Graph Panel

The graph panel shows plots of measured data, impulse responses, filter responses, target responses and the predicted effects of the correction filters on the measured data. The plots are split into groups of traces, selected via the tabs at the top of the graph area.
Trace Selector

The various graph groups are:
Filter Adjust
Impulse
Spectral Decay
Waterfall
Measured
Averaging
Predicted
All Impulses
RT60
Energy-Time
Oscilloscope
Spectrum

Options that affect the appearance of the traces can be found in the View settings.

Each trace can be turned on or off via the selection buttons at the left of the trace name. Trace names are in the same colour as the trace itself, whilst the line style for the trace is shown between the label and the trace's value at the cursor position. Corrected response traces use dotted lines (the default) or a lighter shade of the measurement trace colour according to the option selected in the View settings.
Trace Labels

Frequency Axis Log/Linear Button

The Freq Axis button in the toolbar toggles the frequency axis between logarithmic and linear modes. This function is also available via a command in the Graph menu and the associated shortcut keys.
Log/Linear Frequency Axis Button

Horizontal Axis Zoom Buttons

The horizontal axis zoom buttons Frequency Axis Zoom Buttons zoom in or out by a factor of approximately 2 centred around the cursor position.

Graph Limits Button

The Graph Limits button Graph Limits Button allows desired top, left, bottom and right graph limits to be defined. A dialog pops up in which the values are entered, they are applied as entered or by the Apply Settings button.
Graph Extents Dialog

Vertical Axis Zoom Buttons

The vertical axis zoom buttons zoom in and out on the Y axis.
Y Axis Zoom Buttons

Save Graph as JPEG Button

Save Graph as JPEG Button This button, in the bottom left corner of the graph area, allows the current graph view to be saved as a JPEG image. A dialog pops up to set the desired width of the image (default is to be the same width as the graph).

Zoom control grid

This function is also available via a command in the Graph menu and the associated shortcut key.

Variable Zoom

The Wizard provides a variable graphical zoom capability by either pressing and holding the middle mouse button, or pressing and holding the right button then pressing and holding the left button, and dragging the pointer.

When variable zoom is active a cross is displayed, split into quadrants allowing frequency and/or spl to be zoomed in our out depending on the mouse position. The amount of zoom is governed by how far the mouse pointer is dragged from the start position.

Zoom control grid

Zoom to Area

When the Ctrl key is pressed followed by the right mouse button a zoom box can be drawn by dragging the mouse. Measurement cursors are shown on the outside of the box, to zoom to the shaded area click within it. If the shaded area is too small to zoom in to a message will indicate which dimension is too small for zooming and what the limit is to allow zoom.
Zoom to area

Undo Zoom

To undo the last Variable Zoom or Zoom to Area, press Ctrl+Z or select the Undo Zoom entry in the Graph menu. This will restore the graph axes to the settings they had when the right or middle mouse button was last pressed. This Undo feature can be used even if you have not zoomed, just press the right mouse button when the axis settings are to your preference then you can return to these settings (undoing any subsequent movements or control changes) by pressing Ctrl+Z.

Filter Adjust

The Filter Adjust group shows the measured data and corrected response for the current measurement along with the target response, the response of the equaliser filters with and without the target and the Mic/Meter and Soundcard calibration responses (if loaded).
Filter Adjust Graph Group

This graph group, in common with all groups that have a frequency axis, also shows any filters that have been defined, displaying the filter's number along the top margin of the plot at the position corresponding to its centre frequency.

Measurement

The frequency response of the measurement is labelled with the measurement name

Corrected

The Corrected response shows the predicted effect of the measurement's filters.

Target

The Target trace shows the target frequency response for the measurement, including any desired House Curve response shape. If a House Curve has been loaded the symbol will be displayed by the trace value. The response includes the Bass Management curve appropriate to the Speaker Type selected for the measurement. The overall level of the curve is controlled by the Target Level.

