TDA Users

Brandon Wright

Freshman
Sep 19, 2012
75
2
8
For all of you users (or inventor) of Raimonds TDA program out there, I was hoping we could start a sort of informational thread on its use and application. I am thinking about purchasing the program, but would like to know what I stand to benefit.

Currently, I am averaging spatial measurements in Systune and importing them into RePhase or FilterHose to generate FIR filters in the form of .csv files. These then are being implemented in Powersoft's Armonia software.

Can the TDA software truncate the FIR to a specified sample length? In my case a maximum of 384 samples. I see it exports .fir files. Is this a text file or an audio file for direct convolution? In short, can I take advantage of this software without using the APL1 processor?

Any other insights as to its usability?

Thanks,
Brandon
 
Re: TDA Users

For all of you users (or inventor) of Raimonds TDA program out there, I was hoping we could start a sort of informational thread on its use and application. I am thinking about purchasing the program, but would like to know what I stand to benefit.

Currently, I am averaging spatial measurements in Systune and importing them into RePhase or FilterHose to generate FIR filters in the form of .csv files. These then are being implemented in Powersoft's Armonia software.

Can the TDA software truncate the FIR to a specified sample length? In my case a maximum of 384 samples. I see it exports .fir files. Is this a text file or an audio file for direct convolution? In short, can I take advantage of this software without using the APL1 processor?

Any other insights as to its usability?

Thanks,
Brandon

For speaker measurement you need TDA EQ, for room EQ, TDA. TDA uses the power response not an SPL measurement. Raymond's argues that the way our ears/ brain perceive sound in a space is better modelled with the power response and a target curve than the SPL response.

I have both TDA and TDA EQ - I use TDA EQ fir speaker alignment and export the file to an APL1, I then manually transfer this to my Lake LM26. The Lake will not let you import the file directly.... I believe its a text file.
 
Re: TDA Users

I have looked into TDA quite a bit. It would be nice to know how it differs from the CONEQ program from the Real Sound Lab guys. I know CONEQ allows for direct export of FIR coefficients.


Sent from my iPad
 
Re: TDA Users

I have looked into TDA quite a bit. It would be nice to know how it differs from the CONEQ program from the Real Sound Lab guys. I know CONEQ allows for direct export of FIR coefficients.


Sent from my iPad

Its the same in that respect ...

http://www.realsoundlab.com/files/4250_CONEQTechnologyShort.pdf

"The inventor of the CONEQ™ technology is Latvia-born Raimonds Skuruls. He has worked
as a sound engineer, notably with Latvian pop legends Jumprava, and has spent most of his
life looking at better sound reproduction from loudspeakers. Having a degree in electrical
engineering, Skuruls has a unique mix of knowledge and experience that led to creation of
what is now known as CONEQ™. In 2004, Skuruls together with investor Viesturs Sosars
formed Real Sound Lab SIA. Since then Real Sound Lab SIA has been developing the technology
further, manufacturing CONEQ™-enabled products for professional audio market as
well as licensing the technology for use in consumer electronics devices like TVs, personal
audio systems, and virtually any other device that has a loudspeaker."


http://www.soundonsound.com/sos/feb13/articles/coneq.htm
 
Re: TDA Users

Thanks for starting this, I just bought TDA today, and used the demo pack, will be installing tomorrow.

Regards, Jack
 
Re: TDA Users

I too purchased TDA and TDA EQ. I've used it with some success but don't have a clear understanding on how to best implement the testing for my intended use. Looking forward to learning lots.
 
Re: TDA Users

I of course will now have to get TDA EQ. My first impression of the TDA is, YAY! I have a set up in the living room that we use almost only for dance, either with just us, or when we have a dance party, and others are invited over. It is right in the middle of the house, so it is too intrusive to use a lot, just for casual music listening, if this makes sense. Before, I had used LibInsts to set it up. So with just a little bit of dabbling, I had some minor changes. The system has a preamp, that I send the subwoofer out directly to an amp, the lows and mid/hi are on a powersoft M series, 4 ch amp. Right away I could see that I did not have good alignment with the lows and subs, and I changed the lows to mid highs delay about .2ms. It is interesting to me to see the whole spectrum in time, rather than to independent bands, and compare the time difference. The EQ "mellowed" a bit. Not as much HF, which surprised me in such a small, and relatively well behaved, space. All of the comparisons are done with playback, no live mics, but the sound reminds me more of how the Coda boxes I have/had sound. There have been other small changes, but no much so far. When my wife listened to it, she immediately commented that it sounded more natural. And she does not usually have much comment on the sound. (When we saw the Black Swan, the film had so much wow/and or/flutter, that the sound track was warbling the entire night. All the piano stuff was driving me nuts, and then after that, the dialog too was noticeable to me, but she had no idea. I have learned to try and keep my disgust at a low, but usually at movies it is with suspension of disbelief, not something of this nature.)

We very rarely do house concerts, and have one coming up this weekend, using this same system. I will be very interested to see how this set up behaves with open mics. I am very used to doing a lot of tuning with an open mic, to try and keep the mics from interfering with the sound system.

