XXHighEnd
I can't contain myself... this XXHighEnd sound quality is soooo good. Wow you guys have to try it ASAP. I thought Foobar with ASIO4ALL was as good as it gets. I like being wrong. Now the reclocker is making a difference again in my DAC. I can keep tweaking.
I am using Windows 2008 Server, I haven't tried it with VISTA but I expect it to be good there too since that was what it was developed for.
Linux/Mac guys please give it a try and report back. Now that I am not stuck on Foobar anymore I wonder what those OS/players sound like in comparison to XXHighEnd.
Perhaps I/we should really start a new thread to talk about XXHighEnd, but you highend DAC guys are the ones who will really hear the biggest difference, this is a good place to bring it to your attention.
And no I don't work for XXHighEnd, if I could I would
Cheers,
Brent
I can't contain myself... this XXHighEnd sound quality is soooo good. Wow you guys have to try it ASAP. I thought Foobar with ASIO4ALL was as good as it gets. I like being wrong. Now the reclocker is making a difference again in my DAC. I can keep tweaking.
I am using Windows 2008 Server, I haven't tried it with VISTA but I expect it to be good there too since that was what it was developed for.
Linux/Mac guys please give it a try and report back. Now that I am not stuck on Foobar anymore I wonder what those OS/players sound like in comparison to XXHighEnd.
Perhaps I/we should really start a new thread to talk about XXHighEnd, but you highend DAC guys are the ones who will really hear the biggest difference, this is a good place to bring it to your attention.
And no I don't work for XXHighEnd, if I could I would
Cheers,
Brent
Here we are again!!! The same discussions an a "Wonder"
player continues.
I think numerous people did proove that bit perfect output
from a PC is relatively easy to establish and doesn't really need any great SW-player.
If a SW player does generate any difference in sound, while delivering bit perfect data, we need to talk about the limitiations of the digital chain and the DACs capability to process the data properly.
And - yes - on a poor digital chain the player does make a huge difference, I've been through that ghost-chase by myself.
The PC is a major source for inducing non linearities to the data
stream. If it's running as a master towards your DAC you'll experience these non-linearties. If you run it as slave - you 'll have a hard time to find a difference.
If the ultimate "ec-design" DAC would still show a difference in sound, assuming bit-perfect (quantity and quality) PC output, it would be far away from being the "ultimate" NOS-DAC.
At least John told me (us) several times that the source signal wouldn't make any difference, assuming bit -perfection at the PC output.
As I mentioned earlier asynchronous data transfer and also the related clocking/reclocking is the actual problem. You need to slave your upstream source to your master-clock. Beside that you need to make sure that your physical environment is not poluted with EMI/RFI, reflections and other distortions, which might end up in a messy signal on the receiver side.
I'll fix my ecdesign reclocker next weekend. I'll let you know
about the result I am experiencing.
Cheers
player continues.
I think numerous people did proove that bit perfect output
from a PC is relatively easy to establish and doesn't really need any great SW-player.
If a SW player does generate any difference in sound, while delivering bit perfect data, we need to talk about the limitiations of the digital chain and the DACs capability to process the data properly.
And - yes - on a poor digital chain the player does make a huge difference, I've been through that ghost-chase by myself.
The PC is a major source for inducing non linearities to the data
stream. If it's running as a master towards your DAC you'll experience these non-linearties. If you run it as slave - you 'll have a hard time to find a difference.
If the ultimate "ec-design" DAC would still show a difference in sound, assuming bit-perfect (quantity and quality) PC output, it would be far away from being the "ultimate" NOS-DAC.
At least John told me (us) several times that the source signal wouldn't make any difference, assuming bit -perfection at the PC output.
As I mentioned earlier asynchronous data transfer and also the related clocking/reclocking is the actual problem. You need to slave your upstream source to your master-clock. Beside that you need to make sure that your physical environment is not poluted with EMI/RFI, reflections and other distortions, which might end up in a messy signal on the receiver side.
I'll fix my ecdesign reclocker next weekend.
