PlayClassics TRT v3.0 sneak preview

[Mario Martinez]

Hello everyone!  I finally have something new that I would like to share with you.

 

We have spent the last 18 months working on a new version of our Truthful Recording Technology. We are about to launch it to the public but before we do, I would love to share it with you here at the Computer Audiophile.

 

This new version features a new piano with a new placement on the stage, some minor adjustments on the acoustics of the hall and a completely new calibration. The recording gear and the placement of the mics remains the same. Our philosophy has not changed a bit. We are just digging deeper into our original project.

 

 

Edit (April 14 ):

The previous calibration (v2.0) had some issues that we wanted to solve. I do think v3.0 was successful on fixing those issues but I also felt that the "general balance" was somehow altered on this new v3.0. I have been working on it and as a result we now have a new calibration v3.1. I have prepared a zip file with seven samples processed with this new v3.1

- one movement of a String Quartet
- one movement of a Wind Quintet
- the Flamenco sample number one
- the Drums sample number three
- one short piano piece by Mas Porcel
- one short piano piece by Webern
- Liebeslied by Rachmaninoff
 

If anybody wants to try these please post or send me a PM and I will send you the codes to download them.

 

Here is a sketch of the string quartet:

 

here is a sketch of the wind quintet:

 

here is a sketch of the flamenco sample:

 

here is a sketch of the drums sample:

 

here is a sketch of the piano:

 

the person in the red shirt represents the placement of the mics.

 

I hope you find this interesting. I would really appreciate your feedback

[Mario Martinez]

Thanks Ricardo, I am glad to be back

 

most of you who already know about this project are probably wondering why we went through the trouble of developing a new version. Allow me to use our little sketches to try to explain why we thought this step was necessary.

 

Ok, so ideally we would like to have something like this:

The orange area would be the "calibrated area" of the stage. That would mean that the setup would be prepared to transparently capture the tone of any instrument within that area.

 

Our old piano was placed somewhere within this area like this:

We wanted the piano to sound as transparent as possible, so we adjusted the calibration as much as we could. The more we adjusted, the more we narrowed the calibrated area to the position of the piano. So by the time we had our TRT calibration v2.0 we had something like this:

This calibration worked wonders for the piano. But take a look at what happened when we used other instruments:

This other instruments (the drums, the winds, the flamenco singer and guitar and a possible string quartet) will lay outside of the calibrated area. They will still sound good but the results may not be optimal.

 

That is the problem we were trying to solve by developing this new calibration v3.0.

 

So this time we place the new piano in a new position like this:

We then worked on our calibration until we got the most transparent result possible. By the time we finished we had something like this:

This configuration results in a smaller calibrated area, but even though it is smaller we are still able to fit other instruments too:

A smaller calibrated area allows us to be more precise resulting on more transparent results plus being able to fit other instruments within this area allows those instruments to benefit from the same results.

 

 

[Mario Martinez]

Yes

[Mario Martinez]

I am thinking, ...it might be useful if we could use the same recorded material to directly compare both calibrations.

 

The old and the new pianos are sitting in different positions, so there is no point on processing any old recordings with the new calibration or vice versa.

 

But we could use the experimental recordings we all have (the flamenco and the drums) because those recordings do not involve the piano and the instruments in both cases are sitting on the affected area.

 

According to the previous explanation this would be the result using TRT v2.0:

In both cases the instruments lay outside of the optimized area.

 

On the other hand this is what we would obtain if we were to use TRT v3.0 on this same recordings:

Now the instruments are laying within the optimize area which should translate into better (more transparent) results.

 

I have process both recordings the drums and the flamenco with the new version (TRT v3.0) so you can downloaded them too. Here are the links to all the file formats:

 

Flamenco TRT v3.0 Master

Flamenco TRT v3.0 DVD

Flamenco TRT v3.0 CD

Flamenco TRT v3.0 MP3

 

Drums TRT v3.0 Master

Drums TRT v3.0 DVD

Drums TRT v3.0 CD

Drums TRT v3.0 MP3

 

 

 

[Mario Martinez]

Hi Danny, thanks for listening.

