Popular Post mitchco Posted October 30, 2020 Popular Post Share Posted October 30, 2020 Awesome!! Mathematically speaking, aside from one's preferred tonal response, there is an optimal or "ideal" convolution filter for each set of speakers and room combination. You can see an example of that it in the charts for the "ideal" response in this article: Let's have a look at the "ideal" step or timing response in particular: Preringing is mostly a benign artefact. You have to know what to listen for in order to hear it, even though there maybe some preringing showing in the chart, you may not hear it. How "vertical" the spike is determines the "transient" response of one's system. If the vertical spike is tilting towards the right that means the bass is arriving first and then the high frequencies. Some folks can pick up on that and does not sound as transient as with a straight vertical spike. The FDW settings used in my book was for my system which required quite a bit of excess phase correction up top as compression drivers and horns have both frequency and time domain issues. However, also noted in my book is guidance on what the amplitude and excess phase corrections do to the waveform so one can adjust accordingly. If you are experiencing a loss of sound stage/depth, that is typically a sign of too much high frequency excess phase correction. This is the number one mistake that most folks make (including myself) and where some folks give up on further tuning as they don't like the sound. So backing it off to 1 or 2 cycles (total) is a worthwhile experiment. Further, you may find limiting the excess phase correction above a certain frequency will also restore the sound stage. That's in Macro 3 in Acourate and partial excess phase correction in Audiolense's Correction Procedure Designer. Ultimately, it is a balance between how much direct sound versus reflected sound one wants to correct for at midrange and high frequencies. Ideally one is smoothing out the amplitude (i.e. frequency) response of the direct sound, including phase, but letting more room reflections in at these frequencies which results in a more natural sound without reducing the soundstage or depth of field. Unless one is running a desktop setup, most folks find a longer excess phase correction like 4 to 6 cycles in the low frequencies does a great job of eliminating and/or restoring the phase response to the ideal minimum phase response. Here is an example from measuring by current speakers in my room: This demonstrates both concepts as described above. This measurement is made at the LP some 9ft away with the convolution filter in the circuit and using REW's default 500ms window with no smoothing. Meaning all the room reflections are entering the measurement. I did this on purpose to show that at low frequencies, aside from the even frequency response, there are no room reflections entering in and the minimum phase response has been restored. Listening wise, this produces the smoothest, clearest sounding, most transient sounding bass one can get in any given room. Above the room's Schroeder or transition frequency, I am starting to let the room reflections in, again on purpose, which results in a natural sounding midrange and top end, but still smooth frequency response (if one applies some smoothing like 1/12 octave smoothing). Here we can see at about 350 Hz I am starting to let the room reflections in and not correcting for them, even though the direct sound is smooth and a flat phase response as shown in my "accurate sound" article. The equivalent timing response of above: Very close to the ideal step response. The little dip before the vertical step is not preringing, but rather an experiment I was trying which was using a "mixed phase" target moving a bit away from full minimum phase response with some linear phase action down low. Anyway, I hope that helps explain some of the settings to play with and what the resulting charts look like. It can take many filters and listening tests (I have done hundreds on my own system to learn) to correlate what one sees in the charts versus what one is hearing. If you can get close to the ideal, I can guarantee it is going to sound good :-) From there it is a point of departure on what one prefers. The future of DSP is exciting! I am experimenting using binaural mics instead of a single omni mic. I am also looking into the possibilities of applying high resolution hearing correction as well. Modern audio DSP frameworks in software are becoming more sophisticated and easier to program, even if it is still in C++. Happy listening! asdf1000, ecwl, The Computer Audiophile and 2 others 3 2 Accurate Sound Link to comment
Popular Post mitchco Posted November 10, 2020 Popular Post Share Posted November 10, 2020 Fantastic@ecwljob on "bracketing" in on what you prefer. Re: longer excess phase correction of at least 4 cycles in the low frequency makes the bass sound dramatically better and should really be the minimum setting. Excellent! You are training your ears and hearing it! Re: "I ended up selecting the filter setting that has the highest vertical spike in the step response" Well done! Re: I am really appreciating the full range frequency and phase correction for the system because there is something special about the additional phase coherence/high frequency phase correction. Yes, it is quite something isn't it@ecwlI have yet to come across anyone who prefers the partial over the full range correction. Re: "Partly, it’s the concern that the upgrade to an even higher end speaker system might not be as dramatic once a proper convolution filter is in place." Checkamundo!! Re: "But I’m sure whatever I decide, I am still getting this new level of sonic performance." Enjoy the reference sound! ecwl and vavan 2 Accurate Sound Link to comment
Popular Post mitchco Posted November 19, 2020 Popular Post Share Posted November 19, 2020 @jamesg11 and @ecwl I do both. For some folks, it is wherever they have plunked down their system, for others we can work out room modes and try and find an optimal placement that tries to avoid deep nulls or peaks. Tools like REW's Room Simulator or AMROC's Room Mode Calculator can assist. I like the latter calculator as a) one can hover the mouse cursor over a mode and it will output a tone at that frequency so you can hear it in your room (an ear enlightening experience) and b) one can get a great visualization of the room modes in the Room 3D View of the calculator as you hover the mouse cursor over each mode. Of course these are made for rectangular rooms, but close enough for rock and roll if one has an odd shaped room. The other realization is that there is no escaping room modes. One can place the speakers and listener in the least harmful locations, which usually means inbetween the worst of the peaks and dips. Placing sub(s) is usually the hardest part which is where REW's Room Simulator shines. Thankfully most subs have phase controls to help dial it in before DSP. But is is mostly one half dozen or the other in the end. For example, I just plunked down my speakers and subs where I wanted them to go, dictated by the layout of the room and where everything could fit. But the DSP was able to do its thing and ended up being textbook perfect over a wide listening area: https://www.avsforum.com/threads/official-rythmik-audio-subwoofer-thread.1214550/page-1834#post-57390652 Other folks have taken the time to setup the speakers as best as possible before DSP, which is what I recommend if possible, and the results are also excellent. It is not that the DSP has to work more or less, it is more about the better the input, the better the output, as we are talking transfer functions here. jamesg11, ecwl and vavan 3 Accurate Sound Link to comment
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now