Have been experimenting with Room EQ

shug4476

Established Member
Joined
Jun 14, 2016
Messages
1,011
Reaction score
281
Points
190
Age
34
Location
London
Having been completely flamed on here for suggesting the heresy that room EQ should form no part of a good music reproduction system, I've managed to try a few different configurations over recent months.

My core system is in my signature. I have had a lot of stereo and multichannel kit through my living room over the last two years but so far, nothing has tempted me to give up the Arcam AV9 (although see below). Some bits of kit did specific things better, but as a marriage of performance, flexibility, and convenience, nothing has quite managed to persuade me to part with it.

My room is a nightmare - lumpy in the bass, very pronounced HF roll off, not great in the mids either. A good candidate for EQ.

I've had no time to run measurements - I did run a load with Audyssey before I listened and I can confirm purely on the basis of the measurements, it works exceptionally well, especially in cleaning up the measured response of a subwoofer.

So here is what I tried:

DIRAC Live - software plugin to JRiver on Macbook Pro, calibrated using Umik-1, digital stream then sent on to Chord 2Qute and analogue output to Arcam AV9 in bypass mode.

ARC - Anthem D2V 3D, which I currently have installed in the room, calibrated using the Anthem ARC Kit, driving the Arcam P7 and the rest of the system normally.

Audyssey XT (non-32 edition) - Denon AVC-A11XVA, fed by a Denon DVD-2930 over Denon Link 3rd. I realise most people now swear by the XT32 edition, however, it was only properly implemented (i.e. put into a pre with excellent analogue engineering) by Denon on two bits of kit (AVC-A1HD and AVP-A1HD) and in the several years I've been looking I think I've only ever seen one that came up with the XT32 upgrade installed. I am just not prepared to try it on the 'cheap and cheerful' receivers Denon chucks out almost every year. I know there are the Marantz dedicated processor designs but they are also adapted AV receivers rather than purpose-built processors with good analogue bypass stages, although I'd probably give one a try.

My Views

To my ears the most transparent and effective of the three was Dirac Live provided the cutoff frequency was set fairly low. As a full-frequency EQ it totally killed the top end and 'air' in the music.

ARC was reasonably effective and did not completely destroy signal fidelity.

Audyssey added a sort of glare/harshness to vocals that was difficult to listen to for extended periods.

However, with straight two channel music, all of the EQ systems were doing (to both my and my wife's ears) significant harm to the signal. I did A/B comparisons on all three sets of kit, with the EQ in and out of the chain, and preferred the EQ off every time. There were no exceptions.

I know some people have horrible rooms, but certainly with two channel music, in all but the most extreme situations, I would urge not bothering with EQ (although I have not tried Room Perfect, it is conceptually the same as all the other systems).
 
Of the various bits of kit, the Anthem is easily the best. In fact it is the only piece of kit I have tried - including some very nice dedicated stereo amps - that has significantly bettered my Arcam. It has a very well implemented analogue bypass mode, ARC in multichannel is fantastic, and it is incredibly flexible.

The Denon A11XVA offered 90% of the performance of my Arcam Pre/Pro in stereo, and maybe bettered it in multichannel. If it weren't for the horrible remote I'd have kept it.
 
I only EQ subs and use absorption and dispersion to manage RT60’s. I don’t think room perfect is something special, it’s well marketed though...
 
I only EQ subs and use absorption and dispersion to manage RT60’s. I don’t think room perfect is something special, it’s well marketed though...

I tried (again with measured success) a MiniDSP for the sub, which worked fairly well (although phase correction done manually was a ball ache). Subjective differences were minimal.
 
Have you tried a dedicated stereo software solution like Accourate, AudioLense or, my current favourite for two channel, FocusFidelty?

FF does a great job in my office, which is somewhere I never had much success with Dirac.

I'm about to give Dirac another go as I have a spare minidsp that I can upgrade.
 
Next month I intend to try a 10x10 Mini DSP, better A>D and D>A than my 2x4 HD and I can then run the three subs in the same way but have parallel inputs for each channel and add the subs to each channels played as full range speakers independently (So just add what they miss with the subs)
 
Does that allow all pass filters on each channel? If it does you can replicate DLBC with MSO.
 
Have you tried a dedicated stereo software solution like Accourate, AudioLense or, my current favourite for two channel, FocusFidelty?

FF does a great job in my office, which is somewhere I never had much success with Dirac.

I'm about to give Dirac another go as I have a spare minidsp that I can upgrade.

No I haven't - I deliberately limited myself to software likely to be found in consumer home theatre electronics (some stereo now too).

Dirac was good but I wouldn't use it regularly in a stereo system unless the room acoustics were unendurable.

However, it was far more transparent than I was expecting (I had low expectations!). It is ideal for Home Cinema.
 
For 2 channel music I've always found DSP makes it sound horrible, I've tried quite a few too.

I think I'm less susceptible with cinema surround sound as I like DSP with it, ARC does a fantastic job and I was fairy happy with Dirac too.

