OminMic and PCD - the great FRD Conundrum

[Brewski]

I couldn't tell you that part of the equation I find software/hardware bugs for a living. My guess is it's close to write protecting the data that was saved. Both files had protection enabled at the start.

Take it easy
Jay

[Jeff B.]

what am i missing? a zma should have two columes, impedence curve and phase data. a frd file should have two frequency and phase. please explain the third.

Frequency.........Amplitude(dB).........Phase

[bwaslo]

Arlis,

The three columns are:
Frequency (Hz), Magnitude (dB), Phase (degrees).
The last is optional.
---
I'm going to update the FRD writing routine in OmniMic so that it doesn't print the header line and so that if no phase is being measured, it will list as "0" degrees. That may help with compatibility. If anyone can figure out what all is going wrong, please let me know and I can change it in OmniMic.

I specifically changed the routine in OM to not use scientific notation in case that would confuse some file readers, but didn't anticipate these issues!

[Brewski]

what am i missing? a zma should have two columes, impedence curve and phase data. a frd file should have two frequency and phase. please explain the third.

I only worked on the day115 frd file from your email but there's only two columns of data at the start so the third column of phase data needs to be added I think.

Take it easy
Jay

[arlis_1957@yahoo.com]

i got it. so when incorperate phase, they should work. but you would think they would inport to rm as is.
look foward seeing what you two come up with. thanks for trying.

[bwaslo]

Brewski, can you post the file (or 20 or so lines of it) after you've processed it?

It's just text, no way to protect anything in ASCII that I know of. (Though Windows might prevent anything from overwriting it if the file is also still open in Word or some other programs).

[Brewski]

Just pm me your email addy and I'll send you the after files. playing with 1 year old at the same time... but will send it when I can

[bwaslo]

Your attached file and the original both load to the same thing into OmniMic (and into Praxis). Only diff I can see is more spaces between freq column and dB column on most rows, and addition of the phase row. I wonder if it is only the phase row that is messing things up.

Jeff, does PCD need to see a row of phase values in order to import? If it does, does it try to use them or does it still do Hilbert phase according to the mag data?

[Brewski]

Your attached file and the original both load to the same thing into OmniMic (and into Praxis). Only diff I can see is more spaces between freq column and dB column on most rows, and addition of the phase row. I wonder if it is only the phase row that is messing things up.

Jeff, does PCD need to see a row of phase values in order to import? If it does, does it try to use them or does it still do Hilbert phase according to the mag data?

I can tell you in a bit... Wife just got home to take over baby duty

It looks like just phase was able to fix Arlis's DAY115 file.
Same for the dayton FRD file as well. I dismissed this as the cause since Jeff mentioned trying it as a fix.

[bwaslo]

Thanks Brewski

[arlis_1957@yahoo.com]

they need to be inported into response modeler as well for box and baffle modeling.

[Jeff B.]

Arlis,

The three columns are:
Frequency (Hz), Magnitude (dB), Phase (degrees).
The last is optional.
---
I'm going to update the FRD writing routine in OmniMic so that it doesn't print the header line and so that if no phase is being measured, it will list as "0" degrees. That may help with compatibility. If anyone can figure out what all is going wrong, please let me know and I can change it in OmniMic.

I specifically changed the routine in OM to not use scientific notation in case that would confuse some file readers, but didn't anticipate these issues!

I did some experimenting by simply typing into notepad and creating some frd files. I found I that my VBA routine (used in Response Modeler and PCD) will not bring in a file with only two columns of data, otherwise I could pull in some very simple selfmade files. If the output from Omnimic always has a third column then I think we will be OK. However, I am still not sure why the earlier file wouldn't work after I added a column of 0's for phase. Maybe I accidentally did something else to the file without realizing it.

Jeff

[arlis_1957@yahoo.com]

with a frd created by using spltrace, i have imported just fine into both of your programs. i will experiment some tonight.

[exojam]

Jeff,

Not sure if it helps but I am able to load Omni fdr files with only the first to columns into REW with no issues. I have not gone into the advanced features of Omni to generate a fdr with phase so I cannot say if that file would load into REW.

James

[bwaslo]

Jeff,

My only concern with adding a column of zeroes for phase is whether your (or other) programs would take that to mean that the phase really is zero (which is of course wrong and could cause bad simulations). The thought for leaving the last column optional is that not having it would simply mean "unknown" as opposed to giving incorrect data. If PCD always calculates phase from magnitude (and thus ignores any given phase) then no problem in that case (though who knows what other programs might assume).

[johnnyrichards]

I did some experimenting by simply typing into notepad and creating some frd files. I found I that my VBA routine (used in Response Modeler and PCD) will not bring in a file with only two columns of data, otherwise I could pull in some very simple selfmade files. If the output from Omnimic always has a third column then I think we will be OK. However, I am still not sure why the earlier file wouldn't work after I added a column of 0's for phase. Maybe I accidentally did something else to the file without realizing it.

Jeff

I tried getting Arlis' files to work as well. I added the third column, no go here, either. I tried a lot of things, but nothing really worked for me either. I tried converting to .csv and back to tab delimited again hoping that would reformat the entire text file but that didn't work.

I can see where the Omnimic guys might be getting a little TO'ed right now but it is good to see Bill is aware of and addressing the situation.

[exojam]

Jeff,

Again I am not sure if this will help you but as a test I took an fdr that was created from Omni with no phase, added 00 to a third column where the phase would go and REW would open the file. I placed a snip of the screen below.

James

[Jeff B.]

Jeff,

My only concern with adding a column of zeroes for phase is whether your (or other) programs would take that to mean that the phase really is zero (which is of course wrong and could cause bad simulations). The thought for leaving the last column optional is that not having it would simply mean "unknown" as opposed to giving incorrect data. If PCD always calculates phase from magnitude (and thus ignores any given phase) then no problem in that case (though who knows what other programs might assume).

I encourage everyone to extract minimum phase using the HB routine in my Response Modeler, or to use the actual measured phase if you know the distances are correct. However, my HB tool can't extract the phase from these files either. It just locks up my computer and producea 15mB error report file.

If PCD was able to load your raw two column file it would display 0 degrees for phase anyway, so there wouldn't be any difference between that and actually having a file with a third column of 0's.

I would add that any time someone pulls an frd file into a crossover simulator that didn't have any phase data (like most of the people writing me were trying to do) their crossover simulation would be incorrect and invalid anyway, and they may not even know that if they didn't know how important the phase data was for driver summations. Based on that, the additional column doesn't really hurt more, it just helps to make the file more useful. I am not much of a programmer, but I will still see if I can do anything on my end that will help too.

By the way Bill, I really appreciate you jumping in and discussing this.

[arlis_1957@yahoo.com]

ok, i remeasured with phase. deleted headers, first and last return. they still will not import to rm or pcd.

[Jeff B.]

ok, i remeasured with phase. deleted headers, first and last return. they still will not import to rm or pcd.

Send me that file. Let's see why.