Dirac Live on a Mac - DAP doesn't save preferred output - any workarounds?

[flak]

Thanks Flávio, I didn't know you had changed the OS specific licensing protocol. The fact that I had already bought and paid for a Mac OS licence is the single biggest reason I have not moved to a windoze platform!

 

Can I now use my existing licence number to activate a newly downloaded Windows version of Dirac?

 

Yes, we unilaterally changed the license terms

 

You can use the Windows version that you can download here: https://portal.dirac.se

(you'll have to uninstall the Mac one first 'cause the license still is for one machine if you have the Stereo version)

 

Flavio

[hifi_swlon]

Flavio, since this thread is a bit surplus to requirements now (until the new version is released at least ), could I ask a couple of Dirac questions please?

 

I've been using it pretty much solidly for a week (yeah, I know), first with a lesser set of speakers and now with my mains. The results have been really great with both, albeit my mains are slightly more 'finicky' to dial in. Perhaps that's because they're 'better' or I'm just more attune to their sound...

 

First question: how much difference does the spread of the 9 positions make to the final result? In my first set of recordings with the other speakers, I stayed just about the size of the small sofa (actually two chairs next to each other but you get the gist). When I re-did the measurements on my mains, I went further afield left right fore and aft - not crazy but like maybe 1.5m from the 'hero' spot, and also more extreme higher and lower than listening position, maybe >50cm. How much of an imacpt would this have?, and how do you know when you've got it right? I'm not particularly thinking its less effective than before - and as its different speakers and I did a few things wrong the first time, I'd expect them to be different - but its left a bit of uncertainty. I actually listen a bit more casually from various points in the room, although I do have a favourite, and also like to listen when I'm in the (open plan) dining room next door or walking around, so figured a bit wider was better?

 

The next q is Devialet SAM specific - hopefully you're familiar.

 

I made two sets of measurements, with SAM on and off. For the SAM on, I had to lower it to 20% as I felt the limiter was kicking in and reducing the bass output. But all seemed good with the process, and to be honest I can't hear a huge difference with filters made with and without SAM even when using SAM on the amplifier, and usually at higher percentages.

 

But SAM is supposed to adjust frequency and time domain for the speaker output, and so is Dirac. So I'm just a bit confused as to how the two will inexact - perhaps I'm over adjusting as each is unaware of the other in the system (well, apart from the SAM 20% measurement set). Perhaps it's just too confusing to worry about….

 

A couple of suggestions for future versions…. I have to say even as a relative newbie, what would be nice here would be the ability to just make lots of recordings at various positions, and save them all, so that come filter time you had options. i.e. measure a lot of points, and choose later, rather than it having to be a fixed set of 9. I suppose it would be harder to catalogue them, but its tricky in the home environment to get time to get the mic setup again, so this would make things easier if you could just measure a bit more up front.

 

One other thing that would be really nice, is the ability to overlay more than one target curve, even if they weren't active, just so you could compare. And, for anyone with computer graphics backgrounds, have access to curve tangents so you can make them smooth. But I see these are advanced options and you've made Dirac easy to use, which has been great for me as I may not have done it otherwise.

 

I won't write much here about how happy I am with Dirac so far, even early days - as I've already made a few slightly evangelical posts over at Devialetchat (that's where I heard about it, and I have a Devialet). I will say its totally changed my outlook on DSP -it's the only thing in the past two years that I've tried thats got my system sounding the way I like it.

 

All I can say is I wish I would have discovered it earlier - it would have saved me a lot of frustration, time, and money. I can't believe I was starting to obsess over things like USB cables and speaker cables - but it was from pure frustration and lack of ideas about what I could do. In Dirac I found an actual solution! If there are negatives I dont care, and I certainly don't think so at this stage - other than some slight restrictions on source setups. Oh, and the thread title issue.

 

Now, I'll just wait longingly for the new OSX version with the fix.

[flak]

Hi hifi_swlon,

 

I read that you have tried different listening areas (actually volumes) and you are wondering which one is right... that depends on your listening habits so you should choose the one that best fits your requirements, a larger area will imply a less aggressive correction.

