Software options for Tuning - Page 14

Abmolech · 11-15-2006, 08:12 PM

Quote:

So you want a system that adjusts for equal loudness or basically alters the response based on the "Fletcher-Munson curve"? Am I understanding you correctly?

Quite correct "C" weighted is wrong.

There is a new standard,
http://en.wikipedia.org/wiki/Equal-loudness_contour
http://en.wikipedia.org/wiki/A-weighted

http://www.phys.unsw.edu.au/~jw/hearing.html

I suspect the curve that should now be used is the ITU-R 468 noise weighting.

I contest one of the problems with in car audio is the use of RTA and pink noise.
The idea of pink noise is to have short enough bursts of a given frequency, so that the microphone can "tell" a reflection from the principle wave. The use of pink noise in a car environment is open to interpretation because the bursts are too long. I believe MLS and adjustable pulse type noise may be the solution to obtaining correct measured volume.

RedGTiVR6 · 11-16-2006, 07:27 AM

RTA is most likely going to be thrown out of competition in IASCA for the coming years...especially after what happened this year at finals.

durwood · 11-16-2006, 09:34 AM

Good riddins. I have never competed in IASCA where they used the RTA but I don't think it makes much sense in a car.

I think it would be cool if they could come up with some other type of measuring using something closer to a "real human". I worked for some time at a hearing transducer company that worked with these "dummy heads". They were similar to the Bruel and Kjaer.

http://www.bksv.com/4071.asp

Abmolech · 11-16-2006, 10:21 AM

Probably getting too off topic...

Most software programs for audio are designed for far field listening, including the various theories of acoustic manipulation. Put these into a car situation and most ideas, programs will be sub par. It is interesting reading on headphone acoustics because they face similar problems as us, that is the sound recording was/is designed on two monitors, and requires reproduction in a similar situation to produce the results.

I am not suggesting you throw out all room acoustic theory etc, however most of it is not applicable to a car environment.

Near field listening is a much under-studied acoustic field.

durwood · 11-16-2006, 10:32 AM

Quote:

I am not suggesting you throw out all room acoustic theory etc, however most of it is not applicable to a car environment.

If only home stereo/theater people would try to understand this rather than brush off car people. Then maybe it would be studied more. This is a tad off topic but I think in the long run it will help us find/generate more ideas for software tuning...or human hearing in general. Hell...more understanding lead to all this music compression formats...not that it was good for sound quality but makes putting a carputer in more logical for other reasons.

Added: We need some studio engineers in here. They would be very helpful as far as near-field listening.

Abmolech · 11-16-2006, 08:44 PM

Quote:

If only home stereo/theatre people would try to understand this rather than brush off car people. Then maybe it would be studied more.

Lack of demand is the main reason.
I have no desire to start a bun fight, however, car audio enthusiasts are quite well educated when it comes to some things, and woefully ignorant in others.

Possibly the best example is the extreme lengths people go to designing and making sub boxes, and the antipathy to front enclosure design.

First requirement for software is it should be designed for near field listening.

As mentioned pink noise sucks... There are some better impulse style software "floating" around. I will certainly be availing myself of this, probably with suitable mods.

Next up is the use of crossovers.
It is almost entirely possible to replace enclosure design around driver Fs with a proper crossover. (no this does not negate baffle use) This is particularly useful in a car.

Most crossovers cause phase shift, again this can be compensated for with appropriate software. All this adds to the delay signal which can be allowed for with proper timing. (read software)

Volume control is up my list for software.

Room correction for under 500 Hz should be a real possibility (invert frequency correction)

The "holy grail" will be using software for near field listening. With the popularity of headphones, this is now becoming a reality. I am sure we can "piggy back" on DSP logarithms when they become available.

FyreDaug · 11-17-2006, 01:00 AM

Quote: Originally Posted by durwood

Yes, Winamp locks up. Probably just bad MP3's. I don't really know though.

I am not sure about your strange problem. Just remember though that a 31 band EQ makes small adjustments versus a 10 band. If you adjust a bunch of bands that are close together then you should see a more drastic change. If you notice a difference on the FLAC files though then you should notice a difference on your MP3's on any changes you make. What is your software/hardware setup exactly? I need more information and I will try to duplicate your problem.

Well I tried this, I dropped 6 bands from the right, to 0 on the 31band. On any flac song the treble was missing, but on all mp3 songs they sounded just fine, until I turned the treble knob all the way to 0. And I have 2 of the same song on flac and mp3 that it was tested on.