I'll let you knowabout the result I am experiencing.
Cheers
Easy bit-perfect playback with windows
Hi Brent Welke,
Like noted before, it's very difficult to achieve bit-perfect playback with windows, even when using ASIO / foobar, performance depends on hardware used. When using USB it's even more problematic as the win OS always controls the USB port, regardless of drivers / tweaks / applications.
If XXHighend works, it would probably require extensive (exclusive) resources, placing a high load on computer hardware. It would probably have the same "issues" with USB, as the windows OS still controls the USB ports. Bit-perfect playback is probably only possible with very fast computer hardware and lowest possible additional CPU load.
I tried a different approach, instead of going through extremes trying to achieve bit-perfect audio through the most logical "audio route", I used (wireless) network and a dedicated network audio device (similar to squeezebox). This bypasses the windows "audio route", by just sending plain data across a computer network.
I used an Apple airport express module, it connects through a (wireless) computer network, and has a fixed 44.1/16 bit-perfect output:
http://www.stereophile.com/digitalprocessors/505apple/
I added a photograph of the Apple airport express in action, I connected TOSLINK through a 3.5mm adapter, and used a wired network connection.
It works seamlessly with iTunes for windows. Direct comparisons between a 450 MHz IBM system running XP, and a 1.8 GHz iMac running mac OSX tiger, resulted in identical sound quality. It doesn't require any drivers, OS tweaks, or special software / computer hardware.
This basically illustrates the problem, every windows, hardware, software, driver, OS combination seems to produce different results (when playing back digital audio).
Bit perfect playback is as good as it gets.
Linux/Mac guys use these Operating Systems for very good reason. Reverting to windows = downgrading.
Hi Brent Welke,
Like noted before, it's very difficult to achieve bit-perfect playback with windows, even when using ASIO / foobar, performance depends on hardware used. When using USB it's even more problematic as the win OS always controls the USB port, regardless of drivers / tweaks / applications.
If XXHighend works, it would probably require extensive (exclusive) resources, placing a high load on computer hardware. It would probably have the same "issues" with USB, as the windows OS still controls the USB ports. Bit-perfect playback is probably only possible with very fast computer hardware and lowest possible additional CPU load.
I tried a different approach, instead of going through extremes trying to achieve bit-perfect audio through the most logical "audio route", I used (wireless) network and a dedicated network audio device (similar to squeezebox). This bypasses the windows "audio route", by just sending plain data across a computer network.
I used an Apple airport express module, it connects through a (wireless) computer network, and has a fixed 44.1/16 bit-perfect output:
http://www.stereophile.com/digitalprocessors/505apple/
I added a photograph of the Apple airport express in action, I connected TOSLINK through a 3.5mm adapter, and used a wired network connection.
It works seamlessly with iTunes for windows. Direct comparisons between a 450 MHz IBM system running XP, and a 1.8 GHz iMac running mac OSX tiger, resulted in identical sound quality. It doesn't require any drivers, OS tweaks, or special software / computer hardware.
This basically illustrates the problem, every windows, hardware, software, driver, OS combination seems to produce different results (when playing back digital audio).
Bit perfect playback is as good as it gets.
Linux/Mac guys use these Operating Systems for very good reason. Reverting to windows = downgrading.
Attachments
Hi,
The contrary. ...
CPU load is virtually zero, and what remains comes from the driver (compare e.g. SPDIF with USB). Think in terms of 3 seconds for a complete album for SPDIF and less than 1 second for USB ...
Latency is better than ASIO, and above mentioned figures are with the actual useage of 48 samples latency (around 1ms) throughout playback at 2 channel 44.1/16.
96/24 requires 96 samples latency.
192/24 is coming up this week.
Latency can even be far lower, weren't it that timers become unstable.
Does this help, or do you mean something else ?
Peter
PS: And no matter with what I stress the system (like 3 parallel mass copy of data over SATAII) SPDIF as well as USB remains glitch free.
