What is your favorite house curve

[Matias]

I chose my target to be similar to Toole's curve. It's a linear tilt of about 8 dB or 0.8 dB/octave from 20 to 20 kHz for each speaker individually.

Below are measurements (UMIK-1 MMM) before and after the EQ settings showing how effective it was.

It is quite similar to Genelec 8341A estimated in room response, only mine extends more in the bass region, but both share about 6 dB tilt from 100 to 20 kHz.

Here is the living room:

And here is the office:

 

[database]

My Dirac target:

I spent about a week on this and did listening tests with around 20 iterations of target curves, with a specific focus on treble. I didn't spend as much time on the bass, I started with the target curve as shown, which was recommended to me by a user that believed most curves, including the old curves I was running, boost too much midbass which sounds less natural. From there I just increased sub trim by 4-5 dB until it sounded balanced to me, and the result immediately sounded better than any other target curves I've run in the past, including Dirac's default curve and several Harman curves. I had 4 listeners, young and old, verify that the final curve sounded significantly more satisfying and realistic compared to Dirac off, especially in the treble region, in both music and movies.

At first I was tuning the target curve mainly for female vocals, to remove the their edge and make them sound as natural as possible. I had a few test songs and got these to sound good by using a heavy treble rolloff, but the issue was apparent when listening to other genres like rock and metal. With too much treble rolloff, rock and metal loses too much energy and it sounds like everything is submerged underwater. So I adjusted the rolloff to remove the edge from female vocals without dulling rock/metal, and this got me close. But the next issue showed up when I watched movies. Human voice frequencies extend as high as 10 kHz (from 's' and 't' sounds, for example), and with too much rolloff in this region, male and female voices sounded veiled, as if everyone was talking behind a thick curtain. So my final target curve was adjusted to remove the edge from female vocals, but made sure to retain the energy from rock/metal and also keep male and female vocals as clear and crisp as possible. Once I finally reached I target that worked well for all of these cases, I had the results verified by 4 listeners who confirmed a significant improvement in all of these types of material.

I've mentioned in another comment before that most users I've seen have reported that Dirac worked best for them when run full range, despite some theories that assert that EQ should be avoided above the Schroeder frequency. My experience is one of many where customizing the Dirac target curve and giving it a rolloff in the high frequencies, designed specifically to what sounded correct in my room, was absolutely necessary to get all content to sound crisp, clear, natural, and life-like.

 

[Tim Link]

I want thank everybody contributing to this thread, sharing their thoughts and the results of their professional research and personal experience. It has been very informative for me, correcting some conceptual errors I had, and most importantly allowing me to get my speakers sounding incredibly good to my ears. I built some very large corner horns with a friend a couple years back. We opted for a coaxial tweeter/midhorn combination that proved impossible to control for smooth off axis response. I knew going in that it would have dispersion issues but didn't think they would bother me much. Oh but they did. I finally gave up on the coaxial arrangement after trying all sorts of house curves, crossover settings and absorption schemes around the mouth of the tweeter. Now I'm using a separate, much larger horn for the tweeter (BL-409 from Parts Express. You can see my review there.). That allows for a much smoother, down tilting off axis response and cleaner impulse. The attached image is measurements taken 5 feet back on axis with the tweeter, on the floor, and right next to the wall. The off axis measurements are both about 45 degrees. I had to change the driver from the D202ti to the JBL 2426 to really get the 800hz crossover to work but now it sounds fantastic. That horn and driver work really well together. I don't care for its looks much, but it sounds lovely to me.
I am not using a target house curve, just getting the direct on-axis as flat as I can without benefit of a real anechoic measuring environment. I am using room correcting EQ in the bass after watching Earl Geddes Youtube video on using multiple woofers and eq to get smoother bass response over most of the room. What I've got is not quite what he had in mind because my woofers are all stacked in the two front corners, but it still seems to work very well. I'm exceedingly pleased.

