Advice between Kef R3 Meta or Buchardt S400 MKII ?

[Ellebob]

I wouldn't run full range. You will run into the same problems. You will get a different bass response if the left speaker is playing the bass, compared to the right speaker, or whether the sub is playing the bass and different with the many combinations between them

In small home sized rooms we deal with room modes where there will be peaks and dips based on the room dimensions and shape. In the perfect rectangular sealed room it is easy to predict where these happen and there are a number of calculators for room modes. If you have some measuring device even a phone with a signal generator and a Real Time Analyzer and you can play piano noise through your receiver or process and walk the room and you will see a different bass response in the various seating locations. Move the sub and those responses will change.

It takes a little work to find good sub and seating placement as well as integrate the sub with the speakers. Once those all come together you will have great sound.

 

[CK.]

Update on the customer we lent the R3 speakers. He brought them to a frineds house and compared them to the Focal Aria 906. Their set up was not ideal as the R3 was placed upside down on top the the Focals with a pad between them. They were using a Lyngdorf 1120 and turned off the room correction and did not use the subs (pair of Martin Logan, not sure model). I am pretty sure they didn't volume match so this is no way scientific. They both thought the Kef was a little better not night and day, blow me away better. The one that brought the R3s was surprised because from his memory he thought the Aria would be better. Of course none compared to his more expensive Ultras so his search will continue (might take years ). Again, don't shoot the messenger I am just passing on comments from average people doing their own type of comparison.

I wouldn't run full range. You will run into the same problems. You will get a different bass response if the left speaker is playing the bass, compared to the right speaker, or whether the sub is playing the bass and different with the many combinations between them

In small home sized rooms we deal with room modes where there will be peaks and dips based on the room dimensions and shape. In the perfect rectangular sealed room it is easy to predict where these happen and there are a number of calculators for room modes. If you have some measuring device even a phone with a signal generator and a Real Time Analyzer and you can play piano noise through your receiver or process and walk the room and you will see a different bass response in the various seating locations. Move the sub and those responses will change.

It takes a little work to find good sub and seating placement as well as integrate the sub with the speakers. Once those all come together you will have great sound.

Ok I see thanks! Basically i have REW and umik can this be done ? Also, could you elaborate a bit more as to the integration between speakers and sub? Everyone recommends a cross over frequency of 80hz so if thats a given pretty much then what would a good integration look like? Not having dips/peaks near the cross-over frequency?

 

[YSC]

Ok I see thanks! Basically i have REW and umik can this be done ? Also, could you elaborate a bit more as to the integration between speakers and sub? Everyone recommends a cross over frequency of 80hz so if thats a given pretty much then what would a good integration look like? Not having dips/peaks near the cross-over frequency?

if you've done proper cross over (align phase, set the roll off properly on both side), it should show no weird dip/peak at the cross over region, and the phase measurement should be smooth, you will inevitably have some room modes if you use single sub, say, from ceiling reflection, but generally the bass should be smooth

 

[CK.]

if you've done proper cross over (align phase, set the roll off properly on both side), it should show no weird dip/peak at the cross over region, and the phase measurement should be smooth, you will inevitably have some room modes if you use single sub, say, from ceiling reflection, but generally the bass should be smooth

Thanks I will be posting the frequnecy responce for conments

 

[Ellebob]

Ok I see thanks! Basically i have REW and umik can this be done ? Also, could you elaborate a bit more as to the integration between speakers and sub? Everyone recommends a cross over frequency of 80hz so if thats a given pretty much then what would a good integration look like? Not having dips/peaks near the cross-over frequency?

YSC pretty much answered it. But a couple of notes. Look at your subwoofer only response and and your speaker only response (it is good to check both speakers or front 3 in surround system) if you have any dips near the crossover point the crossover could be moved to allow good integration. Most fall between 80-100hz rarely lower and 80hz is a good starting point. For those that don't have REW and a mic here is a basic procedure that is often overlooked in subwoofer set up.

Play a test tone at your crossover frequency. 80hz sine wave for example. Phone generator is fine and can be done via BT or a cable. Play through one speaker
Use an SPL meter or SPL meter app. Doesn't need to be accurate as it is a relative measurement. Place at listening postion.
Adjust the phase switch or knob on the sub or adjust distance in a processor until the SPL meter has its loudest reading.
Verify with other speaker.
Done

 

[CK.]