Filters

The Filters trace shows the combined frequency response of the filters for this measurement.

Filters+Target

The Filters+Target trace shows the frequency response of the filters overlaid on the desired Target response. Selecting the filter responses to be drawn inverted and adjusting the filters so that this curve matches the measured response will result in the corrected response matching the target.

Invert Filters

When this box is selected the responses of the filters are drawn inverted. This is useful for graphically matching the shape of a filter to the shape of the peak it is being used to correct, when the shapes match the overall response in that region will be flat.

Mic/Meter Cal

The Mic/Meter Cal trace shows the frequency response of the Mic calibration data (if a calibration file has been loaded via the Mic/Meter settings). If the Type has been set to C Weighted SPL Meter this curve will show the effect of C weighting (only outside the range of the calibration data file that has been loaded). The trace is not shown if cal data has not been loaded. The trace is drawn relative to the Target Level.

Soundcard Cal

The Soundcard Cal trace shows the measured frequency response of the soundcard relative to its level at 1kHz (if a calibration file has been loaded via the Soundcard settings). The trace is not shown if cal data has not been loaded. It is drawn relative to the Target Level.

Impulse

The Impulse graph shows the impulse response for the current measurement, the left and right windows and the effect of the windows on the data that is used to calculate the frequency response. It also shows traces and data related to the reverberation time (RT60) and allows octave or one-third octave filters to be applied to the response.
Impulse Graph

The Y axis used for the impulse response can be selected as % FS or dB FS (FS = Full Scale) via a control in the bottom right corner.
Impulse Response Y Axis Selector

Dashed vertical black lines show the extents of the impulse response windows, a dashed red line shows the reference position. If the window settings are changed the region outside the new area is shown shaded until the settings are applied. It is best to set the Y axis to dB to adjust the windows as it is then much easier to see where the response has decayed into the noise.
Impulse Response dB Scale

After each measurement the left window width is automatically set up. For full range measurements (and down to end frequencies of 1kHz) the width is 125ms, below that it increases to allow for pre-ringing effects of using a limited sweep range.

The impulse response is that of the whole system, including the mic/meter and the soundcard, unless the measurement was made with the "Use Left Channel As Calibration Reference" option selected on the soundcard settings, in which case the soundcard's contribution is removed. The mic/meter and soundcard calibrations are only applied when calculating the frequency response.
Tip: To apply or remove a mic/meter or soundcard calibration for a measurement after it has been taken, simply load or clear the cal data as required and press the Apply Windows button to recalculate the frequency response.

Impulse Graph Controls

The impulse response may be plotted with or without normalisation to its peak value according to the setting of the "Plot Normalised" control. When normalised plotting is selected the peak will be at 100% or 0dBFS. This setting applies to both the Impulse graph and the All Impulses graph.
Plot Normalised Control

The response may be plotted inverted according to the setting of the "Invert Impulse" control. Note that this has no effect when the Y axis is set to dB FS.
Invert Impulse

The "Scale Response" control re-scales the impulse response to achieve a desired maximum SPL figure in the corresponding frequency response. The target peak SPL figure (in dB) is updated to reflect the current frequency response maximum SPL whenever the "Apply Windows" button in the IR Windows control is hit.
Scale Impulse Response Controls

Impulse Response Envelope

The envelope of the impulse, also called the energy-time curve or ETC, is useful to identify reflections and see the overall shape of the impulse response, particularly if the ETC Smoothing control is used to filter the envelope. The plot below shows the envelope (the darker trace) after applying 20ms smoothing.
Impulse Response Envelope

Octave and One-Third Octave Filters

Octave and 1/3 octave filters can be selected from the box to the right of the "Calculate RT60" button, the selected filter is applied to the Impulse Response and the corresponding frequency response upon selection. The filter remains active until "No Filter" is selected. The response traces for the measurement are drawn with a dashed line when a filter has been applied.
RT60 Controls

Impulse Response Decay/Reverberation Time

The Impulse graph has several traces and controls that are used when looking at the way the response decays. The rate at which sound decays in the room is affected by many factors including the amount of absorption the room's surfaces have. It is It is captured in a measure called the room's reverberation time, RT60, which is the time taken for sound to decay by 60dB from its initial level. It is useful to examine how the decay time changes in different frequency bands so that acoustic treatments can be selected to address the room's problem areas.

