|
Post by igorzep on Jun 4, 2015 7:47:01 GMT -5
Hello. This question appears then and there so I've decided a dedicated topic is worth it. Please vote and comment...
The XMC-1 has a great Loudness feature. Unfortunately as great the feature is as it is unusable. That is confirmed by many posts on forums by people that refuse to use it because they find it negatively affects the sound (over-boosting low frequencies usually). This is all because recordings are not made to follow one loudness standard and every recording is made at different level. Without the ability to account for that level difference it is simply works well only on a small part of the content. Once optimised for one type of content it beaks for another.
What I suggest is a feature request that should be quite trivial to implement but make this feature not only a nice line in the specs but make it really useful and powerful tool for many owners of the XMC-1. The feature is basically the same as it exists in the Audyssey (and they also created it because users were asking!) and is called a “Reference Level Offset” or RLO for short. This means a parameter defining a level of the MV knob at which no correction is needed. Then all correction is done in relation to that level.
So, let say if RLO is set to 0 - there is no correction at MV 0dB. But it is engaged as soon as MV is less than 0dB adding compensation proportionally to reducing the volume. So, let say it boosts low frequencies by 10dB when MV is at -20dB.
Then RLO can be set to some other level, like -15dB. Then there is no correction at -15dB MV and it starts to engage once volume is reduced bellow that. And the amount of correction is appropriately shifted, so, now, it applies the same amount of correction at -35dB as it was done before at -20dB in the previous example, meaning boosting low frequencies by the same 10dB.
This RLO can be cycled-over by the Loudness button by some increments instead of just current On/Off. For example “0dB (Very High)”, “-5dB (High)”, “-10dB (Medium)”, “-15dB (Low)”, “-20dB (Very Low)”, “Off” settings…
Currently it seems the ‘reference’ level for the XMC-1 is fixed to -10dB (the point when no correction is applied yet).
But, in practice, even on softly recorded content I find the amount of correction imbalanced - when listening at -10dB MV there is not so much bass, but when I turn MV down to -40 or -50 voices basically disappear already, but bass - still every note is heard and distinguishable, meaning it adds correction too fast, faster than needed. May be this is room/target curve/preference dependent, but at least I find it is perceptibly off and quite significantly. So, I suggest another setting that will let user adjust not only the reference point but the aggressiveness of correction. This option could be hidden somewhere in the setup menu.
Firs one (RLO) I consider a must for Loudness feature to be usable, second one (aggressiveness) is a nice, but still very useful addition. Please, can it be implemented?
|
|
|
Post by markc on Jun 30, 2015 1:32:48 GMT -5
I, too, would welcome something like this, as well as a clear explanation of how the XMC "thinks" it is working.
I find the Loudness function absolutely unusable at normal listening levels., which is a shame because I do believe that the natural ability of the ear to hear different frequencies at different volumes is an inescapable fact of how our bodies work and we should not worry about adjusting the purity of recorded sound to give a more flat sounding spectrum at low and high volumes. I previously have used loudness on my old amp and it was transparent in use meaning that I never questioned if it was too weak or too boomy, it just sounded right at all volumes.
When Loudness is engaged whilst listening to a movie or music at normal levels (XMC-1 volume set to -35dB) (I have the high gain XPA-5 Gen 1 amps and the 65dB XMC-1 internal test tone plays at 64dB)) it adds a phenomenal amount of thumping, pounding bass.
I can then turn the volume down, but when I subsequently change to an input without loudness engaged, the overall volume is too low.
I haven't quite worked out how the loudness function is working - I note Igor believes that -10dB is the null correction point, but that is too loud for me to test it!
|
|
|
Post by millst on Jun 30, 2015 1:57:54 GMT -5
Currently, I'm fairly sure that anything louder than -16dB is uncorrected. I didn't do anything extensive, but ran a few trials with REW a while back.
Anybody bored and want to create some graphs?
-tm
|
|
|
Post by JKCashin on May 9, 2016 17:51:03 GMT -5
Surprised this thread died... I find the loudness button is too "aggressive". I would love to see something tunable.
|
|
|
Post by JKCashin on May 9, 2016 17:58:52 GMT -5
|
|
stiehl11
Emo VIPs
Give me available light!
Posts: 7,261
|
Post by stiehl11 on May 9, 2016 20:55:40 GMT -5
I thought it was corrected in a firmware fix last year. I could be wrong, though.
|
|
|
Post by JKCashin on May 9, 2016 22:16:36 GMT -5
Nope, it appears that it does implement the "Fletcher – Munson", loudness compensation curve correctly, but I , as well as many others by my reading, find it too "boomy"
|
|
|
Post by mickseymour on May 10, 2016 0:45:41 GMT -5
I've never been fond of the loudness control on any equipment. The one on the XMC-1 may be a perfect FM curve but to my ears it is horrible.
|
|
|
Post by millst on May 10, 2016 15:09:25 GMT -5
Yeah, I've been listening at lower volumes more than usual lately and agree that it's a bit overzealous.