YSC pretty much answered it. But a couple of notes. Look at your subwoofer only response and and your speaker only response (it is good to check both speakers or front 3 in surround system) if you have any dips near the crossover point the crossover could be moved to allow good integration. Most fall between 80-100hz rarely lower and 80hz is a good starting point. For those that don't have REW and a mic here is a basic procedure that is often overlooked in subwoofer set up.

Play a test tone at your crossover frequency. 80hz sine wave for example. Phone generator is fine and can be done via BT or a cable. Play through one speaker
Use an SPL meter or SPL meter app. Doesn't need to be accurate as it is a relative measurement. Place at listening postion.
Adjust the phase switch or knob on the sub or adjust distance in a processor until the SPL meter has its loudest reading.
Verify with other speaker.
Done

Thanks again, I will try a few different sub positions and measure the FR separately for sub and speakers without any eq at 75db. Now my question is, if i find the optimal position of the sub and there is no dip at 80hz, after I set that crossover and speakrs and sub play together i could notice a dip at 80hz due to phase mismatch that wasnt there in the sub only measurement? If so, i can then try to change the sub phase or change crossover?

 

[Ellebob]

Let me give an example. Let's say your speaker shows a dip at 90hz but your subwoofer doesn't show a dip at 90hz. It would better to raise the crossover to above 90hz so the subwoofer handles that frequency. This can also be vice versa. Placement (speaker/sub/seating) also comes into effect. If you have no major dips with speakers or sub then set your crossover at 80hz. Now you can do a sweep going to both the speaker and sub and see if you have a dip at the crossover. If so the phase will need to be corrected so there is no longer a dip.

 

[CK.]

Let me give an example. Let's say your speaker shows a dip at 90hz but your subwoofer doesn't show a dip at 90hz. It would better to raise the crossover to above 90hz so the subwoofer handles that frequency. This can also be vice versa. Placement (speaker/sub/seating) also comes into effect. If you have no major dips with speakers or sub then set your crossover at 80hz. Now you can do a sweep going to both the speaker and sub and see if you have a dip at the crossover. If so the phase will need to be corrected so there is no longer a dip.

Right….ok i think i got it now! Cheers will revert back with some final measurements

 

[CK.]

Right….ok i think i got it now! Cheers will revert back with some final measurements

@Ellebob
@YSC
@everyone
ok guys, first of all have a nice month, I've managed to get a few measurements at MLP going (see attached), and I would appreciate your input. Attachment titles are pretty much self-explanatory I hope. Initially I measured 4 possible sub (rel 1003 mk ii) positions and I chose the red one thinking it is the best. Then I measured L+R channel separately (KEF R3) (without EQ or sub). Smoothing for all measurements is "var smoothing" except for the one having (1_6) in the name meaning 1/6 smoothing. Then I set the xover to 80Hz and enabled audyssey auto-correction upto 300Hz (I also have the option of manual correction). Finally I also enabled Audyssey dynamic EQ (auto bass boost in low volumes). SPL was 75dB.

Based on these, I have a couple of questions:

1. What do you think of the results?
2. Has Audyssey done a good job or is it worth performing manual EQ (from REW)?

PS. I know I have to take multiple measurements around MLP and I will do that I soon as I know I am on the right track..
update: I suspect that auto-eq applied some boosts also but given that I do not listen above 75dB (-17dB volume) I also suspect this wont be a problem.

Thanks!

 

[dominikz]

@Ellebob
@YSC
@everyone
ok guys, first of all have a nice month, I've managed to get a few measurements at MLP going (see attached), and I would appreciate your input. Attachment titles are pretty much self-explanatory I hope. Initially I measured 4 possible sub (rel 1003 mk ii) positions and I chose the red one thinking it is the best. Then I measured L+R channel separately (KEF R3) (without EQ or sub). Smoothing for all measurements is "var smoothing" except for the one having (1_6) in the name meaning 1/6 smoothing. Then I set the xover to 80Hz and enabled audyssey auto-correction upto 300Hz (I also have the option of manual correction). Finally I also enabled Audyssey dynamic EQ (auto bass boost in low volumes). SPL was 75dB.

