Your subjectivity of "Room" type of corrections

[Pareto Pragmatic]

Any number I give is really close to meaningless, since I have made so many incremental changes over time. And I have not listened to uncorrected since I got my last amp, which was a while ago. So it's really hard for me to judge. But I'll play.

Ok, I will assume that 100% improvement would mean taking it from unlistenable to perfect. 10% would mean a slight improvement, but obvious. Below 10% means you can hear it, but have to listen to hear it. 50% would mean I would not want to listen without it, but I could, and could probably get used to it over time if I only listened to that for a week. On that scale, I would put recapping 40 year old speakers at 25%. Running a dedicated power line to get stuff off my 100 year old wiring more like 40%. Those were both noise reductions of course.

I am not sure if we are talking about all room correction, or just "automatic" room correction as with an AVR. For automatic, it really depends on the curve. My AVR does a good job of fitting to a curve, I just hate the curve. Too bassy, too rolled off on top. But if I drop the bass 3dB and up the treble 1dB and roll the bass off below 40, on top of whatever it does for correction, it's pretty good. Flat enough, but still issues with some room modes at higher volumes.

So, thinking about what I have now and how I have corrected it (still have some fine tuning to do on some of my options for playing sound since I have not had a calibrated mic for very long), I would put it at 50%. BUT I have a very problematic room, with multiple modes including a 30hz mode that is crazy huge due to a very bouncy floor. And a sub 20hz mode at very high SPL. When those get going, they get GOING. So for me, the improvement comes from being able to play a bit louder before the room modes go crazy. And more importantly, knowing when they are likely to go crazy so I can keep things just below that level.

Pre calibrated mic, using first a phone app then a non-calibrated mic jacked in to my 8 year old laptop and REW, probably more like 25% improvement. .

OTOH, my office, small speakers and a sub, correction gives me something more like 10%. Much lower SPLs, much less extensive corrections applied.

So final answer, 50%, but with a defined meaning of what 50% means for me.

 

[Sokel]

It's highly depended on what I listen and how I listen to it.





Uncorrected Psy



Corrected for low level,late night,etc Psy



Corrected for loud Psy

(sorry for the scale,I have posted them here long ago,before I learn the 50db rule)


To all the above the percentages highly relate to material I play.
There also no improvement at all that I can hear if allow correction over 200-300Hz,the times I tried it it was the opposite.

 

[Waxx]

So some more comments:
1. I am very happy to see how people contribute to the discussion feeling free to add their input.

Now bit more from my side, without being able to give an answer to the question I have asked.

This is my room
View attachment 330866

I am not going very low on the lowerfrequencies, since I have small subs.

Then my music is like piano, or music like this
View attachment 330867

then Arias,

I would not think that these songs go too low (I would like to hear your opinion on this). At the same room correction might cost me up to 700-800 euros more so I would expect to benefit the lower ranges of my "whole system" which is not probably i am really using.

Please keep your input coming
Regards,
Alex

only based on the measurements, room correction is needed in your case i think., and your speakers could also benefit from eq. This on the condition that the measurments are made right off course. I would tame the peaks in the bass and raise the treble a bit with eq to start.

 

[Snarfie]

So this is a very subjective topic by nature and we can keep it like that.
Assume that you have used some room correction software for your room and your specific setup how much in percentage would you rate that your listening experience has improved?

I know, I get it, it is very subjective but still, think that we are around drinking beers and someone asked this.

So my experience with room correction software was around 10% improved listening experience.

So what would be your percentage?
Regards,
Alex
P.S I can explain bit later why I am asking but lets focus on the "beer" part of this discussion first

Depends how good or bad you roommodes are. Let use my measurment of my mancave. An attic V shape.

It's that horrible close to 70% improvement using DSP. No gear speakers could resolve above room modes at any price besides room treathment. An yes i used several measurment mic's to see if the measurements were ok. Different speaker brands (colume speakers of same size) gave same sort of measurments. White horizontal target curve took care of whole frequency spectrum did resolve the problems adequate happy with the result. DSP used Mathaudio Room EQ.

 

[FrantzM]

Correcting low frequency peaks generally works well and is almost always an overall improvement, though reducing very large peaks with many dB will also lead to less dynamics. Correcting room related (or other non minimum phase behaviour anomalies) peaks and dips in especially the midrange and treble generally leads to a worse result. Those auto-correction softwares that does a lot of the latter, has the result of really weird and unatural sound to my ears.

I'm not sure I understand the reasoning.

+1

Hi

I am a fan of room correction ... as with most everything, moderation is key..
I use Audyssey's Windows Application MultEQ-X and it is an ear/eye opener. It takes a while to learn it and some time to master. I use it in conjunction with MSO, another Digital Room Correction software... and REW... Improvement are of the 80% orders... Since you asked me for some numbers...
My room has a nasty peak at 37 Hz, and another nasty one at 70 Hz followed by quasi null at 120 Hz... and ... Perhaps a few measurements later. I am working.
I would say the improvements makes 80%. I no longer can imagine a system without those.. Simply cannot.
It requires work. Lot of it. It took me several months, perhaps half a year... REW, nor MSO or MultEQ-X are plug and play. You need to understand how they work and what you are measuring. At the end of these convoluted processes, I have an audio system that I am trying to wrap my head around, so good it is and so little it cost me in term of financial outlay; especially after having been for more than 40 years an ardent subjectivist, immersed in the HEA and spent serious amount of money. My system in my room is a full range (20 to 20 KHz no issues) , full scale (can reach 110 dB peaks at my 2.4 meters listening distance) ... The main speakers are a pair of $250,oo/each monitors, the subwoofers are $250 each, a $1000 AVR, a DSP for $250, the MultEQ-X software cost me $150 on a special.. REW and MSO are free ....and a lot of time ... I've heard 1/2 million dollars system that mine (at circa $2500), would wipe the floor with...
So for me, DRC is mandatory...

