B&W DM601 S3 In Room Response Measurement / Suggestions?

[mattzildjian]

Hi All,

I recently inherited a pair of B&W DM601 S3 Bookshelf speakers. I am using them with my TV optical out, into a Loxjie A30.

Listening position is about 2.5 meters from the speakers.

I made an in room measurement using RTA method to get an average from the listening position.

I am unsure if the 50hz peak is from the speakers or the room but it's very noticeable and ruins the overall balance.

Would you suggest using something like a miniDSP to correct this or would I be better off selling these speakers and buying something else?

Thanks!

EDIT: Corrected the measurement

 

[staticV3]

That measurement looks way off to me. It's severely down tilted to a point where if the speakers really did measure like that, then they would surely sound extremely broken.

I suspect it's a measurement error.

 

[Steve Dallas]

Can you repost the measurement using a 50dB scale? That is the standard we use here.

Under 1000Hz is influenced by the room. Under about 300Hz is heavily influenced. Most speakers will look very similar down there, so swapping speakers likely will not help. (There may be 1 notable exception, which is KEF, who shelves bass down starting at 100Hz to help reduce such peaks.)

Your best bet is digital room correction such as Dirac, Acourate, Audyssey, etc. You can get some of the way there with a good PEQ solution, one of which is supported by miniDSP.

I am not sure I trust your measurement, however. The downward slope is too extreme, indicating the speakers should sound very dark, which is not typical of that model B&W.

 

[staticV3]

Can you repost the measurement using a 50dB scale? That is the standard we use here.

I don't think that would get us anywhere really.

OP's frequency response measurement drops by 50dB from 50Hz to 20k. Something is definitely not right.

 

[mattzildjian]

Thanks for the suggestions so far.
For clarity I made the measurement using pink noise / moving mic method.
I think maybe I missed a step to correct for the slope that pink noise has to begin with.. Does anyone know how to correct this post measurement?

EDIT: I can see an option "Adjust RTA Levels" which mentions altering the levels of pink noise RTA to be equivalent to a sweep. I didn't have this selected!

 

[staticV3]

For clarity I made the measurement using pink noise / moving mic method.
I think maybe I missed a step to correct for the slope that pink noise has to begin with.. Does anyone know how to correct this post measurement?

When you use Pink Noise, then you need to set the RTA mode to 1/48 octave.
You probably left it at Spectrum, which will result in that error.

EDIT: I can see an option "Adjust RTA Levels" which mentions altering the levels of pink noise RTA to be equivalent to a sweep. I didn't have this selected!

That setting correct absolute SPL levels, but does not help in this case.

 

[kemmler3D]

I have some 602 S2s and I can confirm that the bass is pretty boomy with those guys. It may be similar in your case. However, if you are able to get a good measurement and do some correction with EQ, I think you'll end up pretty happy with the results. I would definitely do that before getting new speakers.

StaticV3 is right though, the measurement you've shown here is definitely not right. This would sound like listening to your music from the next room with the door closed under a blanket.

 

[mattzildjian]

UPDATE

I have corrected the measurements

Listening position:

Close measurement 1 inch from speaker:

It would seem that mega 50hz peak is all from the room... wow.

 

[HarmonicTHD]

UPDATE

I have corrected the measurements

Listening position:
View attachment 330111

Close measurement 1 inch from speaker:
View attachment 330112

It would seem that mega 50hz peak is all from the room... wow.

Low frequency is better. Depression at 3kHz and Plateau at 10k could use some TLC.

 

[mattzildjian]

Low frequency is better. Depression at 3kHz and Plateau at 10k could use some TLC.

Incase you mis-interpreted, both those measurements are corrected measurements. Top one is listening position, 50hz peak is still there.
The second measurement is right up agaisnt the speaker so I could see if that 50hz peak is from the speaker or the room.. turns out it's 90% the room.

Not much I can do about that with simple tone controls.
Would you suggest a minidsp to correct an optical signal?

 

[HarmonicTHD]

Incase you mis-interpreted, both those measurements are corrected measurements. Top one is listening position, 50hz peak is still there.
The second measurement is right up agaisnt the speaker so I could see if that 50hz peak is from the speaker or the room.. turns out it's 90% the room.

Not much I can do about that with simple tone controls.
Would you suggest a minidsp to correct an optical signal?

Sorry. Overlooked that. My bad.

 

[kemmler3D]

Incase you mis-interpreted, both those measurements are corrected measurements. Top one is listening position, 50hz peak is still there.
The second measurement is right up agaisnt the speaker so I could see if that 50hz peak is from the speaker or the room.. turns out it's 90% the room.

Not much I can do about that with simple tone controls.
Would you suggest a minidsp to correct an optical signal?

Thanks for posting the new measurements.

I would definitely suggest a MiniDSP, or another solution with similar functionality. That ~50hz peak (and the ~90hz one) is pretty simple to tame with EQ and your listening experience is going to be massively better when you do.

Whether you go with MiniDSP or something else - It mostly depends on how DIY you want to get. Room correction is like many things, where time and effort can get you similar results with a lower cost.

On one end you have a $50 solution (thread) that does as much or more than a MiniDSP unit, but is obscure, doesn't come with a case, and requires a lot of confidence in navigating an unfamiliar and not-consumer-oriented DSP platform. Similarly, Camilla DSP on a Raspberry Pi can do it, but you have to mess with Linux a little.

On the other end, you have receivers with Dirac or similar room correction products that do most of the work for you out of the box, but cost at least a thousand bucks.

MiniDSP lands you somewhere in the middle... it's not exactly cheap, but it works, they are considered good quality and reliable, and AFAIK the interface isn't too bad.