Room treatment made a huge difference for me and something I will always recommend.
 
I'm currently experimenting with JRiver's built in EQ for all 7 channels up to 2khz based on REW's generic filters. Seems to be doing a fantastic job.
 
You should try the trinnov optimiser. Switching it on and off is literally day and night.
 
I'm currently experimenting with JRiver's built in EQ for all 7 channels up to 2khz based on REW's generic filters. Seems to be doing a fantastic job.
I started with that. The FIR filters in things like Focus Fidelity are a huge step up and much closer to the Trinnov.
 
Dead easy. You take measurements in REW from around your listening position and upload them into FF. It's then wizard based to generate the filters. Once you have the measurements you can generate multiple sets of filters with different slopes, ranges, etc. and just A/B them in Jriver.


I'm going to have the opportunity to A/B/C test Dirac in a minidsp, FF, and the Trinnov in the same room with the same system over the coming month or so, I'll report back.
 
I'd be interested in your findings, keep us posted. :smashin:

I'll try and make a start with FF in the meantime, wish me luck!
 
David (the developer) has been very helpful and supportive, or you can PM me and I'll try and help.
 
Thanks man, just realised I need a license, thought it was a freeware like REW lol. I take it they don't have a trial version?
 
No. But if you buy it, try it, and don’t like it, he’ll refund you.

I think audiolense is the same. You buy it and then if it doesn’t work you can return it.

it’s tough as they generate filters that you could just keep using.

one of them will take your measurements and some sample audio files and process them for you so you can hear the difference in room.
 
Surprised REW don't do linear phase filtering, seems to do everything else.
 
I have used all 3 also. I would agree with the OP findings thu, not too sure about killing the top end, but then again you don't really need much correction in this area.

Arc that come with the sub I have is much easier than Dirac to setup and provides a very good level of Bass correction (I only have Paradigm sub).

Coming in 3rd is Audyess XT32 that I had on my Denon amp. While plug in and go and by far the best dummy guide to setting up, the results are not really in the class as the Arc or Dirac but then again value for money maybe the best. Results can be improved by downloading an app and tweaking things, which improve the mid-range somewhat.
 
No. But if you buy it, try it, and don’t like it, he’ll refund you.

I think audiolense is the same. You buy it and then if it doesn’t work you can return it.

it’s tough as they generate filters that you could just keep using.

one of them will take your measurements and some sample audio files and process them for you so you can hear the difference in room.

I might just keep it simple and get a DIRAC based multi-channel MiniDSP unit like @IWC Dopplel at some point. The only problem with JRiver based EQ I have at the moment is I have to watch/listen to everything via the HTPC. Pretty much disqualifies most streaming services unless I source the content from elsewhere. o_O

All you guys with Trinnov processors must be laughing! :D
 
Dead easy. You take measurements in REW from around your listening position and upload them into FF. It's then wizard based to generate the filters. Once you have the measurements you can generate multiple sets of filters with different slopes, ranges, etc. and just A/B them in Jriver.


I'm going to have the opportunity to A/B/C test Dirac in a minidsp, FF, and the Trinnov in the same room with the same system over the coming month or so, I'll report back.
I look forward to your findings!
 
I might just keep it simple and get a DIRAC based multi-channel MiniDSP unit like @IWC Dopplel at some point. The only problem with JRiver based EQ I have at the moment is I have to watch/listen to everything via the HTPC. Pretty much disqualifies most streaming services unless I source the content from elsewhere. o_O

All you guys with Trinnov processors must be laughing! :D
If you use the web based stuff for netflix, amazon, etc then you can route all the audio through Jriver and use it as a processor.

That said, the convolver plug-ins that FF generates give a delay (which I believe can be variable), so they're only really good for audio. The dev is working on quicker filters and multichannel so that we can use them for video as well, but that's a way away yet.

Trinnov doesn't have the all pass filters on all channels that DLBC and MSO require to generate the sub integration that they're capable of. My guess is that'll drop later this year, then I'll really be set.
 
Does that allow all pass filters on each channel? If it does you can replicate DLBC with MSO.
Hi Conrad,

the 10x10 has filters on the input and output so I intend to keep the EQ for the subs as a group on three of the outputs then input all channels apart from the LFE in combination (so the full range signal goes direct to the speakers with no pass through or additional A>D, D>A) then EQ the input to blend with the sub for each channel. So I just ask the subs to fill the missing frequencies for each speaker without needling to process them at all
 
Ok. The all pass routine applies an all pass to all channels to improve the integration with mains and subs.

Basically you time align the subs as you normally would, line that up with the centre to get the best integration, then apply all pass filters at various frequencies on each main channel individually to align the phase with that channel and the subs. The critical difference is that you're not misaligning each main with the centre, which is what you'd have to do to do this without the all pass filters.

But if you're not passing the mains channels through the 10x10 and then on to the speakers you can't do that.
 

The latest video from AVForums

TV Buying Guide - Which TV Is Best For You?
Subscribe to our YouTube channel
Back
Top Bottom