It's been a long time getting to this point, almost a year, but the first version of the WinPCD, the .NET version of Jeff Bagby's PCD spreadsheet, is available for download. A few of you may have followed the progress posting at my web site, but for those who have not, you'll find some details there. I made the final changes late last night, so it's likely that a bug or two will still exist.
Please go first to my blog, I'm trying to work that in since it will allow users to post comments. This does require a request for a subscriber account, however. What I've read about spamming prevents me from making it a fully open blog, but please consider doing this so that partly as help to others who may consider using it.
I've updated my signature with a link to the blog (the link in my signature). From there you'll find a link to the download area.
I've already got a couple of additions in mind that are generic, but feel free to suggest changes. No guarantees on that, but all feedback will be carefully considered. There's still work to be done, but the core is done as are both the 2-Way and 3-Way sections. Some right-click context and mouse hover utility has been built-in, more to come over time (especially with usable feedback ).
Also, consider taking the poll found at my blog. I think I've got it set up so that anyone can take it, no account necessary.
Dave
Please go first to my blog, I'm trying to work that in since it will allow users to post comments. This does require a request for a subscriber account, however. What I've read about spamming prevents me from making it a fully open blog, but please consider doing this so that partly as help to others who may consider using it.
I've updated my signature with a link to the blog (the link in my signature). From there you'll find a link to the download area.
I've already got a couple of additions in mind that are generic, but feel free to suggest changes. No guarantees on that, but all feedback will be carefully considered. There's still work to be done, but the core is done as are both the 2-Way and 3-Way sections. Some right-click context and mouse hover utility has been built-in, more to come over time (especially with usable feedback ).
Also, consider taking the poll found at my blog. I think I've got it set up so that anyone can take it, no account necessary.
Dave
First I've heard of this, and very much looking forward to using it. Thank you Dave, will give it a go presently.
Have you tried it out yet? I've got some changes made towards another release, but would like to get more feedback from users.First I've heard of this, and very much looking forward to using it. Thank you Dave, will give it a go presently.
Dave
Dave,PM me and I will provide the weekends feedback and samples of the xml and frs files.
I like this thing. Just need to help you wring the bugs out.
I have the better part of my quick start done for HOLM and SimpeS as those are the tools that will feed you the frs and zma files. I have an overall list of steps for all the tools needed, lists of where to get stuff, additional reading, and tips. It looks like about 20 pages.
I like this thing. Just need to help you wring the bugs out.
I have the better part of my quick start done for HOLM and SimpeS as those are the tools that will feed you the frs and zma files. I have an overall list of steps for all the tools needed, lists of where to get stuff, additional reading, and tips. It looks like about 20 pages.
The frd input is due is the file size. When I deleted sub 10 Hz data, it shifted way over to where it went to zero. These stupid programs that calculate to ten decimal places when two is more than accurate! This file had "only" 1000 steps, but all three columns are 10 decimal places. If I were a coder, a perl script to clean it up would be really easy. I guess the best way is on ingest.
I'd like to check out the files at issue. Would you send them to me? I'm not aware of that issue with my program. Mine ignores any data below 10Hz.The frd input is due is the file size. When I deleted sub 10 Hz data, it shifted way over to where it went to zero. These stupid programs that calculate to ten decimal places when two is more than accurate! This file had "only" 1000 steps, but all three columns are 10 decimal places. If I were a coder, a perl script to clean it up would be really easy. I guess the best way is on ingest.
Dave
Have you tried it out yet? I've got some changes made towards another release, but would like to get more feedback from users.
I need to recover my FRD and ZMA files for my HF drivers off an older computer that I now use for audio measurements. I'll try to get to that tomorrow. I was able to load up the woofer data fine. It's been a good while since I used Jeff's version, but this one looks quite promising so far.
I will send an email so you can subscribe me to your blog. I got a lot of things working, but am unclear which other ones are enabled at this point.
Last edited:
I'd like to check out the files at issue. Would you send them to me? I'm not aware of that issue with my program. Mine ignores any data below 10Hz.
Dave
We had a couple issues here.
First, HOLM was putting "#" as comments, so short term, delete the comments from a HOLM file. Dave identified this and will update the future version.
You can cut down the number of points it exports. 1000 is the default. 500 is just fine for this. Either works.
Second is that HOLM puts out files referenced to 0dB. Dave expects absolute values. Here is how I achieve that:
1: I have a reference driver that I know, or at lease assume the efficiency of mounted in a baffle. I set up the distance to the mice with my meter-stick exactlly, fiddle with all the levels until I get a nice happy plot in Holm. I use a short gate time and smooth 1/3 octave. I then record what the level is at 2Khz. Adding this to the reference SPL gives me the offset I need to add to the HOLM export of my drivers under test.
2: Measure all my drivers, not touching anything in the setup.
3: Import into EXCEL as text.
4: Insert a column and an equation to add the offset, 107.2 in this case, to all the amplitude values.
Move the original amplitude column out of the way. You can't get rid of it due to the equitation reference.
5: Copy the three columns into WordPad. It won;t complain about it being a table.
6: Save as TEXT. This gets rid of the table leaving yo with a nice three column tab delimited .frs file. These also import into PSD-Lite and plot with the "real" amplitudes.
