Hi Doug and Ethan,

It's a pleasure finally have the RandD software. I'm having some trouble getting it to work out of the box, however. I will keep debugging, but some guidance might be helpful in the event that you've come across a similar problem in the past.

I have the software loaded on an old Dell Inspiron notebook computer running XP. It is one of the rare notebooks that comes with a line in and a line out jack.

I've wired the sound card in the loop-back configuration shown in your CH1 video (mini-plug to RCA adapters, connected via an RCA cable).

The problem is that I get a very noisy response when I run the test to characterize the sound card. Typically, I get a speaker-mic distance of zero (as expected), but my SNR is only about 0.2 to 0.4 db. Sometimes, for some reason, it jumps up to 18 or 19 db, but it doesn't stay there. There appears to be something noisy in the system.

Suspecting a bad internal connection, I ran out and bought a USB soundcard, the Turtle Beach Roadie. I loaded the drivers, installed the "card" and tried again.

This time, after checking connections multiple times, I got only 0.4db SNR best case (speaker-mic distance of 0).

I could not get this card to do any better than this. I checked my mixer settings. I had "Line In" selected for the record path and "MIDI" selected for the output path. Everything else was de-selected.

I checked levels, and tweaked the input/output volume settings to just avoid clipping. The SNR improved slightly as I approached full scale, but never got better than 0.4db.

My next step will be to try this on a different computer.

Any ideas what the problem might be?

Thanks!

Barebones,

I had similar problems with my Dell Latitude (though I managed to get everything working fine on my desktop) -- adding an external soundcard fixed the signal levels (which should be in the range of 0 to -10 dB), but the signal was still very noisy.

It was one of those "too obvious to think of until all complex possibilities have been explored ad nauseum" type things: the AC power was leaking through the USB port. When I disconnected the AC, running solely on the battery, the problem went away.

Hope this helps.

Cheers,

Wayne

Doug,

I know what you mean about ETF vs. RPlusD -- I've been getting into the new version for several days now, and the interface is quite elegant, not to mention RPlusD obviously offers many added features and a level of precision you don't get with ETF5.

There are so many ways of looking at the data, I'm sure it will be very helpful in analyzing my room.

Couple of questions, though:

In RPlusD, on the FR graph, is there any way to pinpoint the exact frequency and amplitude you are interested in? In ETF, you can click on the window at any point, and the frequency and amplitude are displayed to the right of the graph.

Does the right mouse button (dashed square) work the same way in the FR window as it does in the Schroeder plot display?

That's all for now. Thanks.

Wayne

Well shoot. Doesn't that mean that the sound card's SNR is really aweful, then, if it has such bad power supply rejection? Unfortunately, the batteries in this notebook are total crap at this point and can't hold a charge.

Switching computers seems to be the best choice. Thanks for the reply!

Wayne Rose,

I just finished testing out your idea. I set up a test and unplugged the AC just long enough to complete a measurement cycle. Then I plugged the AC back in and saved the screenshot.

If attaching the screenshot works, you can see that the problem only marginally improved. SNR still stinks.

I thought your suggestion was brilliant, however, and I would welcome any others.

EDIT: I can't seem to upload the image. Here are the details:

Mic-Spkr Distance = 0 m
Input Signal = 6.7 db
Input Signal S/N = 0.4 db

Excitation time: 3 s
Pre-excitation: 1.3 s
Auto Range
Hybrid Test signal

The impulse looks right; it's just that the noise is too high.

Barebones,

Your input signal is too hot. It should be between 0 and -10 dB. Then you will get a green "Level OK" message in the Audio Data Gathering display.

I just re-read your original message, and don't understand why you are selecting MIDI as your output port type.

Try this:

First, in RPlusD, turn Mixer Off from the Options/Mixer menu. Then delete AutoMixer Settings from the File/Clear menu. Then close RPlusD.

When the popup asks if you want to save your current mixer settings, click No.

