.
I would suggest you avoid the cheap ClassD switcher amps for both bass and mids, altho the Ncore ones from Hypex do a pretty good job for bass, particularly with their traditional type of power supply. Unless you know what you're doing with these classD amps, it's not at all easy to optomise these amps any better than they come "outa-da-box" and the sound is difficult to match up with the higher efficiency FR drivers as they din't seem to dampen the often quite audible cone resonances that are a part of thes FR drivers - maybe okay for the more highly damped cones but no experience to offer there.
I'm still using passive 12dB Xovers and the chokes are quite daunting things - commercially, rather expensive, and at the low cost of good amp kits, it's actually cheaper to go down the active Xover path, as Vix has indicated, and "qusp" is developing/investigating the computer source with Xover into 2 dac chips (HP and LP dac chips) and is an indication of 'things to come'
Sorry Bud, I haven't kept up with the new version of the EnABL but I did get a Fostex 167 from Dave (Planet 10) and the process/driver does work extremely well and Bob has a lot more info on his site.
I haven't heard much about the "Rethym" speakers from India lately, but these have achieved remarkable results with the often difficult Lowther drivers and is definitely worth a study as some of the ideas outlined there can be directly applied to our own speakers with immediate results.
IMO, a well setup/designed FAST system can match the better dynamics of some horn systems for overall sound balance, particularly in a smaller room, but both require some work with room acoustical control methods, so our FAST speaker systems aren't so simple any more but include amp selection, driver selection/matching plus room acoustics to complicate the design.
Good sound reproduction isn't at all easy, but what a challenge, eh!
May I add that a true sub driver is useless much above 100Hz. They are designed to go REALLY low and that makes mid-bass impossible. A woofer that is going to work in a FAST has to be able to to mid-bass and then some. Drivers that I am considering must have a usable bandwidth above 1kHz. Suppose you want to do a 1st order XO at 300Hz. The woofer is only going to be 12dB down at 1.2kHz. Regardless of XO order, you should look for a driver that is usable two octaves above (or below) the XO point. That is a tall order and not really practical, but you get the drift.
I can second the prohibition of cheap D-amps, particularly for bass. Get a decent class-D amp that has input caps big enough to go as low as you need and a power supply that is big enough to handle the power you need. There is nothing inherently bad about SMPS's, but they have to be big enough. On the high end, a $20 D-amp board is going to have parts that are consistent with the price. For a ready to play D-amp, expect a MSLP of at least $300.
Bob
I can second the prohibition of cheap D-amps, particularly for bass. Get a decent class-D amp that has input caps big enough to go as low as you need and a power supply that is big enough to handle the power you need. There is nothing inherently bad about SMPS's, but they have to be big enough. On the high end, a $20 D-amp board is going to have parts that are consistent with the price. For a ready to play D-amp, expect a MSLP of at least $300.
Bob
+1
Subwoofer as most of us imagine it (up to 100 Hz) , isn't really suitable for FAST. On the other hand, there are quite some good 8" midbass drivers, which, when put in a proper ML-TL box can go quite low (40 hz or even lower) and on the other hand can go high enough, 2 K or more (enough for a 1st order lo-pass at 300-400 Hz) so it makes them perfectly acceptable for FAST duties, as shown in my mock-up experiment. 3-4" fullrangers can be safely hi-passed 1st order 300-400 Hz and some better ones can go up to 20Khz.
The added benefit is that with the slim and tall ML-TL boxes baffle step occurs right there (300-400 Hz) and it is nicely compensated for by crossing at that point.
yes Bob, ... and avoiding the voice coil inductance peak about 1kHz in many of the 12 and 15" drivers and the need to dampen the driver's dynamic range (power compression) - who ever thought bass driver/box was a simple thing!
An interesting driver, Harry Olson's 8" RCA LC-1A - link via ZenMod's post #2326 on the 'F6 Amplifier' thread - and this one started life over 50 years ago. Wouldn't mind trying this one in the Fast system in a Polyconcrete teardrop, tapered resistance chamber .....
Years ago, in the R&R heyday, I used to think the venerable Altec system (15" + m/cell horn) sounded just great - it's a pity the overall quality of recording hasn't progressed since then either, and super-line compression isn't the worst of it by any means!
(sorry for the 'Off Topic' subject - couldn't resist it!)
An interesting driver, Harry Olson's 8" RCA LC-1A - link via ZenMod's post #2326 on the 'F6 Amplifier' thread - and this one started life over 50 years ago. Wouldn't mind trying this one in the Fast system in a Polyconcrete teardrop, tapered resistance chamber .....
Years ago, in the R&R heyday, I used to think the venerable Altec system (15" + m/cell horn) sounded just great - it's a pity the overall quality of recording hasn't progressed since then either, and super-line compression isn't the worst of it by any means!
(sorry for the 'Off Topic' subject - couldn't resist it!)
Well, we don't have to use a 1st order crossover. It would be nice, but not necessary.
Vix, im interested to know which 3" fullranges you have used with a 1st order HP at 300hz. I have TB W3-1285SG and MA A6, and NEITHER worked well at all high passed 1st order. I found 2nd order HP and 1st order LP worked far better at 300hz or thereabouts. Even with 47uf single cap giving a higher F3 didnt work well. Unless the driver Z at Fs is flattened i dont have much confidence in what you claim.
http://www.diyaudio.com/forums/full-range/144099-thread-tysen-variations-fast.html#post1827952
We went active with Tysen, but a similar mid-TL worked well with a 1st order series XO at a bit above 300 Hz.
dave
Dave, that is cool. I wonder though if the actual acoustic slope would have been closer to 150hz on that one. As the filter engages the z peak at 200hz would resist the filter and then as it comes off the peak toward 150hz the filter would fully engage. I'm not sure that's how to technically describe it, but it's how my thought process works.
Edit - nevermind. You said series. I'm not sure how things would be effected.
Edit - nevermind. You said series. I'm not sure how things would be effected.
Last edited:
Of course, you don't have to do 1st order and you don't have to cross at 300Hz. Where you cross is dependant on all of the other parameters of you speakers/room. You do need substantial overlap, even 4th order and you want the woofer cone breakup as far down as possible. But then again, you want to unload the FR as much as possible, and you need the phase rotations as low as possible. These are conflicting requirements. You simply play with it until is sounds right. Hence the beauty of DSP.
Bob
Bob
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.