Simplified REW Setup and Use (USB Mic & HDMI Connection) Including Measurement Techniques and How... - Page 8

 

I think we will have a lot of discussion around this topic.  My current understanding is that the further you go up the frequency, the less important these narrow peaks and valleys are.  A graph of 10K-20K with no smoothing is useless because of the "whiskers".  Smoothing must be applied in this range, and I don't think we hear any of those narrow variations.

 

In the lower frequencies, I believe we are more sensitive, therefore the recommendation to always present frequencies below ~300Hz with no smoothing.

Moreover, the .mdat files can be shared. Look here: Feri REW.zip 15k .zip file

Just unzip it, click on it and it will open up in your REW. Then you can do whatever you want with it, change smoothing, frequency range, dB range, etc.

It's fun.

P.s.: I had to zip the file coz AVS didn't accept .mdat as an attachement, but zip yes.

Hi Jerry, thanks for looking, how about this


Just notice my graph limits has changed

Therein is the major problem with REW, AFAIK, there really isn't one place where you can go to find ALL the correct settings and parameters to set before beginning. The REW Help file will help set the basics for REW, but there are too many possible equipment configurations for Windows computers. Different systems may (and do) need slightly different settings.

HTS can be a LOT of help as those folks have lots of experience with using and configuring REW setups, but even there, with my particular setup (Windows 7 PC to Roland Quad Capture external soundcard to calibrated EMM-6 mic), I eventually had to figure it out myself by going into every menu (Windows and REW) before I could figure out what had to be set in what manner for everything to work properly.
The higher up the FR range we go, the less audible any dips etc. over narrow frequencies are, but a huge dip around the 970-1kHz region will be audible. Our ears are extremely sensitive to that region.

I know that figuring REW out is already going to be time consuming, but if anyone is interested in how the results of measuring and correcting FR anomalies affects the audible outcome, I highly recommend downloading and trying the Harman How-To-Listen program.

*** edited to add *** I went back to look at the specific graph you were referring to and you're actually right, the 10db difference over that small range would be very difficult to hear for most listeners (who aren't Tom Holman).


Max
Edited by djbluemax1 - 1/11/13 at 1:50am

Jeff and I already did. Here you go: Acoustical Measurement Standards.

Note this is focused on two channel, but extending to home theater is straightforward. For HT use the main change would be to tighten down the LF frequency response targets significantly since you have subs which can be physically placed separate to main speakers.

Any questions I'd be glad to answer them. Note Jeff & my thinking on ETC is a bit different. For more on that see this article: Listening Room Reflections and the ETC.

With respect to frequency resolution, I'd suggest 1/3rd octave above 250Hz, and both 1/3rd AND a higher resolution (e.g. 1/12th or 1/24th octave) below 250Hz.

It all depends what you are trying to understand.

Generically speaking now....our ear's loudness summing is closer to a 1/3rd octave filter, individual resonance discrimination is closer to a 1/6th. 1/12th and higher are useful to understand center frequencies of dips and peaks for addressing room issues such as room modes and speaker boundary interference. But with too high a resolution you can't see the wood for the trees...

Measuring LFE + Main? LFE is too hot then... Remember, if you pass signal to LFE directly (not separated by means of the receiver crossover) it is hot by 10dB. May be this is the reason of such difference under 50 Hz?

I have drafted a step-by-step configuration guide to get the beginner up and running on REW, including the configuration of the Windows audio interface. I will be forwarding my draft to Feri and Keith this evening for their feedback, and hopefully will be able to post it in this thread within a day or two. Like Max says, developing a generic "how to" guide is not that easy, but I hope this document will be useful

I reported yesterday that I had reached an impasse with regards to configuring ASIO4ALL with HDMI. Not wanting to accept defeat, I put in a fresh hard drive into the laptop and did a completely fresh install of everything. It now seems to be working properly, even though I still have access only to two channels. This is evidently a limitation of my laptop hardware (a two year old Acer). At least I can proceed with getting the USB mic to simplify my kit.

The 20-40Hz peak does not look healthy. Ar you using the Audyssey mic as a temporary measurement tool? If yes, then we shouldn't worry too much about the measurements. Otherwise, we should try and understand why the bass looks so peculiar.

HTS has some good tutorials and notes on what they like to see in graphs that get posted to them. I think it would be good to adopt their convention so there is some consistency for people that look at both forums. Thoughts?

Yes using DM-A409 mic to play around for now .
Here is my last testing for this evening .

LF, C, RF, SL and SR at MLP w/Audyssey turned off.
Once again thanks for looking.

 

I would say you have a pretty good handle on how to generate the frequency response graphs.  Now we need to get you a "real" mic so the measurements are more accurate.

