My measurements
In “All measurements“, you will see pictures of all companies and studios. In “My measurements“, you will only see the files that are in your folders but all type of files, wav, txt, calibrations, images and correction files (.wav). And you have the possibility to upload, delete, move, rename and modify descriptions.
The folder “My measurements” will allways be available to a user even if its plan is over.
Here is the folder structure for your files. Folder in red are private and only seen in “My measurements” so not seen by other users.
Company or title
Reference, studio
Date
myrec-p0.png page to check recording validity
myrec-p1.png main graphs
myrec-p2.png time and phase graphs
myrec-p3.png various graphs
myrec-p4.png spectrogram and wavelets
myrec-p5.png bongs timing
myrec-p6.png InterChannels Phase Correction
myrec-p7.png amplitude correction
myrec-p8.png time correction
myrec.text your setup infos from web form
myrec.wav your recording
response.mic mic calibration reformatted
mymic.txt your mic calibration
correction48 corrections for a processor that works at 48kHz
target.drc target with 3 columns : frequency, amplitude and phase
bin with corrections for minidsp processors
myrec-lin-L.bin
myrec-lin-R.bin
myrec-min-L.bin
myrec-min-R.bin
json with corrections for QSys and Xilica processors
myrec-lin-L.json
myrec-lin-R.json
myrec-min-L.json
myrec-min-R.json
REW for analysis in REW
myrec-comlin-LR.wav common correction for all channels
myrec-commin-LR.wav
myrec-hyblin-L.wav hybrid corrections
myrec-hyblin-R.wav
myrec-hybmin-L.wav
myrec-hybmin-R.wav
myrec-MMM-Lrew.wav MMM record
myrec-MMM-LRrew.wav
myrec-MMM-Rrew.wav
myrec-seplin-L.wav separated corrections for each channel
myrec-seplin-R.wav
myrec-sepmin-L.wav
myrec-sepmin-R.wav
wav
myrec-lin-L.wav corrections in wav format
myrec-lin-R.wav
myrec-min-L.wav
myrec-min-R.wav
wav-stereo
myrec-lin-LR.wav correction linphase into stereo file
myrec-min-LR.wav correction minphase into stereo file
– folders correction44 and correction88 with wav and wav-stereo folders
– folder correction96 with bin, wav and wav-stereo folders
At the moment, correction files are only calculated for following FIR processors and softwares in both linear and minimal phases versions :
- generic 1024, 2048, 4096, 6144, 8192 and 16384 taps : mono 24bits 48kHz .wav files
- MiniDSP 2x4HD 2048 taps : mono pcm 32bits float 48kHz .bin files
- MiniDSP OpenDRC 6144 taps : mono pcm 32bits float 48kHz .bin files
- QSC Qsys 8192taps : mono 24bits 48kHz .json files
- Xilica Solaro QR1 or FR1 4096 taps : mono 24bits 48kHz, .json or .txt files (ie with Rephase, just export as float 32/64 bits text file)
For FIR corrections, the number of taps is the number of audio samples of the correction impulse, ie for 2048 taps, the length of the IR is 2048 samples (for any format, wav, pcm, bin, json,…).
I don’t think that I will also calculate IIR (parametric) corrections : it is very depending on the actual processor (number of parametrics, Q type,…) and FIR processors are becoming cheaper and more common (ie MiniDSP, Xilica,…). Maybe for parametrics, I’ll just export a measurement file that could be used directly into REW so to calculate IIR correction.
If you bought a plan with correction included, the correction files are found in My measurements page in the correction folder with following names and in both formats .wav (48kHz 24bits) and .bin (pcm 48kHz 32bits float, big endian):
- …-hyblin-L : linear phase hybrid correction for left channel
- …-hyblin-R : linear phase hybrid correction for right channel
- …-hybmin-L : minimal phase hybrid correction for left channel
- …-hybmin-R : minimal phase hybrid correction for right channel
For QSys and Xilica Solaro, the .json files are also available in the “correction” folder.
It is important to use filters at the same sample rate as the processor or software, otherwise you may get distorted sound, noises, artefacts,…
Loudspeakers.audio does all calculation and outputs convolution filters at 24bits and 44.1, 48, 88.2 and 96kHz. So if you need other formats (ie resolution or sampling rate), you can use an online audio converter such as https://audio.online-convert.com/convert-to-wav or other external tool (available in any DAW or free tools such as sox).