PPS: I didn't read into the matters soundcheck is referring to, and I can't be sure to what he actually responds. Anyway, what he achieves by means of Linux, I do with Windows. We know eachother (personally) and I am sure he will agree with what I said in the above. What we both do is on par, and he too is limited by instable timers (or maybe not anymore ?) (something to solve btw).
This may be a kind of interesting read : Latency of 1/10 of a sample ?
The contrary.
...CPU load is virtually zero, and what remains comes from the driver (compare e.g. SPDIF with USB). Think in terms of 3 seconds for a complete album for SPDIF and less than 1 second for USB ...
Latency is better than ASIO, and above mentioned figures are with the actual useage of 48 samples latency (around 1ms) throughout playback at 2 channel 44.1/16.
96/24 requires 96 samples latency.
192/24 is coming up this week.
Latency can even be far lower, weren't it that timers become unstable.
Does this help, or do you mean something else ?
Peter
PS: And no matter with what I stress the system (like 3 parallel mass copy of data over SATAII) SPDIF as well as USB remains glitch free.
PPS: I didn't read into the matters soundcheck is referring to, and I can't be sure to what he actually responds. Anyway, what he achieves by means of Linux, I do with Windows. We know eachother (personally) and I am sure he will agree with what I said in the above. What we both do is on par, and he too is limited by instable timers (or maybe not anymore ?) (something to solve btw).
This may be a kind of interesting read : Latency of 1/10 of a sample ?
XXHighEnd
I don't know why XXHighEnd on VISTA or Windows2008 Server sounds better then Foobar with ASIO. Many people with Higher End gear are reporting the same. Since my USB DAC and yours are similar in the USB department I am confident many will be very happy once they try it.
I was very skeptical about anything being better then Foobar but I was installing Window2008 server anyway and it cost me nothing (which is rare in HighEnd Audio, ask my wife).
With XXHighEnd the volume control now goes louder without having to turn back the dial and dynamic range is improved. Songs seem softer until the lead singer starts singing and then you notice volume level is higher then expected, but you don't bother to turn it down you just enjoy it. The difference is not huge but enough not use Foobar anymore.
For those who want to experiment here is a link to a 240 day trial of Windows 2008 Server. Download the x86 version or the current version of XXHignEnd will not work. By the way... 2008 Server uses the same GUI as VISTA but the Aero feature is off by default. 2008 is a very fast OS you will love it! http://www.microsoft.com/downloads/...0E-935C-415A-A79C-538E933D5424&displaylang=en
The best way IMHO to install is to use two internal hard drives and change the boot order in the computer BIOS to the second drive before installing. Even better is to also disconnect your first drive before installing. This way you don't make a mess dual booting your computer.
By the way... I still use XXHighEnd in demo mode so don't let the fact you are in demo mode have you think you can't get the good sound or that you can't make use of it on a daily basis.
Cheers,
Brent
I don't know why XXHighEnd on VISTA or Windows2008 Server sounds better then Foobar with ASIO. Many people with Higher End gear are reporting the same. Since my USB DAC and yours are similar in the USB department I am confident many will be very happy once they try it.
I was very skeptical about anything being better then Foobar but I was installing Window2008 server anyway and it cost me nothing (which is rare in HighEnd Audio, ask my wife).
With XXHighEnd the volume control now goes louder without having to turn back the dial and dynamic range is improved. Songs seem softer until the lead singer starts singing and then you notice volume level is higher then expected, but you don't bother to turn it down you just enjoy it. The difference is not huge but enough not use Foobar anymore.
For those who want to experiment here is a link to a 240 day trial of Windows 2008 Server. Download the x86 version or the current version of XXHignEnd will not work. By the way... 2008 Server uses the same GUI as VISTA but the Aero feature is off by default. 2008 is a very fast OS you will love it! http://www.microsoft.com/downloads/...0E-935C-415A-A79C-538E933D5424&displaylang=en
The best way IMHO to install is to use two internal hard drives and change the boot order in the computer BIOS to the second drive before installing. Even better is to also disconnect your first drive before installing. This way you don't make a mess dual booting your computer.