Check your settings in Sound and Audio Devices Properties (usually accessed by Right clicking the volume icon in your system tray and selecting Adjust Audio Properties, or from System/Hardware/Device Manager through Control Panel -- make sure your new soundcard is selected, and that the appropriate audio ports are active. Close the Device Manager.

Then, open the soundcard Mixer and leave it open while you launch RPlusD.

In RPlusD, check to make sure Mixer Off is still selected in the Options/Mixer menu. From the File/New menu, select Sound Card Measurement.

In the Audio Data Gathering window that appears, click Check Levels. Now begin adjusting both the Output and Record levels in your Windows Output Mixer, as you continue to Check Levels (the levels should change appropriately after each adjustment).

In your Windows Mixer:

In the Output Mixer Window, Master Volume and Wave should be unmuted (both of these levels need to be adjusted to obtain the correct signal level in RPlusD). All of the other data types (SW Synth, MIDI, CD Player, Line In, Microphone, etc.) need to be muted. In the Record Mixer Window, make sure the Line In module is selected, and again, adjust the level to ensure the correct settings in RPlusD.

The adjustments are pretty subtle, so continue to make slight adjustments in the Windows Mixer, and Check Levels in RPlusD, until you get the Level OK (0 to -10 dB) message in the Audio Data Gathering window.

Once everything looks ok, close the ADG window and select Yes in the Save Current Mixer Settings Window.

Open the ADG window again. You should be ready to Start your soundcard test.

Note that there's no point starting an actual test until the levels are correct, as the data will be meaningless.

It's also a good idea to perform continuity checks on all your cables, and make sure everything is plugged in securely. Also, check to make sure that wdmaud.drv is in your Windows/System32 folder.

The setup process should work even with your AC plugged in. You can tell if you're getting noise out the Line Out port by connecting it to a line level preamp, amp and speakers (keep the levels low so you don't blow anything, including your ears!), and listening to the signal during Check Levels. You should hear the hybrid signal (pink noise and a sine sweep) and nothing else!

If that works, it means you're getting the right signal through the Out port, so if your In port is selected correctly, everything should be ok.

Let us know what happens ...

Cheers,

Wayne

Barebones: "Any ideas what the problem might be?"

Some cheap cables have left and right internally mixed up. I have no idea how this happens but believe it or not - it does happen. It took me a while to realize this.

The good news is that once your mixer settings are right the software will remember them for you the next time you use it and you will not have to set up the mixer again.

Thank you Wayne and Doug. I have some news to report. I could not get the new "Roadie" sound card to give me an SNR better than 0.5db (even with adjusted mixer settings), so I reverted back to the original card that came with the notebook. I found a can of tuner cleaner in my workshop and gave the Line In and Line Out jacks a thorough blasting. Several hours later, there was still no improvement in performance, but, seemingly miraculously, two days later the situation improved markedly. I'm now seeing 20 or 21 db SNR consistently, regardless of whether I'm using battery or AC. After convincing myself that the improvement was repeatable, I saved my sound card settings. I hope to make some preliminarly measurements later today--almost in time for the new season of Rome!

Wayne,

Thank you especially for your reply. As it turned out, I had selected "Wave" rather than "Midi." I just failed to remember it correctly when I submitted my post above. So the mixer settings were always correct.

Despite the good luck I had with internal sound card finally giving me a decent SNR, I started having other problems with the system (memory faults), so I've abandoned that computer and am now working with my wife's brand new IBM ThinkPad. The new computer does not have a "Line In" input, so I'm back to using the Audio Advantage Roadie. The Roadie is, by all online accounts, an excellent card. Yet I still can't seem to do any better than 8 db SNR. A typical reading is closer to 2 or 3 db.

I've tried unplugging the computer, moving to electronic "dead spot" in my house (no lights, no electronics, etc.), and shortening my cables. Nothing seems to make this card behave consistently well (or even well at all).