If you want the best in both situations you can create two (or more) different filters for smaller/larger listening areas that you can istantly switch in the DAP (also you may alternatively want to sometimes listen from a listening point that is asymmetrical)

One world of caution is to anyhow not measure at a distance of less than 40/50 cm. from the MLP in order to avoid possible overcorrection.

 

With reference to Devialet's SAM unfortunately I have no hands on experience with it... my understanding is that it applies a real time correction that is meant to be a speaker correction based on previous manufacturer's measurements of the specific speaker.

Dirac Live is a real time correction that is meant to be a room correction instead... even if speaker correction is somehow built into the process.

As a result I expect that if Dirac Live finds a speaker that has already been corrected by SAM it will take those improvements into consideration and will need to apply less correction, on the other hand if SAM is disabled I expect that at least some of those corrections (phase and frequency response) will be applied by Dirac Live itself.

 

In a few words on paper SAM and Dirac Live together may produce excellent results but I have no factual data and hands on experience... on this subject you are a better source of info than I am

 

Flavio

[hifi_swlon]

Hi hifi_swlon,

 

I read that you have tried different listening areas (actually volumes) and you are wondering which one is right...

 

Yes, thanks Flavio. In my initial test I'd done a smaller volume, but with my other speakers did a bigger volume since I like to move around as well. Like you say, I think I need to do two sets, one smaller volume ad one larger volume, and see for myself the difference. I'm definitely not one to sit in 'exactly' the same place and prefer it to sound good over a wider area.

 

Is auditorium advised for a lounge - would that be better?

 

Hi hifi_swlon,

 

With reference to Devialet's SAM unfortunately I have no hands on experience with it... my understanding is that it applies a real time correction that is meant to be a speaker correction based on previous manufacturer's measurements of the specific speaker.

Dirac Live is a real time correction that is meant to be a room correction instead... even if speaker correction is somehow built into the process.

As a result I expect that if Dirac Live finds a speaker that has already been corrected by SAM it will take those improvements into consideration and will need to apply less correction, on the other hand if SAM is disabled I expect that at least some of those corrections (phase and frequency response) will be applied by Dirac Live itself.

 

In a few words on paper SAM and Dirac Live together may produce excellent results but I have no factual data and hands on experience... on this subject you are a better source of info than I am

 

 

Yes, thats right, similar ideas - amplitude and time domain adjustments, at the lower frequencies only, and based on their measurements of a single speaker (or pair), I guess they assume reasonable tolerances.

 

The difficulty is SAM also has a limiter built in to protect the drivers, and so its not possible to run the measurements at SAM 100%. Due to the fairly high measurement volumes (loudness I mean), and the pure bass signal sweep, the highest I could get was 20% before it sounded like the bass was being scaled back by the limiter. So Diracs filters are based on the SAM 20%, but in reality I use SAM higher (50-100%), so its possible that things are being compensated twice if using SAM above 20%. Of course, I dont know the full technical details of SAM so I suppose its impossible to know - will just go with what sounds best!

 

In theory its better to use SAM as its calculations aren't room based, so should be 'truer', and I think the two of them together are great, so far….

 

 

I had a really good listen last night, I think Im getting to curves I really like as my tweaks are smaller and smaller, but preference does depend on music so I really want to find a 'set and forget' curve as I run headless - most likely will be more back towards your default curve to be honest, although I do like a bit of extra bass but that can sound slightly 'boxy' on some material.

 

My trial is about to end, but I'll back back for the full version as soon as the OSX fix is out. Thanks for the chance to discover Dirac - the trial's a great idea - its quite an investment in learning and mic etc, but its opened up a new world to me!

[hifi_swlon]

Have just come come back from holidays and am really missing Dirac, but just can't justify buying a license without knowing this will definitely be fixed.....

 

I know we're not into Q1 2016 yet, but perhaps we're into beta??

[hifi_swlon]

OK, well, we are 'officially' in Q1 now

[wgscott]

Thanks wgscott. I did take a look at that, but it seems like a bit too much of a hack for me - I don't really like relying on such things in my audio path.

 

To my mind Dirac should address the issue with their DAP application on OSX. There's no point having a virtual sound card that can automatically start on boot up/login, but then doesn't store the default output device (and default filter). A big oversight in my opinion, and a real shame as I like everything else about it, but as implemented its unusable beyond testing.