By the way... I still use XXHighEnd in demo mode so don't let the fact you are in demo mode have you think you can't get the good sound or that you can't make use of it on a daily basis.
Cheers,
Brent
Hi.
I doubt that you loose or change bits while sending it over any digital interface. If so they would be corrected. Otherwise it would mean that none of standard file transfers would work reliable.
Ah Yes - of course - we're talking about a realtime data stream
here. That could be something different - perhaps the USB audio controller avoids resending packages. In this case
I wouldn't bet on a bit perfect data stream either.
On the other hand you can read in 100reds of posts that people looped back the digital data stream and received exacly the same data. We'd be back to a pure jitter discussion over here.
@ ecdesigns: Are you telling us that non of USB audio output is usable, because there is no way to get bitperfect data out of it?
BTW: USB is always controlled by drivers even under Linux.
As I wrote in my Linux thread. There are processes taking care on dynamic bandwith alloction, they are dynamically changing number of packets per URB (USB request block), which all cause sound degradation by introducing non-linearities to the data stream. There's is even a rather poor HW based controller taking care on audio signals over USB. ( Wavelength Audio reported its impact on jitter at Audio-Asylum some time ago.)
With a bit of tweaking you can achieve great sound improvements.
To repeat myself: This is IMO valid for asynchronous chains only.
Is there anybody around who prooved that the sample you play with your application is exactly the one ending up on my I2S bus right in front of the DAC?? How could I measure this?
Back to the ecdesign product:
I fixed the 5V- powersupply for the reclocker and the Linux setup over the weekend.
Now the reclocker has to proove its potential.
@XXHE: What I am saying is that we discuss this weired issue since a while. My approch was and still is to squezze most out of known technically very limited i(a)synchronous solution ( which is also vaild for firewire solutions). And yes, exactly what you're saying I am saying. You don't have reliable clocks inside a PC.
Running a PC as a clock master for your datastream is just nonsense from an audio perspective. That's why I am counting
on the ec-designs reclocker - even though products such as a Benchmark DAC1, which should be rather immune to jitter according to the marketing message, have shown its limitations.
Perhaps we should open up a thread "building up the ultimate pc based digital chain"
Cheers
I doubt that you loose or change bits while sending it over any digital interface. If so they would be corrected. Otherwise it would mean that none of standard file transfers would work reliable.
Ah Yes - of course - we're talking about a realtime data stream
here. That could be something different - perhaps the USB audio controller avoids resending packages. In this case
I wouldn't bet on a bit perfect data stream either.
On the other hand you can read in 100reds of posts that people looped back the digital data stream and received exacly the same data. We'd be back to a pure jitter discussion over here.
@ ecdesigns: Are you telling us that non of USB audio output is usable, because there is no way to get bitperfect data out of it?
BTW: USB is always controlled by drivers even under Linux.
As I wrote in my Linux thread. There are processes taking care on dynamic bandwith alloction, they are dynamically changing number of packets per URB (USB request block), which all cause sound degradation by introducing non-linearities to the data stream. There's is even a rather poor HW based controller taking care on audio signals over USB. ( Wavelength Audio reported its impact on jitter at Audio-Asylum some time ago.)
With a bit of tweaking you can achieve great sound improvements.
To repeat myself: This is IMO valid for asynchronous chains only.
Is there anybody around who prooved that the sample you play with your application is exactly the one ending up on my I2S bus right in front of the DAC?? How could I measure this?
Back to the ecdesign product:
I fixed the 5V- powersupply for the reclocker and the Linux setup over the weekend.
Now the reclocker has to proove its potential.
@XXHE: What I am saying is that we discuss this weired issue since a while. My approch was and still is to squezze most out of known technically very limited i(a)synchronous solution ( which is also vaild for firewire solutions). And yes, exactly what you're saying I am saying. You don't have reliable clocks inside a PC.
Running a PC as a clock master for your datastream is just nonsense from an audio perspective. That's why I am counting
on the ec-designs reclocker - even though products such as a Benchmark DAC1, which should be rather immune to jitter according to the marketing message, have shown its limitations.