Perhaps the card simply cannot do full duplex sound very well. Can you or anyone else personally recommend a cheap USB sound card that you know works well in this application?

Many thanks.

I don't know what the specific problem is, but it's not the sound card itself. Even the cheapest sound card built onto the motherboard of the cheapest computer you can buy will have a s/n ratio of at least 50 to 60 dB.

--Ethan

Ethan,

Thank you for your reply. I am also hestitant to blame the sound card, long ago having learned that the part is hardly ever the problem.

I'm running out of ideas, however. The card I'm using is a full duplex card. That doesn't mean there's good channel separation, however. Aren't full duplex capabilities used mostly for Internet phone, where feedthrough isn't such a big deal?

I'll keep picking at this problem, but I've seen on the web that the USB Sound Blaster MP3+ and M-Audio MobilePre USB have worked for others.

Incidentally, the "Roadie" that I've been using comes with a control panel that shows line in signal levels. I can see that my signal is plenty big, just short of the red zone. So, levels don't appear to be the problem.

I'm also considering whether power might be the problem. Maybe my USB ports don't have enough juice to power the card in full duplex mode.

Just to salvage my self respect, I tried R+D on my office desktop with a built-in sound card. I got 21 db SNR right out of the box. Then I plugged in and installed the Audio Advantage Roadie. With the same mixer settings and hookup on the same computer, the best I could manage was 3 db.

I think I'm going shopping.

Barebones,

I found the Creative Labs Soundblaster USB MP+ on CL's eBay store, refurbished for $14.04 USD; it works great.

Your S/N ratio, as determined by RPlusD, will normally read in the low to mid 30's; anything above 20 will work (if I recall correctly from the documentation).

When you switched from the office desktop to the AAR, did you Delete Automixer Settings? If not, you would get a faulty reading when you switched cards?

The fact that you are getting the same problem on several different computers would suggest that you are consistently doing some small thing incorrectly (see Ethan's statement above about even the cheapest built-in soundcard).

One thing I had a problem with was that changes in my Windows mixer settings were not recognized until I selected the correct ports (or in your case, the right soundcard!) from the Device Manager's audio control dialog. Every time you switch computers, you are adding unkown variables to the problem. Try digging into your system's audio configuration, to make sure the underlying settings are correct.

Have you checked the signal at your out port to test whether you are getting audio through your soundcard at all? If not, there's no need to even be looking at the S/N reading!

And of course, the most obvious and often overlooked solution -- try changing out your cables.

I'm sure you've looked at all these things repeatedly. Sorry, but it seems I can add nothing more at this point.

Good luck with it.

Wayne

Wayne,

Thanks for your perseverence.

I may try the same card.

I did not, but I had never saved the settings (I always answered "no" to that prompt), so I think I'm OK.

The one constant across 3 different computers is that the AAR doesn't perform up to RandD's requirements. Both my original notebook (after fixing the contact problem) and my desktop gave better than 20db SNR with their respective internal sound cards. I wish I could use the original notebook, but it's giving me blue screens.

I could have overlooked something, but I think the settings are correct. I've changed 'ports', checked and rechecked mixer settings, and combed through the sound card's doc. No luck.

I've listened to a CD using the Roadie. It sounded normal. I haven't tried recording anything. I may not bother.

I used the same cables for all measurements, including those that gave better than 20db (for internal sound cards) and those that didn't.

Thanks for trying. I think I've reached the conclusion that there's something wrong with my sound card. I'll know better when I try something different.

Barebones,

Thanks for taking the time to respond in detail.

Playing a CD tests your soundcard, but not whether the RPlusD signal is being read. Try running Check Levels from within RPlusD, while listening to the output (keep your external volume low!).

Hang in there ...

Wayne

I've done that, too. I hear a static sound with some tone sweeps toward the end. I think it's normal.