 

I am glad you like it

 

The purpose of doing this new calibration was not so much to improve the sound of the solo piano but to guarantee that we would get those same results across all other instruments when recording chamber music.

 

Have you had a chance to listen to the quintet and/or the drums?

 

I am really interested to see what your think about the new version of the drums since that is a recording we can use to compare v2.0 vs v3.0. If everything went right that recording should show a significant improvement.

[Mario Martinez]

Hi Danny,

 

thank you very much for your feedback.

 

I did think the old calibration had some issues (otherwise we would have not gone through the trouble of making a new one) but you are really not aware of how much the results are affected by those issues until you hear them resolved and I think this drums recording does just that. I was surprised myself when I heard that bass drum under this new calibration.

 

The good thing is that the quintet (or any other chamber music arrangement that fits within this area) should benefit from this results too.

[Mario Martinez]

Thank you all for taking the time to listen

 

I sent Rando a PM with the links to the old version of both flamenco and drums. If anybody else needs that please tell me and I will send them to you too.

 

I did not explain anything about our setup. It has been a long time since our last thread so let me go ahead and give a summary of how we are doing this for those of you who are new to our project.

 

What you are hearing is the take of two mics always placed on the same spot of our auditorium outside of the stage area. The left mic is the left speaker and the right mic is the right speaker so there is no mixing involved.

 

All these recordings have been made using the same recording setup (by recording setup I mean the system formed by auditorium acoustics, stage area, mic position and arrangement, recording chain and calibration). The only thing that changes from one recording to another are the instruments and the musicians themselves. Everything else remains the same so there is no mastering involved.

 

We do not use any dynamic range compression so what you are hearing is the full dynamic range of the instruments.

 

 

[Mario Martinez]

Apologies accepted

 

Do not worry about it, I completely understand that this "file testing" might bring up other subjects too.

 

What I wanted to know in the original OP is whether or not you felt this new calibration was an improvement over the previous one.

 

But besides that, if you feel these files may be useful for any other purpose please go ahead and use them I will be happy to do that. We already used them once on another thread to compare file formats...

 

[Mario Martinez]

I am grateful to have the opportunity to share our work with all of you.

 

Looking at your comments I think it would be safe to say that this new calibration is an improvement so we are going to go ahead and release all our new material with this new sound.

 

Since we started this type of threads, a couple of years ago, your feedback has been both very helpful and motivating.

 

This kind of work takes a long time and it really helps to know that it is appreciated. So thank you all for making it happen

 

[Mario Martinez]

33 minutes ago, gmgraves said:

Hey Mario! Do you mind telling us what microphones you use (make, model, pattern) and how they are deployed? As a sometimes recording engineer myself, I'd love to know. 

Hi George,
 

The microphones are large-diaphragm cardioid microphones with no attenuation or cut-off (I would rather not say the make and model). We use a standard ORTF setup (capsules are 17 cm apart in a 110º angle). They are mounted on an elastic suspension hanging from their own cable (so the weight is on the cable and not on the elastic suspension)

 

[Mario Martinez]

What we are doing with our mics is nothing new. There are lots of great examples of two mic recordings dating back to the 60s and 70s (ORTF itself dates back to the 60s). What is different is the way we see the relations between the acoustics of the hall and the placement of the mics.

 

Back then they took the acoustics of the hall as a given. That is what they had to work with. Getting a good sound was a matter of finding a better spot to place their mics on.

 

If you are not going to modify the acoustics of the hall no spot is going to be perfect. Some spots may have better balance, others may have better dynamics, others better tone or better image, but none of those spots will maximize the outcome of all this different aspects at once.

 

Our philosophy is different. We placed the mics first with just one aspect in mind (image). Then we worked the acoustics of the hall to try to maximize all the other aspects and developed a calibration that took care of the rest.

 

You obviously cannot do this under regular circumstances. It is too time consuming. You have to get the job done and you only have a few days to do the best you can. So you would never have time to do all this experimenting.

 

But we though it was worth a try. What would the results be if this was to be done exhaustively? Well, this is what we came up with. It really did take a long time but I would do it all over again

 

 

[Mario Martinez]

Every time you try to record an acoustic instrument most of the problems you encounter are actually caused by the hall. The standing waves and resonances of the hall cause deformations on the audio spectrum that translate into coloration and altered balance of the recorded sound.