Perhaps we should open up a thread "building up the ultimate pc based digital chain"
Cheers
My experience... XXHighEnd sounds better with the DAC's reclocker on. Without XXHighEnd I was having a hard time hearing the difference between reclocker on/off and was leaning toward it sounding better with the reclocker off. But with XXHighEnd the DAC sounds better no matter if the reclocker on or off, but even better with reclocker on.
I don't know why, just reporting my findings.
Cheers,
Brent
I don't know why, just reporting my findings.
Cheers,
Brent
Win Server 2008 will be very slow when the pathetic aero will be switched on. My friend uses vistah on his work and switched off aero on all pc's to get them workable in normal conditions. Again music is suffering from complicated gear, and win os is very complicated, it is build in all kind of layers like a box of bricks beginning with dos and several other pieces to fragments of xp/vista. Not to mention the 16, 32, 64bit structure.
Sorry for maibe bashing-like comments.
Sorry for maibe bashing-like comments.
Ok but if you are open minded and searching for the best sound it would only take an evening to try Windows2008 Server and XXHighEnd. Ask me as many questions as you like if you need a hand or advice.
For me to get into Linux, find a player and do all the reading/research to get it going, recompile and all the other Linux geeky stuff one needs to do to use Linux would probably take me months.
I would switch to Linux/Apple if one or more Linux guys would take the Windows2008/XXHighEnd vs Linux/Player sound quality test and declare Linux/Player to be the winner for USB interface.
If you take the challenge then please document your test setup as much as possible please.
Cheers,
Brent
For me to get into Linux, find a player and do all the reading/research to get it going, recompile and all the other Linux geeky stuff one needs to do to use Linux would probably take me months.
I would switch to Linux/Apple if one or more Linux guys would take the Windows2008/XXHighEnd vs Linux/Player sound quality test and declare Linux/Player to be the winner for USB interface.
If you take the challenge then please document your test setup as much as possible please.
Cheers,
Brent
It's the other way round: we are forced to get a sort of "along" with the awkward functioning win stuff all those years, your eyes and pc manors will open up again if you learn to work on linux, and discover to work with a pc how it should be. I want never a ms os again.
To get the best sound of the pc you will have to tweek some, yes ofcourse. But the tweeks as all linux settings i did (marble mouse, video card) are for a good purpose with good results in the end, with win stuff you allways get into another problem. I had that at least.
LRM2 schematics
Hi anbello,
I added the LRM2 schematic. The LRM2 uses a 48 MHz low jitter oscillator, just like the LRM1. However, a clock doubler (U2 + U3) is added to create a 96MHz low jitter clock. Due to the doubled reclock frequency, two 8-bit shift registers are required to enable locking to the 2.8224 MHz BCK input timing signal.
The series resistors in clock and I2S signals are required to reduce the effects of ground bounce, and result in lower jitter. Both R7 and C1 must be tuned to the PCB lay-out and gate propagation delays.
The LRM2 is a prototype currently being tested, and not a final design.
Clean power supplies are also important for best clock circuit performance. Battery power supplies work pretty well, but aren't very practical. I tested super regulators (6nV), but they simply couldn't beat the reference battery power supply. I used a switch to toggle between battery and mains power supply, so I could immediately hear the difference in sound quality during playback. I used 3* 1.5V alkaline batteries, and 4 * 1.2 NiMH rechargeable batteries (AA-size) for reference,
The mains power supply resulted in a slightly darker smeared sound. Battery power supplies result in a slightly "thinner", cleaner sound. This effect is probably caused by changed master clock jitter spectrum.
After lots of experimenting, I was just about to give-up on mains power supplies (for the reclocker) and was was planning to connect a rechargeable battery pack to the DI8M, but then I got an idea, a passive filter with back-up capacitors.
I used 10 ... 20mH ring core inductor and a Panasonic 0.1F (100,000uF) / 5.5V back-up capacitor. To my surprise this filter actually "outperformed" the battery reference supply. These filters are placed after each shunt regulator.
