[Photos and Measurements Inside] Dirac Live Distorts Bass JBL 705P

[mdsimon2]

Agree that this is likely a combination issue related to DSP, amplifier input sensitivity and volume expectations.

Which miniDSP product is this? SHD, DDRC-24, DDRC-88A? I am guessing DDRC-24 as that is probably the most common and most likely to have volume issues due to lower output voltage.

There is a lot of good information in this thread but there is also a lot of "noise".

DDRC-24 has an output voltage of 2V, this will be achieved when output level is at 0 dB. If output level ever goes above 0 dB DDRC-24 will clip. It does not matter how this 0 dB is reached, you can program +50 dB of boost from a combination of input gain, output gain, PEQs and Dirac and if you set the volume control to -50 dB the output will be at 0 dB and not clip. If you look at the miniDSP output levels you can see how close you are to 0 dB and clipping. If your output levels are below 0 dB your distortion problems are not caused by the miniDSP regardless of settings in the miniDSP.

+4 dBu input sensitivity setting on 705Ps corresponds to a voltage level of 1.23V. This means that that the DDRC-24 should be able to drive the 705Ps to full power at this input sensitivity setting. However the actual output voltage from the DDRC-24 will depend on the level of the recording, the processing in the DDRC-24 and the DDRC-24 volume level.

-10 dbV input sensitivity setting on 705Ps corresponds to a voltage level of 0.32V. In this case the DDRC-24 will almost certainly be able to drive the 705Ps to full power (and beyond in to clipping). This setting essentially offers +10 dB gain compared to the +4 dBu setting. Using this setting with the DDRC-24 is OK as long as you are not experiencing hiss (due to higher gain) and you understand that it is possible to clip the 705P amplifier if the miniDSP output level exceeds -16 dB.

I also assume that you have not adjusted any other settings in the 705Ps like input trim, PEQs, etc?

Michael

 

[mdsimon2]

Yes but if, say, an 86dB/watt speaker has a 10dB dip in its in room FR and the software corrects for this the speaker will need the same power to achieve the corrected FR as a 76dB/watt speaker at that frequency would if it did not need compensation, which will be a lot, even if it can take it.
I know the correction has to reduce level before manipulation in order to avoid clipping the digital system but the analogue amp having to supply the power to achieve a correction of 10dB really is 100x the power and it is very likely not to be technically achievable, either because the speaker can't take that much or the amp can't supply it.
It may be able to cope at very low levels.

I suppose that is so why many advise not to try to correct dips.

Might want to check your math on the relationship between dB and power. +10 dB is 3.16X on voltage and therefore 10X on power, not 100X. I agree that a 10 dB boost is still a lot though and likely results in your overall listening level being lower to avoid amplifier clipping.

Michael

 

[Frank Dernie]

Might want to check your math on the relationship between dB and power. +10 dB is 3.16X on voltage and therefore 10X on power, not 100X. I agree that a 10 dB boost is still a lot though and likely results in your overall listening level being lower to avoid amplifier clipping.

Michael

Oops, yes. 10x the power, my mistake in an earlier post but the post you are replying to here is correct isn't it?

 

[SadMonster]

Hmmmm actually I’ve been listening to completely flat and all sorts of other curves and I‘m not really convinced I like any better than flat right now

 

[mdsimon2]

What subs are you using? I agree with others that you should use the miniDSP as the central routing point and not route to the subs first. I see no reason why you cannot implement "stereo bass" with the miniDSP.

The subs may have a limit on input voltage that further complicates your gain structure / input sensitivity issues.

Michael

 

[SadMonster]

What subs are you using? I agree with others that you should use the miniDSP as the central routing point and not route to the subs first. I see no reason why you cannot implement "stereo bass" with the miniDSP.

The subs may have a limit on input voltage that further complicates your gain structure / input sensitivity issues.

Michael

To be honest I don’t really know how I would use the subwoofers with the minidsp. Wouldn’t know how to set crossovers and stuff…

I have JBL LSR310s

Currently best liking 4-6DB bass boost with no mid or hf tilt

 

[abdo123]

perhaps you can start by sharing with us which model of MiniDSP you're using.