 

Say you have been hired to make a recording, you take your equipment to the hall where the recording is going to be done and you do the best you can. The only thing you are in control of is your equipment and where you place that equipment. Working on the acoustics of the hall is out of the question, that is major work that would take up too much time and too many resources.

 

You are not allowed to fixed the source of the problems but you are expected to get good results anyway. No wonder there are so many different recording techniques (anything is good to try to avoid picking up the problems that are messing up the sound). So you record the material as well as you can and then use your regular mixing and mastering to try to get everything to sound "good".

 

What we are doing is the opposite of that. We are trying to address the problem at the source. We gave ourselves the opportunity to work on our own acoustic setup trying to resolve everything there (in the "real world"). It is not physically possible to resolve every issue but you can solve as much as you can and develop a calibration that takes care of the rest.

 

[Mario Martinez]

Here is an article that talks about the birth and evolution of stereo recording:

 

http://www.overgrownpath.com/2013/03/music-to-listener-or-listener-to-music.html

 

It talks about two schools of mic placement. The American School (that favors the multiple microphone approach) and the European School (that favors simpler microphone placement):

 

"The simpler British/European school of microphone placement produced the most realistic and purist sound, as can be heard in Decca and EMI’s iconic Kingsway Hall recordings. But these recordings were subtle rather than showy and could sometimes sound thin and unimpressive to untutored ears when replayed on cheaper stereo systems. There was also the disadvantage that the recording engineers needed specialist knowledge to determine microphone placement, and expensive studio session time was required to balance the sound as there was little scope for correcting errors at later remixing sessions. Predictably the American school produced the opposite results: the sound was less refined but more ‘in your face’, microphone set-up was faster and less skilled, and technical and music balance problems could be ironed out by post-session remixing.

In recent years the classical recording landscape has changed completely. Financial pressures and the consequent advent of ‘live’ concert recordings by orchestra owned record labels mean the more flexible and lower cost American model dominates, to the extent that the geographic labels are now redundant as multi-miking has virtually become standard on both sides of the Atlantic. This has, inevitably, resulted in compromises in sound quality"

 

These two schools are just two different approaches to solving the same problem: every time you try to record something (no matter how good you think the hall is) the microphones seem to pick up the worst of it. It just never sounds quite the way the life event did. They seem to work like a selective audio microscope picking up and augmenting all the defects in the hall. This is what I was referring to when I said:

 

"Every time you try to record an acoustic instrument most of the problems you encounter are actually caused by the hall. The standing waves and resonances of the hall cause deformations on the audio spectrum that translate into coloration and altered balance of the recorded sound."

 

With the European approach we are just trying to find a clean spot within the hall. A place where all those standing waves and resonances somehow balance out to cancel each other. If we are successful in doing so, we will end up with a recording that benefits from the cleaner sound produced by fewer microphones but we will never get the "presence" in the sound produced by the American approach.

 

With the American approach we are just planting mics all over the scene. We do not have to be concerned about the sound of each one of those mics. We just want to make sure we have plenty to choose from. We will deal with the sound latter at the mixing table. The result will be a recording with a lot of "presence" but we will never get close to the clean sound produced by the European approach.

 

So what is it that we are trying to do?

 

We are trying to develop a technology that would be able to combine the best of these two worlds. A technology capable of  producing recordings with the cleanest possible sound while retaining all the "presence" of the real event.

 

The "clean sound" part seemed pretty obvious. What ever we came up with would have to be done using just two mics and no mixing.

 

The "presence" part was the tricky one. That was the challenge. That was the part that had never been achieved before with a two mic setup. We thought the only way to achieve this would be to deal directly with the source of the problem (the acoustics of the hall). So, instead of looking for a "clean spot" within the hall, we decided to place the mics first and then work on the acoustics of the hall to control the sound that was being picked up by the mics.