Athlon 2800 512ddr, auzentech x-plosion pci card, asrock k7s41gx mobo

edit: by 0 I meant the lowest they could go, -10db I believe

RedGTiVR6 · 11-17-2006, 06:16 AM

Quote: Originally Posted by Abmolech

Lack of demand is the main reason.
I have no desire to start a bun fight, however, car audio enthusiasts are quite well educated when it comes to some things, and woefully ignorant in others.

Possibly the best example is the extreme lengths people go to designing and making sub boxes, and the antipathy to front enclosure design.

I agree. But just like home audio, car audio has it's fair share of ignorant folks. Take into account how many home folks go and buy the $200 theater in a box compared to the folks that are really interested in doing this type of tuning. Subs in a car are a lot easier to work with than the front stage. More people want bass in a car than clarity, and I'd say the same goes for home audio as well. Just look at all the frat boys that go goo goo gaa gaa over big towers for the house, or the number of people who go nuts when you pop the trunk and you've got a single 15, now imagine it for 4 15s...

it's what people can see, not what's behind the scenes...same for home, same for car. It's unfortunate, but it's not just for the car.

I posit that home audio could be considered more of a waste than car audio...after all, I spend 3x more time in the car than I do at home in a state than I could actually appreciate a well set-up home audio system.

Ok - now onto the point of this thread...hehe (and no, I wasn't arguing with you, I was more or less just making a few statements...

)

Quote:

First requirement for software is it should be designed for near field listening.

I agree and I'm hoping to get a bit more info this weekend on this...since I'm here at USACi world finals right now.

Quote:

As mentioned pink noise sucks... There are some better impulse style software "floating" around. I will certainly be availing myself of this, probably with suitable mods.

I'd be really interested in what you come up with regarding this.

Quote:

Next up is the use of crossovers.
It is almost entirely possible to replace enclosure design around driver Fs with a proper crossover. (no this does not negate baffle use) This is particularly useful in a car.

Agreed!!

Quote:

Most crossovers cause phase shift, again this can be compensated for with appropriate software. All this adds to the delay signal which can be allowed for with proper timing. (read software)

We've run into this in our car with the 701....it's another area where I was hoping to overcome problems, glad someone else sees this as an issue in car too!

turbocad6 · 11-17-2006, 10:01 AM

agreed, most build there systems backwards, the bass is the primary concern & the front stage is secondary...., to me the front stage is everything & the bass is just a supliment or support....

& yeah, x-overs do change phazing, but is it as much with a digital solution like an alpine? I was always under the assumption that this was much less of an issue in the digital domain?

durwood · 11-17-2006, 12:42 PM

Quote:

& yeah, x-overs do change phazing, but is it as much with a digital solution like an alpine? I was always under the assumption that this was much less of an issue in the digital domain?

This is a commmon misconception. There are two basic types of digital filters.
There is IIR (infinite impulse response) and FIR (finite impulse response). IIR filters are usually analog modeled filters in the digital domain. They suffer from the same problems as analog filters with phasing issues.

FIR filters on the other hand do not have phase problems. They have zero phase shift-or it has linear phase. This is more ideal. It eliminates one factor out of the equation, unless you want the phase changes an analog or IIR filter produces.

I assume the Alpine is using a IIR filter because they are much easier to design and require less processing power. I do not know for sure though. A butterworth or LW4 crossover is an IIR filter type.

I was actually working on a post about the differences between the filters but got sidetracked.

RedGTiVR6 · 11-17-2006, 01:50 PM

back to work on that post!

I want to know more about em....

durwood · 11-17-2006, 03:05 PM

*whip

Ok I put it in its own home in a seperate thread. There is a TON of info and its all super tech language. I don't want to regurgitate wikipedia so I am still trying to reword things so its easier to read for non techies.

Here it is...or at least what I have so far.

http://www.mp3car.com/vbulletin/show...d=1#post964845

Abmolech · 11-18-2006, 03:03 AM

The search for a suitable loudness compensation DSP continues.

http://www.measure.demon.co.uk/Acous.../loudness.html

Worth a read, these are the latest research papers into music/loudness compensation rather than noise/loudness compensation.

Probably you guys know of this one..

http://www.softplatz.com/Soft/Audio-...-Plug-ins.html

Problem is only windows media etc.

Next up is wave 6, have to check if this can do it on the fly, expense is a particular aversion.

http://www.steinberg.net/955+M52087573ab0.html

Last, is window vista

http://msdn2.microsoft.com/en-us/library/ms678766.aspx

This has it built in as standard.

Now to find out if my crossover programs are supported