Whatever route you decide to go, there are threads on here that should help you get there.

As an aside I would also suggest filling in that 3khz dip a little, unless you decide you like it. It's wide enough that EQ should not cause big issues, and it shows up in both the room and nearfield measurements, so it's probably OK to EQ.

 

[mattzildjian]

Thanks for posting the new measurements.

I would definitely suggest a MiniDSP, or another solution with similar functionality. That ~50hz peak (and the ~90hz one) is pretty simple to tame with EQ and your listening experience is going to be massively better when you do.

Whether you go with MiniDSP or something else - It mostly depends on how DIY you want to get. Room correction is like many things, where time and effort can get you similar results with a lower cost.

On one end you have a $50 solution (thread) that does as much or more than a MiniDSP unit, but is obscure, doesn't come with a case, and requires a lot of confidence in navigating an unfamiliar and not-consumer-oriented DSP platform. Similarly, Camilla DSP on a Raspberry Pi can do it, but you have to mess with Linux a little.

On the other end, you have receivers with Dirac or similar room correction products that do most of the work for you out of the box, but cost at least a thousand bucks.

MiniDSP lands you somewhere in the middle... it's not exactly cheap, but it works, they are considered good quality and reliable, and AFAIK the interface isn't too bad.

Whatever route you decide to go, there are threads on here that should help you get there.

As an aside I would also suggest filling in that 3khz dip a little, unless you decide you like it. It's wide enough that EQ should not cause big issues, and it shows up in both the room and nearfield measurements, so it's probably OK to EQ.

honestly I love tweaking stuff so I'll start looking into those DIY methods you mentioned, thanks!!

 

[kemmler3D]

honestly I love tweaking stuff so I'll start looking into those DIY methods you mentioned, thanks!!

Cool! Keep in mind that if you buy that unit from Aliexpress, you also need the USBi to connect it to a PC. And I'm pretty sure you will need to take care converting the EQ parameters from REW to the format it uses.

If you want to dive into the raspberry pi situation, there's a really big thread on it here: https://www.audiosciencereview.com/forum/index.php?threads/rpi4-camilladsp-tutorial.29656/

Lastly, I have used a Windows PC running EQAPO as a room correction EQ successfully. You can get pretty cheap refurbished Lenovo mini-workstations and do that, should run you less than $150 all day. You can add the optical input using one of these: https://hifimediy.com/product/hifime-ur23-spdif-optical-to-usb-converter/ and pretty much any PC can handle 2-channel EQ no problem.

Lastly, definitely share your results here when you get it set up!

 

[mattzildjian]

Cool! Keep in mind that if you buy that unit from Aliexpress, you also need the USBi to connect it to a PC. And I'm pretty sure you will need to take care converting the EQ parameters from REW to the format it uses.

If you want to dive into the raspberry pi situation, there's a really big thread on it here: https://www.audiosciencereview.com/forum/index.php?threads/rpi4-camilladsp-tutorial.29656/

Lastly, I have used a Windows PC running EQAPO as a room correction EQ successfully. You can get pretty cheap refurbished Lenovo mini-workstations and do that, should run you less than $150 all day. You can add the optical input using one of these: https://hifimediy.com/product/hifime-ur23-spdif-optical-to-usb-converter/ and pretty much any PC can handle 2-channel EQ no problem.

Lastly, definitely share your results here when you get it set up!

Hmm, I have a windows laptop that would nice to use just as a proof of concept to see how the speakers take to being EQ'd.
I looked at the product you linked to add optical to the pc, however that is only an input, I would also need an output, optical out perferrably.
Would this work? https://www.amazon.co.uk/LEAGY-External-Surround-Recording-Compatible/dp/B07DGR9M6M

Also if that were to prove successful, could I use the same product with a raspberry pi setup running camillaDSP?

 

[kemmler3D]

Hmm, I have a windows laptop that would nice to use just as a proof of concept to see how the speakers take to being EQ'd.
I looked at the product you linked to add optical to the pc, however that is only an input, I would also need an output, optical out perferrably.
Would this work? https://www.amazon.co.uk/LEAGY-External-Surround-Recording-Compatible/dp/B07DGR9M6M

Also if that were to prove successful, could I use the same product with a raspberry pi setup running camillaDSP?

I *think* that would work. It *should* work, anyway. But, in the original post you mentioned using a Loxjie DAC-AMP so you can just hook that to the PC, no?

One of the CamillaDSP threads mentions an interface with optical in/out that works with raspberry Pi, might be worth digging that up...

 

[mattzildjian]

I *think* that would work. It *should* work, anyway. But, in the original post you mentioned using a Loxjie DAC-AMP so you can just hook that to the PC, no?

That's true, would that also work with the Pi though? Using the laptop would just be temporary.

One of the CamillaDSP threads mentions an interface with optical in/out that works with raspberry Pi, might be worth digging that up...

That one looks very similar, same brand, but more expensive.

 

[kemmler3D]

would that also work with the Pi though? Using the laptop would just be temporary.

I am not sure... @mdsimon2 would have a much better idea... I think it should, but I have never used an RPi so I don't want to lead you astray here.

 

[mattzildjian]

I am not sure... @mdsimon2 would have a much better idea... I think it should, but I have never used an RPi so I don't want to lead you astray here.

I suppose since I'd been hearing that the Loxjie A30 uses an inferior DAC for the USB input I was trying to avoid it and stick with the optical wherever possible.

 

[Steve Dallas]

For POC purposes, why not use the laptop as the music player running EQ APO into the A30? That would be free and immediate. You can also dabble with MathAudio via Foobar2000, a Dirac Standalone trial license, etc. Once you decide on your direction, purchase the bits you need to make it work.