The Schroeder Integral is a curve obtained by backwards integration of the squared impulse response, ideally starting from a point where the response falls into the noise and applying a correction (a starting value for the integral) which assumes the rate at which the Schroeder curve is falling continues for the whole response. REW uses an iterative procedure to estimate the best starting point for the integration, often called "Lundeby's Method" (from the paper by A. Lundeby, T. E. Vigran, H. Bietz, and M. Vorländer, “Uncertainties of Measurements in Room Acoustics,” Acustica, vol. 81, pp. 344–355 (1995)). The slope of this curve is used to measure how fast the impulse response is decaying, deriving a figure for "RT60" which is the time it would take sound to decay by 60dB. RT60 values are calculated when the "Calculate RT60" button is pressed, if a filter is active the results are those for that filtered band. To see RT60 results for all the various octave or one-third octave bands use the RT60 graph. The plot below shows the Schroeder curve of a response which has had a 1 octave wide filter applied, centred on 1kHz. The graph also shows the "Regression Line", which is a line obtained by carrying out least squares linear regression on the Schroeder curve. There are several different variants of this line, depending on which sections of the Schroeder curve are used when carrying out the least squares fit. The selector for which regression line is to be shown is to the right of the "Regression Line" check box.
Impulse RT60

The "r" value shown after each decay measure is the regression coefficient, which measures how well the data corresponds to a straight line. A value of -1 would indicate a perfect fit, values lower in magnitude than -0.98 indicate the corresponding decay figure may not be reliable. The decay measures available are:

EDT
Early Decay Time, based on the slope of the Schroeder curve between 0dB and -10dB.
T20
Decay time based on the slope of the Schroeder curve between -5dB and -25dB.
T30
Decay time based on the slope of the Schroeder curve between -5dB and -35dB.
Topt
An "optimal" decay time based on the slope of the Schroeder curve over a variable range chosen to yield the best linear fit. If the early decay time is much shorter than T30 the Topt measure uses a start point based on the intersection of the EDT and T30 lines, otherwise it uses -5dB. REW tests every end point in 1dB steps to the end of the Schroeder curve and chooses the one which gives the best linear fit.

Distortion Components

A property of the log sweep analysis method is that the various harmonic distortion components appear as additional impulses at negative time, with decreasing spacing as the distortion order increases. For example, this plot shows spikes from distortion components up to the 8th harmonic on a laptop soundcard loopback measurement:
Impulse Response with Harmonic Peaks

Here is a similar measurement for an external USB soundcard, it is a 44.1k card rather than 48k, which limits us to the 6th harmonic in the 1s pre-impulse period - however, only the 2nd, 3rd and 5th harmonic peaks are evident, the 4th harmonic peak is barely visible above the noise floor (which is about 10dB lower than the laptop card). The extended lobes after the impulse are due to the card's much lower -3dB frequency, 1.0Hz versus 22.1Hz (note that the right side of the time axis is 2.0s in this plot compared to 0.5s in the previous plot):
Impulse Response with Harmonic Peaks

Spectral Decay

This group shows spectral decay traces over the region from 10Hz (or the frequency resolution for the selected window width if higher) to the end of the measurement sweep. The plot used logarithmically spaced data at 96 points per octave with 1/24th octave smoothing applied. The time separation of the slices is controlled by the setting just above the Generate Spectral Decay button, the width of the impulse response section that is used to generate the slice is set by the Window control. The corresponding frequency resolution is shown next to the window width.