I like Audyssey MulTEQ-X a lot...

Peace.

 

[DLS79]

interesting point. Can you elaborate a bit more on this?

In an ideal world everything in the signal chain is audibly perfect/flat. This would allow people to easily and consistently eq any devise to meet their sound preference. However since we aren't in an ideal world some gear will need EQ to sound correct.

Please note that correct is a subjective term.

With the signal chain done you have to deal with the space issues. Some issues can be dealt with by placing the speakers in a different positions or at a different angles. Some issues can be addressed with room treatments (sound absorbing materials). Some issues can be addressed with DRC.

 

[Blumlein 88]

Depends upon the room and the speaker. Some speakers can benefit more from Room correction than others. Some rooms have worse problems than others. I'd say most of the time the improvement is a subjective 10-30%.

 

[goat76]

Correcting low frequency peaks generally works well and is almost always an overall improvement, though reducing very large peaks with many dB will also lead to less dynamics. Correcting room related (or other non minimum phase behaviour anomalies) peaks and dips in especially the midrange and treble generally leads to a worse result. Those auto-correction softwares that does a lot of the latter, has the result of really weird and unatural sound to my ears.

After many years of staying away from automatic room correction programs, I decided to try Audyssey in my new Marantz Cinema 50 as I thought it must have improved. But no, and no wonder, it's correcting the full range all the way up to 20 kHz which is wrong, it should have stayed under at least 500 Hz or lower for the corrections or let the listeners decide what range is touched by the program.

I can't put a number on the deterioration it does to the overall sound. If Audyssey could just give me some control over it I would probably love it.

 

[gnarly]

Hey all, can any of you post transfer functions of the electrical processing your DRC is doing?

(Along with the usual acoustic measurements of the speaker(s) before and after applying DRC.)


Or point to posts of others who have done this? ....hard to find any via search, given all the various type DRC threads.
Thx, Mark

 

[anotherhobby]

After many years of staying away from automatic room correction programs, I decided to try Audyssey in my new Marantz Cinema 50 as I thought it must have improved. But no, and no wonder, it's correcting the full range all the way up to 20 kHz which is wrong, it should have stayed under at least 500 Hz or lower for the corrections or let the listeners decide what range is touched by the program.

I can't put a number on the deterioration it does to the overall sound. If Audyssey could just give me some control over it I would probably love it.

You can limit Audyssey's correction range. You can do it with the $20 MultEQ mobile app or the more powerful $200 Windows only Audyssey MultEQ-X program. I limit the Audyssey correction range on my Denon in my living room because I also do not like full range correction. In my home office I limit Dirac to the bass region only as well on my miniDSP Flex. Both corrections noticeably degrade the sound quality in my rooms above probably 600 Hz.

 

[bodhi]

In my room the modes in bass are so atrocious that some songs with low bass sound absolutely terrible. With movies it's kind of OK until there is that WTF explosion that just slams you and fills the room with uncomfortable noise.

But with some songs REQ on/off is not that different. Like "whatever" kind of difference.

I'm using Audyssey XT32 with the MultEQ-X program which I don't use other than limit the correction and remove the default filters.

 

[ahofer]

I might scale these things differently.i would rate the difference between my Harbeth speakers and my Revels as about 10%.

With that as a reference, I’d say EQ has made a 15-20% difference in the two living spaces I use it in.

 

[ahofer]

I greatly appreciate your contribution. Firstly because you are an acoustics professional and secondly because you shed light on an aspect that I hadn't thought of myself. I have emphasized this.

..and it confirms your priors.

One can, of course, shelf the bass at the same time you tame the peaks. I’m not sure making the room boom is a great substitute for “dynamics”. Concert halls are generally too big to produce that effect, so it sounds quite unnatural.

 

[Geert]

In my living room, 80% improvement. In this room, considering practical limitations, there is nothing money can buy that could improve sound as much as my room and speaker correction.

 

[goat76]

You can limit Audyssey's correction range. You can do it with the $20 MultEQ mobile app or the more powerful $200 Windows only Audyssey MultEQ-X program. I limit the Audyssey correction range on my Denon in my living room because I also do not like full range correction. In my home office I limit Dirac to the bass region only as well on my miniDSP Flex. Both corrections noticeably degrade the sound quality in my rooms above probably 600 Hz.

I didn't know it could be done in their mobile app, thank you for the information!

 

[Daverz]

I'm not sure I understand the reasoning.

Reducing the peaks requires digital attenuation (called insertion loss), so you lose the full dynamic range of the DAC or amp (according to whichever comes first, the max output voltage of the DAC, or the max input voltage the amp can handle without clipping) . The rule of thumb (via Mitch Barnett) is to keep insertion loss to less than 10 dB.

Most systems will have so much excess gain, that it's usually not too much of a worry.

 

[fpitas]

(according to whichever comes first, the max output voltage of the DAC, or the max input voltage the amp can handle without clipping)

Wait...I thought we were only attenuating in specific areas?

 

[fpitas]

The rule of thumb (via Mitch Barnett) is to keep insertion loss to less than 10 dB.

Also, wouldn't that have more to do with the DSP bit depth available? I mean we're sawing a hole in response here.