Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How To Interpret Graphs - Page 283 - AVS Forum
Forum Jump: 
 12Likes
Reply
 
Thread Tools
post #8461 of 10857 Old 01-29-2014, 01:44 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: Quintana Roo ... in my mind
Posts: 24,935
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 32 Post(s)
Liked: 131
Quote:
Originally Posted by andyc56 View Post

A few months back I tried this when doing some measurements on the Emotiva UMC-200 pre/pro delay vs. speaker distance, with no acoustical path. I found that when I applied the signal via HDMI and used a pure analog loopback from the Tascam US144 mkII output back to its input, the measured delay vs. distance setting of the UMC-200 made no sense. In order to get meaningful delay measurements, I had to use HDMI for both the applied input and the loopback source. One Tascam input was connected to the output channel of the UMC-200 under test, while the other was supplied from another output of the UMC-200 (the one to which the HDMI loopback signal was applied). IOW, the loopback path went through the UMC-200 itself. Here's how I described it then.
The problem with intermittent bad data described above turned out to be one that others had experienced, and could be at least temporarily fixed by choosing a different HDMI output channel in REW, then switching back to the original one before doing the measurement.
Food for thought. I suppose I should try this with my present kit before concluding to keep it or go all HDMI (and USB mic).

Thanks,
Jeff
pepar is online now  
Sponsored Links
Advertisement
 
post #8462 of 10857 Old 01-29-2014, 02:00 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 6,831
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 244 Post(s)
Liked: 650
Quote:
Originally Posted by cfraser View Post

Totally different topic. smile.gif

Are we sure that the UMIK-1 noise ratings now are the same as what they used to be? I don't understand where these ratings came from, sure don't seem to match my experience with it. i.e. the 50dB noise level. I noticed that mentioned somewhere back when I got the UMIK-1, but didn't query it because I had more important mikes to fry. In a quiet room with all my appliance junk switched off and the wind not blowing ("organ pipe" chimney lol) I'd say it's more in the ballpark of 30dB.

I also have never gotten the "spikes" or something like that (since I've never seen it...) that somebody mentioned you'd see in the higher frequency response that was due to the way the USB mike was powered, or something like that.

Is the difference due to a UMIK-1 design change, my laptop, a driver change (ASIO4ALL for me)? Or am I totally misinterpreting what people said about the UMIK-1?

(My laptop is about as low as you can go on the HDMI-output scale, I think, but works wonderfully with every single thing I've done with it = totally shocked, I got it to use outdoors mostly and they get broken easily, usually, lol)

 

I guess a good place to start answering your question is to ask you how you determined that the S/N for the UMIK-1 is 30dB.  Describe your measurement, please.

AustinJerry is online now  
post #8463 of 10857 Old 01-29-2014, 03:31 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,265
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 46
^ I asked first! smile.gif And I "determined" nothing.

I was questioning what was said elsewhere: the noise is 50dB. What (TF...) does that mean? It was never stated. I said how it "looks like" it's closer to 30dB, and under what circumstances, and I was asking what are the others talking about...you can see what I said... People responded previously to the "50dB noise" as though you couldn't measure anything below 50dB (the noise level). I'm saying that is false for my UMIK-1, the noise is ~20dB below 50dB and I can easily/cleanly/repeatedly measure to down there*, just as I'm saying there are no USB noise spikes with my UMIK-1*.

* AFAIK...why I asked the Q
cfraser is online now  
post #8464 of 10857 Old 01-29-2014, 04:15 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 6,831
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 244 Post(s)
Liked: 650
Try using the REW RTA tool to measure your listening room's noise floor (page 47 in the Guide). Make sure you calibrate your mic before taking a measurement. What do you measure?
AustinJerry is online now  
post #8465 of 10857 Old 01-29-2014, 05:27 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,265
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 46
^ Yes, that's where I was coming from: RTA measurements. Let's just say that if my contention is incorrect somehow and I'm off by 20dB, then I am doing my REW measurements at ~115dB and my room's worst noise floor (all appliances/furnace on and very windy) is at ~75dB. I don't think so...I only need relative measurements to know that's not true. Not to mention Audyssey is set at 75dB (wish they used at least 85dB, 75dB is oh so domestic-harmony-like).