 

 

The beauty of the AU plugin is that it isn't something that starts at boot time. Turn your DAC on, fire up (for example) Audirvana Plus, and press play, and then it loads. The problem then simply doesn't arise. The DAP always seemed clunky and ad hoc to me. The AU plugin, by contrast, works seamlessly for me, every time.

[hifi_swlon]

Couldn't agree more - the spanner in the works is no AU/VST in Roon, which is the only software I use nowadays….. ;(

 

Which lead me to post this!

 

http://www.computeraudiophile.com/f23-dsp-room-correction-and-multi-channel-audio/dirac-live-and-roon-full-integration-27179/#post499439

 

i.e. a desire for Dirac to be integrated in Roon.

[wgscott]

But then you wouldn't be able to use it in other players.

 

It would be better if Roon allowed AU plugins.

[hifi_swlon]

Well, having spent years looking for the right solution, I personally don't see myself using any players other than Roon for the foreseeable future, but I get the point….

 

Roon developers have briefly explained why AU plugins won't work generically in their architecture. So I get the impression its less down to them simply not allowing it, and more down to technical limitations with AU on their 'Core' (server) and 'endpoint' type model.

 

I am a very long way from being an AU/VST expert…..

[wgscott]

You don't need to know anything about them (I don't).

 

I just updated my blog post with step by step instructions:

 

Use AU plugins for free with any Apple audio software - Blogs - Computer Audiophile

[hifi_swlon]

I bought a license anyway - so I'm hoping Dirac come through with this fix ASAP.

 

I have 30 days I guess so fingers crossed…..

[hifi_swlon]

I re-measured my Dirac after a Devialet firmware change. Used similar curves to before and made some new filters. All good, sounds very similar to previously, and the same delay was suggested.

 

After a bit of listening, last night's revelation was moving the listening position forward about 20-30cm from where I generally sit and where the chair was when I did the first 'hero' measurement.

 

Wow, that's a really convincing holographic sound! Probably the best sound I've had to date in this house (3 years). A big leap in fact.

 

Ideally I'll work out how to shift it back a touch as the chairs a bit too close a table in the lounge but will just enjoy it for a while!

 

Going to bed happy tonight, as can live with this arrangement just to enjoy music, but I would like to understand how/why?

 

I didn't do the 9 positions very scientifically - maybe the fore/aft distances weren't even and this affects the outcome? Maybe that doesn't matter and Dirac can't override the speakers natural sweet spot? Maybe it was like this last time but I just never found the sweet spot?

 

Regardless, I'd appreciate tips on the best method for shifting this sweet spot back slightly. Should I try and do it without Dirac first or adjust and re-measure each time? The latter would be a bit laborious.

 

I'm sitting pretty close to the speakers (about 220cm from ear to left speaker) which are 250cm apart.

I'm very slightly offset but Diracs dealing nicely with that.

Maybe I'm too close now anyway? But it feels perfect, in fact I can even move forwards slightly and it's v slightly better.

 

Speakers right now only have a fractional toe in, but I'm assuming I'd need to reduce that to almost zero to move the image back?

[wgscott]

I'm simply guessing, but I think this has to do with the degree of toe-in. If the image is best when you move forward, it suggests that decreasing the toe-in would help. I would find the best position before applying the Dirac filter, then calibrate and apply. If the sweet spot is still too far forward, move the chair back an equivalent distance and then optimize and create a new filter, but hopefully that won't happen...

[hifi_swlon]

I'm simply guessing, but I think this has to do with the degree of toe-in. If the image is best when you move forward, it suggests that decreasing the toe-in would help. I would find the best position before applying the Dirac filter, then calibrate and apply. If the sweet spot is still too far forward, move the chair back an equivalent distance and then optimize and create a new filter, but hopefully that won't happen...

 

Thanks. I kind of thought that, but just wondered if Dirac could somehow 'move' the sweet spot forward/backwards depending on where my 9 positions were. I'm sure it was a little further back, but maybe not.

 

I'll have a play - I'm assuming there's zero tolerance in Dirac - i.e. changing the toe in angle by even a degree will probably need a re-measure. That doesn't go down too well in my house, my wife hates the process (she likes the final sound though).

