Jump to content
IGNORED

Acourate+JRiver or Amarra Symphony IRC on Mac?


Recommended Posts

I don't know what to choose... I am actually using "normal" Amarra 2.5.1 and really like the sound of it, but I would like to experience with Room Correction in my set. What is drawing me in the direction of Acourate+JRiver are the following points: 1) From what I've read, Acourate seems to be more "transparent" and better sounding than Dirac (not integrated in Amarra),, 2) It seems that JRiver works more reliable than Amarra (I've experienced quite a few "problems" with Amarra with the last versions), 3) Without having tried, I think JRiver works nicer as a Music Library compared to ITunes (don't really like ITunes...) and it seems that JRemote is a very nice remote app. What is holding me back to go the Acourate+JRiver route, is the fact that Ammara without IRC seems to be better sounding than JRiver for Mac. I can wait until there have been loads of reviews, but by that time the Amarra offer to upgrade to Symphony IRC will be finished... Perhaps I have overlooked some important facts which can influence my choice?

Link to post
Share on other sites

Get the free two week trial of Dirac and see if you like how it sounds with Amarra. I've only looked at a screenshot of Dirac+Amarra, but I think it is essentially the same thing. You could also try it with Jriver on the Mac.

 

I can't get Audrivana Plus to work with Dirac in iTunes integrated mode, but I can get Dirac to work with BitPerfect, and that combination sounds quite good.

 

Does Acourate allow you to create AU (or similar) plug-ins that can be used with any software? That is the big shortcoming of Dirac in my opinion, not the sound quality.

Link to post
Share on other sites

I have not seen any test comparing Dirac, Acourate, Audiolense, DRC-sourceforce, and others.

 

You can find many positive reviews of either software, but very few people seems to have tried more than one.

 

They all seem to do frequency correction with a custom target. This is a major improvement over non-corrected sound, so that will explain the many happy users.

 

Dirac is a commercial success correcting audio in several expensive car models. It offers a free trial and seems to have the least difficult user interface of the four I mentioned earlier. But it also seems to be the least advanced not offering XO, time domain, phase correction and pre-echo correction.

 

Audiolense and Acourate have all those features (not sure about Acourate and pre-echo). Neither is any huge commercial success - I don't think the authors can do without a job on the side, but that is just a guess. I believe Audiolense has a money back gurantee if not satisfied. Both have excellent support directly from the author. Audiolense is probably a little easier to get started with, but if you got a detailed guide and someone to ask for help it doesn't matter much.

 

DRC-sourceforge is all free and pre-dates both Acourate and Audiolense. But it has no user interface whatsoever, so you will have to use the command line and edit your target in a text editor: xy-coordinates. This is not really a problem if you know what you are doing.

 

But that is your problem at this point: You want to get started and don't know much about it. There is a learning curve no matter which solution you choose.

Link to post
Share on other sites

Maybe I should go for DRC-sourceforge. I compiled it on OS X without any issue, and prefer the unix command-line to most GUI interfaces, so it probably makes the most sense to invest my efforts in that.

 

Today is the last day for the Dirac discount, and I probably would have bought it if it enabled me to create AU plugins. But it doesn't, and for $420 I would want it to do exactly what I want. I also someday want to bi-amp my speakers properly...

Link to post
Share on other sites

Bought Amarra Symphony + IRC update today because of the reduced price option. I have really liked the SQ of Amarra for the last 2 years of using it, so I hope the incorporation of Dirac will have stunning SQ also. Need to wait a little for the Microphone though....

Link to post
Share on other sites
Bought Amarra Symphony + IRC update today because of the reduced price option. I have really liked the SQ of Amarra for the last 2 years of using it, so I hope the incorporation of Dirac will have stunning SQ also. Need to wait a little for the Microphone though....