 

[Robbo99999]

I've recently bought JBL 308Mkii speakers for a 2 channel music & movie/TV experience, and after experimentation and measurement I use a full on Harman Curve for music listening:

And then I use a "Half Harman Curve", which is half the bass boost and half the treble loss of a normal Harman Curve for Movie & TV watching (I since surmise movies/TV are recorded with more bass & less treble than music recordings, I used the movie Prometheus as my reference track to base my testing around, as I think the sound is excellent and also 'modern/forward thinking'):

....but it was also a toss up with the following flat curve with slight downward tilt after 1K to match natural high frequency roll off of my speakers:, (but with the above Half Harman Curve being slightly better for greater bass extension into the lower frequencies)

I think it's always gotta be a downward curve of some description, but how much is the question/choice!


EDIT: I only recently created these curves, and I'm watching a film now "Bumblebee" and there's way too much bass and too muffled using the Half Harman Curve, so I've switched to the "Flat curve" (last graph) for watching that. I think I'll be switching between the last two graphs for movie watching depending on how the source material has been recorded, they obviously don't all follow the same curve for their movie production process. I'm fairly confident that flipping between my "Flat Curve" and the "Half Harman Curve" will cover all bases in terms of films...means I just have to flip between curves by loading saved profiles in miniDSP.

P.S. I added 5dB Low Shelf Boost at 110Hz prior to measurement if you're curious about the natural frequency response of my speakers, I did that to help eliminate bass holes, and these speakers are bass light in my room generally.

 

[Senior NEET Engineer]

I like the curve from my room

 

[RayDunzl]

I'm exceedingly pleased.

Is it really that loud in there?

 

[QMuse]

EDIT: I only recently created these curves, and I'm watching a film now "Bumblebee" and there's way too much bass and too muffled using the Half Harman Curve, so I've switched to the "Flat curve" (last graph) for watching that. I think I'll be switching between the last two graphs for movie watching depending on how the source material has been recorded, they obviously don't all follow the same curve for their movie production process. I'm fairly confident that flipping between my "Flat Curve" and the "Half Harman Curve" will cover all bases in terms of films...means I just have to flip between curves by loading saved profiles in miniDSP.

P.S. I added 5dB Low Shelf Boost at 110Hz prior to measurement if you're curious about the natural frequency response of my speakers, I did that to help eliminate bass holes, and these speakers are bass light in my room generally.

You can't make any conclusions from predicted response curves, you need to re-measure again with acitve filters to see what is really happening.

 

[Hephaestus]

I have this kind of a curve in my small office. Pair of 8341A, no subs, no acoustic treatments. Few dips here and there but I have been enjoying these very much.

Thanks to this forum I found properly engineered speakers and learned how to eq sensibly.

 

[Robbo99999]

You can't make any conclusions from predicted response curves, you need to re-measure again with acitve filters to see what is really happening.

Yeah, good point, but I've done that before for previous EQ curves with these speakers in this room and it's been very accurate, so I don't bother to re-measure again after seeing the predictive EQ's. For example here is an actual measurement after EQ had been applied when my speakers and listening position were in a different position to my current setup (with my current listening position providing a better response than the one posted below), and this was essentially identical to the predicted EQ response which can also be seen as to how well the actual measurments are hugging the target:

 

[QMuse]

Yeah, good point, but I've done that before for previous EQ curves with these speakers in this room and it's been very accurate, so I don't bother to re-measure again after seeing the predictive EQ's. For example here is an actual measurement after EQ had been applied when my speakers and listening position were in a different position to my current setup (with my current listening position providing a better response than the one posted below), and this was essentially identical to the predicted EQ response which can also be seen as to how well the actual measurments are hugging the target:
View attachment 58579

Ok, I see. This curve should sound fine.

 

[Dimifoot]

I think it's always gotta be a downward curve of some description, but how much is the question/choice!

7-10db’s

 

[QMuse]

I think it's always gotta be a downward curve of some description, but how much is the question/choice!

Pick one of these and you'll do fine:

 

[Robbo99999]

7-10db’s

Pick one of these and you'll do fine:

View attachment 58586

Yeah, I saw those curves earlier. In my experience I like the flatter curves for Movie & TV, whereas I prefer the full on Harman Curve for music listening, as I said earlier. The 7-10dB downslope you talk of Dimifoot, that applies to my Harman Curve for music listening. But for TV/Movie the flat curve I made which is only about 2dB loss seems the most universally applicable to various TV & Movies, whereas the Half Harman Curve I created which is 5dB loss is sometimes the best for some Movies, but for others the Flat Curve I did is better and also sound pretty great on even the movies where the Half Harman Curve is the best....so the Flat Curve is kinda the "set it & forget it" curve for TV & Movie, but I'll be switching it to Half Harman Curve if a movie is noticeably lacking bass or is too bright. I watch movies broadcast over terrestrial TV here in the UK and also Amazon Prime, so perhaps that format says something about the curves I created, I don't know.

