|
Post by markusv2 on Jan 2, 2014 4:50:45 GMT -5
^ Prepare for a steep learning curve.
|
|
|
Post by neo20013 on Jan 2, 2014 11:46:05 GMT -5
REW does indeed have a complicated learning curve. But after a few frustrating hours, at some moment it begins to make sense, and after that point its simple to preform the basic measures ...
|
|
|
Post by neo20013 on Jan 4, 2014 13:33:02 GMT -5
Following Markus instructions, i think i've done everything right this time. I've used direct mode, disconnected all the inputs in the UPA700, and used the left speaker RCA to connect to the channel input i wanted to measure. I had to disconnect all the gear before i change the inputs, because even in standby, there was a sound coming for the speaker. So here are the measures for all the speakers, including Soundcard and Loopback measure. SOUNDCARDLOOPBACKLEFTRIGHTCENTERSUBRIGHT SURLEFT SURAfter this, i think my next step should be, using the EQ functionality of REW in order to calculate PEQ settings. For example, to calculate the PEQ for the LEFT speaker, i would select the LEFT measure, and select EQ:As the LEFT chanel has 11 bands, i disabled 12 to 20 REW filters:Now, using REW default settings, you could see that the LEFT / Predicted / Target chart values aren't correct ( LEFT / Predicted are the same). I'm totally lost in this settings For LEFT Speaker (Crossover at 80Hz) what should be the correct settings ?Equalizer settings - Generic Target Settings- Speaker Type : ? - Crossover : ? - Cutoff (Hz) : ? - LF Slope : ? - LF Cuttof (Hz) : ? - LF Rise Start (Hz) : ? - LF Rise End (Hz) : ? - LF Rise Slope (db/.): ? - HF Fall Start (Hz) : ? - LF Fall Slope (db/.): ? - Target Level : ? Filter Tasks- Match Range : (Start) TO (End) - Individual Max Boost: 0 - Overal Max Bost : 0 - Flatness Target : ? Markus, could you please help ? MDAT file can be found here : meocloud.pt/link/c8e0b019-027d-4f8b-bc52-bcc0a071bb3c/measures_UMC200.zip/
|
|
|
Post by markusv2 on Jan 4, 2014 16:26:30 GMT -5
^ Is the mic calibrated? The drop above 6kHz is fishy.
|
|
|
Post by neo20013 on Jan 4, 2014 16:53:47 GMT -5
^ I'm not using a mic, i'm using a cheap Radioshack clone : www.amazon.co.uk/Silverline-633937-Sound-Level-Meter/dp/B0015NSTLIAlthough there is a calibration file for the Radioshack, i suppose that it isn't accurate to use with the clone model, so i didn't use any calibration file. Please note that this process is mainly for learning, because i'm waiting for the delivery of several panels of rockwool and mineral wool, to build a few bass traps, in order to correct the nulls, and lower the reverberation times. In the future i intend to buy a calibrated mic, but for now, i will have to use the SPL meter. Regarding the EQ settings, would you mind to provide me some guiding? I'm a little lost
|
|
|
Post by markusv2 on Jan 4, 2014 17:04:42 GMT -5
^ In your room reverberation is way too high but you already know that. It will be addressed by room treatments in the future. In the meantime I would concentrate on "damage control" and tame the worst peaks in the bass. Could you please provide two additional measurements of the sub about 10cm to the left and the right?
|
|
|
Post by neo20013 on Jan 4, 2014 17:16:55 GMT -5
^ In your room reverberation is way too high but you already know that. It will be addressed by room treatments in the future. In the meantime I would concentrate on "damage control" and tame the worst peaks in the bass. Could you please provide two additional measurements of the sub about 10cm to the left and the right? Sorry, tonight it will not be possible, but i will take those measures tomorrow. You can have a better understanding of the speakers placement by this picture :
|
|
antoninus9
Minor Hero
She's the combination called the Wabash Cannonball
Posts: 35
|
Post by antoninus9 on Jan 4, 2014 18:24:47 GMT -5
I suspect the 6 kHz drop is due to the lack of a calibrated mic/meter.
|
|
|
Post by neo20013 on Jan 8, 2014 7:41:32 GMT -5
After measuring each speaker individually, what are the next steps?
1 - Use REW to calculate the filters for EACH speaker in isolation to insert in the PEQ?
BUT, as the final sound is a combination of the interaction between ALL speakers and the room, how do we correct the possible interaction problems between certain speakers?
|
|
|
Post by markusv2 on Jan 8, 2014 8:52:46 GMT -5
^ Bringing each speaker closer to a defined target reduces the differences between the speakers themselves at the same time.
Nevertheless you want to optimize the crossover region between sub and satellites so sub and speaker become a single entity. I'm trying to use the same crossover frequency for all speakers. Without software it's probably not feasible to optimize differing sub/satellite crossover frequencies.
|
|
|
Post by neo20013 on Jan 8, 2014 9:07:19 GMT -5
^ Nevertheless you want to optimize the crossover region between sub and satellites so sub and speaker become a single entity... Is this process BEFORE or AFTER equalizing each individual speaker? I suppose it should be BEFORE, as its necessary to select the crossover in the REW EQ settings. But if so, the only way i see to guarantee the sub/satellites is : 1 - Define crossover (Ex: 80Hz); 2 - Generate PEQ for each speaker 3 - Insert PEQ in UMC200 4 - Measure sub/satellites Repeat for 80/90/100/110/120/130, and compare measures. Is it correct or i am missing something?
|
|
|
Post by markusv2 on Jan 8, 2014 10:05:42 GMT -5
^ Did you fix the 6kHz issue? It would be a waste of time to base equalization on data that is not valid.
Generally the process is as follows: - Measure each speaker and sub full bandwidth at multiple (!) locations around the listening area and perform a general assessment of the room's acoustics - Determine a common crossover frequency and set UMC-200 accordingly - Measure each speaker at multiple locations around the listening area - Average the magnitude responses of each speaker. IMPORTANT: Don't average the frequency response (magnitude AND phase) but only the magnitude response - Optimize the averaged responses with REW's EQ tool - Transfer EQ settings to UMC-200 - Set levels and delay in UMC-200 - Measure response of each speaker with sub(s) at the listening position and fine-tune
|
|
|
Post by neo20013 on Jan 8, 2014 10:28:55 GMT -5
The 6kHz issue is probably due to the SPL meter, so there is nothing i can do right now.
Thank you very much for the procedure, its much clear now !!!
|
|
|
Post by markusv2 on Jan 8, 2014 10:40:42 GMT -5
The 6kHz issue is probably due to the SPL meter, so there is nothing i can do right now. Sure, have it calibrated or get a calibrated one
|
|
|
Post by neo20013 on Jan 8, 2014 11:09:40 GMT -5
|
|
|
Post by markusv2 on Jan 8, 2014 11:24:36 GMT -5
^ I wouldn't do that. The only way to find out how far off it is is to compare the response of your mic with the known response of another mic. Also known as calibration.
|
|
|
Post by neo20013 on Jan 8, 2014 17:19:17 GMT -5
|
|
|
Post by Jim on Jan 8, 2014 20:02:21 GMT -5
|
|
|
Post by neo20013 on Jan 12, 2014 12:18:46 GMT -5
|
|
|
Post by markusv2 on Jan 12, 2014 12:29:26 GMT -5
|
|