Single-Point Digital Acoustic Correction for desktop & HiFi speaker systems Using REW and rePhase P5
ฝัง
- เผยแพร่เมื่อ 11 ต.ค. 2024
- Part 5
In this section, we use rePhase to compile the excess phase data and the magnitude filter banks into the left and right correction filters. These filters are then brought back into REW, where they are combined into a single stereo correction filter.
Notes:
Filter Taps: 65536
Windowing: Blackman-Harris
Sample Rate: 96kHz
Format: 64 bits IEEE mono (wav)
All filter bands below 500Hz should be in minimum phase banks.
All filter bands above 500Hz should be in linear phase banks.
Hey there, just wanted to also thank you for your most informative, thorough lessons I've had since I got back into audio. This is certainly different than setting up active components in my car audio days and just finding time and XO numbers... I have been trying to understand and demystify rephase, so this helped fill in a few blanks and help me figure out an active crossover loudspeaker build- mostly 2-way with two woofers one tweeter, but I am equing the woofers separately just as experiments, so while you were going into other hardware at the end, I took what you explained and more or less made it work for a minidsp 2x4HD, but also adding a linear HP crossover in rephase for the tweeter, and keeping woofers PEQ. And EQ'd without house curve to flat since I will do house curves after XO. Pretty sure you just upgraded my dynamic range by a lot, but letting the in-your-head stuff dissipate first. was really hoping to figure out how to use the mult-banked PEQs developed into the 2x4HD, but haven't figured it out yet and just use a single bank for now since I am switching to another DSP for these (ncore fusion plate amps). 2x4 will move to some bookshelves, so ill re investigate another time. Seriously your channel deserves 1,000 likes from me, as you also helped this nerdy brain understand more without all the hours of combing through the forums. I seriously think its a mix of ocd/undiagnosed adhd or something bercause this I got, but getting, collating, separating fact from fiction on all the forums was getting really tiresome. I believe someone posted your link in the ASR forums, so you do have a few out there that appreciate this knowledge.
Thanks! I know how it is. I was the one combing through everything for over a year myself. I figured out some things, though, and put most of it into these videos. There will be more as I learn and improve as well.
Your tutorial so clear and straight forward. Yes, alot of tedius back and forth implementing and saving but hey its free. One of the best guides I have seen. Thank You, David Brancato.
I was all new to this when I decided to take an audio step beyond buying a DAC and Powered Bookshelfs for my desktop PC. I keep thinking it would be nice to just take a few Microphone measurements, input them into a software program and let it do all the steps you covered into this tutorial. I guess Dirac comes closest to this, but still alot of footwork involved. The person who can write this dreamware will sell millions of copies.
Hello, I would like to thank you for your excellent work.I mixed these filters you explained here, with the Phase Inversion Filter i found in another YT channel, (Mr Gùr , if I can tell), to extend phase correction to 1kHz and my spaker never sounded so good. Thank you so much.👍👍👍
That's great, I tried it and it gives better results than other methods you can find in yt. Thanks a lot!
That's great I'm glad it helped you. Keep an eye out for a new method soon. I just need time to focus on it.
Congratulations for the videos, really good!. I didn't understand one thing, on the phase there is no correction with this workflow, if we measure the speakers after inserting the filter, the phase module remains the same....
I hadn't noticed that, mostly because I was going for more auditory confirmation that the sound had improved. I used filters from Audiolense, Focus Fidelity, and Dirac Live for comparisons when figuring out the steps. I wasn't too worried about how the other screens display info... perhaps that could be something to look into, but the filters sound great, so maybe not :)
@luciano mariotti are you referring to not adjusting the phase using the crossover and paragraphic phase adjustments?
great tutorial.
Why do you work with the excess phase and why dont you just correct everything with linear phase fir filters? What is the point importing the excess phase to rePhase? Sorry that I have missed this point. Best regards
I've just used your method. Thank you very much for sharing your experience in the form of this tutorial! Highly appreciated. I still have a couple of questions:
1. Is there also a written version (for future reference)?
2. I have a 2.2 setup, how to incorporate them correctly in this method? What I did is I had my subs with front speakers time aligned, x-overed and pre-EQd up to 120 hz, based on Serkan's method and routed both subs with both right and left channel when doing measurements to start with your method.
3. I see significant differences in phase response between measured ir and minimum phase fitters between 300-3k and 10k-20k. I can hear the difference but it's hard for me to determine which is "correct", can you please elaborate a bit more? What should I listen for?
4. I applied the filter on EqualizerAPO, and it shows a very high latency of 670 ms which means video and audio are way out of sync, what's the reason and can it be improved? Is there any general way to apply video delay as correction? Unfortunately it's not in my TV settings
1. No, there is no written version.
2. I haven't been able to set up twin subs yet, unfortunately. I don't have the right hardware available yet.
3. Listen for tight controlled low end and clear open top end - wide 3D soundstage.
4. I don't use EQAPO but I do know you can change the settings in rephase to shorten the delay. Change the centering from "middle" to something like 20ms or 30ms.
Thanks for this informative Tutorial. I'm just asking myself, how to correct for Phase shifts of the speakers that you can see in a windowed measurement with REW and how to maybe use the paragraphic Phase eq within rePhase.
If I generate a filter for lets say the left channel based ob this tutorial I can also directly upload it to my miniDSP 2*4 (with the right No. of taps of course)... Right?
I'm not sure. I've never tried. I only use the filters within my operating system via convolver. Try it and see what happens!
Why exactly are you using the Minimum Phase and Excess-Phase Versions? Actually I'm Not sure which information these contain.... 🙄 Is it "just" for time correction in the bass area up to 500hz? Why can't I see any movement of the graph when uploading the EP-version to rePhase in the beginning of this video?
@@andreasheiden7122 For the pure and simple reason that I was very pleased with the way the final filters sounded. I generated many different versions of the filters using various methods and this was simply the best way to do it.
hello david your videos more informative..... thank uuu
Different dsp use different filter formats .txt ,csv etc ....is it possible to export from rephase in csv format
Hi David, can achieve acceptable results using your method and my DSP, which is the MiniDSP FLEX?
When would I use the regular convolution filter vs. the minimum phase version? Just a matter of which one sounds better to me?
Exactly. Use the one that you like best.
Wow. this is awesome. thanks
You're welcome
Very well done and very informative but is there two P4's?
There are two different methods I've gone through. The first method was done before this one, and I had used multiple measurement positions to work from. The second method only uses a single point for the measurements, which I recorded three times. So, yes, there are two P4s, but one is for the first set of videos and one is for the second. Is that what you were asking?
What is the difference between this set of videos and the other? I just watched both series and perhaps I missed it
Another question: can you confirm the possibility to combine gain EQ based on measurements with moving mic method (MMM) and use vector averaged sweeps for phase correction?
I'm not sure about MMM. I measure everything with sine sweeps the same way the professional packages do: from either a stationary single-point main listening position or an array of stationary points around the main listening position.
@@davidbrancato update: I made new filters using your method, but with MMM measurements to create the EQ banks. I used vector average of sweep measurements for developing the excess phase version of the corrected vector avgs, but that was a lot of work just to end up with an excess phase measurement file that did not contain much (relevant) data at all, just a very gentle slope of a couple of degrees of phase at the upper end of the frequency spectrum. I don't fully understand what's going in that step. In the end I chose the minimum phase filter version that sounds very good!
How to prevent ringing and check for ringing?
This method is not quite that sophisticated. Remember, REW wasn't designed to compete necessarily with the state of the art software packages. If anyone knows of a way, I'd be interested to see it, though.