Also, I know that the signal is at least partially coming through because I can see the impulse and my mic-speaker distance reads zero. It's just that the noise floor is too high.

Maybe I just have a defective card. Or maybe the card simply was not designed well for this application. If you or anyone else (Doug or Ethan) would like to try this yourself, I'll gladly send you the card (as long as you promise to return it).

Bare Bones;

When a sound card does not work it almost always means that it does not have WDM drivers (standard windows drivers). You can check to see if it does in the documentation.

If it does not have WDM drivers then you can always get 3rd party WDM drivers for download.

See if that card is on your standard windows mixer and if it can be controlled by the mixer.

Sound cards that will not work for RPlusD are hard to find, except the sound cards that are built into laptops - they almost never work.

I can see and control the Roadie from my standard Windows mixer. Does that mean I have the WDM drivers? I don't have the doc with me now.

You should know that this card is packed with features. It does 7.1 channel surround sound as well as basic stereo. It has an internal, 2-channel mic, a microphone jack, a headphones jack, a combo line-in/SPDIF-in, and a combo line-out/SPDIF-out. It provides a host of digital effects and special processing. But it doesn't seem to be able to do this simple thing.

Possibly, it's a software/firmware issue. The card is so complex, I wouldn't be surprised if they never tested it in this simple mode.

Bare Bones, screen grab your mixer settings and detail your relevant connections.

Watch the video posted for RPlusD describing this.(Ch1)

There are some pics of properly set mixers on this thread.

"I can see and control the Roadie from my standard Windows mixer. Does that mean I have the WDM drivers? I don't have the doc with me now.
"

- Yes.

The new soundcard arrived yesterday. I got a refurbished Creative Labs Soundblaster USB MP+. I plugged it in, installed the software, and it just worked!

My first loopback test gave me 30db SNR. With some levels tweaking, I got it just above 50db. So, I'm on my way.

From my point of view, it appears that the problem has been with the Roadie. It simply does not do this kind of thing well. I don't know why--whether it's hardware, software, or what have you. I don't think the problem has been cockpit error.

I don't want to spend any more time fiddling with Roadie, but my previous offer still stands. If you want to test it for yourself, I'll gladly send it to you. Just promise to return it (It's actually of little use to me now, but who knows).

In any event, you might want to consider cautioning new users not to buy the Audio Advantage Roadie for use with RandD software. In my experience, it doesn't work.

Now that the soundcard problems are behind me, I've been able to tinker with the software. So far, so good. It took a few minutes to figure out how to get started, but I think I'm finding my may.

One question: Doug, you say in the doc that the left and right signal levels should be lower than 0db. Why is that? When I adjusted my mixer settings, I set the levels the maximize SNR. The optimal levels gave me about 5 or 6db. Reducing them so that my reference channel is less than 0db will reduce SNR. Do I really need to do this?

"One question: Doug, you say in the doc that the left and right signal levels should be lower than 0db. Why is that? When I adjusted my mixer settings, I set the levels the maximize SNR. The optimal levels gave me about 5 or 6db. Reducing them so that my reference channel is less than 0db will reduce SNR. Do I really need to do this?
"

You do not really need to do this. It is a good idea in a case where you resample your results. Sometimes the resampled sequence can have a peak that is too high,

As far as the Roadie goes I would not write it off just yet. It will be the first sound card that would not work with the software. Perhaps its just your computer with that sound card or the WDM driver you are using.

I would help you get it working by email.

I too am very new to all of this, and Doug has been especially helpful via the telephone. I apologize in advance for being naive. I don't want to keep bugging Doug.

Right now, I set-up my R+D on a laptop using an M-Audio Firewire Solo interface that was recommended to me by a musician friend. It seems to work well, with S/N in the 50+ range when doing the loop-though test and using an XLR to RCA adapter for the mic input. I do have to re-adjust the mic level when I switch over to it.