Because correction can be positive in amplitude at certain frequencies, it is important to check the global level of the correction to avoid any overload. This won’t generally be a problem inside the processor itself but you need to check at the processor output, before D/A conversion : because many music tracks are mastered at a very high level near digital fullscale, you need to check that level of music with correction never exceeds fullscale. It is wise to reduce digital level so that the max level of correction is under 0dB, ie -1dB is fine. If you really want to avoid any intersample peak, you can even set max at -4dB.
Also when comparing before/after correction, you have to adjust level to match : with typical setup, you will generally need to reduce uncorrected level by some dB for a valid comparison.
The correction process tries to remove any level difference between channels but for QSys and Xilica Solaro, the level correction may be a bit wrong due to a conversion step from audio (the convolution wav file) to a json or text data file which may modify those level. It is necessary to redo a measurement with correction to really check channel levels.
Especially if your processor has a limited amount of taps, ie 2048 or less, it is wise to start with parametric EQ in the lowest frequencies, under 100Hz with manual EQ, and then measure again and automatically calculate the FIR correction. To ease this procedure, you can use REW sofware, import into REW the files named xyz-MMM-(LR)rew.wav in DRC folder as audio data, apply a 1/48 octave smoothing, and use the REW EQ function to find the best low frequencies EQ.
Even for higher taps value, this strategy can be a good one : first do manual EQ in bass before to calculate FIR.
In your measurements, you’ll get an L+R curve which corresponds to a mono signal send to both L and R loudspeakers. This L+R black curve is important because most recordings are mono in lower frequencies and it better represents low frequencies perception.
In some cases, the black curve is lower than each blue L and red R responses. This means that L and R signals are for some frequencies out of phase and instead of adding power, those signal destructing each other ! This has to be avoided and generally you will need to find better positions for your loudspeakers. With one subwoofer this cannot happen because summation is done electronically without any phase opposition. The hybrid correction computed by loudspeakers.audio tries to optimise L+R but sometimes it cannot completely solve this important problem.
The easiest way is to go to “My measurements” and for each folder, you see “Add files” with + button to upload your pictures.
You can also use the “select” button at bottom of “upload recording” form and choose a picture to upload.
If you want a thumbnail for your “Company or title”, keep the field “Reference, studio” blank. And if you want a thumbnail for a certain “Reference, studio”, just fill this field, save and upload the picture. Only .png and .jpg pictures are accepted.
After applying correction EQ, you can enjoy listening but it is very recommended to also re-measure after EQ to check validity of the applied correction. You may also check at different sound levels.
If you change your loudspeakers positions, you should redo measurements and calculate another EQ. Most changes will affect frequencies under 300Hz.
If you do important changes to your room, you should redo measurements and calculate another EQ.
If you cannot get your measurements, there are many possible reasons :
- if your folders names (company or title, studio,…) contains accents or special characters, graphs may not be calculated
- the audio file may be missing the sync signal at start or may be too short at the end. There should be one second or more after the sync signal at the end.
- microphone calibration file may not be conform : ie take the umik file from scratch, do not add any line even blank !
- calibration file must also be in right order : frequencies must be increasing in all lines, with lower frequencies first
- too long a path may be a problem because we use ftp copy for some files : so name of your company + name of studio or ref + name of your measurement file (or mic file if it is longer) must be less than about 180 characters
- please avoid accents and special characters in folders or measurement names
- no dots (.) in file names : another dot than the one before the extension will prevent uploading. This is to avoid security problems with double extensions.
- another problem seen quite often : right level is measured much lower than left level ! This may come from your computer or software. When the level is quite high, a level compression is automatically set and the recorded level of right channel (and also center channel) is lower because the compressor has been activated by the left channel signal !
- for USB glitches, read FAQ about USB problems
- If uploads don’t work, try another web browser (ie sometimes Chrome works better than Firefox)
- do not send multiple recordings one after the other, preferably wait for complete graphs before sending a new one
- if you don’t get graphs for a measurement, uploaded it another time but give it another name because successive uploads with same name are not processed
- if problems when viewing in french, change language to english : seems that with some browsers, it may help
For any problem, please get in touch with support(at)loudspeakers.audio
I did a comparison between the report you get here and the GRADE report you get with Genelec speakers (those equipped with SAM) :