 

Hi Mark, I'm in the same boat as you regarding the Microphone, 2 weeks wait for me (I'm in the USA) on the XTZ. I upgraded to Symphony yesterday .. gorgeous SQ. Hope to read your impressions after your mic arrives and you've done the measurements.

Link to post
Share on other sites
Hi Mark, I'm in the same boat as you regarding the Microphone, 2 weeks wait for me (I'm in the USA) on the XTZ. I upgraded to Symphony yesterday .. gorgeous SQ. Hope to read your impressions after your mic arrives and you've done the measurements.

 

For me it doesn't take that long to get the microphone (I hope...).

 

I am with you about the SQ of Symphony 2.6 without using the IRC option. Sounds really good. I always have used the Parametric Equalizer in the previous versions of Amarra (not the Symphony version but "normal" Amarra). This doesn't work in Symphony... After scrolling to another song the EQ settings are gone... The "special" features of Symphony also shows another sort of equalizer (with 4 bands instead of 3 bands I had before), that one works just fine.

Link to post
Share on other sites

 

Dirac is a commercial success correcting audio in several expensive car models. It offers a free trial and seems to have the least difficult user interface of the four I mentioned earlier. But it also seems to be the least advanced not offering XO, time domain, phase correction and pre-echo correction.

 

 

Well, please note first that I'm partial to Dirac... but I have to say that the reason why Dirac is a commercial success is the fact that Dirac Live works most effectively in the time domain, and consequently handles phase-correction and pre-echo correction as well as frequency response correction.

Most probably you did not have the time to test it yet, if you'll do it you will see the pulse response correction (phase).

 

0000228.jpeg

 

Dirac also successfully corrects phase in a wide listening area, which makes an important difference from the other similar applications, but you do'nt have to trust what I'm saying... you can measure and test it yourself by downloading the two weeks free demo here: Dirac RCS | Dirac Research

 

Also I'm glad to read Bibo01's posts in this Forum, he knows Denis's DRC well and his comments are interesting :-)

 

Flavio

Warning: My posts may be biased even if in good faith, I work for Dirac Research :-)

Link to post
Share on other sites
For a DRC interface here it is from the same Italian developers:

 

Digital Room Correction

 

Just Google translate the text. The software is in English...well, not all.

 

Enjoy!

 

Ciao Bibo :-)

I'm new to the forum, how can I make it transparent that I work for Dirac?

Are there "industry" qualifications for a nick?

 

Flavio

Warning: My posts may be biased even if in good faith, I work for Dirac Research :-)

Link to post
Share on other sites
Ciao Bibo :-)

I'm new to the forum, how can I make it transparent that I work for Dirac?

Are there "industry" qualifications for a nick?

 

Flavio

Hi, :-)

 

I see quite a few industry related people here. You just add it into your signature to qualify your position.

RE those comments above by sshd were rather inaccurate, I just did not feel like posting about...

Link to post
Share on other sites
Ciao Bibo :-)

I'm new to the forum, how can I make it transparent that I work for Dirac?

Are there "industry" qualifications for a nick?

 

Flavio

 

Hello Flavio,

 

Welcome to CA.

 

Search for Barry Diament and observe how he handles his affiliations.

Best,

Richard

Link to post
Share on other sites

Did the first steps in using Amarra with IRC! Measured the room and made up 3 different filters, where one of them was close to the suggested target curve. Did small modification to this curve at high frequency (around 10 K) so it does follow better the uncorrected frequency spectrum of my speakers. Without this modification done by me, it tried to increase the level in this region quite a bit and I don't like the idea of that, as my system was already sounding really well. This filter prooved to be the best sounding of the 3, however I've made up some plans for a new target curve already. ;) It does sound really well this way, but it's not a massive difference. I need to say that I spend a lot of time in the past to setup my loudspeaker position in the room and the position of the couch to achieve a decent sound. I still need to play with the target curve, but the low frequency region is a lot more "even" as before, but it's still a bit "loud" in total so that needs to be corrected. With big orcherstral classic music, the difference is most obvious. There is a more "see through" sound stage with more depth. Did some listening to choir music, it's more easy to hear that a choir is made of quite a few men and women singing together, rather than one blurred singing sound. I am happy with most things I heard this evening and I am sure that the few remarks I had, can be solved by changing the target curve a bit and calculate some new filters with that.