Where I was partly coming from is many laptops, perhaps the vast majority in the case of somewhat older PC laptops, have USB ports that are quite unsuitable for quality audio use of any kind. Power supply noise and worse (including the USB chips and lowest-level drivers), plus that they totally weren't designed for audio as a prime app. I think this is part of the problem some people have with USB mikes. I would say my UMIK-1's noise floor on my laptop is under 30dB, based on as quiet as I can reasonably get it here.

Edit: just finished a "null" Audyssey run and am going to try doing the PEQ after Audyssey, instead of before like usual for me. I should have done this before for this room arrangement, but running Audyssey is just so boring compared to REW.

So if you want me to check something else re the mike, now's a good time to ask! smile.gif BTW, I just got my UMIK-1 a couple months ago at most, early Dec. IIRC, so it's a fairly recent version (if they've changed them at all).
cfraser is online now  
post #8466 of 10857 Old 01-29-2014, 06:28 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: Quintana Roo ... in my mind
Posts: 24,935
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 32 Post(s)
Liked: 131
Quote:
Originally Posted by cfraser View Post

Where I was partly coming from is many laptops, perhaps the vast majority in the case of somewhat older PC laptops, have USB ports that are quite unsuitable for quality audio use of any kind. Power supply noise and worse (including the USB chips and lowest-level drivers), plus that they totally weren't designed for audio as a prime app. I think this is part of the problem some people have with USB mikes.

Is there analog signal traveling over the USB bus?
pepar is online now  
post #8467 of 10857 Old 01-29-2014, 06:33 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 6,831
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 244 Post(s)
Liked: 650
I live in a single-family home in a very quiet neighborhood, and I am single, living alone, no pets, etc. When I measure for the noise floor, my environment is dead silent.

With my legacy REW kit, which has a calibrated EMM-6 mic, the noise floor measures 40dB. In the same environment, both the UMIK-1 and the UMM-6 mics (I have both) measure the noise floor at 50dB. This is why I claim that the USB mics have a worse S/N than the non-USB EMM-6. Other USB mic owners in this thread have also noticed that the noise floor measures higher.

If you are saying that your noise floor is 30dB, that would be world-class low. Not that I doubt you, but that would be an unusually low reading. That is why I asked whether you had actually checked your mic's calibration using an external SPL meter. If you are relying on the Sensitivity parameter in the calibration file without checking it against an external reference, this could explain why your measurements are lower than just about everyone else.

Your thoughts?
AustinJerry is online now  
post #8468 of 10857 Old 01-29-2014, 07:45 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,265
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 46
^ It sometimes goes above 30dB below 20Hz. It is hard for me to get it that quiet here...and now I'm "freezing" lol. And I apologize if I was bandying about level and floor: I actually meant noise "floor", which is of course quite different than overall noise level. Floor such as we would use when determining what is worth looking at in waterfalls etc. Yes, my overall calibration could be somewhat off, I have the old RS analog SPL meter.

My living conditions sound similar to yours, except re pets: I have a parrot who does a really excellent imitation of the REW test signal, and then throws in a touch of that Audyssey test signal "snap" at the end. He calls it "fixing the stereo, make better"...little does he know. IOW it's much easier for me to do measurement things when it's dark.

Jeff: no, not an analog signal in the way you mean (all signals and everything else in the real world are analog), but there is a lot of noise potential from the USB bus. People who use their PC for a USB DAC sometimes go to great lengths to minimize this noise, especially from the USB power lines, or you can buy a rather expensive DAC that has means to minimize it...partly why some people can get great results with relatively cheap USB DACs with some USB ports and others have to go to more extremes/cost.

As for my PEQ before or after Audyssey: the PEQ before is slightly better in frequency response flatness, and noticeably better in the waterfall especially at the low end. This is what I've been using. But to tell the truth, it is just fine without any PEQ, and Audyssey by itself does an IMO great job. We're talking +/-1dB flatness below the XO frequency without PEQ, so there's not a lot to fix. And my PEQ* is totally useless for doing anything worthwhile I think, it only goes down to 31Hz and my first needed correction point is closer to 20Hz. Conclusion: before or after Audyssey, or just forget PEQ, there's not a lot of diff here. I have to be wary of chasing the graphs, though it is kinda fun, lots of "work" for little return.