Earlier this evening, I tried using the Omnimic on REW.

Seems that it would work but loading the calibration file was problematic.

It doesn't use the file extention that REW is looking for. It has something like .omm

Would like to know if changing the file extention to .TXT would be enough for it to work correctly.

Does anyone know?

Does anyone know how to test it?

I would be surprised if changing the file extension would work. Is it even a text file?

I'm not that familiar with the OM microphone. Does it work without a calibration file? If yes, you could always send the mic to Cross-Spectrum Labs and they can create a calibration file. That might cost as much as buying a REW-compatible mic, though.

Hey Feri, how's it going? BTW, just wanted to mention that the graph you ;posted above is still not 1/24 octave smoothing. It's still 1/3 octave smoothing. Not sure how you measured the graph or captured it, but the post below shows what a 1/24 octave smoothing graph looks like.

BTW, here are two graphs of my own taken from the exact same measurement. Graph a) is the 1/3 smoothing graph
a.jpg 71k .jpg file

and graph b) is the 1/24 octave smoothing graph.
b.jpg 81k .jpg file





Max

P.S. personally, I prefer to use 1/6 octave smoothing for graphs to 20kHz and No smoothing for graphs focusing on the 5-250Hz range.

The same graph above with 1/6 smoothing


And a bass region only graph with No smoothing

Edited by djbluemax1 - 1/11/13 at 2:05am

I think you may be driving the LFE channel (channel 4 on HDMI) as well as a full range channel? Only run channel 4 in isolation, the other channels include the sub due to AVR bass management.
Regards, Mike.

Edit, sorry didn't see the earlier relies.
Edited by AV_mike - 1/11/13 at 2:11am

 

I often find that when someone replies to a question that has already been answered, that the additional replies, being worded slightly differently or explained from a different perspective, can actually illuminate.

With all respect to AustinJerry, what Nyal says is pretty much the industry standard. My suggestion is follow that std so you can compare apples to apples outside of this thread.


Sent from my 32GB iPhone4 using Tapatalk

I just renamed the .omm file and was able to open it in a text reader.

It looks fairly similar to a mic correction file with some added information.

I've asked John over at HTS to take a look at it.

 

 

How about following the HTS guide, for compatibility?

1st, thanks for all involved for putting this together. Now my issue, I have REW configured, and I can see all my channels, but for some reason I only can get to do one test tone, then I always get the below dialog box.

Capture.PNG 8k .PNG file


Sorry for the attachement, couldn't figure how to just get the message to show.


If I exit REW, go back in, I can do one test tone, then when I try another, I get the same message, any ideas?

Thanks

I am all for it, Keith. This is a good suggestion from Mike. The important thing is that we have consistency when sharing measurements.

I think I found the issue, not sure if this is the way its suppose to work, but after I do a test tone, my reciever(Onkyo TX-NR818), drops back to All Ch Stereo(which is why I recieved the sound card message). I have to wait till it goes back to Multich, then it works. It is suppose to drop out from the Multich mode?

It could easily be a HDMI handshake issue between your laptop and the Onkyo. Try to disconnect the HDMI cable and plug it in again. Check for a consistent Multi CH IN on the Onkyo display before measuring.

Also please make sure no other app in your laptop is active (media player, etc.) that can suck away resources from REW. Look into TaskManager to verify no such app is running in the background.

Hope this helps.
Edited by mogorf - 1/11/13 at 7:26am

Yes, it does. Thanks..

When solved please come back with the solution. It will help others with same or similar issues for doing a quick troubleshooting.

Thank you.

Agreed. Very important.

Well, I can't tell you how close I came today to jacking the whole REW thing in. I spent 5 hours getting it to work (I think it works) using my OmniMic for testing purposes. Even following the brilliant step-by-step guide Jerry is preparing it took me the best part of the day to get it working. There are just so many things to configure and fiddle with. Eventually I think I have it running - but there are still a few things that bother me and give me reason to wonder if I am actually configured properly. I won't go into them now as I intend to spend a little more time over the next few days on it. But Plug and Play.... well, it ain't.  By contrast, with OmniMic, I ran the s/w, plugged in the mic and started measuring.... now THAT is PnP.

I ended up re-installing all of my audio drivers, and that seems to have fixed it, I no longer get that dialog box, between taking test tones.

Next step is to try using my Audyssey Mic, and see what I can do with that.

Thanks.

Can one use the mic that comes with the Omni-mic system with REW? I mean the OMM cal file is nothing more then a text file at its heart, at least right clicking it and editing it with Notepad++ shows it just like the Radioshack generic cal file that you can download for use the REW.

Am I off base here?

Bill