 

[Dimifoot]

Yeah, I saw those curves earlier. In my experience I like the flatter curves for Movie & TV, whereas I prefer the full on Harman Curve for music listening, as I said earlier. The 7-10dB downslope you talk of Dimifoot, that applies to my Harman Curve for music listening. But for TV/Movie the flat curve I made which is only about 2dB loss seems the most universally applicable to various TV & Movies, whereas the Half Harman Curve I created which is 5dB loss is sometimes the best for some Movies, but for others the Flat Curve I did is better and also sound pretty great on even the movies where the Half Harman Curve is the best....so the Flat Curve is kinda the "set it & forget it" curve for TV & Movie, but I'll be switching it to Half Harman Curve if a movie is noticeably lacking bass or is too bright. I watch movies broadcast over terrestrial TV here in the UK and also Amazon Prime, so perhaps that format says something about the curves I created, I don't know.

I am the other way round. I prefer more slope for movies, (more bass/less treble).

 

[Robbo99999]

I am the other way round. I prefer more slope for movies, more bass-less treble.

That's strange isn't it. I'm guessing we both like the Harman Curve for music? You'd think if two people liked the same curve for music then the same would follow for movies. I think movies are recorded with a lot less treble than most music, hence me liking the flat curves more for movies, that's my theory. A lot of the "SciFi sparkle" of effects is lost if you use Harman in movies, that's what I've found, also dialogue is less intelligible.

 

[QMuse]

That's strange isn't it. I'm guessing we both like the Harman Curve for music? You'd think if two people liked the same curve for music then the same would follow for movies. I think movies are recorded with a lot less treble than most music, hence me liking the flat curves more for movies, that's my theory. A lot of the "SciFi sparkle" of effects is lost if you use Harman in movies, that's what I've found, also dialogue is less intelligible.

Movies soundtracks typically have louder HF as theaters absorb more HF than rooms.

 

[Dimifoot]

That's strange isn't it. I'm guessing we both like the Harman Curve for music?

Yes, my preferred target is similar to the Harman, a touch less boost in the bass

 

[Robbo99999]

Movies soundtracks typically have louder HF as theaters absorb more HF than rooms.

From my research it's the opposite, large movie theatres result in the HF not being absorbed as quickly as it is in small rooms (but I'm entirely sure about this as can't find the resource again now that I look). And the traditional X-curve for movie making showed an extreme lack of treble because in the history of audio in movies there was much hiss due to the technology of the 70's, etc, so they reduced the treble to remove some of the hiss. From what I read movie sound hasn't come on that far in terms of target curves, and in my experience movies do not have an overabundance of high frequency, it's the opposite, they're "high frequency poor". And I'm quite certain that is why I prefer a flatter curve for movies due to their lack of treble. In my experience movies are recorded with a lot less treble than music.

Yes, my preferred target is similar to the Harman, a touch less boost in the bass

Interesting, I probably listen to music with a higher target curve for the bass than you considering I use full on Harman for music listening, yet you prefer the opposite to me when it comes to movies with more bass....you'd think it would be the opposite for us both. At the moment you turn down the bass for music and turn it up for movies, and I turn up the bass for music and turn it down for movies. Interesting dichotomy, I'm not sure how to explain how that would be flipped between us.

 

[QMuse]

From my research it's the opposite, large movie theatres result in the HF not being absorbed as quickly as it is in small rooms (but I'm entirely sure about this as can't find the resource again now that I look).

Well, your "research" is wrong. As I said, movie audio tracks have boosted HF as theatres tend to absorb them and for that reason most of the AVRs have some kind of "cinema EQ" to reduce HF on the movie tracks to avoid it sound hursh and with accentuated sybilants.

 

[Dimifoot]

Interesting dichotomy, I'm not sure how to explain how that would be flipped between us.

Different taste, different systems, different rooms