• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Neumann KH80 DSP Monitor Measurements #3

BYRTT

Addicted to Fun and Learning
Forum Donor
Joined
Nov 2, 2018
Messages
956
Likes
2,452
Location
Denmark (Jutland)
hey amir, im just wondering, would you be able to post the PIR slope graph for #3?

In below attached zip-folder there's a bunch of txt-files for Amir's analyze of KH 80 sample 3 including the PIR curve, rename extension to frd if software dont like txt-files :)..

matt3421.PNG
 

Attachments

  • matt3421.zip
    79.5 KB · Views: 109

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
thanks. is there compatible software available for Mac?

REW works fine in Windows -- maybe it works fine in a mac(?) I can just drag and drop the files. For some reason it does not like opening text files directly.
 

napilopez

Major Contributor
Forum Donor
Joined
Oct 17, 2018
Messages
2,109
Likes
8,420
Location
NYC
You can open them in Room EQ Wizard. You should probably get one of the more recent betas for easy drag and dropping.

Note however that the files @BYRTT shared use a corrected calculation for the PIR. The Klippel NFS calculates the PIR incorrectly due to ambiguity in the spinorama standard document. The differences usually aren't ginormous though ( and the preference score calculations use the proper PIR), but worth noting.

Below, in red is the PIR as calculated by the NFS, in blue is the correct PIR per Harman:

Snag_bb67df3.png
 

BYRTT

Addicted to Fun and Learning
Forum Donor
Joined
Nov 2, 2018
Messages
956
Likes
2,452
Location
Denmark (Jutland)
thanks. is there compatible software available for Mac?
REW works fine in Windows -- maybe it works fine in a mac(?) I can just drag and drop the files. For some reason it does not like opening text files directly.
As ernestcarl suggest thanks :) get REW for Mac and in it sounds you a first time user it mayby can help a bit see below presentation of the 7 times txt-files from zip-folder that will form you a spinorama graph summary, note in that ERDI Total & Directivity index curves is presented down low at zero dB then in below presentation i have used REW's feature to offset them +55dB to get them lifted up :)..


matt3421_2.PNG
 

usern

Senior Member
Joined
May 31, 2020
Messages
487
Likes
499
I did not run distortion tests then. I did run them for the KH310A I am testing now. And since I own a KH80DSP, I can go back and capture that at some point.
I would like to see KH80DSP distortion tests and compare them with newly released Rokit 5 Gen 4 results.
 

Newman

Major Contributor
Joined
Jan 6, 2017
Messages
3,449
Likes
4,209

Matias

Master Contributor
Forum Donor
Joined
Jan 1, 2019
Messages
5,029
Likes
10,794
Location
São Paulo, Brazil
I borrowed a pair of these from a friend to test them out. I really liked them! I will probably buy myself a pair.

Below I did a quick sweet spot measurement and EQ correction for each individually. They are in lowest level and gain, using Small Desk setting, and hiss is inaudible at arms length.

IMG_20210314_220843.jpg


IMG_20210314_221642.jpg




Real life is a lot messier than the nice flat anechoic curves that the monitors have...

kh80dsp.jpg




And below I set 1 EQ band manually to up the bass to my target, and let REW calculate the remaining bands. The response is a LOT better now.


80L EQ.jpg


80R EQ.jpg



It really is a pleasure to listen to music with these small speakers. Of course bass extension is limited, and I don't want to add the bulk of a subwoofer in my office. But other than that, the imaging and resolution is really pleasant. Highly recommended! :)
 
Last edited:

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
Real life is a lot messier than the nice flat anechoic curves that the monitors have...

Hopefully, you are not EQing using single static measurements. I strongly advise using MMM or an average of multiple sweeps, say, 9 or even up to 30+ per speaker if you have a larger listening area!

The "transfer function is defined by the position of the source and the listener" so the room and setup placement will always have a marked effect to what is measured.
 
Last edited:

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,737
Likes
16,159
And below I set 1 EQ band manually to up the bass to my target, and let REW calculate the remaining bands.
Mind you that a +9 dB! at 65 Hz PEQ will increase the distortion, IMD and limit the SPL a lot in such small and already bass SPL limited loudspeakers.
Also I would remove hard reflecting surfaces like the chair when measuring and like @ernestcarl writes better use MMM especially when you correct above transition frequency (which I personally might consider not doing at such anechoically flat loudspeakers). https://www.audiosciencereview.com/...ut-room-curve-targets-room-eq-and-more.10950/
 

Joaquinín

Member
Joined
Sep 10, 2020
Messages
68
Likes
45
I have just measured my pair of Kh-80 at my desk (90 cm equilateral triangle) with Dirac Live, and there is one oddity. The measured impulse response of both speakers is phase inverted. Has anybody checked this with their Kh80? May be my DAC (Gustard X16) is inverting the signal?
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
Hopefully, you are not EQing using single static measurements. I strongly advise using MMM or an average of multiple sweeps, say, 9 or even up to 30+ per speaker if you have a larger listening area!

