A few month ago I had to made a decision wether to use a complete hardware like MiniSharc or to build a PC for convolution. My requirement was a "black box" which acts as a FIR crossover with 8 output channels. I ended up with a small PC on Linux basis. It was cheaper and much more powerful than the MiniSharc. And it can directly stream my music. No jitterful S/PDIF or AD/Da conversion is involved.
As sound card I use a Xonar Essence STX II 7.1. I wanted an internal solution and no USB involved. So I can turn-on and shut-down the "black box" with one push the power button. And my little daughter can not pull out any USB cable.
PS: forgot to mention that the PC was cheaper than a MiniSharc with case.
Excellent solution overall. My plans is rePhase to implement filters and Jriver to run the filters. In the STX-II 7.1, You have pointed me to a very compelling option instead of an external 8ch DAC i am considering (http://www.diyaudio.com/forums/grou...kbird-multi-channel-es9018-dac-group-buy.html)
My platform is going to be a fanless HTPC running Win 7. That sound card might not suit the smaller form factor HTPC cases i was considering.
But my concern is how convenient volume control is going to be. I do want to avoid using a monitor and keyboard
Thanks a lot for all the details. Will get back with questions once i start trying it out.
I agree with BYRTT. Using JRiver's WDM driver is very easy, it is stable, and it allows all audio playback incl. netradio to be processed by DSP built into JRiver.
For remote control it is very convenient to use an iPad, a Surface unit or another PC to replicate content of the computer screen and for navigation. In addition to this I use a classic HTPC remote control, which is simple to setup through the remote control section of JRiver. This remote can control volume and the usual playback (start, stop pause, skip etc.).
For remote control it is very convenient to use an iPad, a Surface unit or another PC to replicate content of the computer screen and for navigation. In addition to this I use a classic HTPC remote control, which is simple to setup through the remote control section of JRiver. This remote can control volume and the usual playback (start, stop pause, skip etc.).
You're lucky it's stable for you. For lots of us it doesn't work without a lot of crackling. The devs have just buried third head in the sand over the issues.
I do use it bit wish there was smoother solution that offered a sonar audio engine with WDM input but alas I don't think there is.
FoLLgoTT. Seems like you've got it going on.
Would you consider starting a thread or even writing an article for the forum?
It would be awesome, since a linux based PC is clearly one of the best solution to perform active filtering
Have you optimized your PC for audio playback? I have helped several audiophiles with setup, and this usually solves the problems. Earlier in this thread, I have linked to Sweetwater's guide to windows optimization which is a good description. Be sure to turn off all energy saving options on the PC including USB bus power management. Also deactivate a possible second graphics card and activate playback from memory. Have you tried increasing the buffer size to the DAC? Also I find the WASAPI playback option in JRiver to more stable than ASIO.
orangeheart,
Agree tips by Laotzu1 and Pano, also if you on a desktop one can change priority electric hardware routes. Example my dedicated audio PC that run on low cost AMD platform (Liano APU A8-3870) have normal prioty electric routes turn upside down. All onboard hardware turned off to free IRQ's and memory is only one 4Gb module in single mode because think sound is better than running dual mode by two modules on this particular platform. Dirty jobs is moved to electric route via chipset south-bridge that is VGA GT720 (8xPcie) physical have a 4xPcie electric slot and a special PCI USB3 card for mouse/keybord and flash disks and OS HDD is down here too. Then have 3 free Pcie slots with direct electric route to CPU where normal VGA would have been mounted, at those three run network adapter for streaming and a AHCI controller for dedicated audio HDD's and a Renesas USB3 controller for USB audio device or pull that USB3 controller out if a dedicated Pcie audio device is used. Below you see MB the three Pcie slots at right (two x1slot and one x16 slot) have direct electric route to CPU the other slot's goes via chipset south bridge. Setup in this way is probably not nessasery but worked pretty good for this platform, on the fly can tick convolution engine and IRR filters in JRiver on/off to compare sound before/after without any crackling. Also have no problems let program TrueRTA build in signal generator send square waves via JRiver virtual sound device/Jriver DSP/Hardware sound device-DAC/AMP/Speakers/microphone back to TrueRTA oscilloscope presentation.
Agree tips by Laotzu1 and Pano, also if you on a desktop one can change priority electric hardware routes. Example my dedicated audio PC that run on low cost AMD platform (Liano APU A8-3870) have normal prioty electric routes turn upside down. All onboard hardware turned off to free IRQ's and memory is only one 4Gb module in single mode because think sound is better than running dual mode by two modules on this particular platform. Dirty jobs is moved to electric route via chipset south-bridge that is VGA GT720 (8xPcie) physical have a 4xPcie electric slot and a special PCI USB3 card for mouse/keybord and flash disks and OS HDD is down here too. Then have 3 free Pcie slots with direct electric route to CPU where normal VGA would have been mounted, at those three run network adapter for streaming and a AHCI controller for dedicated audio HDD's and a Renesas USB3 controller for USB audio device or pull that USB3 controller out if a dedicated Pcie audio device is used. Below you see MB the three Pcie slots at right (two x1slot and one x16 slot) have direct electric route to CPU the other slot's goes via chipset south bridge. Setup in this way is probably not nessasery but worked pretty good for this platform, on the fly can tick convolution engine and IRR filters in JRiver on/off to compare sound before/after without any crackling. Also have no problems let program TrueRTA build in signal generator send square waves via JRiver virtual sound device/Jriver DSP/Hardware sound device-DAC/AMP/Speakers/microphone back to TrueRTA oscilloscope presentation.