Link to post
Share on other sites
RE those comments above by sshd were rather inaccurate, I just did not feel like posting about...

 

The world would be a much better place if people with real knowledge would crawl out from the bush they are hiding under and correct any misunderstanding and inaccuracies.

 

I admit I have not tried Dirac. In fact I wrote "it seems" in my post.

 

It is very good of flak to come and tell us about Dirac. Most of us in here are pretty clueless to DSP/DRC. A few of us have successfully tried one or two software solutions, which in present company makes us the closest thing to experts.

 

flak: Please shed some light on these two questions:

 

1. On dirac.se I read the following: "At Dirac we use a proprietary filter structure. In essence, maximum performance at much less processor usage than regular FIR filters. This filter structure also has great numerical properties, not causing any digital noise."

My question is: What kind of digital noise is regular FIR filters introducing and what does it sound like? Since I am using FIR I should be able to hear this noise?

 

2. Does Dirac offer a XO (crossover) solution? I see this nowhere on the site. If the answer is yes, then I am very eager to try Dirac. If the answer is no, then you will agree that Dirac is the least advanced solution? For example: How can do control the time domain if you do not control the drivers indepentenly? Say the accoustic centrums the drivers does not align. You need to time-align the drivers. Easy if you control them independently...

Link to post
Share on other sites
The world would be a much better place if people with real knowledge would crawl out from the bush they are hiding under and correct any misunderstanding and inaccuracies.

 

I admit I have not tried Dirac. In fact I wrote "it seems" in my post.

 

...

Sincerely it's not a question of hiding, but rather giving priorities and having time at one's disposal.

Link to post
Share on other sites

Hi SSHD,

I'm waiting for some info about your first question, and today is sunday, so you'll have to forgive me if I answer your second question first:

 

How can do control the time domain if you do not control the drivers indepentenly? Say the accoustic centrums the drivers does not align. You need to time-align the drivers. Easy if you control them independently...

 

Let me say first that I agree with you that active crossovering is a really good solution, I'm using it myself as I'm using Dirac Live and Linkwitz-Riley 48dB/octave crossovering in multiamping a pair of transmission line subwoofers and a pair of electrostats.

 

In your post you are correctly saying that you need to time align the drivers by aligning the acoustic centrums of the drivers.

As you know this was "new science" a long time ago, the problem has been solved in the 70's by the Dalquist DQ10s that addressed this problem by aligning the acoustic centers of the drivers (you can see a pair of open ones here)

 

homepage.jpg

 

Of course vintage speakers that are even older than DQ10s do not address this problem, the following is a measurement of the impulse response of a Klipsch speaker that clearly shows the results of very different path-lengths between the drivers:

 

CHR1A-TIM.gif

 

By the way this also relates to the discussion about impulse response evaluation that I've posted here:

http://www.computeraudiophile.com/f23-dsp-room-correction-and-multi-channel-audio/listening-rooms-and-pulse-response-17633/

 

Dirac Live perfectly time aligns a speaker system as a whole as you can see here, where applied time delays are shown:

 

0000230.png

 

On the other side time-aligning the individual drivers by aligning the acoustic centers is the responsibility of the speaker manufacturers and designers, they are perfectly capable of doing it and they actually do it even in low cost speakers of reasonable quality.

So we can normally forget about crossovering and time-alignement problems of drivers...

if we have to, it means that we have the knowledge and tools to select the best analog or digital, hardware or software, crossover solution that best fits our requirements.

 

But here we are talking about Digital Room Correction and in my opinion we should try to choose the best Digital Room Correction solution for its merits as a digital room correction.