* There are two optional PEQ filters in the sub, so they are useless in practice above the XO frequency.
cfraser is online now  
post #8469 of 10857 Old 01-29-2014, 07:51 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 6,831
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 244 Post(s)
Liked: 650

I use an old analog RS SPL, so nothing wrong with that.  Sorry, you didn't answer my question--do you calibrate the mic following the exact procedure outlined in the guide?  As I said, a 30dB noise floor is exceptionally low.

AustinJerry is online now  
post #8470 of 10857 Old 01-29-2014, 07:59 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,265
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 46
^ Let me check tomorrow, wrapped up the REW for tonight. I believe I followed all your instructions, with allowances for differences because of the 8 channel HDMI output here. IOW I forget, it is not something I've done (or read) since the mike was new, so need to refresh myself on the process. Unlike Keith, I do not calibrate my mike before each set of readings. I don't go by "absolutes", I go by "relatives", if that makes sense, since I don't have a "standard" here. [For some of my instruments I do have actual standards, and without them, calibration itself is a relative term IMO.]
cfraser is online now  
post #8471 of 10857 Old 01-29-2014, 08:13 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 6,831
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 244 Post(s)
Liked: 650

I understand what you are saying, but I believe your original question had to do with whether something was different with your mic, and to be able to answer that we need to use absolutes for a moment.  Calibrating the USB mic to a known external reference like the RS SPL gets us to absolute long enough to answer the question.

 

I appreciate you working with me on this.

AustinJerry is online now  
post #8472 of 10857 Old 01-29-2014, 10:22 PM
AVS Special Member
 
Cowboys's Avatar
 
Join Date: Jun 2012
Posts: 1,346
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 106
In my case with the UMIK-1, my noise floor with the sensitivity line (down loaded from miniDSP with my mic serial #) added in my cross spectrum calibration file is between 58-60. I had a radio shack meter but cannot find where I place it, so I have not calibrated REW against a spl meter. I have my HTPC used for REW measurement and it is quite noisy with CPU/case fans plus the noise from my projector mounted about 4-5' above my UMIK-1 position. I also noticed when I did Audyssey calibration with Audyssey mic and the UMIK-1 side by side (within 2"), Audyssey shows an spl of 74-75db while doing the subs level match while REW spl shows 84/85dbs. What does that mean? Thanks.
Cowboys is offline  
post #8473 of 10857 Old 01-29-2014, 10:36 PM
Senior Member
 
3ll3d00d's Avatar
 
Join Date: Sep 2007
Location: London, UK
Posts: 302
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 59 Post(s)
Liked: 26
Quote:
Originally Posted by cfraser View Post

Re Keith's posts yesterday re "PEQ" after running Audyssey...I know a lot of you use Audyssey too. Posting here instead of Audyssey thread since you need REW or equivalent to do this...

How have you found the benefits of doing the PEQ before or after Audyssey? Let's say we're just talking about the sub-300Hz region here. There are also some ideas and considerations in the Audyssey FAQ.

Doing it before Audyssey requires a bit of a measurement dance with REW and Audyssey to make sure that Audyssey isn't "undoing" much of what you did with the PEQ, and that your PEQ is actually "helping" Audyssey to give you more of what you want. Not difficult, just a little more time-consuming. Part of the objective to doing it this way is to minimize what Audyssey has to do too, plus that we have more control with the PEQ than we do with Audyssey at specific frequencies (not speaking of Pro since I don't know).

The advantage of doing the PEQ after Audyssey is it's more straightforward. There are some pitfalls too, but I think if you say "I generally like what Audyssey does" and your objective is just to tune it up a bit, then this might be the way to go for most people and setups. (It isn't the way I've gone so far, for a personal system reason that occasionally gives a tiny bit more operational convenience but in retrospect not enough to be really useful.)
I have done it both ways. It depends what you are trying to achieve basically. For example, I have applied eq first to tame a couple of particularly large modal responses. This gives audyssey (I have multeq xt) less work to do and it produces a flatter response as a result whereas, on its own, it can't handle the peak and gets confused about what to do.

