Folks,

Those of you who follow my magazine articles and forum posts know that I'm a huge fan of the ETF software. ETF provides all of the tools needed to do a thorough room analysis, yet it's very affordable. ETF author Doug Plumb has been developing a new program called R+D - for Resonance and Distortion - that will eventually replace ETF.

In an effort to move people over to the superior functionality of R+D this newer program is available at no additional cost to all registered ETF users. Doug is working hard to expand the documentation and example files for R+D, but its operation is very different - and improved - compared to ETF that we're now used to.

I know that many of you here use ETF, and probably R+D too, so as a service to the MusicPlayer.com community I have arranged for Doug to visit on a regular basis to offer technical support. Indeed, the idea for this thread came from my constant pestering of Doug by email! He was spending so much time answering my questions and explaining things that we both realized it would be useful for others to benefit from his answers too.

Besides Doug's valuable involvement, having a single sticky thread where all questions and suggestions are posted helps all of us have our questions answered more efficiently. Therefore, I have started this thread to serve as a single repository for all R+D questions and answers. All I ask is that you please keep your posts on topic, and be as brief and to the point as possible. This helps Doug, and it helps the rest of us too by making it easier to see if a question has already been asked and answered.

Also, be sure you have the latest version of the program, which is updated regularly, and have viewed all of the demo videos, example files, and Application Notes posted on the R+D site:

http://www.etfacoustic.com/RPlusDSite/index.html

Finally, please understand that this is not the place to ask sales-related questions! The purpose of this thread is to provide technical support for R+D only, and possibly to discuss current and future features, so we can all learn to use this terrific program to its fullest capability.

Welcome Doug, and let the exchange begin!

--Ethan

Doug,

I'll kick things off:

Since R+D is still in development, it's great that you post updates on your site as you make them. But what's not clear is whether we need to uninstall the existing version first, or if it's okay to just install again which is more convenient. I've done it both ways with no problems, but I figured its worth asking you just to be sure.

--Ethan

Thanks for creating this Ethan. I think it will be a great benefit.

Oh, and Doug - thanks in advance for stopping by.

Thye installer will give you the option of only repairing or removing the software only if it is currently installed.

When you run the installer on a new download then when prompted remove the old.

When you re run the installer the installer will give you the option of installing the software.

It will not let you install the software if it is already installed without removing it first.

I followed Ethan here from AVS Forum and the "ETF shortcuts" thread . .

I'm a bit confused - is the new app ready yet? How much re-learning will I need to do? My plan was to re-position my sub and re-do the EQ tonight. Should I just stick with ETF for that and look at the new app for some upcoming RT60 stuff I want to do?

R+D was designed to be better than ETF in every way and one weakness with ETF is its lack of documentation.

I have made a real effort with R+D to produce better docs and there are videos that show you how to use the software.

App notes are being written to show how to take and interpret the measurements.

I think R+D is even easier for seasoned ETF users. You could get more from R+D than ETF just from learning it from the videos - about 40 minutes in tot for 4 videos.

The videos show an older version of the software, many features have been added but the videos. The videos will be updated when the project is done.

R+D uses a new method for RT60 so its much better than ETF but some people are having problems getting it to work well for them and I haven't figured out why yet.

I'm currently finishing the emulation for the FBQ 2496 and I'm doing a "snap to fit" feature for the filters making exact emulation easier.

R+D is the best way to EQ that is available anywhere at any price in my opinion and thats what the app is really all about. It's better than ETF for measurements and I did this so ETF users would switch and I wouldn't have two apps to support.

R+D is free when you buy ETF.

The MLS test signal has the primary advantage of speed. Computers today are very fast and this increase in speed isn't really necessary.

I'll be using MLS in the real time analyzer because it is fast.

The real time analyzer is primarily used to measure EQ's and set them to the R+D solution.

I am currently working on a DSP application note to explain what R+D does with DSP and how to use it.

Doug,

Welcome to the site.......

It's great to see you here and this will probably be a huge benifit to both you and the users of your product.


Folks - I am excited about the developement of R+D - in fact - it was being developed as I was writing my new book - and although ETF is featured in the bare room testing section of the book - it is R+D that is featured in the final testing and room treatments section.

Doug has some great things happening here - so ask your questions and learn - you'll find it's well worth the investment - both from a cost point of view as well as the overall benifits.

Rod

I'm just finishing the FBQ 2496 emulation.

Its important that DSP1124 users note that BW in the DSP 1124 is not the same is in the FBQ 2498.

A bandwidth of 10 in the DSP 1124 is actually the same as a bandwidth of 20 on the FBQ 2496

Hey Doug,

Thank you for all your help over the phone.. I most tell about 10 people a day to buy your product so I hope I have helped sell a few here and there..

Glenn

Thanks Glen.

When I put a link up on my site later today this thread should become much more active.

I'm just still testing version 1.0.28. Any users will want to download this after I post it. Its got GUI improvements for measurements and for EQing. It now emulates the FBQ 2496.

Doug, I'm having a problem with R+D. I have it set up in loopback configuration to set levels. I click the Check Levels button and get the following messagebox: immediately followed by followed my a repeat of the first messagebox. The vertical meter then shows some activity and the level readouts show -22 dB.

I'm using a M-Audio Revolution 7.1 card. No problems with it under ETF. R+D version 1.0.28.

It's working now. It was one of two things, probably the second. 1) I loaded the RSAnalog.cal file, 2) I had the Shockwave tutorial on pause while doing the loopback level set. Shut down the film, re-tried, and it worked.

Sound levels are right about where I had them set for ETF, so I think we're okay now.

The software and sound card interface is perfect as far as I am concerned. Any reported errors turn out to be user errors. This is not surprising since the same code since 1996 is being used. It is bug free.

