Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
Dirac and room correction software
#11
(27-May-2014, 22:40)thumb5 Wrote: I think perhaps you're just missing the [img] and [/img] tags round the link?

Thank you! Sorted! Smile

Guillaume
Industry disclosure: UK distributor for Shunyata Research

220 PRO, totaldac d1 server with additional external power supply, totaldac d1-seven, Echole PSU for Totaldac, Wilson Audio Sasha 2, Shunyata Research cables, Shunyata Hydra Alpha A10 + DPC-6 v3, Various Entreq ground boxes and cables, Entreq Athena level 3 rack, 2 X SOtM sNH-10G with sCLK-EX + 10MHz Master Clock input + sPS-500 PSU, i5 sonicTransporter w/ 1TB SSD

UK
Reply
#12
Hi Guys,

Thanks for sorting this. They are indeed kef ls50s. I have moved them to my other room now for a second system. Have to admit the ls50s are a truly magnificent speaker - through the devialet they are like magic. As long as you don't want uber potent bass - I go next door for that and let the maestro take my head off Wink
Reply
#13
(27-May-2014, 20:56)Bill Wrote: My personal route has been via Acourate using the convolution engine in jriver. I can stream this via AIR to the Devialet or via my USB / spdif connector, the latter being my preferred route as it sounds a fraction smoother and more real.

.....

When I added the acourate filter (64bit filter in jriver) the soundstage dropped back significantly and the imaging just snapped into focus. I played around with the target curves and realised that I like quite a steep slope shelving down in the treble.

Anyway - would never go back now and see it a must for any system. There are no drawbacks other than I can't yet do this for vinyl - this is what disappointed me most about the latest announcements from devialet - i.e. Not supporting full correction.

If anyone wants room pics and/or target crurves and stuff - let me know. Not sure how I add them to the forum though so will need a lesson on that.

Hi Bill and All,
I love the idea of this forum, a place to really swap experiences without (I hope!) "Disbelievers" droning on. I too use acourate in the JRiver 19 convolution engine, and I echo your statements about soundstage and imaging. I intend putting my collection on my NAS using AcourateNAS to alter the tracks so I don't need to use the PC. I 'm trialling various iterations of EQ, pre-ringing, measurement window etc., to find one I can apply to a large library.
My room is small - 15 feet by 12 by 9.5 and is full of my better half's stuff, so bass trapping, extensive diffusion, and placement flexibility are off the agenda. I made inquiries of Dirac, but found that I would have to buy a multi license installation because my main PC is in another room and can't be moved - I run an ethernet cable and USB cable to the listening room - so I needed to install it on a laptop as well but only for measurements. There was no such problem with acourate because the measurement software is free.
I cannot listen to music without room correction, and I would love it if Devialet allowed a convolution plugin - I don't even know if that's possible - so I could listen to LPs.
Cheers
Graeme
Roon Lifetime, HQPlayer with Acourate RC,Devialet 220 Pro,Synology NAS,AcourateNAS, MinimServer,Moon Mind 180 AES/EBU currently w. Canare Cable,W7 PC JRiver 20, AcouratePro, 15' Belkin USB. Linn LP12,Cetech,OL1,Denon 103R,VAF Signature i66 Mk.11,VAF Gravitas SW
Reply
#14
Interesting recent interview with Mathias Johansson, CEO of Dirac:

Audioholics - Dirac

And this is a link to a technical description of Dirac Live

Dirac Live


I was a bit surprised by their recommendations for 'computer performance' I would have thought that higher performance would be required for real-time DSP and 24/192 source files? I haven't seen any reference to the use of multiple cores.

• Windows Vista, 7, 8
• Mac OS X 10.6 Snow Leopard - 10.9 Mavericks
• Intel Pentium III or later, AMD Athlon XP or later (SSE required)
• 2 GB RAM or more

Dirac offer trial versions of their software for Mac or PC. Maybe I will try it if I have time over Christmas. EDIT - I see from Thumb5s posts above that it will work with AIR which is a surprise to me Smile

Looking at the changes in room response that Thumb5 measured I was a bit disappointed to be honest as there are still plenty of potential 'issues' (From the Audioholics article I can understand why this is the case). I understand that you can set up Dirac to work for only part of the frequency range - perhaps using it only below 200 Hz might be a sensible thing to try.
IMac macOS 10.15.3 (no link to Devialet Sad ) / MacBook Pro Retina OS X 10.14.4 / Linn LP12 / Devialet 200 Wilson Benesch Discovery. 
Qobuz Desktop Latest Version / Audirvana 3.2.18 / Audirvana Remote / iTunes 12.9 / AIR 3.0.4 / Wi-Fi / FW 8.1.0 / SAM 50%
Cambridge, UK (Updated 27th February, 2020)
Reply
#15
Hello Phil,

It seems a long time ago since I did those experiments, and I've learned a bit since that might make me do things a little differently now!

To reply to a couple of the points you mentioned above:

I was also a bit surprised that Dirac "just worked" with AIR on my Mac - both for measuring the room and (more so) for correcting it. Whether this would also be true on a PC, I'm not sure.