 

[mdsimon2]

To be honest I don’t really know how I would use the subwoofers with the minidsp. Wouldn’t know how to set crossovers and stuff…

I have JBL LSR310s

Currently best liking 6DB bass boost with no mid or hf tilt

Looks like those have straight analog pass through so it should be fine but the manual is unclear (for example does the input sensitivity setting change the output behavior?). miniDSP has some decent application notes on setting up crossovers and we can definitely help you out as well. Can you confirm you have a DDRC-24?

I run stereo subs myself with a miniDSP and although I no longer use Dirac I like the idea of using stereo subs with a 2 channel Dirac setup as it can apply more dedicated correction to each sub.

Here are the miniDSP application notes.

https://www.minidsp.com/applications/digital-crossovers/subwoofer-integration-with-minidsp

https://www.minidsp.com/applications/digital-crossovers/stereo-2-way-xover

The subwoofer integration app note shows a mono bass setup but it is easy to change this in the routing matrix.

Michael

 

[raindance]

Wait, you aren't high-passing your speakers at 80Hz? That is the entire point of owning subwoofers, to take the load off your speakers.

Explain this? Isn't the mini DSP doing this high pass already? What would you gain?

But I think you have it figured out. The mini DSP units are challenged for headroom at the best of times, so it takes a bit of fiddling to get it right. Typically they introduce some loss, so compensating by changing the speaker sensitivity seems like a winner.

 

[SadMonster]

Alright for now I have decided to be an objectivist. As there is no curve that can suit all sources or preferences and I don’t want to fiddle with tone controls based on content.

I am aware some studies have been done that suggest there are some general curve guidelines that see to increase preference ratings across the board but I have decided to not try to make things sound “good” but only “accurate” for better or worse and view the audio system as not something meant to bring joy but to be used as a clinical reproduction tool.

I might update soon with my minidsp details in case I can improve reproduction quality by having my minidsp select optimal crossover and stuff. Can dirac do this?

 

[SadMonster]

Agree that this is likely a combination issue related to DSP, amplifier input sensitivity and volume expectations.

Which miniDSP product is this? SHD, DDRC-24, DDRC-88A? I am guessing DDRC-24 as that is probably the most common and most likely to have volume issues due to lower output voltage.

There is a lot of good information in this thread but there is also a lot of "noise".

DDRC-24 has an output voltage of 2V, this will be achieved when output level is at 0 dB. If output level ever goes above 0 dB DDRC-24 will clip. It does not matter how this 0 dB is reached, you can program +50 dB of boost from a combination of input gain, output gain, PEQs and Dirac and if you set the volume control to -50 dB the output will be at 0 dB and not clip. If you look at the miniDSP output levels you can see how close you are to 0 dB and clipping. If your output levels are below 0 dB your distortion problems are not caused by the miniDSP regardless of settings in the miniDSP.

+4 dBu input sensitivity setting on 705Ps corresponds to a voltage level of 1.23V. This means that that the DDRC-24 should be able to drive the 705Ps to full power at this input sensitivity setting. However the actual output voltage from the DDRC-24 will depend on the level of the recording, the processing in the DDRC-24 and the DDRC-24 volume level.

-10 dbV input sensitivity setting on 705Ps corresponds to a voltage level of 0.32V. In this case the DDRC-24 will almost certainly be able to drive the 705Ps to full power (and beyond in to clipping). This setting essentially offers +10 dB gain compared to the +4 dBu setting. Using this setting with the DDRC-24 is OK as long as you are not experiencing hiss (due to higher gain) and you understand that it is possible to clip the 705P amplifier if the miniDSP output level exceeds -16 dB.

I also assume that you have not adjusted any other settings in the 705Ps like input trim, PEQs, etc?

Michael

Thank you so much for this post!!

