Remember that inevitably, the power to a PCI sound card will come from the same rails that power the rest of the motherboard, with all the noise that is injected by the other components. A USB DAC has the advantage of running from clean rails.
However, the USB standard isn't really designed for audio. The Firewire standard is far superior. I won't explain why here, as I'm sure you've had people preach about it to you before.
I was just wondering, Peter, if you'd considered developing a FireWire DAC? There are plenty of commercial products, but most are designed for recording and monitoring and are thus focused around the ADC section, and very few could be considered Hi-Fi designs.
However, the USB standard isn't really designed for audio. The Firewire standard is far superior. I won't explain why here, as I'm sure you've had people preach about it to you before.
I was just wondering, Peter, if you'd considered developing a FireWire DAC? There are plenty of commercial products, but most are designed for recording and monitoring and are thus focused around the ADC section, and very few could be considered Hi-Fi designs.
I'm also under impression, that soundcard is a weak link in outputting signal from a computer; as bad as some try to picture USB, it has certain advantages, and it certainly shows in my current system.
I'd like to come up with firewire DAC, but I still don't know what chip to use, any recommendations?
I'd like to come up with firewire DAC, but I still don't know what chip to use, any recommendations?
Rick,
I'm checking this manual: http://photos.imageevent.com/cics/cmp/Installation Guide and User Manual.pdf
He still uses same CPU and Biostar Mobo, which is basiaclly same as previous version. Are there any other manuals?
Also, what's the best source for Juli@ card? I wouldn't mind trying it. My DAC does not accept higher formats, but I also have a different one that does, although I still can't get excited about it
I got some improvements after I finally disconnected SRC resampler few days back.
I'm checking this manual: http://photos.imageevent.com/cics/cmp/Installation Guide and User Manual.pdf
He still uses same CPU and Biostar Mobo, which is basiaclly same as previous version. Are there any other manuals?
Also, what's the best source for Juli@ card? I wouldn't mind trying it. My DAC does not accept higher formats, but I also have a different one that does, although I still can't get excited about it
I got some improvements after I finally disconnected SRC resampler few days back.
Hi,
i´m using the 120-WI-25 PICO and i´m feeding it with 15 Volts out of a
Notebook PSU. But i already planned to build an PSU on my own but originally
planned toroidcore transfo with msr and 10000 mü elna silmic.
But maybe i´ll try your lt1083 psu.
Peter and could you please post a picture of your modded pico?
Greetings Ulf
i´m using the 120-WI-25 PICO and i´m feeding it with 15 Volts out of a
Notebook PSU. But i already planned to build an PSU on my own but originally
planned toroidcore transfo with msr and 10000 mü elna silmic.
But maybe i´ll try your lt1083 psu.
Peter and could you please post a picture of your modded pico?
Greetings Ulf
I'm afraid I don't know of a simple solution. Since products exist, there is obviously a chip to do the job, but I'm afraid I don't have the spare funds to go purchasing Mark of the Unicorn gear just to see what chips are inside.
I know TI's TSB43CB43A can be used for the job, but it's about a million times more complex than a common USB-I2S chip like the PCM2707. The reason for that is that while USB devices are all slaves to the host (the computer) all Firewire devices are on equal footing, and as such, Firewire interface chips require an embedded micro-controller.
But still, if M-Audio and MotU and DigiDesign and whatnot can bring products to market for a few hundred dollars, it's got to be possible to create something, even if it is pricey.
I know TI's TSB43CB43A can be used for the job, but it's about a million times more complex than a common USB-I2S chip like the PCM2707. The reason for that is that while USB devices are all slaves to the host (the computer) all Firewire devices are on equal footing, and as such, Firewire interface chips require an embedded micro-controller.
But still, if M-Audio and MotU and DigiDesign and whatnot can bring products to market for a few hundred dollars, it's got to be possible to create something, even if it is pricey.
JULI@ iis only used for SPDIF
My DAC is completely outboard and with a transformer input for galvanic isolation. There is no difference in audible noise with computer on or off. I do not have a scope so who knows what going on way up there. But with the clean high frequencies, and the fact that my amplifiers stay cool makes me think that the transformer is doing its job.
look here:
http://www.audioasylum.com/forums/pcaudio/messages/2/27041.html
and here:
http://www.audioasylum.com/forums/pcaudio/messages/3/31286.html
There is much that has happened since these threads were started.
It is true that the components on the MB would do much to spoil a perfect power supply but this is the case with almost anything.