On my 2008 MacBook Pro (2.5 GHz Core 2 Duo) I measured a CPU load of around 3-5% due to the Dirac Audio Processor when playing 16/44 material from iTunes. The load due to iTunes and AIR is each also about the same, making a total of 12-15% or thereabouts. I guess the loading would go up with higher sample rate material.

I agree with your comment that the corrected response still looked a bit rough. Subjectively the corrected system did sound noticeably better, mostly in the bass (not surprisingly). I should probably point out that this was pretty much my first go-round with Dirac and FuzzMeasure so I was more interested in getting some quick results to have an idea what might be possible, rather than trying to fine tune for the best possible outcome.

I've since done quite a bit of work on integrating my sub-woofer more carefully, and have added the DSPeaker Anti-Mode 2.0 which subjectively does a better job than Dirac of sorting out the sub-150 Hz part of the spectrum. (Much as it pained me to add another box, it does mean the correction gets done for all sources without having to run software on a general-purpose computer. As Kari pointed out in another thread, you can now buy a dedicated box from MiniDSP to run the Dirac correction software.)

My impression so far is that full-range computer-based room correction is probably not a "set and forget" operation, but something that one might want to continually tweak in an attempt to flatten out ever smaller kinks in the measured response. For what it's worth, I think I've decided that I don't have the inclination to do that at the moment, now things are sounding broadly right (to my ears). I might come back to it one day, though.

Ian
Roon (Mac Mini), Wilson Benesch Full Circle, Expert 1000 Pro CI, Kaiser Chiara
Warwickshire, UK
Reply
#16
I have a few miniDSP units , the Acourate (openDRC) , 4x10 and on Friday my Dirac unit arrived.
I have tried Acourate and open DRC on my D-premier and G1's and now want to biamp them and try DIRAC only on the bass section (I couldnt limit Acourate for the minidsp to do bass only , it does the whole spectrum.)
Initially , today , I intend to use Dirac full range and compare the results to Acourate and thereafter will do the biamping thing for bass only..
I will report back.
In the interim , I was using my trusty Z-sys RDP-1 parametric to take out some big bass nodes etc .. and it sounded pretty good.
AIR is unstable for me , so I use a Squeezebox touch..the DIRAC doing full range will be inserted after the Touch and into a digital input.
When I do biamping , I will use the minidsp on a variable digital output and have bought the board to convert my D to D to D to DA ($75) and will use it as a DAC as well (into a Crown CDI4000 just for the bass)
Roon/tidal > Squeezebox touch  > Trinnov St2 or DIRAC (minidsp ddrc-22d) > Dual mono D premiers > Vivid Audio Giya G1 Spirits  ...fully treated  dedicated 6x8m room
Reply
#17
(08-Dec-2014, 08:40)Rodney Gold Wrote: I have a few miniDSP units , the Acourate (openDRC) , 4x10 and on Friday my Dirac unit arrived.
I have tried Acourate and open DRC on my D-premier and G1's and now want to biamp them and try DIRAC only on the bass section (I couldnt limit Acourate for the minidsp to do bass only , it does the whole spectrum.)
Initially , today , I intend to use Dirac full range and compare the results to Acourate and thereafter will do the biamping thing for bass only..
I will report back.
In the interim , I was using my trusty Z-sys RDP-1 parametric to take out some big bass nodes etc .. and it sounded pretty good.
AIR is unstable for me , so I use a Squeezebox touch..the DIRAC doing full range will be inserted after the Touch and into a digital input.
When I do biamping , I will use the minidsp on a variable digital output and have bought the board to convert my D to D to D to DA ($75) and will use it as a DAC as well (into a Crown CDI4000 just for the bass)

Very interesting - I look forward to your next update. I'm using my system in my study at the moment which measures approx 4.8m x 3.7m. I have the system firing along the room now because there are major bass issues when I fire across. However, firing across would be far better practically because of the position of windows, door and radiator. If Dirac could solve the bass problem for me and allow me to do this I would be very happy ;)

(07-Dec-2014, 22:34)thumb5 Wrote: ... On my 2008 MacBook Pro (2.5 GHz Core 2 Duo) I measured a CPU load of around 3-5% due to the Dirac Audio Processor when playing 16/44 material from iTunes. The load due to iTunes and AIR is each also about the same, making a total of 12-15% or thereabouts. I guess the loading would go up with higher sample rate material.

If the processor load is proportional to bit rate (which it may well not be ???) then that would imply 20-30% processor load for 24/192 files just for Dirac which is pretty high.iTunes and AIR load would increase as well.

I agree with your comment that the corrected response still looked a bit rough. Subjectively the corrected system did sound noticeably better, mostly in the bass (not surprisingly).

I've since done quite a bit of work on integrating my sub-woofer more carefully, and have added the DSPeaker Anti-Mode 2.0 which subjectively does a better job than Dirac of sorting out the sub-150 Hz part of the spectrum.

That's disappointing because I had high hopes of Dirac based on what I read on their site. It will be interesting to hear Rodney's conclusions. I just looked at the DSpeaker Anti-Mode 2.0 Dual Core spec on the web-site and it seems to have integral DACs. Are these always in the signal path?