Based on these, I have a couple of questions:

1. What do you think of the results?
2. Has Audyssey done a good job or is it worth performing manual EQ (from REW)?

PS. I know I have to take multiple measurements around MLP and I will do that I soon as I know I am on the right track..

Thanks!

How come the sub has such a steep slope below ~40Hz? Specs say it should go down to 24 Hz anechoically (-6dB point). Is some kind of high-pass filter enabled?

Regarding the EQ correction - the target slope seems too flat to me, i.e. it looks like it might sound bass-deficient. If I draw a quick trend line it seems to me a -0,6db/octave target slope might be more appropriate:

You can see the measured response is significantly below the trend under 100Hz for both channels. Of course some people prefer less low bass, but I'd suggest to try both options and see what you prefer.

I'd also try to play a bit with subwoofer crossover frequency, level and phase settings to see if I can get a fuller response in the crossover region for both speakers.

In case it helps for comparative purposes, this is what I was able to get in my acoustically untreated living room with 2x Revel M16 and a single SVS SB-1000, measured with MMM (~55Hz crossover in my case, since I can't high-pass the M16):

(Images from this post, EQ#2 variant shown)

Have you tried moving the R3s a bit as well to minimize the 100-200Hz suckout (still visible in the 'no EQ' measurement), without relying on EQ to fill it in?

 

[CK.]

How come the sub has such a steep slope below ~40Hz? Specs say it should go down to 24 Hz anechoically (-6dB point). Is some kind of high-pass filter enabled?

Regarding the EQ correction - the target slope seems too flat to me, i.e. it looks like it might sound bass-deficient. If I draw a quick trend line it seems to me a -0,6db/octave target slope might be more appropriate:View attachment 315826View attachment 315827
You can see the measured response is significantly below the trend under 100Hz for both channels. Of course some people prefer less low bass, but I'd suggest to try both options and see what you prefer.

I'd also try to play a bit with subwoofer crossover frequency, level and phase settings to see if I can get a fuller response in the crossover region for both speakers.

In case it helps for comparative purposes, this is what I was able to get in my acoustically untreated living room with 2x Revel M16 and a single SVS SB-1000, measured with MMM (~55Hz crossover in my case, since I can't high-pass the M16):

(Images from this post, EQ#2 variant shown)

Have you tried moving the R3s a bit as well to minimize the 100-200Hz suckout (still visible in the 'no EQ' measurement), without relying on EQ to fill it in?

Thanks for the comments! Regarding the sub, i dont have any filter and for the settings and connections in the back please see attched pic. I lowered the volume as per audysseys recommendation when i did the previous auto eq (the vol on receiver is +5). Anything weird? The measurements are with autoeq disabled. Have you looked at the measurements with dynamic eq enabled that boosts the bass region? Regarding the suckout at 140hz, i tried to move it a bit but did not made much difference, unfortunately i cannot make big placement moves due to my room setup.

 

[dominikz]

Thanks for the comments! Regarding the sub, i dont have any filter and for the settings and connections in the back please see attched pic. I lowered the volume as per audysseys recommendation when i did the previous auto eq (the vol on receiver is +5). Anything weird?

Ah, the crossover between the sub and mains is managed by Audyssey, I see! Then I guess you probably don't have a lot of control over it. Still I find it quite strange that you don't seem to have a lot of bass below 40Hz, not sure what might be happening there.

The measurements are with autoeq disabled. Have you looked at the measurements with dynamic eq enabled that boosts the bass region?

I did but I believe the "dynamic EQ" feature is meant to bring the bass up when listening at a low level (due to the psychoacoustic phenomena commonly illustrated with the equal loudness contours). So I'd still work towards having (what I feel is) a more balanced profile, that way dynamic EQ might also work better.
However if you're happy with the sound now you're done!

Regarding the suckout at 140hz, i tried to move it a bit but did not made much difference, unfortunately i cannot make big placement moves due to my room setup.

That's too bad, but understandable!