Attachments
Last edited:
Based on inputs, here are my 3 choices for DAC for the DSP stream from the HTPC
1. Asus Essence STX II 7.1 PCIe soundcard : approx $400
The card seems to use top notch components. I dont know if the power provided from a potentially noisy HTPC smps affects SQ in any way
2. Motu Monitor 8 : Uses Sabre32 DAC: Approx $1000
3. Black bird DAC http://www.diyaudio.com/forums/grou...kbird-multi-channel-es9018-dac-group-buy.html. Approx $1500
Comments, views etc please.
1. Asus Essence STX II 7.1 PCIe soundcard : approx $400
The card seems to use top notch components. I dont know if the power provided from a potentially noisy HTPC smps affects SQ in any way
2. Motu Monitor 8 : Uses Sabre32 DAC: Approx $1000
3. Black bird DAC http://www.diyaudio.com/forums/grou...kbird-multi-channel-es9018-dac-group-buy.html. Approx $1500
Comments, views etc please.
Hi Jojip,
There are more options, depending on your target configuration: 2.0 stereoup to 7.2 +CABS.
http://www.bodziosoftware.com.au/AES_EBU_System_Evolution.pdf
http://www.bodziosoftware.com.au/AES_EBU_Amplifier_Construction.pdf
http://www.bodziosoftware.com.au/Hi_End_Audio_Server.pdf
http://www.bodziosoftware.com.au/AES_EBU_24Bit_96kHz_System.pdf
http://www.bodziosoftware.com.au/Hybrid AES_EBU_Analogue_System.pdf
http://www.bodziosoftware.com.au/Home_Theatre_Updates_2015.pdf
Best Regards,
Bohdan
There are quite some good multichannel DACs out there. The Motu Monotor 8 is from the AVB-series, which should all sound identical (except Ultralite AVB). I use the Motu 1248, and can confirm that it is better than Roland OctaCapture. I believe it is hard to find a better sounding multichannel DAC at about 1000USD. For revies see JRiver forum or Gearslutz
Lynx Aurora is more expensive, but also good sounding - it uses the same chip as Motu: ESS 9016S. There are claims on the net that Motu sounds better, but I have never compared Lynx and Motu directly. The ultimate industrially produced DAC may be Cranesong, Prism or the like - prices rise very quickly.
The Blackbird project may be able to develop an even better sounding DAC. I "love" my Motu 1248 and I am completely satisfied with it, but anyway I am involved in a project to build an ESS 9018S based DAC from scratch. It will be controlled by an Arduino board.
Lynx Aurora is more expensive, but also good sounding - it uses the same chip as Motu: ESS 9016S. There are claims on the net that Motu sounds better, but I have never compared Lynx and Motu directly. The ultimate industrially produced DAC may be Cranesong, Prism or the like - prices rise very quickly.
The Blackbird project may be able to develop an even better sounding DAC. I "love" my Motu 1248 and I am completely satisfied with it, but anyway I am involved in a project to build an ESS 9018S based DAC from scratch. It will be controlled by an Arduino board.
Right. The possibility of using an anlog input was a primary reason for my choice of the 50% more expensive 1248. The intention was 1) to be able to playback analog sources, 2) to be completely free to choose any meaurement mic. I have however never seriously used the analog inputs, bur I can confirm that it works.
My measurement mic is MINIDSP U-MIK1, and it contains it's own ADC, so it connects to a USB-port and does not need a mic preamp. One advantage is that the calibration file does contain calibration for the entire recording chain, because the signal does not pass an external (and for that reason variable) preamp and ADC circuit. It is a qiute cheap for a calibrated microphone.
My measurement mic is MINIDSP U-MIK1, and it contains it's own ADC, so it connects to a USB-port and does not need a mic preamp. One advantage is that the calibration file does contain calibration for the entire recording chain, because the signal does not pass an external (and for that reason variable) preamp and ADC circuit. It is a qiute cheap for a calibrated microphone.
I'm using a Behringer fca610.
Software wise I use Reaper as a vst host, hosting various convolvers for FIR crossovers (designed in rephrase)/room correction.
Reaper has an asio loopback driver which I use for playback from foobar, or vb-audio cable from everything else.
Vinyl playback is just handled in reaper, I use the mic inputs on the Behringer, RIAA is handled in the digital domain as another vst plugin.
Id like to try a higher end interface such as the motu, but for now, the $200 behringer suits.
Software wise I use Reaper as a vst host, hosting various convolvers for FIR crossovers (designed in rephrase)/room correction.
Reaper has an asio loopback driver which I use for playback from foobar, or vb-audio cable from everything else.
Vinyl playback is just handled in reaper, I use the mic inputs on the Behringer, RIAA is handled in the digital domain as another vst plugin.
Id like to try a higher end interface such as the motu, but for now, the $200 behringer suits.
In ASIO mode ARTA requires the input and output channels to be on the same soundcard.
With Essence STX II 7.1 or Motu, both ADC and DAC are handled through the same driver and may suit tools like ARTA better.
With blackbird DAC, the output channel is an XMOS ASIO driver, while the Mic preamp input driver will be my focusrite 2i2 driver. I dont think i can get ARTA to work with that