Toodles for now, Jack
 
Re: TDA Users

I of course will now have to get TDA EQ. My first impression of the TDA is, YAY! I have a set up in the living room that we use almost only for dance, either with just us, or when we have a dance party, and others are invited over. It is right in the middle of the house, so it is too intrusive to use a lot, just for casual music listening, if this makes sense. Before, I had used LibInsts to set it up. So with just a little bit of dabbling, I had some minor changes. The system has a preamp, that I send the subwoofer out directly to an amp, the lows and mid/hi are on a powersoft M series, 4 ch amp. Right away I could see that I did not have good alignment with the lows and subs, and I changed the lows to mid highs delay about .2ms. It is interesting to me to see the whole spectrum in time, rather than to independent bands, and compare the time difference. The EQ "mellowed" a bit. Not as much HF, which surprised me in such a small, and relatively well behaved, space. All of the comparisons are done with playback, no live mics, but the sound reminds me more of how the Coda boxes I have/had sound. There have been other small changes, but no much so far. When my wife listened to it, she immediately commented that it sounded more natural. And she does not usually have much comment on the sound. (When we saw the Black Swan, the film had so much wow/and or/flutter, that the sound track was warbling the entire night. All the piano stuff was driving me nuts, and then after that, the dialog too was noticeable to me, but she had no idea. I have learned to try and keep my disgust at a low, but usually at movies it is with suspension of disbelief, not something of this nature.)

We very rarely do house concerts, and have one coming up this weekend, using this same system. I will be very interested to see how this set up behaves with open mics. I am very used to doing a lot of tuning with an open mic, to try and keep the mics from interfering with the sound system.

Toodles for now, Jack


If you try to EQ a speaker in a room “flat” with Smaart or Systune, especially with respect to the low frequencies, it will not sound right. Smaart is a great program and will tell you a lot, but it must be used with care and understanding. You must still listen and trust your ears. Your ears and brain do an excellent job of selecting from the background noise what you want to hear. What sounds great will not necessarily look great on a Smaart plot.

What TDA will do for you is more or less what you are doing by ear, but much more accurately.

For example, if you hear a resonance around 100Hz, TDA will tell you the exact frequency, the amount and band width of what you need to adjust to keep the power response in the room balanced. It will also show any time alignment issues you have. The end result will sound natural and balanced... you still need to listen of course :)

The adjustments can be automated by sending a correction curve to a FIR / DSP engine such as APL1.

TDA EQ on the other hand is not dissimilar to Smaart in many ways; it tries to measure the performance of the speaker independent of the environment and provides some great information about the time domain performance of the speaker.

Here is a TDA EQ plot ... an almost perfect text book response I made today of a BMS 4594He (and the B&C 12s ;-) in the DIY using the tricks discussed in the FIR thread.

Note: 24dB LR IIR filter at 100Hz
 

Attachments

  • TDA 60 degree 2016 2.jpg
    TDA 60 degree 2016 2.jpg
    303 KB · Views: 18
Last edited:
Re: TDA Users

Here is a TDA EQ plot ... an almost perfect text book response I made today of a BMS 4594He (and the B&C 12s ;-) in the DIY using the tricks discussed in the FIR thread.

Note: 24dB LR IIR filter at 100Hz

Peter, was this with the 90 deg horn? What was your test set up for the measurement, your usual 4M high by 2M out front? When you move the test mic in the grid-like pattern, how do you maintain a consistent distance from the DUT during the test tone?
Sorry for these basic deployment type questions but I've struggled some with getting consistent measurements with TDA EQ.

Your results look spectacular! The FIR thread still has my head spinning.
 
Re: TDA Users

Peter, was this with the 90 deg horn? What was your test set up for the measurement, your usual 4M high by 2M out front? When you move the test mic in the grid-like pattern, how do you maintain a consistent distance from the DUT during the test tone?
Sorry for these basic deployment type questions but I've struggled some with getting consistent measurements with TDA EQ.

Your results look spectacular! The FIR thread still has my head spinning.

It looks the same for both the 60 and 90 degree version. I just used one position for the mic which was about 4.0 to 4.5m high. The speaker was about 3.5m high pointing up. Its hopefully a free field measurement with no reflections.

I used Systune initially, and fine tuned the response with TDA EQ. I also tried a different approach this time. I EQ the natural roll off of the double 12" to match an exact 12dB 100Hz Butterworth to about -9dB. I then applied a 12dB Butterworth high pass. Two Butterworth responses in series is a 24dB LR :). This gave me much better low frequency time domain behaviour than I had before. The DIYs now integrate perfectly and seamlessly with the subs.

It was interesting to hear the difference from what I did with Systune, and then with some slight adjustments from TDA. It went from really nice to very real.
 
Let me heat this thread up a bit :).
Ok so I just ordered a Crown 4-3500HD (it has some FIR on board) for a good price, which I'm going to use to drive just the 2x12" B&Cs in each of four PM60s. I'm currently a bit in regret-mode, but not sure if for a good reason. While I'm just starting to get a grip on FIR I do understand that I need to (and want to) use it for squeezing the best sound out of the tops. First I wanted to get a Lake LM26 and 2xCrown I-Tech 6000 for this, so I can have the Lake as a central brain for doing FIR, also for the Coax drivers, at least for 2 with the 6 outputs of the Lake LM26.
But now there are a couple of questions, the first one regarding TDA and the second one FIR for my use case:

1) Can I transfer optimized FIR results from TDA somehow to this Crown amp? Or do I have to import the correction curve into a DSP engine, for example APL1 first and can then manually transfer FIR parameter that I somehow obtain from the APL1? I have to admit I'm not sure how exactly FIR parameters look like, I'm not there, yet.
Or in other words, can TDA help me to find the right FIR settings for my Crown.

2) Is it a valid idea to just use FIR via the amp on the 12"s and not at all on the Coax's, when aiming for best PM60 sound or would I miss out?