[EA Support]

EA Support

5 January 2022 at 6:53 am #394

Hi Lukas,

Our apologise we haven’t kept the Target file formats matched. We’ll remedy that in the next update.

Best regards,
EA Support

[EA Support]

EA Support

21 December 2021 at 8:23 pm #391

The best placed to start is the “FIR Designer 3 Overview” video on https://eclipseaudio.com/fir-filter-design-tutorial/. After that, take a look at the “FIR Designer M (Multi-way) Overview” video.

Best regards,
EA Support

• This reply was modified 4 years, 2 months ago by EA Support.

[EA Support]

EA Support

4 December 2021 at 6:14 am #387

The required number of taps increases for lower frequencies and with higher Q filtering. Consequently LF drivers will need longer filters than HF drivers. There isn’t any clear rule of thumb. Just increase the filter length until the processing works for your design.

Best,
EA Support

[EA Support]

EA Support

3 December 2021 at 10:05 pm #385

Your original 48 kHz filter was 27 ms long. In the most recent screenshots your filter is only 6.25 ms long. I think you’ll need to increase your filter length. (See my previous comment above about doubling the tap length, when changing from 48 kHz to 96 kHz, to maintain the filter length in ms.)

Regards,
EA Support

[EA Support]

EA Support

3 December 2021 at 11:47 am #379

Could you send your System and Channel files to our “info” address and we’ll take a look.

In 2.2.3 we fixed a bug where the top “Enable” checkboxes on some tabs were sometimes not refreshing correctly when a channel project file was loaded, but we haven’t seen this with Systems.

Best regards,
EA Support

[EA Support]

EA Support

3 December 2021 at 6:21 am #375

No you don’t need to remeasure. (The sample rate of the measurement does not need to match the FIR Designer sample rate.) Just change the sample rate in FIR Designer to 96 kHz and take a look at the Export tab – you’ll probably need to double the filter delay and length settings to match the time intervals at 48 kHz – then export the filter again.

Regards,
EA Support

[EA Support]

EA Support

2 December 2021 at 5:27 pm #373

>> i was doing 48khz sample rate while the xta is native at 96k.
>> Do i need to redo my measurement as well in 96k?
No, just change the sample rate in FIR Designer to 96 kHz and check the Export tab – you’ll probably need to double the filter delay and length settings to match the time intervals at 48 kHz – then export the filter again.

>> …. are these correct?

Yep, I think that’s a good starting point. We have tutorial showing a 2-way design with only FIR at

Designing FIR Crossover Filters for 2-Way Loudspeaker

This is just one approach to doing a FIR 2-way. Like I said above, another way is to do a regular IIR crossover design, then measure the whole cabinet and create a single FIR correction filter for the whole speaker. Some processors can only do a single FIR before crossover. Some can run a FIR for every output.

Best,
EA Support

[EA Support]

EA Support

2 December 2021 at 2:43 pm #371

Since you’re trying to flatten the phase from about 200 Hz upwards, I think it’s ok to leave the 100 Hz HPF engaged.

The bigger issue is the crossover and the relative physical placement of the drivers in the cabinet. Generally you can’t ignore these and expect to linearise the phase of the LF driver from 200 Hz up to crossover. The crossover filtering itself and the relative physical placement of the drivers both impart phase change through the crossover region. And so you need to measure the system as a whole. And like I said above, the manufacturer supplied crossover would have been designed with dispersion characteristics in mind.

The only time it may be possible to linearise the drivers separately is if you use a linear-phase brick-wall type crossover (like in the Lake processors). However you’ll likely still have a narrow frequency phase jump at crossover.

The 300 Hz versus 600 Hz issue could be that you’re designing the filter for 48 kHz sampling but the processor is running it at 96 kHz. Try changing the sample rate (in FIR Designer) to 96 kHz and see what happens.

Regarding question 3, FIR filters can be minimum phase (with no coefficients prior to the peak), linear phase (with coefficients symmetrically either side of the peak), maximum phase (with the peak at the end) or a mixture of all. Phase processing, as you are doing, generally results in a mixed phase filter. The “Filter delay” parameter enables the peak to be placed to minimise the error between the ideal FIR filter, and the filter windowed and truncated to the number of taps available (e.g. 1300). For more details, check out the links and tutorials referenced above.

When phase linearising using a FIR filter, the net effect is that the loudspeaker will have a bulk delay of approximately this “Filter delay” value. And if you are doing this on just one driver, the other driver will need to be delayed by the same amount. However see my previous comment about measuring the speaker as a whole.