On the other hand I used to run a house curve, this meant applying a wide boost to the low end after audyssey had run. If you tried to do that before then audyssey would undo it. Actually in that case I was eq'ing before and after but the point remains, it depends on what you are trying to achieve.
3ll3d00d is online now  
post #8474 of 10857 Old 01-30-2014, 02:07 AM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 16,636
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 581 Post(s)
Liked: 1280
Quote:
Originally Posted by cfraser View Post

^ Yes, that's where I was coming from: RTA measurements. Let's just say that if my contention is incorrect somehow and I'm off by 20dB, then I am doing my REW measurements at ~115dB and my room's worst noise floor (all appliances/furnace on and very windy) is at ~75dB. I don't think so...I only need relative measurements to know that's not true. Not to mention Audyssey is set at 75dB (wish they used at least 85dB, 75dB is oh so domestic-harmony-like).

Where I was partly coming from is many laptops, perhaps the vast majority in the case of somewhat older PC laptops, have USB ports that are quite unsuitable for quality audio use of any kind. Power supply noise and worse (including the USB chips and lowest-level drivers), plus that they totally weren't designed for audio as a prime app. I think this is part of the problem some people have with USB mikes. I would say my UMIK-1's noise floor on my laptop is under 30dB, based on as quiet as I can reasonably get it here.
 

 

I am having the same difficulty in understanding as Jerry seems to be. " I would say my UMIK-1's noise floor on my laptop is under 30dB, "  What I’d like to know is the procedure you used to determine that.

kbarnes701 is online now  
post #8475 of 10857 Old 01-30-2014, 02:12 AM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 16,636
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 581 Post(s)
Liked: 1280
Quote:
Originally Posted by 3ll3d00d View Post
 
I have done it both ways. It depends what you are trying to achieve basically. For example, I have applied eq first to tame a couple of particularly large modal responses. This gives audyssey (I have multeq xt) less work to do and it produces a flatter response as a result whereas, on its own, it can't handle the peak and gets confused about what to do.

On the other hand I used to run a house curve, this meant applying a wide boost to the low end after audyssey had run. If you tried to do that before then audyssey would undo it. Actually in that case I was eq'ing before and after but the point remains, it depends on what you are trying to achieve.

 

That is it in a nutshell. Depends entirely on the objective.

kbarnes701 is online now  
post #8476 of 10857 Old 01-30-2014, 07:20 AM
Advanced Member
 
toofast68's Avatar
 
Join Date: Nov 2007
Location: NorthEast Ohio, USA
Posts: 938
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 19

Ok, so I think I have been following everything....

 

Since I can't get a AVR now that has better bass mgmt. than my MultiEQ Audyssey  - is appears this is my next best option.  I first thought it was the MiniDSP - but then is appears it was wrong.

 

http://www.amazon.com/gp/product/B0051TZBO0/ref=pe_152980_112299340_em_1p_1_ti

 

See me My Setup in Signature for more details.

toofast68 is online now  
post #8477 of 10857 Old 01-30-2014, 12:05 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,265
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 46
Quote:
Originally Posted by AustinJerry View Post

I understand what you are saying, but I believe your original question had to do with whether something was different with your mic, and to be able to answer that we need to use absolutes for a moment.  Calibrating the USB mic to a known external reference like the RS SPL gets us to absolute long enough to answer the question.

I appreciate you working with me on this.

OK, I reviewed your procedure, been more than *2 months* lol since I last read/did it, may as well have been a decade for me. Yes I did that...once. FWIW the way *I* check everything is working OK is I do a measurement and compare with known good results, first thing for every session. The SPL curves better overlay perfectly, less a null I have where depth can vary. I do not care about audio absolutes here, I only care about relatives/changes. Makes a huge diff when you don't have a calibrated standard. As I was once a "professional measurer" in the field and lab (not audio), the distinction is massive and important. Making two "random" things the same is not a calibration and tells you nothing absolute. Lecture over. I had it many times, especially when I started working with scientists lol (I'm not one) and there were legal implications.