 

I'm really hoping the fix for the USB output not getting lost (as per thread title) comes very soon - having to go via VNC every time I want plays music is becoming a bore. In fact, if its not fixed by the time my 30 days nears an end, I'll probably get a refund and find another solution. A real shame as I really like Dirac, and don't want to have to do anything in Windows or get extra hardware, or add fudges to my setup, but its just too annoying to live with - let alone inflict on other users in the house...

[wgscott]

The USB issue is most likely Apple's fault, so Dirac can't fix it.

 

The DAP preferences problem is likely a permissions problem. Try setting it up in another user account, and if it works, that tells you it has to be a permissions problem or some other issue specific to your user directory.

 

I wouldn't want to reboot into windows to run it either, FWIW.

[hifi_swlon]

The USB issue is most likely Apple's fault, so Dirac can't fix it.

 

 

I think Flavio already acknowledged it, and said that Dirac would release a fix in Q1. It certainly doesn't happen with Roon or any other software player I've used via USB...

 

Here's hoping...

[flak]

I think Flavio already acknowledged it, and said that Dirac would release a fix in Q1. It certainly doesn't happen with Roon or any other software player I've used via USB...

 

Here's hoping...

 

Yes we aknowledged it, the updated DAP reconnects automatically when a USB device is attached (it was designed to "wait" for the selection of an output device if none was connected) and sleep/hibernate is handled properly... I imagine that the next question would be "when is it going to be available?"

Within Q1 is the scheduled release date, if it has not been released yet within the expiry date of your 30 days money back guarantee you may get the refund and purchase it again later on if you want.

 

Flavio

[Peter Myers]

Hi Flávio...thanks for the information.

but the next question isn't "when is it going to be released?"

The next question is..."if it's already fixed why is it not already released, maybe as a patch ?"

 

Thanks

[flak]

Hi Flávio...thanks for the information.

but the next question isn't "when is it going to be released?"

The next question is..."if it's already fixed why is it not already released, maybe as a patch ?"

 

Thanks

 

Hi Peter,

 

because we are working on other aspects and issues also, the update will include the DLCT in addition to the DAP and they have to be beta tested... in a few words they are not ready yet.

 

Flavio

[hifi_swlon]

Yes we aknowledged it, the updated DAP reconnects automatically when a USB device is attached (it was designed to "wait" for the selection of an output device if none was connected) and sleep/hibernate is handled properly... I imagine that the next question would be "when is it going to be available?"

Within Q1 is the scheduled release date, if it has not been released yet within the expiry date of your 30 days money back guarantee you may get the refund and purchase it again later on if you want.

 

Flavio

 

Thanks for the follow-up Flavio.

 

I have to say I'm with Peter on the software updates front - its much nicer (for any product) to get more frequent updates for things like small issues/bug fixes, than to wait longer for 'hero' releases encompassing all new features…. especially since you've fixed them

 

I'd hate to have to try and find an alternate, but if we're looking at late Q1 then I'd probably have to do that, or just not use RC for a while, which would be a real shame…. it seems like a small detail, but with a headless server and other people using the stereo its a real problem.

 

What's DLCT?

[flak]

Hi hifi_swlon and Peter,

 

I expect to have available for you a working beta of the new 1.2 Dirac Live update anytime now... and if beta testing will be successful as we hope the official release should come very soon after.

 

(DLCT is Dirac Live Calibration Tool, the application that is used for measurements and filter creation)

 

I'll keep you posted

Flavio

[flak]

Well... "anytime now" has become "now" for the Windows version, the Mac version has not been packed yet (it will be soon)

If interested you can get the DLCT and DAP downloads of the 1.2.0 release candidates at helpdesk.dirac.se

 

Flavio

[hifi_swlon]

Hi hifi_swlon and Peter,

 

I expect to have available for you a working beta of the new 1.2 Dirac Live update anytime now... and if beta testing will be successful as we hope the official release should come very soon after.

 

(DLCT is Dirac Live Calibration Tool, the application that is used for measurements and filter creation)

 

I'll keep you posted

Flavio

 

Now that is exciting!!!! Thanks Flavio - I sense you helped push this along....

[flak]

I'm pleased to inform that release candidates of Dirac Live 1.2 for both Mac and Windows are now available for testing from helpdesk.dirac.se

 

Flavio