New loudspeakers & room digital correction service

[Celts88]

That's my biggest worry. Not that Thierry's software won't give good results, but that I'll take inferior measurement readings via REW that will affect the end result. I've never used REW or taken measurements using any software before, so I'm worried that I'll do the measurements wrong - or not so much wrong, but not as accurate as they can be.

Don't worry. Please check my earlier post. I think the measurements were easy to take with these instructions.
https://devialetchat.com/showthread.php?...5#pid65395

If anyone is facing a difficulty making his/her first measurements with REW, please forward me a link to the measurement files and I will be happy to help if I can.

Its relatively easy to make reasonable measurements. It's not harder than to make the measurements for Dirac or any other application IMHO.

Petrik

Thanks for the info.

SwissBear

Appreciate the offer. May have some spare time tomorrow, so will go it a go doing the measurements - here's hoping 

[petrik]

First impressions. I have been listening with these HAF filters about 30 minutes now. I have to say that I am really positively surprised about these. My system has never sounded like this although I have used digital room correction before (parametric EQ, REW, Dirac, and Acourate). These HAF filters do something really differently. I had to enter -9 dB to the Roon's headroom management to prevent digital clipping. Earlier with Acourate's filters, -6 dB was enough.

As said, I have listened only about 30 minutes with these and I'm quite amazed. I guess I need to listen way more before doing any final conclusions but I thought to share my first impressions about these. First of all, everything sounds so coherent, natural and powerful. I noticed that listening is now more joyful and more emotional experience than when compared to listening without these HAF filters. Everything sounds so balanced and 'just right' that it is even hard to find anything negative to say. These filters mainly just make me want to listen more music and forget everything else. I rarely feel this way when doing listening tests. I wonder why I feel so positive this time. Perhaps I just had a bad day at the office and listening music just makes me feel good now? I suppose that could be one possible explanation. Alternatively the HAF filters are doing something right

Well, if I have to say something negative, then the sound balance might be a little bit dark and bass-heavy for me so I'm probably going to ask for a bit lighter target curve. I have to say that these filters sound so much fun already that I have not yet decided what to do with the target curve. I have never heard such a mid-bass from my system than now. I love it. And this three-dimensional imaging is something I have never heard before. It sounds so different that I might need some more time to get used to it. The bass is perhaps not as tight as it could be, but it's probably only because the target curve is having too much bass for my taste. My listening room is having the biggest problem around 40 Hz region as you may have noticed from the pictures that I posted earlier. I might want to reduce a decibel or two from the target curve in that region.

I briefly compared these HAF filters with parametric EQ that I had set in Roon earlier. There was no comparison, the PEQ sounded almost awful in comparison. The HAF filters sounded more natural and fun.

I definitely need to continue listening and think about adjusting the target curve a little bit. Later on when the target curve is right, I'll will compare results with the Acourate's filters.

[Hifi_swlon]

Sounds like he did a good job!

I felt the curve would be a little bass heavy for some - it would probably be perfect for me though!

You could ask him do do some 'classic' curves. I'm sure he will have B&K and a 'Dirac style' straight slope of about 6db falloff at the top end. You could also get him to do a dead flat curve - which would sound awful directly - but would allow you to do try different house curves in Roons PEQ which is harder if it's a curve on top of a curve if you see what I mean.

One real advantage Dirac has in its measurement system is it sends left, then right, then both together as a triplet of sweeps that it deals with without intervention. I wish others would copy this idea as it speeds things up a lot. REW is a decent free tool, but it's not the friendliest.

I agree Thierys website is sparse, but hopefully he takes the bait - I think some good customer communication about his services would really help his business. Assuming his written English is ok of course. My French certainly wouldn't be up to it.

[petrik]

I felt the curve would be a little bass heavy for some - it would probably be perfect for me though!

I think you would like this target curve and these filters  I can recommend trying these.

[Celts88]

If anyone is facing a difficulty making his/her first measurements with REW, please forward me a link to the measurement files and I will be happy to help if I can.

Its relatively easy to make reasonable measurements. It's not harder than to make the measurements for Dirac or any other application IMHO.

SwissBear
Appreciate the offer. May have some spare time tomorrow, so will go it a go doing the measurements - here's hoping 

SwissBear

Gave a try taking the measurements, not sure if there good or bad (or more importantly, did I do them correctly)???

https://www.dropbox.com/sh/g3az7aigjf8de...3jC0a?dl=0

Left Speaker

   

Right Speaker

   

[SwissBear]

SwissBear

Gave a try taking the measurements, not sure if there good or bad (or more importantly, did I do the correctly)???
https://www.dropbox.com/sh/g3az7aigjf8de...3jC0a?dl=0

Hi Celts88,

Your measurements look very well done to me. The impulses are clean. Nothing to say. Congratulations on the measurements !
I noticed that you had only 8 measurements and not 9. Did you not make the sweet-spot ?
As far as presentation is concerned, it is probably easier for Thierry to manipulate one file with the 16 measurements labelled L1..L8 and R1..R8 inside it. Just open a new file, import (open) each measurement and rename it.