+1 for the idea of some configurability in the next firmware.
-tm
|
|
|
Post by fludimir on May 11, 2016 9:59:52 GMT -5
>Nope, it appears that it does implement the "Fletcher – Munson", loudness compensation curve correctly It cant implement that feature correctly while it looks only on volume level in XMC, as real SPL level depends on speakers sensitivity + amps gain(you can configure this in xmc) and on record volume, which may be 10-20-30db different and there is no way to configure it in xmc. Absolutely need smth like that igorzep proposed to make feature usable, but how to persuade Emotiva to make that?
|
|
|
Post by Priapulus on May 11, 2016 13:46:05 GMT -5
I find the Loudness button unusable. Even at very low volume levels it boosts the bass way too much. I love bass, but this is unnatural. There needs to be a way for user calibrating. /b
|
|
|
Post by alfredo on Apr 6, 2017 14:23:55 GMT -5
Currently, I'm fairly sure that anything louder than -16dB is uncorrected. I didn't do anything extensive, but ran a few trials with REW a while back. Anybody bored and want to create some graphs? -tm These measurements are from Systune, XMC loudness at different levels, I think a simple addition to offset the level in db would help getting better results.
|
|
|
Post by AudioHTIT on Jun 12, 2018 14:26:47 GMT -5
I think this thread deserves a bump, I too would like to see the Loudness effect be made useable with the XMC and RMC. Listening at lower volumes requires some kind of compensation, but if it doesn’t account for the current/correct SPL then it can’t apply the curve properly. Like the others above I find the current implementation too boomy and mostly unusable. They went to a lot of trouble to put it in, how about doing what’s left to get it right? Disclaimer : I do not think working on this should distract from getting the XMC-1 HDMI and Atmos Boards delivered, nor the XMC-1 G3 or RMC-1 ... but I’d hope it could make a bug/enhancement list as it needs work. igorzep has identified a good solution.
|
|
|
Post by 405x5 on Jun 12, 2018 16:40:43 GMT -5
Hello. This question appears then and there so I've decided a dedicated topic is worth it. Please vote and comment... The XMC-1 has a great Loudness feature. Unfortunately as great the feature is as it is unusable. That is confirmed by many posts on forums by people that refuse to use it because they find it negatively affects the sound (over-boosting low frequencies usually). This is all because recordings are not made to follow one loudness standard and every recording is made at different level. Without the ability to account for that level difference it is simply works well only on a small part of the content. Once optimised for one type of content it beaks for another. What I suggest is a feature request that should be quite trivial to implement but make this feature not only a nice line in the specs but make it really useful and powerful tool for many owners of the XMC-1. The feature is basically the same as it exists in the Audyssey (and they also created it because users were asking!) and is called a “Reference Level Offset” or RLO for short. This means a parameter defining a level of the MV knob at which no correction is needed. Then all correction is done in relation to that level. So, let say if RLO is set to 0 - there is no correction at MV 0dB. But it is engaged as soon as MV is less than 0dB adding compensation proportionally to reducing the volume. So, let say it boosts low frequencies by 10dB when MV is at -20dB.