I had asked Paul Hynes to come up with a shunt reg for a computer and when he gets time he is going to see what he can come up with. He is going to try out cMP and cPLAY but seems to be so dedicated to LP he is not terribly motivated to drop everything and DO IT. But, to get back on point something conceptually makes me think the shunt reg could maybe absorb some of the returned garbage unlike any other regulated approach. One would think a shunt reg for the 3.3 and 5 volts rails would be much easier to implement that one for the 12 volts rail which brings me back to thinking that a BIG batter would be best for this. Most power is required for the 12 volts rail.My idea was to use a power diode to give a little voltage reduction but then you would lose the give and take aspect. Of course, my concept could be completely ridiculous.
I would like to hear what others think.
There is much good in this and much more possible.
My DAC is completely outboard and with a transformer input for galvanic isolation. There is no difference in audible noise with computer on or off. I do not have a scope so who knows what going on way up there. But with the clean high frequencies, and the fact that my amplifiers stay cool makes me think that the transformer is doing its job.
look here:
http://www.audioasylum.com/forums/pcaudio/messages/2/27041.html
and here:
http://www.audioasylum.com/forums/pcaudio/messages/3/31286.html
There is much that has happened since these threads were started.
It is true that the components on the MB would do much to spoil a perfect power supply but this is the case with almost anything.
I had asked Paul Hynes to come up with a shunt reg for a computer and when he gets time he is going to see what he can come up with. He is going to try out cMP and cPLAY but seems to be so dedicated to LP he is not terribly motivated to drop everything and DO IT. But, to get back on point something conceptually makes me think the shunt reg could maybe absorb some of the returned garbage unlike any other regulated approach. One would think a shunt reg for the 3.3 and 5 volts rails would be much easier to implement that one for the 12 volts rail which brings me back to thinking that a BIG batter would be best for this. Most power is required for the 12 volts rail.My idea was to use a power diode to give a little voltage reduction but then you would lose the give and take aspect. Of course, my concept could be completely ridiculous.
I would like to hear what others think.
There is much good in this and much more possible.
Hi Rick, thanks for the info.
I checked last night Asylum thread and you guys put a lot of work into that thing. I have some questions:
Is switching to GIGABYTE MB due to nonavailability of Biostar or because of better performance, if so, how much better?
How much improvement comes with E7200 comparing to previously used E2140?
How would you compare Memoery Player to playback directly from HDD?
And finally, what's the last word on RAM size and type?
I would be adding separate 5V supplies for mouse and HDD to see if there's any improvement here.
I checked last night Asylum thread and you guys put a lot of work into that thing. I have some questions:
Is switching to GIGABYTE MB due to nonavailability of Biostar or because of better performance, if so, how much better?
How much improvement comes with E7200 comparing to previously used E2140?
How would you compare Memoery Player to playback directly from HDD?
And finally, what's the last word on RAM size and type?
I would be adding separate 5V supplies for mouse and HDD to see if there's any improvement here.
Hi Peter,
Another software to try is XXhighend. I love the sound of this memory player.http://www.phasure.com/index.php..Im using the same Biostar MB , you just need to add more ram
Another software to try is XXhighend. I love the sound of this memory player.http://www.phasure.com/index.php..Im using the same Biostar MB , you just need to add more ram
Minimizing power usage ...
is a big part of what guides cics. Almost all of his changes revolve around this.
The BIOSTAR MB cannot accept the e7200. The e7200 is made with the latest INTEL 45nm technology.
This set-up allows one to run the CPU at voltages below 0.90 volt. This allows fanless operation with a large heatsinlk. Plus, the CPU is very powerful which is important for using the -145 bB upsampler, which may not make a difference to you, but the other aspects would still apply.
Certainly powering the USB devices off of the board is a good idea.
Wish we could figure out how to power the soundcard off board.
XP PRO is preferred over VISTA since there is more flexibility in turning more of it OFF.
Minimizing need for memory, again, allows for lower power usage. Using low latency memory running at much slower speed, again, allows for lower power usage, still.
cPLAY will automatically split your files so one can use 512 mB memory. KINGSTON HYPER X LL memory (UL better still if you can find it) seems to be the consensus choice. I tried some expensive low latency memory from CORSAIR and found it did not sound as natural as the HYPER X. I know it sounds silly. Same with dual channel memory; single channel sounds better.
One is left with a dedicated device for music playback. Some to find this problematic. I can only assume they have found a way to receive emails with their CD players.
cics has found there is more to "jitter" than the conventional wisdom. Many ways to disjoint the waveform. That is what I hear with his approach: a complex waveform delivered intact.
Bye,
Rick McInnis
is a big part of what guides cics. Almost all of his changes revolve around this.