Spectral Decay Controls

The traces for each slice can be drawn as conventional lines or as filled areas, selected by the "Fill slices" check box. The alternative views are shown below.
Low Freq Spectral Decay, normal traces
Low Freq Spectral Decay, filled traces

These Spectral Decay plots are generated by shifting the impulse response window to the right by the slice interval to generate each succeeding slice. Note that if the slice width and interval are large, fewer than 8 slices may fit within the impulse duration, in which case some slices will not be generated. The default window type for this plot is Tukey 0.25, other types may be selected via the Low Frequency Decay entry in the Analysis settings.

Waterfall

This group shows a waterfall plot over the region from 10Hz (or the frequency resolution for the selected window width if higher) to the end of the measurement sweep. The plot used logarithmically spaced data at 96 points per octave with 1/24th octave smoothing applied. The time separation of the slices is controlled by the setting just above the Generate Waterfall button, the width of the impulse response section that is used to generate the slices is set by the Window control. The corresponding frequency resolution is shown next to the window width. Best results are obtained when the window is smaller than the time span.
Low Freq Waterfall

The waterfall plot is generated in the same way as the Spectral Decay plots, shifting the impulse response window to the right by a proportion of the time range to generate each succeeding slice. The default window type for this plot is Tukey 0.25, other types may be selected via the Low Frequency Decay entry in the Analysis settings.

The labels at the sides of the plot show the time axis values
Waterfall Time Labels

The controls in the centre of the legend area alter the appearance of the waterfall.
Waterfall Controls
The slice slider selects which slice is at the front of the plot - as the slider value is reduced the plot moves forward one slice at a time. The trace value shows the SPL figure for the frontmost slice, the corresponding time for that slice is shown at the top right of the graph.

The x, y and z sliders alter the perspective of the plot, moving it left/right, up/down and forwards/backwards respectively. The buttons next to the sliders allow the perspective to be disabled in that axis. Disabling the x axis can make it easier to see the frequencies of peaks or dips. Disabling the z axis turns off all the perspective effects which makes the plot like a filled spectral decay. Here is the same plot as above but with the x-axis perspective effect turned off.
Waterfall no X perspective

The waterfall allows another measurement's plot to be overlaid on the current measurement. The overlay is generated slice-by-slice, plotting a slice of the current measurement's waterfall, then a slice of the overlay, then the next slice of the current measurement and so on. N.B. before a measurement is available to overlay it is necesary to generate the waterfall data for it.
Low Freq Waterfall Overlay

The overlay is selected using the controls at the left hand side of the legend area.
Waterfall Overlay Controls
Measurements which do not have waterfall data are shown in grey in the selection list. To generate the data for a measurement select it as the current measurement and use the Generate Waterfall button.
Waterfall Overlay Selection

Transparency can be applied to the main plot, the overlay, or both. When transparency is set to 0% both plots are solid. If only one plot is selected for display it is drawn solid regardless of the transparency setting. In the image above the main plot is drawn at 75% transparency, allowing the overlay to show through. The transparency mode can be switched between main/overlay/both to ease comparison between the plots.

Measured

This group shows the measured responses for all measurements on the same plot. The "Separate the traces" check box offsets each trace downwards from the preceding trace to make it easier to distinguish individual features when the traces are at similar levels.

Averaging

The Averaging group allows the average of selected measurements to be generated. Any of the traces which are selected when the Average The Responses button is pressed will be included in the averaging. The frequency range of the average trace covers the region where all included traces have data, for example if one trace was measured to 200Hz, another to 500Hz and a third to 1000Hz the average would range to 200Hz (to the lowest end frequency).

New measurements (those made after the last average was generated) show new next to the trace value, whilst those included in the last average show avg.

Note that Average traces do not have any impulse response data so waterfall and spectral decay plots will not be available.

Predicted

This group shows the predicted effects of the equaliser filters for all measurements on the same plot.