[Mario Martinez]

The article talks about how things developed into two schools of mic placement. But more importantly, it gives a very good explanation of how one of these schools became the model for today's standard:

 

"In recent years the classical recording landscape has changed completely. Financial pressures and the consequent advent of ‘live’ concert recordings by orchestra owned record labels mean the more flexible and lower cost American model dominates, to the extent that the geographic labels are now redundant as multi-miking has virtually become standard on both sides of the Atlantic. This has, inevitably, resulted in compromises in sound quality"

 

There is one other factor that helped things evolve the way they did:

 

The American approach may not produce a clean sound but its flexibility allows for the engineer to "build the sound" at the mixing table. This may not be of much use for classical music but it is essential when recording "pop".

 

Here we have an approach that has an "in your face" sound, requires less time and less skills, allows for correction of technical and musical problems on post-session remixing and is cheaper to implement.

 

If you had to build a professional studio to record all kinds of music (classical and pop) which way would you go? The American approach has become the standard in both worlds because it has proven itself to be the most "practical" one.

 

But we are not trying to build a commercial studio. Ours is a research project. We wanted to pick it up where those Kingsway Hall recordings left it to try to see if that simple approach could be improved.

 

We know that this system is more time consuming (thus much more expensive) and less flexible (the musicians have to sound good, here they are the ones responsible for the balance and color not the engineer) but we thought the results (specially for classical music) would be worth the effort (it has taken us 9 years to get to this point).

 

 

[Mario Martinez]

Hello everyone,

 

I know I did say:

On 3/16/2018 at 9:38 PM, Mario Martinez said:

Looking at your comments I think it would be safe to say that this new calibration is an improvement so we are going to go ahead and release all our new material with this new sound.

 

But I could not help myself...

 

The previous calibration (v2.0) had some issues that we wanted to solve. I do think v3.0 was successful on fixing those issues but I also felt that the "general balance" (which was perfect on v2.0) was somehow altered on this new v3.0 (I did receive some private comments that I think were probably caused by this)

 

So I have been working on the "general balance" of this last calibration (v3.0) to try to restore it back to the balance of v2.0. As a result we now have a new calibration v3.1 (which I hope will be the end of our developing journey)

 

I have processed all our new albums with this new calibration (v3.1). They are all now available on our web. I think we are now ready to open a new master file giveaway thread...

 

 

 

 

[Mario Martinez]

I have prepared a zip file with seven samples processed with v3.1

 

- one movement of a String Quartet

- one movement of a Wind Quintet

- the Flamenco sample number one

- the Drums sample number three

- one short piano piece by Mas Porcel

- one short piano piece by Webern

- Liebeslied by Rachmaninoff
 

If anybody wants to try these please post or send me a PM and I will send you the codes to download them.

[Mario Martinez]

Thank you rando for taking the time to listen.

 

I just sent you the codes to the new samples.

 

If for whatever reason you do not find the Polish Songs (v2.0) tell me and I will send you a code to download that too

 

I understand all this upgrading and re-listening can get a bit tiresome so I really appreciate it that you all endure with me on this.

 

I am the grateful one. I feel it is a privilege to be able to count on your help. Constructive criticism is the only way to improve

[Mario Martinez]

11 hours ago, semente said:

I don't know if you're comfortable disclosing this but does the "calibration" process involve physical changes to the room

 

 

No problem. Actually I would be happy to explain more than that.

 

Lets make a comparison between methods so we can see how it all works.

 

Imagine we have been hired to make a series of recordings (4 or 5 recordings of different chamber music ensembles).

 

The person that hired us wants us to make all these recordings in the same venue and always aiming for the most natural sound possible.

 

This is what we would do if we were to try to accomplish this using multi-miking:
- Things we do prior to each recording session: we place one mic for each instrument. We just need to make sure each mic captures the sound of one instrument as independently as possible from the others. The mic arrangement will be completely different for each recording session (different number of mics, different distances from the instruments, different places within the stage...)
- Things we do after each recordings session: we should have been able to record each instrument on a separate track. The recording engineer can now treat the sound of each instrument independently without affecting the others. He will try to use everything that is at his hand (limiter, compressor, expander, gate, eq, etc...) to try to make each one of those instruments sound as "good" as possible. I say "good" because here the engineer has no objective way to check if the resulting sound is more or less "true". All he can do is aim for what he subjectively thinks sounds more natural. Once he has fixed all the tracks he will go ahead and mix them at the mixing table. He will adjust the level and pan (left and right) each instrument separately until he gets a stereo sound that, again, he subjectively thinks sounds "good". Now, we could stop at this point, or we could send this final mix to the mastering engineer for further processing. The mastering engineer will treat the sound of this stereo file using everything that is at his hand to try to achieve what he subjectively thinks sounds even better.

