I'm using open office 3.3.
I keep getting Error 502
502 Invalid argument A function argument has an invalid value, e.g. a negative number for the root function.
when I try to extract the minimum phase data in the spreadsheet using the manual method. Does anyone have experience using these spreadsheets in Open office?
I tried emailing the author but:
Delivery to the following recipient failed permanently:
The error that the other server returned was:
550 5.1.1 <charlie@claub.net>: Recipient address rejected: User unknown in virtual mailbox table
I keep getting Error 502
502 Invalid argument A function argument has an invalid value, e.g. a negative number for the root function.
when I try to extract the minimum phase data in the spreadsheet using the manual method. Does anyone have experience using these spreadsheets in Open office?
I tried emailing the author but:
Delivery to the following recipient failed permanently:
The error that the other server returned was:
550 5.1.1 <charlie@claub.net>: Recipient address rejected: User unknown in virtual mailbox table
@Zoneschimmel, I sent you a private message about how to contact me. I will try to figure out where the problem is happening, although I do not have Office 3.3 (I assume you mean open office).
-Charlie
-Charlie
Well, some bad news on this topic. I'm afraid that the Blender doesn't work in Calc (e.g. OpenOffice, LibreOffice, etc.). It's been a couple of years since I developed the algorithm. This was done in Excel, and takes advantage of some very funky things that you can do with just worksheet formulas. Although I thought that I had tested this in Calc as well, I can't seem to get any of my development versions to function and obviously the current version doesn't work either in this program. After playing around with a few of the formulas, which I thought were the root cause of the incompatibility, it seems that it is the way in which Calc calculates the spreadsheet that is to blame (as far as I can tell). The method works under Excel, but not for Calc. This means that the Blender doesn't function in Calc. Sorry about that. It's odd that you are the first person to mention this (the Blender was developed in 2012) so most people must be using Excel.
It's possible to use the Play On Linux emulator and get Excel working on your system. The Blender should work under Excel 2003 and later. For info, see:
https://en.wikipedia.org/wiki/PlayOnLinux
https://www.playonlinux.com/en/supported_apps-3-0.html
I'd like to try this myself, actually, since I would like to get Excel 2003 or 2007 running on my Ubuntu machine.
It's possible to use the Play On Linux emulator and get Excel working on your system. The Blender should work under Excel 2003 and later. For info, see:
https://en.wikipedia.org/wiki/PlayOnLinux
https://www.playonlinux.com/en/supported_apps-3-0.html
I'd like to try this myself, actually, since I would like to get Excel 2003 or 2007 running on my Ubuntu machine.
Not a direct answer to Zonneschimmel's issue, but you can also use the OmniMic software (you don't need the OmniMic hardware) to blend FRD curves. Import the two curves (only 2!) into the Frequency Response graph as "Added curves". When you add each, select the frequency range of it to display (and use for blending) in the properties form that appears after addeing. When you add the second one, click on "save spliced to another" to make the blended FRD file.
If the original curves overlap, they will be blended in the new file. If they don't, they will be smoothly interpolated into each other.
I'm not positive that this is the same algorithm as Charlie's though, but worth a try, and free. Look for "Latest Software Version" at Dayton Audio Dayton Audio OmniMic V2
If the original curves overlap, they will be blended in the new file. If they don't, they will be smoothly interpolated into each other.
I'm not positive that this is the same algorithm as Charlie's though, but worth a try, and free. Look for "Latest Software Version" at Dayton Audio Dayton Audio OmniMic V2
If all you need is to blend two FRD curves together, I can give you (that is anyone, just contact me) an early version of the Blender that doesn't include the minimum phase extraction stuff but works fine in Calc. Without the phase part (phase is just set to all zeros) I'm not sure this is all that useful however.
I could try to build on to that some functionality that simply "blends" the phase responses but I believe that is typically not all that accurate except under quite limited circumstances. I was recently contacted by someone who wanted to do something like that, so perhaps it is worth some effort. If you think so, please post here or contact me. I don't want to invest time making something that no one will use...
I could try to build on to that some functionality that simply "blends" the phase responses but I believe that is typically not all that accurate except under quite limited circumstances. I was recently contacted by someone who wanted to do something like that, so perhaps it is worth some effort. If you think so, please post here or contact me. I don't want to invest time making something that no one will use...
Thanks for looking into it Charlie. I'm on Windows 7 so no problems getting Excell to work. I'd like to use the blender for getting a good starting curve to build linear phase filters with rePhase for the woofer. For the mid and tweeter I'm using farfield gated measurements, but for the woofer I'd like nicely blended curves. I'm very new to FIR processing but it would seem phase information is quite important for doing that.
With RePhase, you can just import the measured dB&phase curves (adjust delay for closest to linear phase to start out) and adjust freq points by hand for linear phase and magnitude, generate the coefficients, load to your FIR engine. Actually it's surprisingly easy.
That is how I did it now, I have a linear phase system (between 100 and 20K Hz) running but for further finetuning of the respons below 400 I'd like a better starting measurement. I'll take a look at Omnimic.
- Status
- Not open for further replies.