The "transfer function is defined by the position of the source and the listener" so the room and setup placement will always have a marked effect to what is measured.

why? it is a near field setup, so only one spot matters.

However, it's probably not a good idea to EQ a flat speaker so viciously like that.
 

Matias

Master Contributor
Forum Donor
Joined
Jan 1, 2019
Messages
5,029
Likes
10,794
Location
São Paulo, Brazil
@ernestcarl, @thewas, thank you for your input. Indeed I measured in a single position, single pass for each speaker. I am a "head in a vice" listener, that is, I stay still in the center sweet spot when to listening to music. So I assume that averaging over an area may not be so relevant in my case? (correct me if I am wrong)

Also I don't plan to listen loud in this office system, for that I have my main system, and even there I usually don't listen loud. So max SPL for these KH 80 DSP is not really an issue. I adjusted -9dB on a gain before the EQ, and the speakers have limiting, so I am not worried about clipping. Just for testing, I tried maximizing the volume of the RME ADI-2 DAC, using the lowest gains in the KH 80 DSP as I wrote before, and even listening to bass heavy electronic dance music, still it was loudish enough for me (should have measured how many dBs, I forgot, may do so later). But the point is: I rather have a fuller sounding bass at mid to low levels than worrying about loud volume distortion. Good enough compromise for me. :)

And finally, I checked by ear playing around with sine waves and the frequency slider while seated in the sweet spot, sounds pretty linear to me, so I am happy.
 
Last edited:

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
why? it is a near field setup, so only one spot matters.

We have two ears on a fully mobile head and body even if there were only a single seat to consider. It is very easy to over-EQ frequencies above Schroeder if only one spot is measured. The fact that there is such a huge discrepancy in the high frequencies between the ultra flat anechoic response of the KH80 and the single-point measurements at the listening position should give one pause.
 

abdo123

Master Contributor
Forum Donor
Joined
Nov 15, 2020
Messages
7,423
Likes
7,940
Location
Brussels, Belgium
We have two ears on a fully mobile head and body even if there were only a single seat to consider. It is very easy to over-EQ frequencies above Schroeder if only one spot is measured. The fact that there is such a huge discrepancy in the high frequencies between the ultra flat anechoic response of the KH80 and the single-point measurements at the listening position should give one pause.

I agree, he is EQing the poor things to death. However, the room continues to influence the speakers till 2 KHz, the shroedenger frequency is the boundary in which below it you hear the room and the room only.
 

Matias

Master Contributor
Forum Donor
Joined
Jan 1, 2019
Messages
5,029
Likes
10,794
Location
São Paulo, Brazil
In a few weeks I will try to measure in a small cube say 20 cm (8") around the sweet spot to check if it changes much. Thanks again.
 

ernestcarl

Major Contributor
Joined
Sep 4, 2019
Messages
3,106
Likes
2,313
Location
Canada
Extract the files posted by BYRTT above and apply the EQ to the LW just to see how much the auto-generated EQ mangles the original flat HF of the KH80.

1615898418828.png


Just think about it.
 

dominikz

Addicted to Fun and Learning
Forum Donor
Joined
Oct 10, 2020
Messages
803
Likes
2,626
In a few weeks I will try to measure in a small cube say 20 cm (8") around the sweet spot to check if it changes much. Thanks again.
I'd also advise caution when EQing above the transition frequency, as a lot of the deviation (vs anechoic) you may measure will be a result of reflections, which the ears and brain can separate from the flat direct sound - i.e. EQ will improve the measured graph, but not necessarily perceived SQ.
In my experience the easiest and best sounding results with REW are obtained by using MMM for measurement, then applying variable smoothing, and lastly EQing to a desired target with cut-only filters (and for nice speakers limited to below the transition frequency). In some cases certain severe dips in the response require boost filters too, but that should be added in moderation, after listening and experimentation.

Actually @thewas opened my eyes to how well suited and effective MMM is for EQ corrections Before that I was always struggling with single and averaged multi-point sweep measurements and often got less-than-satisfying results after EQ (and consumed much more time doing so)!
 
Top Bottom