I also used a Neutrik gadget (Minirator), with a signal I know is accurate in level (via my HP spectrum analyzer), to directly input to the RS SPL meter; yes it doesn't test the mike, but it does test the other electronics. My few-decades-old RS meter is "accurate enough" with direct input to give confidence in what it shows...so the electronics/meter movement aren't buggered at least.

I do regularly use the RTA and its little level (??) display, and I think that relates to our "disagreement" as I am unsure exactly what it's showing at this point (will try to find out). If I thought it was particularly useful, it would be in line with what you're saying and then I would 100% agree with you.

Perhaps the "disagreement" is in the terminology of noise floor vs noise level, they are not the same thing, and each has different practical uses with REW. For waterfalls etc., we are interested in the noise floor, not the measured noise level. We only want to look at everything above the noise floor. The measured noise level of course will be higher and is mostly useful for setting the calibration signal level to get a good S/N. IOW signals between the noise floor and the measured noise level are perfectly valid to look at in the waterfall etc. if you want to. Using good judgment, since the UMIK-1's noise floor does vary with frequency in my setup, mostly at the low end.

From your document, your RTA noise floor looks roughly similar to mine in shape, except mine shows ~20dB lower and doesn't have the noise spikes (almost certainly due to the particular USB interface). The Audyssey test signal measures within a couple dB of 75dB with the RS SPL meter beside the UMIK-1 at the MLP, both pointed up. When I first ran Audyssey on the X4000 it adjusted the test signal level: it is the closest thing I have to an "audio standard" here in that it's probably quite close to 75dB. FWIW I use the REW test signal at ~95dB according to UMIK-1/REW and RS SPL meter. Again, just because they "agree" does not mean they are correct. It's the relatives that are of interest and practical use to me.

So, my UMIK-1's noise floor is no more than 30dB across the whole audio spectrum (i.e. no USB noise spikes), when the room is quiet, with occasional rises to just above 30dB below 20Hz when the wind lightly blows (best I can command). However, the noise level is ~50dB at the same time. I guess this is the same as everybody elses UMIK-1 then, if you think quoting 50dB is important. I do not see that the 50dB thing quoted is particularly meaningful for our purposes except to setting a decent S/N and nothing more, the higher the better. Doesn't everybody measure as loud as they can get away with or are comfortable with? It does mean you probably shouldn't use the UMIK-1 for many purposes though (broader spectrum absolute noise levels).

Anyway, I'll redo the "calibration procedure" in the REW 101 document tonight and see what happens.
cfraser is online now  
post #8478 of 10857 Old 01-30-2014, 12:21 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,265
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 46
Quote:
Originally Posted by kbarnes701 View Post

I am having the same difficulty in understanding as Jerry seems to be. " I would say my UMIK-1's noise floor on my laptop is under 30dB, "

Why? Look at Jerry's document. If he didn't have the USB (presumed) noise spikes and the slight issue below 20Hz, he could reasonably quote a 30dB noise floor.

I don't have the USB noise, so basically you're questioning the below 20Hz thing for me. It's tough to control here, my bugaboo is the fireplace in the room and the wind. Hope I said where I was coming from (tediously) enough above...
cfraser is online now  
post #8479 of 10857 Old 01-30-2014, 12:34 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: Quintana Roo ... in my mind
Posts: 24,935
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 32 Post(s)
Liked: 131
Are youse guys measuring the noise floor of the mic or of the room? Absent an anechoic chamber, I don't see how you can measure the mic's noise floor.

Jeff
pepar is online now  
post #8480 of 10857 Old 01-30-2014, 01:02 PM
AVS Addicted Member
 
kbarnes701's Avatar
 
Join Date: Jan 2008
Location: Main Listening Positon
Posts: 16,636
Mentioned: 5 Post(s)
Tagged: 0 Thread(s)
Quoted: 581 Post(s)
Liked: 1280
Quote:
Originally Posted by cfraser View Post
 
Quote:
Originally Posted by kbarnes701 View Post

I am having the same difficulty in understanding as Jerry seems to be. " I would say my UMIK-1's noise floor on my laptop is under 30dB, "

Why? Look at Jerry's document. If he didn't have the USB (presumed) noise spikes and the slight issue below 20Hz, he could reasonably quote a 30dB noise floor.