All Impulses

This group shows the impulse responses for all measurements on the same plot. If the "Plot Normalised" check box is selected all responses will be plotted normalised to their peak value, placing the peaks at 100% or 0dBFS.

RT60

The RT60 Reverberation Time curves at each filter centre frequency are displayed on this graph. The plots can use with horizontal bars centred on each filter frequency and spanning the filter's bandwidth, or lines joining the filter centre frequencies, according to the View settings.
RT60 curve controls

The "Time Reversed Filtering" control applies the octave band filters backwards in time, this reduces the filter's own contribution to the measured delay. When using 1/3 octave filters at low frequencies the filter decay time can be significant, over 200ms for a 100Hz 1/3 filter, for example. Applying the filter in reverse reduces this delay to less than 50ms, but it does affect the response somewhat, such that Early Decay Time (EDT) figures using Time-Reversed filters may not be valid.

If the "Show Correlation Factor" box is checked the graph legends show the quality of the line fit for the selected decay measure, if the correlation magnitude is less than 0.98 the corresponding decay measure is shown in italics to indicate it may be unreliable. The various decay measures are explained here.
RT60 curves

The reverberation times for the current measurement can be written to a text file using the File -> Export -> "RT60 data as text" menu entry.

Energy-Time

The full range energy-time curves are displayed on this graph. The energy-time curve is the magnitude of the analytic function whose real part is the impulse response and whose imaginary part is the Hilbert transform of the impulse response. The energy-time curves can be smoothed by a moving-average filter of the selected duration.
Energy-time curve controls
Energy-time curve

Spectrum

The Spectrum group allows spectrum analyser or Real Time Analyser (RTA) plots to be generated, updated live as the input signal is analysed. It is activated by pressing the record button in the bottom right hand corner of the graph area, after which it will continuously analyse blocks of input samples and display the frequency spectrum of each block. Sometimes the analyser would be used without a test signal, for example to look at the frequency content of background noise, but more often it would be used together with the REW generator or an external generator or signal source. If the generator is playing a pink noise signal (or even better, pink Periodic Noise) the RTA display will show the frequency response of the room, updated live so that the effects of changing EQ settings can be immediately seen. Playing a test tone on the generator allows the levels of the tone and its harmonics to be observed on the analyser and distortion percentages to be calculated.

The Spectrum plot shows the currently selected measurement tab as a reference (the first entry in the plot list) and the live spectrum. If Inverse C compensation is being applied the icon is shown after the trace value. If Mic/Meter calibration file or soundcard calibration file have been loaded they are applied to the results. Below the Spectrum selection the current Input RMS value is shown, in dB SPL or dB FS according to the setting of the Y axis. This figure excludes any DC content in the signal. If clipping is detected in the input a Clip! indication will be shown by the input RMS label.

Spectrum Plot

Spectrum controls

The controls for the spectrum plot are at the right hand side below the graph.
Spectrum Controls

The Mode can be set to Spectrum for a spectrum analyser plot or to various RTA resolutions from 1 octave to 1/48 octave. In Spectrum mode the plot can either draw lines between the centres of the FFT bins or draw horizontal bars whose width matches the FFT bin width, this is controlled by Use Bars on Spectrum Plot in the View settingss.

In RTA mode the frequency from which the display is accurately reflecting an RTA is indicated to the right of the Mode selector, ">37Hz" in the example below. An RTA sums all the power within the fractional octave span it covers, at low frequencies and high RTA resolutions this span may be very small. If it becomes smaller than the width of an FFT bin (which is the sampling frequency divided by the FFT length) REW shows the FFT bin widths rather than the fractional octave width at that frequency to indicate the true resolution of the display. The RTA plot can either draw horizontal bars whose width matches the octave fraction or lines between the centres of the bins, this is controlled by Use Bars on RTA Plot in the View settings.