Hi anbello,
I added the LRM2 schematic. The LRM2 uses a 48 MHz low jitter oscillator, just like the LRM1. However, a clock doubler (U2 + U3) is added to create a 96MHz low jitter clock. Due to the doubled reclock frequency, two 8-bit shift registers are required to enable locking to the 2.8224 MHz BCK input timing signal.
The series resistors in clock and I2S signals are required to reduce the effects of ground bounce, and result in lower jitter. Both R7 and C1 must be tuned to the PCB lay-out and gate propagation delays.
The LRM2 is a prototype currently being tested, and not a final design.
Clean power supplies are also important for best clock circuit performance. Battery power supplies work pretty well, but aren't very practical. I tested super regulators (6nV), but they simply couldn't beat the reference battery power supply. I used a switch to toggle between battery and mains power supply, so I could immediately hear the difference in sound quality during playback. I used 3* 1.5V alkaline batteries, and 4 * 1.2 NiMH rechargeable batteries (AA-size) for reference,
The mains power supply resulted in a slightly darker smeared sound. Battery power supplies result in a slightly "thinner", cleaner sound. This effect is probably caused by changed master clock jitter spectrum.
After lots of experimenting, I was just about to give-up on mains power supplies (for the reclocker) and was was planning to connect a rechargeable battery pack to the DI8M, but then I got an idea, a passive filter with back-up capacitors.
I used 10 ... 20mH ring core inductor and a Panasonic 0.1F (100,000uF) / 5.5V back-up capacitor. To my surprise this filter actually "outperformed" the battery reference supply. These filters are placed after each shunt regulator.
Attachments
Well I agree that clean power supplies are very important for clocks but actually clean supplies are also important for every aspect of the audio chain.
If the chokes are added to EVERY other stage in your design and done one at a time, as well as getting a minor eureka with each stage you will probably notice a very big difference when the last "noisy" supply is replaced.
. . . for maximum effect you also have to replace cheap resistors in critical positions ( those in signal path or feedback path ) with good quality low noise types e.g. caddock TF020 or vishay bulk foil
All this may seem like a huge effort & expense but when it is all done I expect your ears & emotions will conclude it was worth it.
( This is assuming that you do not have cheap or electrolytic capacitors in the signal path. )
If the chokes are added to EVERY other stage in your design and done one at a time, as well as getting a minor eureka with each stage you will probably notice a very big difference when the last "noisy" supply is replaced.
. . . for maximum effect you also have to replace cheap resistors in critical positions ( those in signal path or feedback path ) with good quality low noise types e.g. caddock TF020 or vishay bulk foil
All this may seem like a huge effort & expense but when it is all done I expect your ears & emotions will conclude it was worth it.
( This is assuming that you do not have cheap or electrolytic capacitors in the signal path. )
-ecdesigns- wrote:
With all due respect and risking to be called "mono-thematic", something like that can also be achieved with the TeddyReg:
1) The RC filter (R= 75K; C= 33uF tantalum, in my units) part of the Reg has a corner F of a fraction of a Hertz.
2) The final capacitance of of the gyrator is 33uF times the gain of the home made Darlington, composed by BC550 + D44H11 (around 500*60) ...a little short of 1F
I have used it only on analog circuits with predictable benefits. I used it as pre-reg on Doede's USB module with noticeable improvement (I have not replaced the stock shunt-reg for fear of damage )
Lets hope that Soundcheck succeeds with his project...
http://www.pinkfishmedia.net/forum/...ead.php?t=39990
Update: I no longer find any fault on the highs of my DI16 (with TeddyRegs on DAC's power supplies) ... it was only a bad choice of interconnects. Now I feel very extended highs and better detail. Sound is even more open. By extension, I doubt the stock DAC had any trouble there because when swapping the stock Reg for the TeddyReg I heard less extended highs
Cheers,
M
With all due respect and risking to be called "mono-thematic", something like that can also be achieved with the TeddyReg:
1) The RC filter (R= 75K; C= 33uF tantalum, in my units) part of the Reg has a corner F of a fraction of a Hertz.