I don't have the USB noise, so basically you're questioning the below 20Hz thing for me. It's tough to control here, my bugaboo is the fireplace in the room and the wind. Hope I said where I was coming from (tediously) enough above...

 

OK - I guess you're not going to tell us your verification procedure. You have the only USB mic with a noise floor of 30dB. No problem.

kbarnes701 is online now  
post #8481 of 10857 Old 01-30-2014, 02:28 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 6,831
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 244 Post(s)
Liked: 650
Quote:
Originally Posted by pepar View Post

Are youse guys measuring the noise floor of the mic or of the room? Absent an anechoic chamber, I don't see how you can measure the mic's noise floor.

Jeff

 

We are saying that the S/N ratio of the mic prevents the mic from measuring anything below ~50dB.

AustinJerry is online now  
post #8482 of 10857 Old 01-30-2014, 02:38 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,265
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 46
Quote:
Originally Posted by AustinJerry View Post

We are saying that the S/N ratio of the mic prevents the mic from measuring anything below ~50dB.

^ That is untrue, depending on what you're doing, as I explained. The noise level is NOT the noise floor. Check your own document for the ~30dB noise floor you'd have with a better USB interface. edit: less the sub-20Hz stuff anyways. I believe this "50dB stuff" is extremely misleading, REW-wise, which is why I first "challenged" it. It is virtually irrelevant for our purposes unless measuring in a monastery.

And Keith: seriously, did you not read my previous post where I said that? Did you not look at Jerry's document? Did you not understand it? I can (possibly) clarify...

Anyway, it's dark now so I can do the "mike cal" again.
cfraser is online now  
post #8483 of 10857 Old 01-30-2014, 02:52 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: Quintana Roo ... in my mind
Posts: 24,935
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 32 Post(s)
Liked: 131
Quote:
Originally Posted by cfraser View Post



And Keith: seriously, did you not read my previous post where I said that? Did you not look at Jerry's document? Did you not understand it? I can (possibly) clarify...

Getting popcorn ...
pepar is online now  
post #8484 of 10857 Old 01-30-2014, 02:58 PM
Wireless member
 
pepar's Avatar
 
Join Date: Jul 2002
Location: Quintana Roo ... in my mind
Posts: 24,935
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 32 Post(s)
Liked: 131
Quote:
Originally Posted by AustinJerry View Post

We are saying that the S/N ratio of the mic prevents the mic from measuring anything below ~50dB.

Both balanced and USB mics are rated with a 70dB S/N ratio. I am very rusty ... how does that prevent the mic from measuring below 50dB?

Just askin' , not arguing!
pepar is online now  
post #8485 of 10857 Old 01-30-2014, 03:03 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,265
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 46
Quote:
Originally Posted by pepar View Post

Getting popcorn ...

Don't bother, I'll soon be done with it (for myself anyway) and may not ever mention it again since it has no relevance or practical use for me. Making random things "the same" is not a calibration, so seeing that my "random calibration" is maybe better than Jerry's "random calibration" in our rooms has no value for me.

But a Guinness makes doing useless "calibration" a lot easier to swallow...lacking popcorn. smile.gif
cfraser is online now  
post #8486 of 10857 Old 01-30-2014, 03:48 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 6,831
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 244 Post(s)
Liked: 650
Quote:
Originally Posted by cfraser View Post

Don't bother, I'll soon be done with it (for myself anyway) and may not ever mention it again since it has no relevance or practical use for me. Making random things "the same" is not a calibration, so seeing that my "random calibration" is maybe better than Jerry's "random calibration" in our rooms has no value for me.

