• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). Come here to have fun, be ready to be teased and not take online life too seriously. We now measure and review equipment for free! Click here for details.

Harman target curve calculator

Snarfie

Addicted to Fun and Learning
Forum Donor
Joined
Apr 30, 2018
Messages
552
Likes
299
Location
Netherlands
#1
I was wondering is there a Harman target curve (HTC) calculator. I guess the HTC is a known calculation. Enclosed a website with HTC's https://mehlau.net/audio/dirac-live-2/
where you can download target curvers based on HTC form 10 to 4 db guess what i need a 2 db curve. Did try to figuer out the similarities/calculation between the curves but could not find a conclusive answer.
Anybody how to calculate a 2 db HTC. Attached also the target curves where you can find the data..
 

Attachments

Last edited:
OP
Snarfie

Snarfie

Addicted to Fun and Learning
Forum Donor
Joined
Apr 30, 2018
Messages
552
Likes
299
Location
Netherlands
Thread Starter #3
The one I use, and it has been discussed elsewhere around here is +2 at 20 hz, -8 at 20khz. You could take the 4db curve and halve the boost or cut.
I'm using Mathaudio room eq. This week a new version came out where it is possible to add target curve's. For that i need a .txt file.
If i'm correct you suggesting to take the 4db curve an halve the data so i can add it into mathaudio room eq?.
 

Ron Texas

Major Contributor
Joined
Jun 10, 2018
Messages
3,023
Likes
3,227
Location
A mysterious place with no name.
#4
I'm using Mathaudio room eq. This week a new version came out where it is possible to add target curve's. For that i need a .txt file.
If i'm correct you suggesting to take the 4db curve an halve the data so i can add it into mathaudio room eq?.
Yes. You might have to clean up comments at the top and bottom. I'm not sure why you want 2db. Search for house curve on the board to see some of the suggestions. Mathaudio's built in house curve is similar to the B&K curve used by many recording studios.
 
OP
Snarfie

Snarfie

Addicted to Fun and Learning
Forum Donor
Joined
Apr 30, 2018
Messages
552
Likes
299
Location
Netherlands
Thread Starter #5
Yes. You might have to clean up comments at the top and bottom. I'm not sure why you want 2db. Search for house curve on the board to see some of the suggestions. Mathaudio's built in house curve is similar to the B&K curve used by many recording studios.
I was using the 4db curve but it was still to loud(the low's). Basicly my mid's an high's reveal with certain music less like Frank Sinatra live at the sands the timbre in his voice is for 50% gone (with the 4db curve) but the lows where just fine so i thought of a compromise using a 2db curve. Also music from the 70ties like Al Stewart - Year of the Cat cleans up a lot with a 4 db curve this is all quite subjective from my part also considering my specific room acoustics.
 
Last edited:
OP
Snarfie

Snarfie

Addicted to Fun and Learning
Forum Donor
Joined
Apr 30, 2018
Messages
552
Likes
299
Location
Netherlands
Thread Starter #7
The B&K curve has no bass boost.
The target curve's for my taste works better OR worse with certain music/recordings. In average (till now i'm stille testing) i like the sound without target curve's using Mathaudio. Could be that the 2db boost could make a difference.
 
Joined
Mar 14, 2016
Messages
234
Likes
157
Location
Corpus Christi, TX
#8

Blumlein 88

Major Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
9,860
Likes
13,317
#10
I've found curves flat to 200 hz and then with the gentle roll off to work better for me.
 

Vovgan

Member
Forum Donor
Joined
Jan 9, 2019
Messages
92
Likes
160
Location
Moscow, Russia
#11
I'm using Mathaudio room eq
I started using Mathaudio via Audio Hijack on my Mac yesterday. I’ve soon noticed that even a slight bass correction at 50 and 70 Hz (where I have two small peaks) make the sound and in particular vocals sound duller. Have you noticed something similar?
 
OP
Snarfie

Snarfie

Addicted to Fun and Learning
Forum Donor
Joined
Apr 30, 2018
Messages
552
Likes
299
Location
Netherlands
Thread Starter #12
I started using Mathaudio via Audio Hijack on my Mac yesterday. I’ve soon noticed that even a slight bass correction at 50 and 70 Hz (where I have two small peaks) make the sound and in particular vocals sound duller. Have you noticed something similar?
No not duller. The acoustics of my room let mids en highs dominante the sound so Barry White voice sounds like a highhat. When using Mathaudio the voice sounds dark as it shoud be. Ik don't have to make any corrections despite i (also) have peaks.
 

Vovgan

Member
Forum Donor
Joined
Jan 9, 2019
Messages
92
Likes
160
Location
Moscow, Russia
#13
No not duller. The acoustics of my room let mids en highs dominante the sound so Barry White voice sounds like a highhat. When using Mathaudio the voice sounds dark as it shoud be. Ik don't have to make any corrections despite i (also) have peaks.
Thanks for your reply! In my room and with my speakers (B&W 703 S2) even a tiny correction makes the sound duller...

Sorry, didn't understand your last sentence: do you mean that you're not using Mathaudio to make corrections despite having peaks? Then what are you using it for? Isn't applying a target curve in Mathaudio is cutting peaks and that's it?
 

Attachments

OP
Snarfie

Snarfie

Addicted to Fun and Learning
Forum Donor
Joined
Apr 30, 2018
Messages
552
Likes
299
Location
Netherlands
Thread Starter #14
Thanks for your reply! In my room and with my speakers (B&W 703 S2) even a tiny correction makes the sound duller...