In other words a crossover is a crossover, and a DRC (digital room correction) is a DRC...

I would not blame a crossover for not applying a room correction as well as I do not blame a DRC product for not crossovering.

 

I will not get in this post into the reasons why Dirac Live, as a standalone application, and Amarra with iRC, are unique and could even be defined as "new science"... this a long post already :-)

 

Thanks for your time,

Flavio

Warning: My posts may be biased even if in good faith, I work for Dirac Research :-)

Link to post
Share on other sites

Great info so far, thanks to all on this thread! Here's a question or three for Flavio or others on Dirac or other solutions...

 

My questions are a bit more basic, I have Amarra Symphony 2.6 (w/o iRC yet), but I listen to a wide variety of Mac-sourced content (movies (high def or Netflix, iTunes, Amazon, etc.), iRadio, Spotify, various websites in general, apps, all sorts of content) and I want ALL of this audio content to be room corrected. My understanding is that the Amarra iRC option ONLY works for content that can play in Amarra directly (basically music I have in iTunes). Is this true? And is it true that the "Dirac Live Room Correction Suite - Stereo Version" can still work with Amarra as a stand alone app (and also with any other audio source)?

 

Also, specifically with the Dirac Live Room Correction Suite stereo version, can it store multiple correction filters for use in different locations? My same MacBook Air is used at different times to source a full living room system and at other times a desktop system with KEF X300A speakers. Each would of course require different correction settings.

 

Lastly, I'm looking for more detail on the feature set of Dirac solutions - I'm a bit unclear what the "Auditorium" mode is and what I give up by accessing only "Average Measurements" instead of "All Measurements" and whether or not I can tweak the correction to taste (i.e. additional EQ either in Dirac and/or Amarra) or not.

 

Thanks for your insights,

Bob

Link to post
Share on other sites

Flavio: Thank you for confirming that Dirac does not implement a crossover. For me this is a missing key feature and I cannot test/use Dirac.

 

 

I would actually have stopped posting if it wasn't for this:

So we can normally forget about crossovering and time-alignement problems of drivers...

 

- Sure there are a few speakers produced in the past four decades that addresses time-alignment of drivers. The vast majority however does not.

 

- Additionally the vast majority uses passive crossovers. Passive crossovers causes massive distortion sometimes much worse than the accoustics of the listening room.

 

- Electronic crossivers/analogue signal processors are an alternative. Much better but quite expensive.

 

- Another alternative is digital crossovers, mostly used in pro-audio. Some models does only takes analogue input, so you have an additional AD/DA conversion. Others are expensive as hell and the cheap ones have extremely bad reviews.

 

Combining DRC with XO is the next logical thing!

- It is FREE - all you need is a multi channel audio interface.

- The drivers can be corrected individually and we can make all frequencies in the crossover point arrive at the listeners ear at the same time.

- We can easily change the crossover points and steepness. It can be as steep as we like.

 

But most important: We can build speakers that look great without caring much about time alignment and other things that are trivial to correct in the crossover. Just like we can use DRC to have great sound in a listening room that isn't perfect.

Link to post
Share on other sites

Thanks Bob :)

 

Dirac Live acts as a "virtual soundcard" so all sound from the computer regardless of whether is a media player (including Amarra obviously), YouTube, video game, Skype conversation, etc., is optimized... as far as Amarra with iRC is concerned I have to ask because I'm a Win user even I think that your description is correct.

 

You can store the correction filters for different locations or systems as you suggest, as well as the correction filters for different listening areas, musical genres or listening volumes (i.e. you may want to have a different target curve for low volume listening to compensate for what was once called the Fletcher-Munson effect)

You can save as many as you want and load them as necessary, but only four of them can be concurrently loaded so that you can instantly switch from one to another using the DAP panel (the four buttons on the left):

 

0000230.png

 

The Auditorium mode is targeted at large listening areas as the name implies, we then have the Sofa and the Chair options.