 

This is what we would do if we were using ambient-miking (just two mics):
- Things we do prior to each recording session: we look for a good sounding spot to place our mics. We try several positions that we think are reasonable and listen to figure out which one sounds best. We should do this for every different ensemble because we might end up using different spots for different ensembles.
- Things we do after each recording session: we should have been able to capture all the instruments in one stereo track. We cannot treat instruments separately. Anything we do will have to be applied to all instruments at once. The recording engineer will now do whatever it takes to try to make this stereo track sound as "good" as possible. Again, I say "good" because it is still a subjective matter. Different engineers would probably end up at different places. After this, we could stop here or we could send this files to the mastering engineer so he can again treat the sound just like he did in the previous scenario.

 

Here is what we would do if were using TRT:
- Things we do prior to ANY recording session: we choose a spot to place our mics in but, at this point, the only thing we should be concern with is image (as if we were placing a video camera). This will be the place of the mics for every recording. Once we have our mics in place we run tests to determine what we can do to the acoustics of the room to try to improve the sound at the pick up point. The more resources and time we have the better the results will be but even with endless resources and time we will never completely fix everything. What ever we end up with will be the room configuration we will use for every recording. Then we work on an eq filter that will hopefully be able to fix the things that could not be fixed on the actual room. We will use this exact same eq filter in all our recordings (this eq filter is what we call the "calibration").
- Things we do prior to each recording session: nothing.
- Things we do after each recording session: simply run the calibration file over the stereo file we recorded. No subjective decisions involved.

 

 

[Mario Martinez]

6 hours ago, danadam said:

In Stories - The Romantic Spanish Guitar, what is this other sound, beside the guitar? Is this the player (i.e. Dimitri) breathing?

 

Dimitri and I talked about this during the recording sessions. He has a severe asthma condition. Even right after treatment you can still hear his breathing...

 

Because of the way we record there was nothing I could do to avoid picking it up. Our mics are not directed towards the instruments. Everything is configured to record the whole room. Any noise produced with in the room will be picked up just the same as the instruments on stage. Had there been a listener in the room he would have heard the same thing.

 

 

[Mario Martinez]

I have been asked about something that I think would be good to post:

 

Say you donate to download an album (for example one of these new albums on calibration v3.1).

 

If we ever came up with a new calibration (no matter how many times we upgrade) we will always contact you to let you know about the upgrade and give you a gift code so can download the new version.

 

I just thought I should clarify this since it does not say it any where on our website but that is just the way we do it. We have done it before and we will keep doing it as long as we keep developing this technology.

[Mario Martinez]

Hello everyone,

 

I would like to update those of you who are following our developments.

 

It has almost been a year since we posted our last calibration (version 3.0). At that time we had two different calibrations: 2.0 on the albums recorded with the old piano and 3.0 on the albums recorded with the new piano.

 

The results obtained with calibration 3.0 were not as good as expected. Some of you contacted us privately to give us you feedback (for which we are very grateful)  and told us about the issues that you had with this new calibration.

 

We tried to work on those issues and developed a couple of fixes but it did not seem to work. It looked like we had overlooked something that was throwing us off. For that reason we decided to drop version 3.0.

 

Since then, we have been working on a new calibration but this time we have followed the same path that took us to calibration 2.0 This time we did not end up with a different calibration for each piano. Instead we have created a calibration that works for both pianos and everything else on the stage. Now we have one calibration for all 12 albums (the old ones and the new ones).

 

I would like to have the opportunity to present you with these  results. It would be my pleasure if you could (once again) take the time to listen and give your feedback. I will open a new thread to offer a giveaway...

 

[Mario Martinez]

Hi Dieter,

 

Here is a link to the new thread:

PlayClassics TRT Sound master file giveaway for CA members

 

Hope to see you there :)