 

[YSC]

@Ellebob
@YSC
@everyone
ok guys, first of all have a nice month, I've managed to get a few measurements at MLP going (see attached), and I would appreciate your input. Attachment titles are pretty much self-explanatory I hope. Initially I measured 4 possible sub (rel 1003 mk ii) positions and I chose the red one thinking it is the best. Then I measured L+R channel separately (KEF R3) (without EQ or sub). Smoothing for all measurements is "var smoothing" except for the one having (1_6) in the name meaning 1/6 smoothing. Then I set the xover to 80Hz and enabled audyssey auto-correction upto 300Hz (I also have the option of manual correction). Finally I also enabled Audyssey dynamic EQ (auto bass boost in low volumes). SPL was 75dB.

Based on these, I have a couple of questions:

1. What do you think of the results?
2. Has Audyssey done a good job or is it worth performing manual EQ (from REW)?

PS. I know I have to take multiple measurements around MLP and I will do that I soon as I know I am on the right track..
update: I suspect that auto-eq applied some boosts also but given that I do not listen above 75dB (-17dB volume) I also suspect this wont be a problem.

Thanks!

I am no expert in this but a few things in my own naive observations:

1) there is quite a suckout between 60-80hz, it could either be having some sort of reflection cancellation around 1-2m from the subs, or there are phasae issues, try flip the phase and see if it becomes a bass boost instead of a dip, if so maybe the auto EQ didn't boost it but aligned the phase and cut it down

2) boost is a bad idea if you don't apply EQ pre-amp shelving down the same peak boost db, coz say if you set a boost at 100hz for 2db, if the track contains material calling for 100hz at full volume you will be using EQ to push that to 102%, which the dac itself can't send out

3) ignoring the 60-80hz dip which can be crossover issue or phase issue+ sub boundary effect, when I eyeball it it isn't too bad for a general trend line with a bass boost ~6db below 100hz and shelving down from 100hz to 200hz, I would start off with only taking out the two circled peaks to see if that sounds satisfactory

 

[CK.]

Ah, the crossover between the sub and mains is managed by Audyssey, I see! Then I guess you probably don't have a lot of control over it. Still I find it quite strange that you don't seem to have a lot of bass below 40Hz, not sure what might be happening there.

I did but I believe the "dynamic EQ" feature is meant to bring the bass up when listening at a low level (due to the psychoacoustic phenomena commonly illustrated with the equal loudness contours). So I'd still work towards having (what I feel is) a more balanced profile, that way dynamic EQ might also work better.
However if you're happy with the sound now you're done!

That's too bad, but understandable!

The sub is basically 10’ so not sure if i could get much below 40hz in untreated room? Not sure

The xover is controlled by the avr but i can set it to any value and also the phase.

Dynamic eq does boost the bass for low lsitening levels yes and if you boost the sub manually then it just does not apply the same level of boost but i cam try to boost it manually.

 

[CK.]

I am no expert in this but a few things in my own naive observations:

1) there is quite a suckout between 60-80hz, it could either be having some sort of reflection cancellation around 1-2m from the subs, or there are phasae issues, try flip the phase and see if it becomes a bass boost instead of a dip, if so maybe the auto EQ didn't boost it but aligned the phase and cut it down

2) boost is a bad idea if you don't apply EQ pre-amp shelving down the same peak boost db, coz say if you set a boost at 100hz for 2db, if the track contains material calling for 100hz at full volume you will be using EQ to push that to 102%, which the dac itself can't send out

3) ignoring the 60-80hz dip which can be crossover issue or phase issue+ sub boundary effect, when I eyeball it it isn't too bad for a general trend line with a bass boost ~6db below 100hz and shelving down from 100hz to 200hz, I would start off with only taking out the two circled peaks to see if that sounds satisfactoryView attachment 315870

Thanks for the comments

1. Yeah not really sure what the auto eq has done but i will try flipping the phase.

2. Got it but tbh audyssey should be smart enough to consider this right? Personally i will not manually boost a frequency.

3. I will try it but does that mean you do not prefer the result shown with auto eq and dynamic eq? I mean assuming no signals has been boosted.

PS: is the issue in (2) still valid even if listening volumes are nowhere close to reference?

 

[YSC]

Thanks for the comments

1. Yeah not really sure what the auto eq has done but i will try flipping the phase.

2. Got it but tbh audyssey should be smart enough to consider this right? Personally i will not manually boost a frequency.

3. I will try it but does that mean you do not prefer the result shown with auto eq and dynamic eq? I mean assuming no signals has been boosted.

PS: is the issue in (2) still valid even if listening volumes are nowhere close to reference?