My favourite one, and the one that I suggest, is the Sofa option which I like both for listening quality and flexibility in listening area setting.

 

When we talk about "average measurements" instead of "all measurements" you should take into account that Amarra with iRC and Dirac Live use nine measurements... "average measurements" means that you will see the most important measurement which is the average of the nine measurements in the nine positions (both impulse and frequency response) while "all measurements", as you expect, means that you can see the individual nine measurements.

 

Lastly you can certainly tweak the correction to taste (as well as you can later istantaneously compare two or more additional EQs) and this is especially easy 'cause you can modify the target curve on screen by simply pulling the orange control points in the target curve that you see here (you can add as many control points as you want)

 

0000227.jpeg

 

If you have a measurement microphone and you will download the demo I'm sure that you will enjoy it and have fun :-)

Flavio

Warning: My posts may be biased even if in good faith, I work for Dirac Research :-)

Link to post
Share on other sites

Hi SSHD,

 

as you correctly noted there are different routes to reach the general architecture that you are pursuing...

a multiamplified solution with room correction.

 

Let me outline first that while I personally like it, it is not a common one as it involves higher costs because of the requirement of as many amplifiers as the drivers that you want to individually drive and a high quality multichannel audio card that is not cheap.

 

Physically aligning drivers is no big deal, and many out of the box speakers systems take care of it... also when you say:

-The drivers can be corrected individually and we can make all frequencies in the crossover point arrive at the listeners ear at the same time.

- We can easily change the crossover points and steepness. It can be as steep as we like.

you are talking about aspects that require a good amount of knowledge i.e. type of crossover, xover frequency and slope, dispersion and sum response so that delay, phase and magnitude response add optimally in addition to other considerations about the characteristics of the drivers themselves, distorsion, dampening and so on and on...

 

As a result most persons prefer to leave these design choices to speaker designers, and I do'nt think they are wrong.

 

Let me stop here 'cause this is not a thread about crossovering and I have not answered your other point where you outline that DRC is free (I think that you refer to Denis Sbragion's DRC)

I've personally suggested it to persons who are not afraid of the intricacies of using it, have the necessary knowledge and are not willing to spend money for a software application.

 

Now let me also explain why it may be worth considering an alternative to your DRC and crossover solution even if they are free...

DRC has been designed for a single fixed listening point and it uses one measurement (it's not the only one, Acourate for example uses one measurement too)... both are excellent applications.

 

But these applications use one measurement point only... as you will see from the picture below using more measurements at various points in the listening space, even small changes in location create large differences in the impulse response and magnitude response.

 

ScreenShotA.png

 

ScreenShotB.png

 

Basing a phase correction on one point in space only is a very risky strategy as it may give audible pre-ringings in other positions.

In our opinion the only way to adequately deal with phase correction and magnitude response correction is to use several measurement positions and have an algorithm that does not just invert those responses but actually evaluates and optimizes the sound in all positions, just as Dirac Live does. As far as I know Dirac Live is the only Mac and PC application with true phase correction in a listening space of arbitrary size.

 

Dirac Live RCS is very user friendly and can be used by anyone. You are guided step by step through the process. Those applications are a set of complicated tools with a large number of parameters. Dirac Live RCS does not require the user to be an expert in the algorithm. The user should not have to worry about algorithmic details. At the same time, we allow an advanced user to change the target response suggested by the software, to suite his or her taste.

 

Compatibility... Dirac Audio Processor is a virtual soundcard, that is all sound from the computer regardless of whether is a media player, YouTube, video game, Skype conversation, etc., is optimized. Other room correction solutions use VST plugins which only work with certain media players. Also both Mac and Windows are supported by Dirac RCS.