Some nice features about using the Solo are that it has a phantom power switch on it so that microphones requiring phantom power can be used, as well as having physical knobs to adjust the levels when using the front inputs and the headphone output [via an adapter]. I'd post an image of the arrangement, but I don't know how yet.

I picked up a cheapo Behringer ECM-8000 microphone since I am only using this stuff to make room measurements for acoustic treatment. I thought this would suffice since I am not EQ'ing.

However, I seem to get a lot of noise when I do my real measurements. I tried using the Radio Shack meter and that was tons worse on the impulse [with a really bad S/N too], but in some ways better on the waterfall. It is this noisy waterfall plot with the ECM-8000 that makes me suspicious of it. I get a reasonably good S/N with the ECM-8000 of about 30, with the soundcard S/N in the mid to upper 50s. I did not buy the mic new either, so maybe it has been abused.

Does anyone know where to get a Behringer ECM-8000 checked and/or calibrated, or is there another microphone that is really good for this application? I never asked Doug what the IBF-EMM8 would cost on its own. I would presume it would work with the Solo.

I am also confused on some other things, but I am so new that I will review more documentation before discrediting myself with more silly questions.

Thanks in advance.

Doug,
I'd just like to add one more data point for you: After a frustrating hour trying to get the loopback to work, I tried (on a whim, thinking your documentation might have L/R switched (he he))I switched the L and R cables on one Y adapter and presto - 50 db S/N, perfect impulse etc. So do please continue to use this as a possible first fix for someone who thinks they have it all set up properly and it JUST WON'T WORK!!!

Do you know, in a stereo mini-plug which part of the plug is R and which is L? I'd like to find the culprit.

Thanks

The tip is Left and the ring is Right.

--Ethan

Ethan, you do a lot of recording don't you? Do you have a preferred mic for these kinds of measurements?

Can the Audix TR-40, the Behringer ECM-8000, the Superlux ECM-999 or the DBX RTA-M get the job done OK. Better yet, are all these mics actually the same thing?

David,

> Ethan, you do a lot of recording don't you? Do you have a preferred mic for these kinds of measurements? Can the Audix TR-40, the Behringer ECM-8000, the Superlux ECM-999 or the DBX RTA-M get the job done OK. Better yet, are all these mics actually the same thing? <

I do record, but mostly myself. I recently bought a DPA 4090. DPA is a child company of acoustic and microphone giant B&K. The 4090 is one of those "tiny diaphragm" omni condensers that are very accurate to beyond 20 KHz. The diaphragm is so small that its self-resonance is higher than the more usual 1/2 inch type. DPA promotes these in magazine ads as drum mikes, and I'm sure they're fabulous for that too.

I bought my DPA for acoustic measurements, but I'm now working on a very large project - dozens and dozens of tracks of me playing many different instruments - and I'm using my one DPA microphone for everything. I have been thrilled with the clarity and lack of coloration. Highly recommended, though expensive.

As for the others you mentioned, I have no direct experience with any of them. Basically, any small diaphragm omni mike is fine for low frequency testing. Where microphone quality starts to really matter is at high frequencies.

--Ethan

Thanks a lot Ethan. As I am just toying with this stuff as a hobbiest at the moment, the DPA looks out of my price-range...unfortunately.

Ray Rayburn has been very helpful with Superlux questions.

BTW - where would we ever get to hear one of your recordings?

"Doug,
I'd just like to add one more data point for you: After a frustrating hour trying to get the loopback to work, I tried (on a whim, thinking your documentation might have L/R switched (he he))I switched the L and R cables on one Y adapter and presto - 50 db S/N, perfect impulse etc. So do please continue to use this as a possible first fix for someone who thinks they have it all set up properly and it JUST WON'T WORK!!!

Do you know, in a stereo mini-plug which part of the plug is R and which is L? I'd like to find the culprit."

Some of those crappy cables have the channels confused and switching left to right and visa versa can fix things. I just recently put the mic Left/Right switch in.