A few thoughts on your room if you allow me:

• you look to have a slightly elevated RT60 (reverberation). You might want to think about putting an acoustic ceiling or some sound absorbent
  [img][/img]
  
• in some places, you look to have reflections which can affect the sound you listen to:
  [img][/img]
  To capture this information, you need to look at the Energy Time Curve (ETC). You would not be happy to have reflections above -15db. On the right channel, you have one at -10db and 5ms, which means that the additional distance run by the sound is 5 x 0.340 = 1.7m compared to the direct sound. This might be the wall behind the speaker or a wall on the right of the speaker or the ceiling. You might want to put some absorbent there.
  
• Finally, you have probably noticed some loud resonance in the very low end of the spectrum. [img][/img]
  If this is a nuisance, and instead of installing ugly bass-traps, you can discuss with Thierry the possibility to put a brick-wall filter to not produce sounds below say 20Hz. Up to you.
  

Hope this helps.

[SwissBear]

On the subject of correcting the reverberation, I had a very ugly situation at my home (see red curve). I installed a sandwich of 8cm isolation, 12cm air and an acoustic ceiling and this largely improved the situation. It is not yet perfect, but given the room size, it is very acceptable.

[img][/img]

[img][/img]

[petrik]

SwissBear - Do you know is it better to take all 9 measurements or do 5 measurements be enough? Will more measurements give better results?

[SwissBear]

SwissBear - Do you know is it better to take all 9 measurements or do 5 measurements be enough? Will more measurements give better results?

Hi Petrik,

The goal of having multiple measurements is to make 'spatial averaging', ie to separate the signal from the noise (see page 16 of this document: https://www.dropbox.com/s/r9ipzfiq46givt...2.pdf?dl=0 ). The more measurements you have, to a certain extend which has been defined as 9 measurements by Dirac as well as by Thierry, the better.

What is important in any case is to have a measurement made in the sweet-spot and to identify it for Thierry.

[Celts88]

SwissBear

Gave a try taking the measurements, not sure if there good or bad (or more importantly, did I do the correctly)???
https://www.dropbox.com/sh/g3az7aigjf8de...3jC0a?dl=0

Hi Celts88,

Your measurements look very well done to me. The impulses are clean. Nothing to say. Congratulations on the measurements !
I noticed that you had only 8 measurements and not 9. Did you not make the sweet-spot ?
As far as presentation is concerned, it is probably easier for Thierry to manipulate one file with the 16 measurements labelled L1..L8 and R1..R8 inside it. Just open a new file, import (open) each measurement and rename it.

A few thoughts on your room if you allow me:

• you look to have a slightly elevated RT60 (reverberation). You might want to think about putting an acoustic ceiling or some sound absorbent
  
• in some places, you look to have reflections which can affect the sound you listen to:
  
  To capture this information, you need to look at the Energy Time Curve (ETC). You would not be happy to have reflections above -15db. On the right channel, you have one at -10db and 5ms, which means that the additional distance run by the sound is 5 x 0.340 = 1.7m compared to the direct sound. This might be the wall behind the speaker or a wall on the right of the speaker or the ceiling. You might want to put some absorbent there.
  
• Finally, you have probably noticed some loud resonance in the very low end of the spectrum.
  If this is a nuisance, and instead of installing ugly bass-traps, you can discuss with Thierry the possibility to put a brick-wall filter to not produce sounds below say 20Hz. Up to you.
  

Hope this helps.

SwissBear

 
Really appreciate you having a look at my measurements, and all the explanations (even if I don't really understand them  ).
 
Firstly, and unfortunately the easiest one to answer is that I won't be able to put any physical Acoustic fixes into the room. Where my hi-fi set-up is in the living room that is open when first entering the house and as such there's Buckley's chance of me being able to change anything without a divorce being on the cards (the Aussies will know what I mean by Buckley's - it's another way of saying 'not a hope in hell').
 
The roof in the room is part recessed, and as such no chance of being able to change this (again will just have to live with it).
 
Also unfortunately I am a complete beginner with this Room Correction stuff, and have never used REW (or any software) before. I really don't have any idea where you got the charts from (look really good, but still trying to get my head round what they mean).
 
I took 8 measurements as Thierry says on his website to do between 5 & 8, and the diagram is pretty random measurements (a bit like what I did, with no real logic behind them except the Left & Right ones always being in the same place). Also when you say about combining all the files into 1, do you mean to zip them, I had a look at REW and can't see how to combine all 16 measurements into 1 file. L1 & R1 are my sweet spot measurements, with 2 to 8 being taken randomly (with one each for Left and Right Speakers each).
 
Another big no no to surely a lot of people is that I normally sit offset from the middle of the speakers and have actually did the measurements with me being in the offset (1/3 to the left side of centre, i.e. 0.5m from the left speaker axis, and 1.8m from the right speaker). Would it be better if I did the measurements sitting exactly in the middle (not where I normally sit unfortunately)
 
Due to not being able to change anything physically in the room, and being a bit of a Luddite with Room Correction software, hence why I was really happy to see the options being given by Thierry. Will forward on your comment about putting in a brick-wall filter to not produce sounds below 20Hz
 
Cheers
John