R+D is different from other analyzers in that it uses decorrelated L&R test signals. Sometimes L&R get mixed up at the output and it makes no difference in other analyzers.

Anayzers that have correlated L & R test signals interpret sound card interchannel crosstalk as signal. R+D uses decorrelated signals and this crosstalk becomes noise in the results. Its better this way but its also slower to compute.

> It is bug free.

I've been writing software for nearly 30 years and I've learned to take that kind of claim with a grain of salt.

You have to consider the software as a whole package. Your users certainly will. I've seen way too many packages that are "perfect" if used in a certain way but which fail mysteriously if used in a way that the author didn't intend or test for.

The first part of the Ch 1 tutorial is just asking to be used stepwise, as I did. If that causes the package to fail, at the minimum you should warn about that in the opening of the tutorial.

If I was developing the product I would document this as a package bug and put it in the task list.

R+D will never be bug free but this part of the code that accesses the sound card does so in only one way and uses a fixed set of functions to accomplish the task. Every possible usage gets tested each time the software is run.

During the first few years up until yr 2000 or so I was finding little errors that caused problems on some sound cards. Now I do not find these anymore. When RPlusD was first released I changed the audio code from ETF a little and some people were reporting problems.

It is now the same as in ETF and user problems with specifically getting the sound card to work are always user errors.

The same cannot be said for the automatic mixer but it too has a fixed set of commands and this code only performs two different functions in the same order.

"If that causes the package to fail, at the minimum you should warn about that in the opening of the tutorial.
"

What exactly is the problem that you would specify in the documentation ?

There is a limit to the amount of code that I want to put into the soundcard control - adding more checks can lead to a lot more very confusing problems that are not so easily diagnosed.

At the beginning of Ch 1 I'd say something like, "In this chapter we'll discuss how to set up the sound card for use with R+D. Of course, if like most people you only have one sound card, you won't be able to do this while the movie is playing. Watch the movie, then close your browser to do the setup."

No additional code required.

First of all, thanks for coming here to answer technical questions on our behalf.

My background is someone who used tone generators and SPL meters to manaully plot room responses (which took a long time) and then trial-and-error set filters on my BFD. I've since been using Room EQ Wizard for some simple tuning issues (the recommendation of exact filters for the BFD 1124 is nice).

However, I am really interested in "going advanced" with ETF. I came here because the next step for me is room treatments and careful positioning of my mains (dipole planar speakers) -- definitely outside the realm of REW. Looks like R+D is going to fit the bill.

If I want to get started with it now, I just need to get ETF and it's included? Or do I need one of the add-ons.

If my laptop can do line out (I can reconfigure the audio ports to be line/mic/digital/etc), would I still need an external USB sound card? It looks like I would not due to the auto-correction, but as long as we're asking questions here

Thanks,
AC

Hi Anth,

You may not need the external card. Try the demo. Watch the videos to see how to hook it up.

You only need the basic version of ETF to get the basic version of R+D for free.

There is no trial and error work with the filters in R+D. It does a full filter emulation.

A mixer bug was fixed. This would prevent some single sound cards from not working correctly with the automatic mixer.

This is version 1.0.30

Hi Doug, hope you are doing fine ;-)

I just got 1.0.29 and it reflects wrong speaker distance / delay measurements, 1.0.30 too.
And subsequent measurements change the value too, must be a bug ?
And a saved measurement will not display the distance...
I did the measurements at a level of about-3 dB / -0.4 ( Gathering window ), on a DCX 2496 ( x - over ) Lo freq output.
Shall I send you a screenshot ?
It'd be great to work on it today, as I am busy

Thanks and best regards

Klaus

"I just got 1.0.29 and it reflects wrong speaker distance / delay measurements, 1.0.30 too."

Thats the electronic delay of the DCX 2496 Xover being added to the physical distance. It should be equal to about a meter or so. I haven't tested the DCX 2496 so I don't know what it should be.

I would like you to email me a zipped file. Screen shots don't help.

A saved measurement does display didtance you just have to click the single curve display to see it.

The distance shown in ETF + the impulse shift from zero (usually none) reflect the true distance + any electronic delays.

R+D files can be compressed. This compression occurs by chopping off the noisy tail of the impulse. This compression facilitates emailing files on the internet.

You should compress so that the compressed file has a time period greater than twice the longest gate time you intend to use.

I cannot receive files with attachments other than zip.

I *want* to measure the delay introduced by the DUT, but the results are somewhat esoteric...
751.31 m speaker distance is really not what it should be, but 4.22 m, measured with another tool.
I managed to get the distance shown of a measured file by reverting from freq response to impulse response and pushung the left arrow, although only 1 curve is in the file...
But 751.31 m ? Whoah
Definately something wrong.
I send you the screenshot ( zip ), please look at the SN ratio, at these levels, it should be a lot better.
It is a lo freq spectrum, that behaves like this... as I said
Thanks a lot
best

Klaus

Thanks Doug. As soon as I get my taxes done, I will talk the wife into letting me spend some money!

Klausrm,

This file is practically nothing but noise.

It really looks like L&R are mixed up or something.

There is no identifiable impulse- just all noise.

If you look at the freq response ( log ), it reflects pretty much the x - over output including some filtering at around 100 Hz.
Don't know what's happening in the impulse window, but clicking on
the left arrow button in impulse mode shows the impulse for me
?


Best

Klaus

This measurement is all noise. Look at the S/N ratio.

I've scanned the impulse response.

If anyone is having trouble with the latest download in getting the sound card to work, 1.0.30 and you have had a version installed before then delete your saved mixer settings from Function->Clear -> Delete AutomixerSettings. This is absolutely critical.

The automixer settings were modified before saving (a bug) which allowed the software to work once and then on subsequent tries not work. This happens very occasionally.