Then RLO can be set to some other level, like -15dB. Then there is no correction at -15dB MV and it starts to engage once volume is reduced bellow that. And the amount of correction is appropriately shifted, so, now, it applies the same amount of correction at -35dB as it was done before at -20dB in the previous example, meaning boosting low frequencies by the same 10dB.This RLO can be cycled-over by the Loudness button by some increments instead of just current On/Off. For example “0dB (Very High)”, “-5dB (High)”, “-10dB (Medium)”, “-15dB (Low)”, “-20dB (Very Low)”, “Off” settings… Currently it seems the ‘reference’ level for the XMC-1 is fixed to -10dB (the point when no correction is applied yet). But, in practice, even on softly recorded content I find the amount of correction imbalanced - when listening at -10dB MV there is not so much bass, but when I turn MV down to -40 or -50 voices basically disappear already, but bass - still every note is heard and distinguishable, meaning it adds correction too fast, faster than needed. May be this is room/target curve/preference dependent, but at least I find it is perceptibly off and quite significantly. So, I suggest another setting that will let user adjust not only the reference point but the aggressiveness of correction. This option could be hidden somewhere in the setup menu. Firs one (RLO) I consider a must for Loudness feature to be usable, second one (aggressiveness) is a nice, but still very useful addition. Please, can it be implemented? Unlikely you will see this I suspect. The loudness feature on the XMC1 fails for a number of reasons but not because it wasn’t a great idea. It was a bonus I was initially excited about when I replaced my Sunfire Theater Grand II with this processor. Being a Processor rather than a preamp or receiver is the first black eye 👁 because the majority of XMC owners have a subwoofer, which is the ultimate “loudness” compensation. The other problem I see is morphing the kind of control you’re looking for into the mix with automated room correction and the manual settings. I once had loudness compensation that was attenuated separately from the main volume control. A Sherwood Receiver from 1963. Bill
|
|
|
Post by pknaz on Jun 12, 2018 22:57:07 GMT -5
I just tuned my system with Speaker Preset 1 for low level listening 65-75db (which amounts to about 10-15db of boost at varying levels depending on frequency, below about 150hz). On the rare occasions that I do turn things up loud (95-100db) its usually rock music, which still sounds fantastic with the extra bass anyawy Speaker Preset 2 I have configured with nothing other than left/right speakers and default to Reference Stereo (or whatever its called) for those times I want unadulterated sound at any level
|
|
|
Post by AudioHTIT on Jun 13, 2018 6:46:20 GMT -5
... Being a Processor rather than a preamp or receiver is the first black eye 👁 because the majority of XMC owners have a subwoofer, which is the ultimate “loudness” compensation. The other problem I see is morphing the kind of control you’re looking for into the mix with automated room correction and the manual settings. Using the subwoofer, tone controls, and eq for ‘loudness compensation’ can help, but it’s a manual process that doesn’t really take level into account. Though it’s what I suspect many are doing to get something that’s listenable. Not sure if you mean that the loudness could morph into Dirac, or be part of it, but it seems the Dirac measurements could be useful in setting the RLO as they should have some information about the system’s efficiency or sensitivity, thereby letting the loudness compensation know the appropriate SPL at a given setting of the volume control.
|
|
|
Post by TechChallenged on Jun 18, 2018 18:40:45 GMT -5
This is likely low priority for Emotiva, since they haven't delivered on the Atmos/DTX-S upgrades. I agree that the Loudness feature isn't something I like to use either. I have used it on both the McIntosh MX-30 and Yamaha C-60, but they featured a variable loudness control. Worked much better because it was very gradual. Another thing I prefer, if using loudness as the XMC-1 features, is a little lower frequency selection. Selecting a lower target frequency, such as 50, 60 or even 80 Hz could result in less boomy sound. If this is handled digitally, the frequency could be changed. I remember Aiwa mini-systems that used a multi-level loudness control through the BBE and T-Bass settings. The amount of boost would increase with each push of the button for 3, or 4, different amounts of bass using the T-Bass and the mids and highs were influenced more with the BBE feature. If Aiwa could put multi-level compensation into a cheap mini-system, don't see why it can't be added to the XMC-1. Each push of the loudness button can boost a little with each push and back to "Off".
|
|
|
Post by 405x5 on Jun 18, 2018 19:41:30 GMT -5
I see this thread was started 3 years ago with but a handful of posts, which tells us everything we need to know about Loudness compensation. Modern high power electronics and much more capable and low frequency loudspeakers have rendered the feature useless and unnecessary, but 40 years ago (longer actually) it was s cool feature to have. (RIP)
Bill
|
|
|
Post by AudioHTIT on Jun 18, 2018 20:49:33 GMT -5
I see this thread was started 3 years ago with but a handful of posts, which tells us everything we need to know about Loudness compensation. Modern high power electronics and much more capable and low frequency loudspeakers have rendered the feature useless and unnecessary, but 40 years ago (longer actually) it was s cool feature to have. (RIP) Bill As long as the human ear works the way it does, something like Loudness compensation will be beneficial. If you compensate with additional sub power, and high frequency boost (it’s not just bass), then your system will be unbalanced at normal levels.
|
|
|
Post by TechChallenged on Jun 19, 2018 1:27:40 GMT -5
Agreed AudioHTIT. Naturally, the ear is attenuated to the mid-band of frequencies. The extremes are much harder to hear. As we grow older the extremes are likely further diminished, making a useful "loudness" control even more beneficial. A multi-stage, or variable, "loudness" could be the answer, but may require more than a discussion here to get results. This is unlikely a deal breaker for anyone doing business with Emotiva. It is one of those things that can be good PR if enough of us find it a useful enough tool to have in the XMC-1. I've had receivers that offered a decent, standard, loudness option. The Pioneer Elite receivers I've owned were quite effective, but I don't know the parameters. I believe they typically would boost the bass response more than the treble. Whatever they did seemed to work better. .
|
|