The BIOSTAR MB cannot accept the e7200. The e7200 is made with the latest INTEL 45nm technology.
This set-up allows one to run the CPU at voltages below 0.90 volt. This allows fanless operation with a large heatsinlk. Plus, the CPU is very powerful which is important for using the -145 bB upsampler, which may not make a difference to you, but the other aspects would still apply.
Certainly powering the USB devices off of the board is a good idea.
Wish we could figure out how to power the soundcard off board.
XP PRO is preferred over VISTA since there is more flexibility in turning more of it OFF.
Minimizing need for memory, again, allows for lower power usage. Using low latency memory running at much slower speed, again, allows for lower power usage, still.
cPLAY will automatically split your files so one can use 512 mB memory. KINGSTON HYPER X LL memory (UL better still if you can find it) seems to be the consensus choice. I tried some expensive low latency memory from CORSAIR and found it did not sound as natural as the HYPER X. I know it sounds silly. Same with dual channel memory; single channel sounds better.
One is left with a dedicated device for music playback. Some to find this problematic. I can only assume they have found a way to receive emails with their CD players.
cics has found there is more to "jitter" than the conventional wisdom. Many ways to disjoint the waveform. That is what I hear with his approach: a complex waveform delivered intact.
Bye,
Rick McInnis
Thanks again for that valuable info. I have to try that memory playback, it seems like an interesting approach, and I found some local suppliers of Hyper LL memory.
With my E2140 and low profile Freezer heatsink, I can also run it without fun, not even lowering CPU voltage. Problem is however, that the small heatsing on MB runs extremely hot without any air movement
I'm using this keyboard and mouse: http://www.tigerdirect.ca/applications/SearchTools/item-details.asp?EdpNo=1578906&CatId=1482 and they are very convenient without a need for any other remote function devices or specific software.
With my E2140 and low profile Freezer heatsink, I can also run it without fun, not even lowering CPU voltage. Problem is however, that the small heatsing on MB runs extremely hot without any air movement
An externally hosted image should be here but it was not working when we last tested it.
I'm using this keyboard and mouse: http://www.tigerdirect.ca/applications/SearchTools/item-details.asp?EdpNo=1578906&CatId=1482 and they are very convenient without a need for any other remote function devices or specific software.
le´flu said:
This is how it looks. As you see, one cap is removed completely, the other is BG N 100/16. Presently I'm using Panasonic Audio Grade 6800/63 before regulator and BG N 100/50 after. I decided for LT1083-12 as this seemed to be the only integrated regulator available for 12 V and reasonably high (7.5A) current output. Arrow still has it.
Attachments
3875 I decided to reduce some noise or something by switching to star grounding.
I soldered PG+, PG-, CG-Right, CG-Left, and with a 100ohm resistor in the path wire that connects to the chassis.
When I turn my amp on I have a tiny bit of noise which I am afraid is just from my transformer being so close to the rectifier board. Which reminds me anyone know anything to isolate this thing from that? I tried copper foil grounded to neutral but that just made the FM worse, did nothing for killing noise.
I have Hawthorne Silver Iris speakers. I would like for them to be dead silent with no source on but with amplifier on. At least I suppose this would equal overall better sound, less distortion.
Lastly I have some subwoofer amps getting their information from high level in. I tried turning them all the way off to see if it helped but it did nothing but kill their transformers noise.
I soldered PG+, PG-, CG-Right, CG-Left, and with a 100ohm resistor in the path wire that connects to the chassis.
When I turn my amp on I have a tiny bit of noise which I am afraid is just from my transformer being so close to the rectifier board. Which reminds me anyone know anything to isolate this thing from that? I tried copper foil grounded to neutral but that just made the FM worse, did nothing for killing noise.
I have Hawthorne Silver Iris speakers. I would like for them to be dead silent with no source on but with amplifier on. At least I suppose this would equal overall better sound, less distortion.
Lastly I have some subwoofer amps getting their information from high level in. I tried turning them all the way off to see if it helped but it did nothing but kill their transformers noise.
When doing star ground, please refer to this post: http://www.diyaudio.com/forums/showthread.php?postid=1518369#post1518369
CG should not be used at all, your main ground point on amp board is OG
Transformer usually is not a cause of noise and it can be very close to the rest of a circuit.
If the noise you are getting is of radio type, check this post: http://www.diyaudio.com/forums/showthread.php?postid=1516522#post1516522
CG should not be used at all, your main ground point on amp board is OG
Transformer usually is not a cause of noise and it can be very close to the rest of a circuit.
If the noise you are getting is of radio type, check this post: http://www.diyaudio.com/forums/showthread.php?postid=1516522#post1516522
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.