For PS first, yes, cause what you’re doing with volume is cutting the signal down as a whole after it’s eq boosted, so if no pre amp eq to tune down the the whole spectrum by a margin, say subtract all by 6db then boost back the dip in FR it at some point will exceed the dac’s call for 100% and the boost won’t show up, making the experience inconsistent

I don’t prefer the auto EQ as that is basically in room flat with mid bass chopped off even more due to room mode or phase issue, it will sound very bass shy. Since a neutral speaker in a good room have in room slop down target, and mixing is done with that in mind, what you’re doing will make every properly produced recordings sound bright

 

[CK.]

For PS first, yes, cause what you’re doing with volume is cutting the signal down as a whole after it’s eq boosted, so if no pre amp eq to tune down the the whole spectrum by a margin, say subtract all by 6db then boost back the dip in FR it at some point will exceed the dac’s call for 100% and the boost won’t show up, making the experience inconsistent

I don’t prefer the auto EQ as that is basically in room flat with mid bass chopped off even more due to room mode or phase issue, it will sound very bass shy. Since a neutral speaker in a good room have in room slop down target, and mixing is done with that in mind, what you’re doing will make every properly produced recordings sound bright

Ok I see your point thanks. I will try just cuttin those two peaks without any other eq but I also have the option to set a target curve in the autoeq parameters, which could be limited to the bass frequencies and follow the trend you are describing, would that be something you would consider as a good option?

 

[YSC]

Ok I see your point thanks. I will try just cuttin those two peaks without any other eq but I also have the option to set a target curve in the autoeq parameters, which could be limited to the bass frequencies and follow the trend you are describing, would that be something you would consider as a good option?

I would try auto EQ with Harman target curve to begin with, which is available here somewhere. Try limit the correction to below 200hz and the auto eq should have option to limit it to no boosting, usually that would give the best result

 

[CK.]

I would try auto EQ with Harman target curve to begin with, which is available here somewhere. Try limit the correction to below 200hz and the auto eq should have option to limit it to no boosting, usually that would give the best result

Hello again,

Ok so I did a bit of trial and error with the sub direction and REW EQ and it turns out that the sub woofer direction, not "position", also plays an important role although have not seen this advice going around. Personally I turned the sub 90 degrees facing a side wall (about 1.5m away) instead of my MLP.
I also manually tweaked the sub delay and sub trim (+2 on my avr, 30% vol on knob) in order to achieve a harman like response and smoother response.

In any case, I am attaching my results and would once again appreciate your input/advice. I have included both 1/6 smoothing and var smoothing but also a third graph that compares the various cross-overs 80,90,100,110. As XO goes up so does the 50-80Hz region but I opted for 90Hz cross-over because did not want to mess up the bass localisation.

Is the 90Hz XO ok or should I keep the 80Hz?
Is the bass region too exaggerated? This is measured at 75dB and I usually listen to around 70dB.
The narrow null at 100Hz is it really audible?

 

[YSC]

Hello again,

Ok so I did a bit of trial and error with the sub direction and REW EQ and it turns out that the sub woofer direction, not "position", also plays an important role although have not seen this advice going around. Personally I turned the sub 90 degrees facing a side wall (about 1.5m away) instead of my MLP.
I also manually tweaked the sub delay and sub trim (+2 on my avr, 30% vol on knob) in order to achieve a harman like response and smoother response.

In any case, I am attaching my results and would once again appreciate your input/advice. I have included both 1/6 smoothing and var smoothing but also a third graph that compares the various cross-overs 80,90,100,110. As XO goes up so does the 50-80Hz region but I opted for 90Hz cross-over because did not want to mess up the bass localisation.

Is the 90Hz XO ok or should I keep the 80Hz?
Is the bass region too exaggerated? This is measured at 75dB and I usually listen to around 70dB.
The narrow null at 100Hz is it really audible?

The facing could be due to how the driver itself relative to the reflecting wall, basically changing the boundary effect as you've moved the sub closer to the wall

it should be fine for 90hz X over, generally if you can't perceive where the sub is (feeling the bass coming from the sub location instead of from the stereo pair)

narrow nulls (High Q) isn't audible to ears so it should be fine, from measurements it looks decent, so how did you subjectively feels? somehow more enjoyable now?