If you measure the loudspeaker as a whole, you can still design the FIR for, say, 200 Hz to 1 kHz. The important part here is that you’ve measured WITH the crossover in the measurement/s.

Best,
EA Support

[EA Support]

EA Support

2 December 2021 at 9:11 am #369

Hi santodx5.

It sounds like you might be measuring either the XTA outputs or the XTA + LF driver on its own, and not the whole speaker.

The loudspeaker’s overall response is a combination of both the processor, the loudspeaker driver characteristics and their relative physical position in the cabinet (w.r.t. delay/phase). So to flatten the phase of a loudspeaker, you need to measure the loudspeaker itself with all processing enabled.

(If you try to measure and correct the individual XTA outputs, you’ll be undoing the processing – particularly the crossovers – that is/are necessary to align the drivers. Also the loudspeaker dispersion, through the crossover frequency range, is controlled by the existing processing, and so if you change individual driver processing, you’re potentially changing the dispersion characteristics of the cabinet.)

When you have a measurement you’re happy with, take look at

FIR Filter Design for Loudspeaker Equalization

and the “FIR Designer 3 | Overview” video at

FIR Filter Design Tutorials

(The video also discusses the filter delay. Maybe also take a look at https://eclipseaudio.com/fir-filter-guide/.)

Linearising a loudspeaker (in mag, phase or both) from a single measurement can be problematic. Whilst the phase might be what you want at that particular measurement point in space, the phase response in other directions could be very different. Generally this kind of thing should be done by taking measurements at many different angles (within the loudspeaker coverage pattern) and averaging the measurements (and preferably all in a clean measurement environment or anechoic chamber). FIR Designer includes a measurement average function – see the Overview video mentioned above.

Most of what we’re discussing here goes beyond the scope of how to use our software and relates more to the fundamentals of loudspeakers. These forums are mostly for how to use our software but we’ll help where we can.

Regards,
EA Support

[EA Support]

EA Support

2 November 2021 at 7:46 pm #366

That’s it. 🙂
Glad you found it.

[EA Support]

EA Support

28 October 2021 at 6:28 am #361

The problem is the 1st line starting with a number. If the line starts with a letter or the line is removed, the file loads fine.

Kind regards,
EA Support

[EA Support]

EA Support

28 October 2021 at 6:12 am #359

Hi DomDomm,

Take a look at the various file format options on

Loudspeaker Measurement Import File Formats

If this page doesn’t help, please email us the file at info@eclipseaudio.com.

Note: SMAART recently changed their TRF file format to support MTW+. We only recently released updates for this so if TRF files are the problem and you use MTW+, check for FIR Designer updates. Alternatively, in SMAART change from MTW+ to MTW before taking the measurement.

Kind regards,
EA Support

[EA Support]

EA Support

19 September 2021 at 9:01 am #354

Hi hulkss,

We haven’t yet implemented the system step response but yes, it would be good to see. It’s on the TODO list. As an alternative, the “Wavelet Transform” view shows the combined system behaviour and gives a sense of the overall temporal response.

Best regards,
EA Support

• This reply was modified 4 years, 5 months ago by EA Support.

[EA Support]

EA Support

14 September 2021 at 9:25 am #351

Sounds good. The Help in FIR Designer M is quite lean and focusses mostly on UI controls and their effects.

I’d suggest starting with the tutorials and videos on https://eclipseaudio.com/fir-filter-design-tutorial/

Best,
EA Support

[EA Support]

EA Support

14 September 2021 at 8:09 am #348

FDW is REW’s equivalent of SMAART’s FTW and SysTune’s TFC.

Whether or not to use it is completely up to you, and is based on your measurement environment and preferred workflow.

FDW, FTW and TFC are intended for situations where there might be strong mid and/or high frequency room reflections (which we’d like to remove from the TF with a window) and where shortening a traditional window (to remove these reflections) would cause loss of resolution in the LF. The frequency varying window provides for a longer window length at LF and shorter window length at HF. In a lab environment or anechoic chamber where reflections are minimal, FDW, FTW and TFC aren’t really necessary.

We have customers who use the SMAART FTW by default for measurements that they use in FIR Designer, and it gives them tuning results they are happy with.

See also…
https://www.rationalacoustics.com/download/Smaart-v8.3-Release-Overview.pdf See page 8.
https://www.prosoundtraining.com/2010/03/15/easera-systune/ See Figure 1.

Best regards,
EA Support

• This reply was modified 4 years, 5 months ago by EA Support.