Sorry, didn't understand your last sentence: do you mean that you're not using Mathaudio to make corrections despite having peaks? Then what are you using it for? Isn't applying a target curve in Mathaudio is cutting peaks and that's it?
It looks like you are using no target curve (you are using the bypass signal). Your green line (which is white in Windows) has to be located under the greyline at a level that suits you.
Have a look at mine curve looks like we have more or less the same sort of full range speakers.
The fitrst one has no curve it is neutral or bright. I Favour this curve above the second picture that shows a Harman 2db curve. Using this second curve alters the information in my mid an high range. For instance details in voice's ar less but the whole sound picture is more impressive. Basicly i see using target curves as a trade off between more impressive sound or detaild sound.
Try to pull your green line down to -15 db. If you want to compare the green line with the bypass sound your sound level for both have to be the same for the best result.


 
Last edited:

Vovgan

Member
Forum Donor
Joined
Jan 9, 2019
Messages
92
Likes
160
Location
Moscow, Russia
#15
It looks like you are using no target curve (you are using the bypass signal).
The radio button was accidentally in the “bypass” mode because I was experimenting with turning the curve on and off. I do have a curve, but it just knocks off a couple of dBs off the peaks at 70 Hz in the left and right channels, whereas your curve effectively flattens frequency response across the spectrum.

After reading your post I too tried this radical equalization and after volume correction it did sound good. However, I still liked the sound without Mathaudio better for at least half of the tracks I listened to, and knocking off just a few dBs off the peaks on other tracks sounds better to my ears than equalizing the whole spectrum.

Looking at your graphs it seems that bass response of your speakers is insufficient in your room, have you tried adding a sub before equalization?
 
OP
Snarfie

Snarfie

Addicted to Fun and Learning
Forum Donor
Joined
Apr 30, 2018
Messages
552
Likes
299
Location
Netherlands
Thread Starter #16
The radio button was accidentally in the “bypass” mode because I was experimenting with turning the curve on and off. I do have a curve, but it just knocks off a couple of dBs off the peaks at 70 Hz in the left and right channels, whereas your curve effectively flattens frequency response across the spectrum.

After reading your post I too tried this radical equalization and after volume correction it did sound good. However, I still liked the sound without Mathaudio better for at least half of the tracks I listened to, and knocking off just a few dBs off the peaks on other tracks sounds better to my ears than equalizing the whole spectrum.

Looking at your graphs it seems that bass response of your speakers is insufficient in your room, have you tried adding a sub before equalization?
Yes it is radical but neccecary the mids an highs (around 300 & 18000 Hz) are way to loud sometimes almost 15db so i had to correct this. By correcting them the bass comes alive an is tight no subwoofer needed. The only problem is that you need a damm good amplifier that can cope with this -15db difference for that i changed my ampilfier for the NAD C370 an amp with enough juice. But lets have a look at your speakers found this freq respons lab measurment of your B&W's.



Source: https://www.avhub.com.au/product-reviews/hi-fi/bw-702-s2-loudspeakers-review-test-512096 Compared to your measurments it looks way of with the lab measurment probably done in a anechoic chamber. IMO your green line have to be around -15db's to overcome your specific room acoustic inefficencies just as i have done.

This is what the engineers of your speakers aiming at i suppose a flat respons (± 3,5db) as possible for the best results in perfect conditions. With Mathaudio and or REW you can come close to this result. But still personal taste is in the ear of the beholder/listner;) What type/brand measuring mic are you using.? Did you do a atleast a nine point measurement.? It also help to use the High resolution measurment in stead of the normal resolution measurment. Your mids & highs will sound a bit more bright in high resolution (a bit less dull mabey).
 
Last edited:

Absolute

Addicted to Fun and Learning
Forum Donor
Joined
Feb 5, 2017
Messages
536
Likes
859
#17
If I understand @Floyd Toole correctly, the Harman curve is not a result of a tuned curve after EQ, but rather the approximate resulting natural steady-state curve in a normal reflective room when neutral speakers with smooth on and off-axis is used. Please correct me if I'm mistaken.
 
OP
Snarfie

Snarfie

Addicted to Fun and Learning
Forum Donor
Joined
Apr 30, 2018
Messages
552
Likes
299
Location
Netherlands
Thread Starter #18
If I understand @Floyd Toole correctly, the Harman curve is not a result of a tuned curve after EQ, but rather the approximate resulting natural steady-state curve in a normal reflective room when neutral speakers with smooth on and off-axis is used. Please correct me if I'm mistaken.
Yes this is what i understand too an the B&K curve does more or less the same if i'm correct. An if i use those curves it sound more impressive comfortable. But if i listen more specific to voices for instance i notice that there is a degradation in clarity/information timbre in Frank Sinatra's voice for instance. I have the live cd/recording of Frans Sinatra at the Sands if you use the harman curve it will let sound the orchestra more impressive but the timbre in the voice of Sinatra is for 70% gone. On the other hand if i use the harman curve with Al Stewart - Year of the Cat the whole album comes a live. So basicly using Harman curves (or any other curve) is a trade off IMO it works for some recording better or worse it is quite subjective taking in acount your specific roomacoustic's and or your personal taste. In 90% of the music i listen too the flat/neutral curve as you see here under is my favorit target curve making use of Roomcorrection.
 
Last edited:
Top Bottom