But a Guinness makes doing useless "calibration" a lot easier to swallow...lacking popcorn. smile.gif

TBH, I have lost the train of thought, and no longer understand why we are having this conversation. And I don't have a clue what you mean in your most recent post. Let's just move on.
AustinJerry is online now  
post #8487 of 10857 Old 01-30-2014, 03:57 PM
AVS Special Member
 
AustinJerry's Avatar
 
Join Date: Jan 2003
Location: Austin, Texas
Posts: 6,831
Mentioned: 3 Post(s)
Tagged: 0 Thread(s)
Quoted: 244 Post(s)
Liked: 650
Quote:
Originally Posted by pepar View Post

Both balanced and USB mics are rated with a 70dB S/N ratio. I am very rusty ... how does that prevent the mic from measuring below 50dB?

Just askin' , not arguing!

Measuring the noise floor involve set two steps. First, the mic needs to be calibrated using the REW SPL calibration routine, measured against a known external SPL (in my case a RS SPL). Second, one follows the procedure outlined in the Guide, using the REW RTA tool to measure noise level in the listening room, with it as quiet as possible.

In the same room, following the same procedure, the EMM-6 mic measures a noise floor of ~40dB, but both the UMM-6 and the UMIK-1 measure ~50dB. So, if in the absence of sound the mic goes no lower than 50dB, how can the mic measure a sound that is lower than 50dB? Please help me out here.
AustinJerry is online now  
post #8488 of 10857 Old 01-30-2014, 03:58 PM
AVS Special Member
 
cfraser's Avatar
 
Join Date: Aug 2002
Location: Toronto area, Canada
Posts: 1,265
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 13 Post(s)
Liked: 46
^ Kinda like me Jerry. It's funny how a comment made for the sake of interest somehow becomes "significant" when it wasn't really meant to be.

Anyway, I followed your procedure and I'm getting a solid 38dB. But because the RS meter is bouncing around a bit between the + and - 1dB marks, I'll call it a (convenient, for my memory) round 40dB. I had to set my MV to -9dB (DenonX4000) to get the recommended 80dB (C/S) SPL reading. This is louder than my usual -12dB MV setting, so slightly different than the last (only) time I did this before. This time it said the max SPL I could measure with this cal is 103dB (why I was originally questioning the "20dB out" thing because it would mean I was running a 115dB test signal).

Edit: I was using the "with subs" cal signal. BTW my front speakers are 4 ohm and power hungry, I do not use the X4000 amps at all.

Noise level is across the whole measured spectrum. Noise floor for our purposes is the level of the noise inherent in the measuring apparatus/system. So if the noise measured at one frequency is XdB, the level measured across the whole frequency spectrum has to be more than that, and how much more depends on the weighting etc. We can see that the noise "floor" varies with frequency. (I can't even see below 14Hz with my REW setup, just doesn't work.) It is quite valid to look at signals above the noise floor, but below the noise level as measured across the whole spectrum (RS/REW C/S) since it's an "accumulation" rather than a measure of limitation. Unless of course your whole purpose is to measure absolute level.

And again: I bet the +3dB MV increase (to get 80dB) between this cal and the last cal is due to the bass traps (which seem to do most here "not in the bass" lol). The traps required a 4dB increase in my sub level to get my usual 0dB Audyssey trims.




That very LF noise in the graphs is I'm sure my very noisy new gas hot water heater, and I forgot to close the (nearby!) door to the basement during the cal too...d'oh!
cfraser is online now  
post #8489 of 10857 Old 01-30-2014, 04:52 PM
Advanced Member
 
toofast68's Avatar
 
Join Date: Nov 2007
Location: NorthEast Ohio, USA
Posts: 938
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 19

OK, so I am trying to get some more testing done and for the life of me I am stuck.

 

I am getting this error when trying to do a measurement.  It seems like my mic is "working"...I see a blue light on it and below is also my MIC setup...what the heck am I missing.

 

toofast68 is online now  
post #8490 of 10857 Old 01-30-2014, 05:00 PM
Advanced Member
 
toofast68's Avatar
 
Join Date: Nov 2007
Location: NorthEast Ohio, USA
Posts: 938
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 19

Ok, so I think this might be my issue, but heck why can't I get REW to see my UMM any more ?

 

toofast68 is online now  
Reply Audio theory, Setup and Chat

Tags
Dayton , Dayton Audio , Room Equilizer Wizard Rew

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off