 

The algorithms behind Dirac Live are patented and based on the cutting edge of acoustic and signal processing reasearch. It's a matter of striking the best balance between correcting all that can be corrected, but not more than that. That's why many measurement positions are used, to be able to find out what is position-dependent behavior that can't be corrected without causing problems elsewhere, and what is consistent behavior that can be corrected robustly.

 

That's it :)

Flavio

Warning: My posts may be biased even if in good faith, I work for Dirac Research :-)

Link to post
Share on other sites

Thanks Flavio for the explanations!

 

I wondered about the compatibility issues because I have read some other folks in other forums comment about how Dirac Live didn't work with all modes of various apps (I think it was Audiolense and Acourate and possibly Audirvana in certain "direct" modes. Don't know the details, but they couldn't use their apps with Dirac Live because of this. Maybe this is related to bugs in the various apps?

 

On another issue, I'm wondering about the latency of the Dirac Live app (not the iRC included in Amarra which I've concluded is not the best option for me, but the separate Dirac Live Stereo app) for use with video. Do you have rough info on the magnitude of the latency and if can be adjusted in any way by changing processing resolution or some other adjustment? Obviously this is going to depend on the computer, I have a MacBook Air (most recent version: 1.7GHz Intel Core i7 processor with 8GB 1600MHz DDR3 RAM).

 

Lastly, my understanding is that the current Dirac Live operates up to 24bits/96KHz. Is there going to be a 192K version anytime soon?

 

Thanks again for all your help and insights! I do sense Dirac Live will be quite a good match for my needs!

 

-Bob

Link to post
Share on other sites

Hi Bob :)

 

as far as your first question is concerned I would not worry about compatibility between Acourate or Audiolense with Amarra's iRC or Dirac Live... they are all DRCs so I imagine that one will use one or another.

Your second question about latency and video is an interesting one (and of general interest too)

While latency is uninmportant in audio only reproduction it is of interest when audio and video have to be in sync.

Also you have correctly outlined that processor speed should be taken into account.

 

You will be pleased to know that you can measure the actual latency with your own computer by looking at the impulse responses shown by Amarra iRC and Dirac Live before and after correction.

If you look at the following image you see that there is a delay of appx. 10 milleseconds from time zero to the first spike, the blue one before correction.

This is caused by the distance between the speaker and the measurement microphone and you can deduct that the microphone had been placed at a 3,4 meters appx. distance (the speed of sound is 0,3438 meters/millisecond at 20°C)

Nice to know, but you can completely ignore this detail :) :) :)

 

What is of interest to you is the fact that the green after correction pulse is placed at appx.15 milliseconds on the X axis from the first before correction impulse.... that is the latency with that specific computer.

 

s6t5.jpg

 

So you can precisely measure the latency introduced by DRC in your own computer by looking at the pulse response..

Other info about impulse response evaluation here: http://www.computeraudiophile.com/f23-dsp-room-correction-and-multi-channel-audio/listening-rooms-and-pulse-response-17633/

 

Also important is the fact that in order to accomodate the requirements of audiovideo users there is the possibility, if necessary, to reduce the latency at the expense of some quality.

If you look at the lower right corner of the following image you'll see an Operating mode toggle between "maximum resolution" and "minimum latency"... that is what you were looking for.

 

0000230.png

 

Finally I can confirm that there will be a 192 KHz update of Dirac Live and that it will a free one.

 

Thanks for your interest,

Flavio

Warning: My posts may be biased even if in good faith, I work for Dirac Research :-)

Link to post
Share on other sites

Thanks Flavio, and great news both about the low-latency setting and free 192K update!

 

One last question (hope I'm not overstaying my welcome!) is Apple TV or Airplay compatible with Dirac Live? I know Apple TV resamples all audio to 48K, so it's not the favored option for highend listening, but for some situations, the flexibility of using a completely wireless connection to an audio system is warranted and it would be great if the Room Correction processing could be used in these situations. And most of these situations for me would be for movies which are often 48K audio, so it's a good match of sample rates.

 

Thanks,

 

-Bob

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...