2) The final capacitance of of the gyrator is 33uF times the gain of the home made Darlington, composed by BC550 + D44H11 (around 500*60) ...a little short of 1F
I have used it only on analog circuits with predictable benefits. I used it as pre-reg on Doede's USB module with noticeable improvement (I have not replaced the stock shunt-reg for fear of damage
)Lets hope that Soundcheck succeeds with his project...
http://www.pinkfishmedia.net/forum/...ead.php?t=39990
Update: I no longer find any fault on the highs of my DI16 (with TeddyRegs on DAC's power supplies) ... it was only a bad choice of interconnects. Now I feel very extended highs and better detail. Sound is even more open. By extension, I doubt the stock DAC had any trouble there because when swapping the stock Reg for the TeddyReg I heard less extended highs
Cheers,
M
Hi MaxLorenz.
I finished my TeddyRegs last Friday.
THX a lot for your support.
One of the regulators I am using for the supply of the Opticis USB--Cable, which in turn is delivering the 5V for the 2706 of the reclocker module.
I also built myself a tiny USB-Adapter. In the end I reduced my cable-length down to 2 inches, which is another advantage.
Now I have a perfect galvanically isolated USB connection and a close to perfect 2706 supply.
As I wrote in my Linux thread:The galvanic isolation helps a lot to reduce further distortions.
The other one I am using of course for the reclocker.
I don't know if all the stuff needs some more time to break-in, however - the sound quality improved already. This was clear after the first seconds of listening.
Especially the soundstaging & imaging improved. ( I was very impressed after listening to my favorite test track for these characteristics on Chesky-Stereo Gold Stereo and Surround Setup CD - Track 22!! ) Improvements in all directions. You can see the percussion-crowd running in circles right in front of you. ( These folks around here knowing this CD, know what I a mean )
However. I still have the feeling that something is missing and/or slightly distorted. I am still a bit lacking air between the instruments and also the timbre of some instruments is not 100% as I would expect it. I am still working on it. I am getting there.
One comment to the battery vs. cap discussion. I am using Northstar batteries with a very low resistance (2mR) or ESR.
This equals pretty much the ESR of large caps I'd say.
I used to have smaller batteries, which usually have a much higher ESR. And yes this makes a difference. I am also buffering and filtering the 12V power-rails a bit. I am quite happy with the overall result.
Cheers
I finished my TeddyRegs last Friday.
THX a lot for your support.
One of the regulators I am using for the supply of the Opticis USB--Cable, which in turn is delivering the 5V for the 2706 of the reclocker module.
I also built myself a tiny USB-Adapter. In the end I reduced my cable-length down to 2 inches, which is another advantage.
Now I have a perfect galvanically isolated USB connection and a close to perfect 2706 supply.
As I wrote in my Linux thread:The galvanic isolation helps a lot to reduce further distortions.
The other one I am using of course for the reclocker.
I don't know if all the stuff needs some more time to break-in, however - the sound quality improved already. This was clear after the first seconds of listening.
Especially the soundstaging & imaging improved. ( I was very impressed after listening to my favorite test track for these characteristics on Chesky-Stereo Gold Stereo and Surround Setup CD - Track 22!! ) Improvements in all directions. You can see the percussion-crowd running in circles right in front of you. ( These folks around here knowing this CD, know what I a mean
)However. I still have the feeling that something is missing and/or slightly distorted. I am still a bit lacking air between the instruments and also the timbre of some instruments is not 100% as I would expect it. I am still working on it. I am getting there.
One comment to the battery vs. cap discussion. I am using Northstar batteries with a very low resistance (2mR) or ESR.
This equals pretty much the ESR of large caps I'd say.
I used to have smaller batteries, which usually have a much higher ESR. And yes this makes a difference. I am also buffering and filtering the 12V power-rails a bit. I am quite happy with the overall result.
Cheers