- So the dirac correction never exceeds +10db at anytime.
- I have 0 other filters on.
- The JBL 7 series are set to the louder gain setting and it does get slightly more hiss this way but I have to deal with it.
- I have the MiniDSP DDRC24 set to -10db on the outputs to prevent any clipping from the dirac corrections. Is this correct or can I set it to 0 to get more volume? Sometimes some random shows or movies will be really quiet for some reason ;[

 

[mdsimon2]

Thank you so much for this post!!

- So the dirac correction never exceeds +10db at anytime.
- I have 0 other filters on.
- The JBL 7 series are set to the louder gain setting and it does get slightly more hiss this way but I have to deal with it.
- I have the MiniDSP DDRC24 set to -10db on the outputs to prevent any clipping from the dirac corrections. Is this correct or can I set it to 0 to get more volume? Sometimes some random shows or movies will be really quiet for some reason ;[

I am glad it was useful.

If you are using the DDRC-24 for volume control I would remove the -10 dB on the outputs. While it is possible you could have clipping in the DDRC-24 with the outputs set to 0 dB as you mentioned certain sources have much lower levels than typical and having that extra 10 dB of output is helpful. Again you can always look at the output levels in the DDRC-24 plugin to see what your actual output levels are. If they are below -10 dB you will not have clipping in the DDRC-24 itself and if they are below -16 dB you will not have clipping in the JBL amplifier. Ultimately if it sounds good to you I would not worry about clipping.

Michael

 

[SadMonster]

I am glad it was useful.

If you are using the DDRC-24 for volume control I would remove the -10 dB on the outputs. While it is possible you could have clipping in the DDRC-24 with the outputs set to 0 dB as you mentioned certain sources have much lower levels than typical and having that extra 10 dB of output is helpful. Again you can always look at the output levels in the DDRC-24 plugin to see what your actual output levels are. If they are below -10 dB you will not have clipping in the DDRC-24 itself and if they are below -16 dB you will not have clipping in the JBL amplifier. Ultimately if it sounds good to you I would not worry about clipping.

Michael

THANKS AGAIN!!! I do use the DDRC24 to control the volume.

Is the "-xdBFs" under the output sliders where I have to keep it -16? Even at -10 on music it sometimes goes above 15 v.v

 

[mdsimon2]

THANKS AGAIN!!! I do use the DDRC24 to control the volume.

Is the "-xdBFs" under the output sliders where I have to keep it -16? Even at -10 on music it sometimes goes above 15 v.v

Yes, exactly. Again, if it sounds good to you I would not worry about it, especially if you are below -10 dbFS and not seeing the clipping indicator on the 705P.

Michael

 

[SadMonster]

Yes, exactly. Again, if it sounds good to you I would not worry about it, especially if you are below -10 dbFS and not seeing the clipping indicator on the 705P.
View attachment 133253

Michael

I have literally never seen the red clipping light on the speakers. I think you'd have to be really far from the speakers to get them to hit clipping without causing permanent hearing loss o_o

 

[Ron Texas]

Could it be that Dirac isn't any better than REW and some human intervention?

 

[SadMonster]

Could it be that Dirac isn't any better than REW and some human intervention?

Would need a smart human and even still it would be more work I think

 

[Ron Texas]

Would need a smart human and even still it would be more work I think

Thanks for the compliment. You must prefer the magic box approach.

 

[Soundmixer]

Stereo bass isn't really a thing. Below 100Hz our ears are omni-directional, you cannot distinguish the left/right subwoofer based on placement. They might sound different based on room modes, but that is a bad thing, not a good thing.

Actually, it is not our ears that are omnidirectional, the bass from the speaker(s) is.

 

[RCAguy]

You were definitely right about the bass boost being better than flat! Interestingly when I set the subwoofer volume and crossover by ear before measuring - the result came pretty damn close to the harman curve!! But I think I might be a mutant because I have always been able to tell where the subwoofer is

You were definitely right about the bass boost being better than flat! Interestingly when I set the subwoofer volume and crossover by ear before measuring - the result came pretty damn close to the harman curve!! But I think I might be a mutant because I have always been able to tell where the subwoofer is