Written by cyberix the 4 Oct 09 at 00:42.
Category: Multimedia.
Related project:
Nothing/Others.
Status: New
Rationale
I'm a bass guitar player. Thus I often pay extra attention to the low tones. When I listen to music late at night. or using integrated speakers. the low frequencies are often too silent to be audible over the more dominant high pitched tunes.
I want this implemented EXACTLY as per the mockup.
Darwin Survivor(Brainstorm moderator)
wrote on the 4 Oct 09 at 05:04
This would be very nice. Unfortunately due to hardware limitations in most (all?) sound cards, this would almost certainly need to be done at the software level.
Actually equalizer with PulseAudio already does exist...but without GUI yet unfortunately..(see http://ubuntuforums.org/showthread.php?t=789578 appendix D). It would be great if it was per applications as well (per audio stream).
A spatializer,extra stereo plugins, crystalizer would be great too.
Actually the only player capable of all this plus some advanced crossfade management (really advanced, not like the amarok one) is Audacious ;)
Isn't this a thing applications should take care of? Different media might be mixed in very different ways; I don't think a system-wide setting makes sense.
I'm totally down with the graphic equalizer, with presets (or at least an on/off switch).
This can add a HUGE boost to laptop speakers. I already tested it using vlc player: Turn the bass (300hz or less) all the way down, the upper-mids (1khz to 3khz) all the way up, and leave the rest as-is. For most music, this gives a nice loud-and-clear sound that's distorted but in a good way.
I don't know how PulseAudio is currently implemented, but with the graphic EQ enabled, it would have to be:
The hardware volume control is always at maximum. The Volume Controls and the EQ are both at the software level. That way, the EQ won't distort the sound unless Master Volume is high.
Just to note: some SBLive cards have maximum volume at... 74% not 100%.
For optimal sound quality volume should be set to maximum all the time and you should regulate volume by a knob on your amplifier not by software.
I guess you should try to target zero resistance, zero amplification with your sound card. Because it makes no sense to turn the volume down, if it is going to be amplified again after going out from the sound card, but you'd also not want to use a cheap integrated amplifier that your sound card might have.
PulseAudio uses hardware volume as Master, but includes "preamp gain" in the EQ. That way, if it distorts, you can turn down the preamp gain and turn up the Master Volume.
One problem with equal loudness contours: some speakers will never produce enough of some frequencies no matter how much power goes into them.
For example, small speakers will never be able to produce bass, but only the distorted overtones of bass. In that case, it's best to cut out bass completely, allowing more power for mids and treble that you can actually hear.
I agree with elie. Every set of speakers is different, even each amplifier will have it's own frequency characteristics.
Not many programs in Linux have an equalizer, that's why I miss unified EQ from my sound card drivers in Windows so much.
The ability to mask out certain frequencies is a good one, but is separate and distinct from equal loudness curves. Remember at low volume, you have more head room to play with, so in effect, you just turn some frequencies down faster than others, you don't actually boost anything.
I am glad to see that someone has open the discussion on this issue, I do really think that an equaliser is missing.
I completely disagree with the 4th solution because the idea of the equal loudness contours is that the sound pressure has to be more important at low frequency for the listener to have the same experience of loudness.
But this is the same for everyone and this is already taken into account when the musician plays ! I mean, the aim of the equaliser is/could be to compensate for the frequency response of the loudspeaker, not for the sensibility of the human ear.
If you read the rational at the beginning, the poster is able to hear the low notes of his base guitar at normal volume, but wants still to be able to hear them when he turns down the volume. That's not a problem an equaliser will solve, unless you're prepared to adjust it every time you adjust the volume.
The musician cannot take playback volume into account as you suggested, as this will be different for each listener.
You want an equaliser for a different reason than the original poster. I agree, however, hat an equaliser would be very useful for a number of applications and would like to see it added.
I'm ok with adjusting EQ by hand whenever I need it. In any case manual controls are more important than any automatic function, as automagic can only solve one problem and cannot necessarily solve that one too well.
Some times I might want to turn down the low tones, so I could play music louder in the middle of night without disturbing neighbors too much. It all depends on what I happen to be doing at the moment.
I'm not expecting that an eq would make my speakers play lower tones than they do. I'd just want to remove the high pitched tones to hear the low ones better.
I think there is a fair amount of misunderstanding in the comments posted thus far.
Firstly, a comment on system-wide equalization: in my mind this is a feature that is sorely missed at present. A system-wide equalizer is essential for equalizing (to a limited extent) the non-ideal frequency response of the user's speakers; this solution also allows the user the freedom to tailor the sound of the their system to their personal taste - a case in point being a bass fan with an amplifier lacking tone controls. In addition, a "boost" feature may be useful - it is quite common to find an audio source that is just far too quiet, even with all application/system volume controls at maximum; an additional (more than "100%") boost would be useful in this case.
Secondly, regarding equal loudness curves: as noted previously, it should not be the intent to compensate for the user's perception of sound in an absolute sense - rather a relative compensation, based on volume level, may be appropiate. We should not require that an inverse of the Fletcher-Munson curves be implemented as a standard equalizer preset, however it may be advantageous to use a filter based on the difference between Fletcher-Munson curves of different volume levels - i.e. an amplitude-related "loudness" compensation (where high/low frequencies are boosted more at lower volumes, for example).
Thirdly, saving of audio presets is an easy-to-implement feature that would add value with little GUI clutter. Thumbs up from me on that one.
Lastly - if significant computational impact is imposed by a system-wide equalizer - there should be a warning to the user along the lines of "Enabling the equalizer may impact performance on slower systems. If a drop in performance is encountered, disable the equalizer".
Wow, this is what I'm looking for. I bought a totally new PC, based on Intel Chipset and everything else, to work perfectly with Linux. Everything went fine, but I didn't managed to set up a system wide equalizer, so I had to go back to Windows XP :(, because I'm a music freak. I really like Linux' community, the way it's free and the update system. So, I really hope, that this idea gets implemented as soon as possible, because a system without a working, system wide equalizer isn't a system for music freaks.
Very much agree with this, this one of very few flaws but it's a massive one, which seems to have been talked about for years, but not addressed yet.
Arguably, it wouldn't be so important if the best integrated and most complete Media Player had been suitably equipped, but with the ever-increasing online content where those players are not equipped either, an OS level equaliser would be perfect.