My impression so far is that full-range computer-based room correction is probably not a "set and forget" operation, but something that one might want to continually tweak in an attempt to flatten out ever smaller kinks in the measured response.

Oh no, that's the last thing I want. I just want 'set and forget'

Ian
IMac macOS 10.15.3 (no link to Devialet Sad ) / MacBook Pro Retina OS X 10.14.4 / Linn LP12 / Devialet 200 Wilson Benesch Discovery. 
Qobuz Desktop Latest Version / Audirvana 3.2.18 / Audirvana Remote / iTunes 12.9 / AIR 3.0.4 / Wi-Fi / FW 8.1.0 / SAM 50%
Cambridge, UK (Updated 27th February, 2020)
Reply
#18
(08-Dec-2014, 10:36)PhilP Wrote: I've since done quite a bit of work on integrating my sub-woofer more carefully, and have added the DSPeaker Anti-Mode 2.0 which subjectively does a better job than Dirac of sorting out the sub-150 Hz part of the spectrum.

That's disappointing because I had high hopes of Dirac based on what I read on their site. It will be interesting to hear Rodney's conclusions. I just looked at the DSpeaker Anti-Mode 2.0 Dual Core spec on the web-site and it seems to have integral DACs. Are these always in the signal path?

Well, I would encourage you to try Dirac anyway -- it's fairly easy to get started using the trial version to see what it does in your system/room, and there's nothing to lose except a bit of time.

The Anti-Mode has analog and digital inputs and outputs, so I understand that if you use it in a digital-in, digital-out mode there is no conversion taking place. That could work nicely if you use it in line between a digital source and the Devialet, I guess. In my case I'm using it in analog-in, analog-out mode so there will be a DAC and ADC in the signal path. However since it's in line between the Devialet's sub-out and the sub-woofer input, I doubt the conversion is causing any significant artefacts within the sub's frequency range.

As I recall, the miniDSP Dirac boxes are a bit less flexible: you choose when you buy them whether you want digital in and out, analog in and out, or digital in to analog out. This slightly limited flexibility was one of the main reasons I opted for the Anti-Mode instead.

(08-Dec-2014, 10:36)PhilP Wrote: My impression so far is that full-range computer-based room correction is probably not a "set and forget" operation, but something that one might want to continually tweak in an attempt to flatten out ever smaller kinks in the measured response.

Oh no, that's the last thing I want. I just want 'set and forget'

Perhaps a better way to say this is that the nature of the process and in particular the visibility of the frequency response measurement could encourage frequent tweaking if one is of that bent. Maybe you are more disciplined than I and could resist the temptation!

I hope you don't let my conclusions put you off, but try it out and let us know how you get on.

Thinking aloud...it occurs to me that it would be very useful if the Devialets had a digital pre-out and power-amp in, like an old-fashioned tape loop. This could be the perfect way to integrate an external room correction box. Maybe their architecture makes this impossible?
Roon (Mac Mini), Wilson Benesch Full Circle, Expert 1000 Pro CI, Kaiser Chiara
Warwickshire, UK
Reply
#19
To test the effects of Dirac in the configuration I want I will have to move all the furniture, filing cabinets etc in my room to rotate the hi-fi through 90 degrees so this will probably have to happen some time between Xmas and New Year.

I will also need to buy a decent microphone. I guess it makes sense to use the MiniDSP UMIK-1 that they recommend... Interestingly, there's a place near me where I can hire an Earthworks mic by the day which is tempting.

Another challenge for me is that I really only use my Mac for streaming Qobuz (+some Qobuz Hi-res downloads) - say 50% of my listening. It would be so much easier if Devialet added room correction within their amps Wink
IMac macOS 10.15.3 (no link to Devialet Sad ) / MacBook Pro Retina OS X 10.14.4 / Linn LP12 / Devialet 200 Wilson Benesch Discovery. 
Qobuz Desktop Latest Version / Audirvana 3.2.18 / Audirvana Remote / iTunes 12.9 / AIR 3.0.4 / Wi-Fi / FW 8.1.0 / SAM 50%
Cambridge, UK (Updated 27th February, 2020)
Reply
#20
(08-Dec-2014, 13:48)thumb5 Wrote: The Anti-Mode has analog and digital inputs and outputs, so I understand that if you use it in a digital-in, digital-out mode there is no conversion taking place. That could work nicely if you use it in line between a digital source and the Devialet, I guess.

I have an anti-mode and did try it like this. In fact, even digital to digital it sample rate converts everything passing through it to 48kHz.
Personally I did not find the loss of overall quality of the antimode to be worth the improvement in the bass, for me.
FWIW
Devialet Original d'Atelier 44 Core, Job Pre/225, Goldmund PH2, Goldmund Reference/T3f /Ortofon A90, Goldmund Mimesis 36+ & Chord Blu, iMac/Air, Lynx Theta, Tune Audio Anima, Goldmund Epilog 1&2, REL Studio. Dialog, Silver Phantoms, Branch stands, copper cables (mainly).
Oxfordshire

Reply


Forum Jump:


Users browsing this thread: 2 Guest(s)