Only takes a few seconds really. HOLM does have an offset, but it does not like to go more than a few dB. Maybe it exports OK, but it sure can't display.
DAVE,
Nice program with a well choosen layout.
I have down loaded the program and try it out. The box-tools does not working. Here with the advice to add box-tool1 and box-tool2 (e.g for woofer and a mid enclosure).
It give more services by easy overview and corrections. Inform me why the box scheduler not is activated.
At the present I am working with Jeff's Excel-Designer 7 (2003).
wmbr, ton
Nice program with a well choosen layout.
I have down loaded the program and try it out. The box-tools does not working. Here with the advice to add box-tool1 and box-tool2 (e.g for woofer and a mid enclosure).
It give more services by easy overview and corrections. Inform me why the box scheduler not is activated.
At the present I am working with Jeff's Excel-Designer 7 (2003).
wmbr, ton
Hello Ton,DAVE,
Nice program with a well choosen layout.
I have down loaded the program and try it out. The box-tools does not working. Here with the advice to add box-tool1 and box-tool2 (e.g for woofer and a mid enclosure).
It give more services by easy overview and corrections. Inform me why the box scheduler not is activated.
At the present I am working with Jeff's Excel-Designer 7 (2003).
wmbr, ton
Thanks. This release was just for the 2-way and 3-way crossover to get user feedback. The box section is yet to be done. I'm working now on some user interface changes from some feedback I've received. It's better to get some feedback before everything is done so that I can incorporate changes as I go forward.
Dave
Hi Dave,
Though I might have missed the controls, looks like the driver response and raw driver response are the same colors. Offset colors makes it easier to see.
Thanks,
Jay
Though I might have missed the controls, looks like the driver response and raw driver response are the same colors. Offset colors makes it easier to see.
Thanks,
Jay
Dave, two points regarding Holm:
1. You can choose not to have any headers in the file (in the export dialog box).
2. You can scale the output by upto 99 db. You can choose not to normalize the output, then scale by 99db and then export. This was enough for the old excel version of PCD. All you have be sure of is to scale all drivers by the same amount.
Will check out software later.
Thanks!
1. You can choose not to have any headers in the file (in the export dialog box).
2. You can scale the output by upto 99 db. You can choose not to normalize the output, then scale by 99db and then export. This was enough for the old excel version of PCD. All you have be sure of is to scale all drivers by the same amount.
Will check out software later.
Thanks!
I tried importing FRD files and gives me an unhandled exception error: Index was outside the bounds of the array. Win7, 64bit.
Any ideas?
Any ideas?
Hi Jay,Hi Dave,
Though I might have missed the controls, looks like the driver response and raw driver response are the same colors. Offset colors makes it easier to see.
I assume you're referring to the system display. I purposely left them the same because I see that screen as primarily for the equalized responses. Personally, I never display the raw responses there. The reason I thought it OK is because that is a very, very busy screen.There are many curves that can be displayed simultaneously. It's not likely to get differing colors for everything without getting one or more very close to another, creating confusion this way. You have raw, raw phase, eq, eq phase, total, total phase, overlay and power (eventually). I'll look into that again.
When I work with a raw measurement, I'm always in the tab for the specific driver working on the crossover section. The raw and eq'd response are different in the graph for individual drivers, raw is blue and eq is red. I only included raw display in the system graph to match the original PCD.
I have noted that I forgot to match the midrange colors in its tabbed graph to the W/T versions, so I'll fix that.
Dave
Thanks for the info, I hope that helps other users. I may have to create a "Notes" page for this type of info, probably just at my web site.Dave, two points regarding Holm:
1. You can choose not to have any headers in the file (in the export dialog box).
2. You can scale the output by up to 99 db. You can choose not to normalize the output, then scale by 99db and then export. This was enough for the old excel version of PCD. All you have be sure of is to scale all drivers by the same amount.
Will check out software later.
Thanks!
Not sure, I've seen this when the raw file had more than one extra line at the end of the file (occured when using Excel to manipulate the file for sensitivity by one user). I've fixed an earlier issue. If you'll email the file to me, I'll go into debug to find/fix it, if it's not this one.I tried importing FRD files and gives me an unhandled exception error: Index was outside the bounds of the array. Win7, 64bit.
Any ideas?
Dave
Not sure, I've seen this when the raw file had more than one extra line at the end of the file (occured when using Excel to manipulate the file for sensitivity by one user). I've fixed an earlier issue. If you'll email the file to me, I'll go into debug to find/fix it, if it's not this one.
Dave
Here's the file Dave. Rename to .frd and it should work.
Attachments
The reason it failed is because is it a CSV, comma-separate value file. I have not put in handling for that type of file. I'll look more into it, but for now save it with white space separation (one or more spaces or a tab character).Here's the file Dave. Rename to .frd and it should work.
Dave
Looking at this further, I noticed that the phase looked wrong. Your file is not "wrapped" phase, that is, it exceeds 180/-180 values. I handle this in the crossover, but had not considered that input files might not be wrapped phase.
I have added a change to handle files with phase such as this.
Dave
I have added a change to handle files with phase such as this.
Dave
- Status
- Not open for further replies.
- Home
- Design & Build
- Software Tools
- WinPCD version 1.0 released