RTA Valid Frequency

FFT Length and Window

The FFT Length determines the basic frequency resolution of the analyser, which is sample rate divided by FFT length. This resolution is also affected by the Window setting. Rectangular windows give the best frequency resolution but are only suitable when the signal being analysed is periodic within the FFT length or if a noise signal is being measured. The Rectangular window should always be used with the REW periodic noise signals. Most other signals, e.g. sine waves from the REW generator or test tones on a CD, typically would not be periodic in the FFT length. Using a rectangular window when analysing such a tone would generate spectral leakage, making it difficult to resolve the frequency details - the plot below shows an example of a 1kHz tone from an external generator with a Rectangular window.
1kHz tone, rectangular window

Here is the same tone analysed with a Hann window.
1kHz tone, Hann window

The window allows the harmonics of the tone to be resolved. However, the tradeoff is that windows cause some spreading of the signal they are analysing, which reduces the frequency resolution. The figure next to the FFT Length setting shows the resolution allowing for both the length and the selected window.

The Hann window is well suited to most measurements, offering a good tradeoff between resolution and shoulder height. If very high dynamic range needs to be resolved (very small signals close to very large signals) use the 4-term or 7-term Blackman-Harris windows. If the spectral peak amplitudes must be accurately measured use the Flat Top window, this will provide amplitude accuracy of 0.01 dB regardless of where the tone being measured falls relative to the bins of the FFT. The other windows only show the spectral amplitude accurately if the tone is exactly on the centre of an FFT bin, if the tone falls between two bins the amplitude is lower, with the maximum error occurring exactly between two bins. This maximum error is 3.92dB for the Rectangular window, 1.42dB for Hann, 0.83dB for the 4-term Blackman-Harris and 0.4dB for the 7-term Blackman-Harris.

Averaging

The plot can be set to show the live input as it is analysed or to show the result of averaging up to 64 measurements, according to the selection in the Averaging control. There is also an Exponential averaging mode, which gives greater weighting to more recent inputs. The Reset button restarts the averaging process. Averaging is needed when measuring with pink noise or when there is noise in the signal being measured. Note that if measuring a response using pink noise the best results are obtained using REW's periodic noise signals, which can be exported as wave files from the signal generator to produce a test disc for the system to be measured if direct connection to the PC running REW is not possible.

The final spectrum controls are the Y axis setting, which can be either dB FS or dB SPL, and the Save button. The Save button converts the current display into a measurement in the measurements pane. It is converted in the current mode of the analyser, so if the analyser is in Spectrum mode the measurement shows the spectrum, if it is in RTA mode it shows the RTA result. The saved measurements can be used as references for subsequent spectrum/RTA measurements.

Distortion Measurements

When the Calculate distortion check box is selected the analyser calculates distortion figures for the input, including THD and THD+N and the relative levels of the 2nd to 9th harmonics. The highest peak is used to determine the fundamental frequency of the input, this is displayed with the level of the fundamental. The THD figure is based on the number of harmonics whose levels are displayed and is calculated from the sum of those harmonic powers relative to the power of the fundamental. The THD+N figure is calculated from the ratio of the input power minus the fundamental power to the total input power (note that it is possible for THD+N to be lower than THD using these definitions). The example below shows data for a 1kHz sine input that is being driven 0.1dB into clipping, generating odd harmonics. The positions of the harmonics are shown on the spectrum or RTA plot.

Distortion Results
Distortion Results Plot

Oscilloscope

This group shows the generated sweep test signal and the raw captured system response as acquired via the soundcard, which may be useful for troubleshooting. This is not a live display, it updates with new content after a sweep has completed. Only the signals for the last measurement are shown. The Y axis is the percentage of digital full scale. The generated sweep is shown normalised so that its peak value is 100%. If the captured trace reaches +100 or -100% it is clipping and the sweep level or AV processor volume should be reduced.
Oscilloscope Plot

A check box is provided to invert the captured trace for easier comparison with the test signal if the soundcard input is inverting.

Help Index