PlayClassics TRT v2.0 master file giveaway for CA members

[Mario Martinez]

The main goal of our project is to develop a recording setup that solves the problems artists face when recording their work.

 

To get an understanding of what these problems are you can read this paper "The performer's place in the process and product of recording" published by Professor Amy Blier Carruthers (Royal College of Music, London) over the 2nd PSN Conference on 2013.

The performer's place in the process and product of recording

 

We started working on our setup on 2009 long before that paper was published. Here you can read about our solution:

About - PlayClassics, the art of true music

 

As you can see, there are many aspects to these problems. The quality of recorded sound itself is one of the biggest issues. Our solution to this respect is based on transparency: "In our studios the recording hall is configured so that the sound of the hall is exactly the same as the sound of the take"

 

The purpose of the Truthful Recording Technology is to achieve this transparency. To go about this, we only use two mics always placed on the same exact spot of our auditorium (we do not use any mixing or mastering). Our technology is all about calibrating the hall and the recording chain to be transparent. The better the calibration, the more transparent the setup will be and the more control the artists will have over the recorded sound.

 

Nine months ago we started a thread called: PlayClassics master file giveaway for CA members. The purpose of that thread was to have CA members try the sound of the Truthful Recording Technology on their systems. What we shared with you back then was the most transparent sound we had so far (that was calibration version 1.0). That truly was a great experience. It was a lot of fun being able to share all that with you.

 

Up to that point we had only used our setup with classical music instruments, but thanks to that thread we started trying new things out. One of the things we tried was a drums set and a rock band. The recorded sound of those instruments was not as transparent as we would have expected. That made us realize there was still some room for improvement on our calibration. We tried to do some fixes over calibration v1.0 but that approach did not seem to work. (Here is the thread where we published those results: PlayClassics Truthful Recording Technology v1.1)

 

We then knew we would have to re-calibrate from scratch:

 

I would like to try something new.

 

I think fixing this minor problems by applying little patches over calibration v1.0 is not going to work.

 

In order to get good results we might have to re-calibrate from scratch. Do not worry, we are not going to change the sound. We are always aiming at transparency, so even if we do the calibration 100 times we will always end up at almost the same identical place. But this time, I am aware of this little problems, so I will try to deal with them from the beginning.

 

I might take me a few days, but I will report back as soon as I have something that I am convince of

 

It has been 5 months since then, but I think it was all worth it. We have a new calibration v2.0 that we would like to share with you. This is not a patch or a fix over v1.0 it is a completely new calibration. Our studio is setup so that you can record and playback right away so it is very easy to compare the quality of the live source against the quality of the recorded sound. Comparing the quality of the recorded sound of both calibrations (v1.0 and v2.0) against the quality of the live source it becomes clear that v2.0 is more transparent. What we would like to know is how does that turn out on your systems.

 

We will be happy to send a Gift Code to all CA members who would be willing to try the Truthful Recording Technology (calibration v2.0) sound on their systems.

 

Just post the album of your choice and we will send you a Gift Code to download the full dynamic range 24bit 96kHz Truthful Master.

 

All we want in exchange is to get as much feedback as possible. It would be great if we could get your testimonials to include them on our web page.

 

PlayClassics, the art of true music

[Mario Martinez]

Thank you all. I am really happy you are taking the time to listen!

 

I am thinking maybe we should all get two albums.

 

See, all these recordings are made exactly the same way: It is always the same piano, with the same tuning, sitting on the same spot of the stage. The hall, the recording gear and the calibration is always the same. In fact the only thing that is different would be the performer himself. So if you listen to two recordings of two different pianists you should get a sense of how those pianists use the same instrument to produce sound in a different way.

 

This is actually an important part of this Truthful Recording Technology. Listening to different recordings of different pianists you would get a real sense of how they use the instrument just like you would if you were to listen to all those pianists in a real concert situation.

 

So, if we go ahead and do this, you should all have the Albéniz Iberia by Luis Grané and one of the other three albums by Cabrera (Cabrera plays Debussy, Chopin Polish Songs or Songs of Paolo Tosti)

 

I hope you do not mind me going ahead and sending you all codes to complete your two albums

 

P.S. Debussy Preludes by Enrique Bernaldo de Quirós was recorded before we finished building our setup. Calibration v2.0 has been developed specifically for the current setup (working since April First 2013). Calibration v2.0 cannot be applied to that album. I took this album out of the list for this particular reason.

[Mario Martinez]

Are the tracks newly recorded with 2.0 calibration or they are reworked from previous recordings just like the 1.1 version? If the latter, I would like to have the third track of the drum set solo. It has been my new reference for a while.

 

Hi Francis,

 

What we have done is develop a new calibration on the gear side of things (the physical setup remains the same) We have run that calibration on all our previous recordings (except the Debussy Preludes because that was recorded long ago with a different physical setup)

 

So yes, in your words, they are reworked. but this time, instead of trying to fix calibration v1.0 we have developed a new calibration from scratch.

 

I am posting the albums first. Once everyone has listen to them I will be posting the drums, flamenco and rock. This time we will be hearing all the albums, the drums, the flamenco, and the rock with the exact same calibration. If the quality of the sound is consistent across all audios (which did not happen the first time) then we will know we are at the transparency point.

[Mario Martinez]

I am sorry to hear about this problem with the downloads.

 

Some people have already reported some issues. I do not know if this will help everybody, but this is what we did to solve them:

 

Firefox allows you to resume the download if it stops for any reason (probably Chrome or Safari do that to, but I just have not used them).

 

The size of the Albéniz Master is about 1.5Gb and the size of the Cabrera Master is about 1.0Gb. The files are quite big, so they take a long time to download. If the connection stops at any time you do not have to restart the download. You can resume it right at the point where it stopped. All you have to do is click on "Downloads" under the "Tools" menu and resume the download.

 

I hope that helps...

[Mario Martinez]

The 2.0 Iberia.

It is crispier, cleaner and more lively. A slight veil in the 1.1 version that I did not notice has been removed. It is louder now and I need to reduce -2db in gain compared with the 1.1. The piano is now "perceived" more realistically in the centre of the soundstage.

If I am asked to demo a piano track, this is the one.

Thank you Mario.

 

Thank you Francis for your wonderful feedback!

 

I really like your description about that "slight veil". It is amazing but it always works that way.

 

We have spend the last three years working on different calibrations. With every calibration we were getting closer and closer to transparency. Funny thing is, you never noticed that "slight veil" until you got to the next version.

 

The only thing that kept us going was having a setup that allowed us to directly compare the sound of the latest version with the sound of the live source. You could hear there was a difference and that difference got smaller and smaller with every version.

 

By the time we were at v1.0 the difference was so small we thought we could not go any further. So we published what we had thinking it was the best you could possible get.

 

It was only after this Computer Audiophile experience that I took the courage to try it one more time. And it worked, those differences are now gone. I still cannot believe we got here

[Mario Martinez]

Must be me, but I can't get this to work. When trying to open in Firefox, it takes me to the old version, and trying to copy the path name into a download manager does not work either. I guess I have to pass on this.

 

 

Hi Dieter,

 

I sent you a PM with an alternative way. Hope that works...

[Mario Martinez]

It's not only the clarity and detail, but also the balances across the frequency spectrum from bass to treble that seem more convincing than in v 1.0. When 1.0 was indeed excellent, 2.0 is simply stunning.

 

i am not sure what is meant by the 'calibration' that you have posted about, but whatever it entails, you seem to have perfected it to an extent that is quite uncommonly found in commercially available piano recordings.

 

 

Hi Guido and thank you very much for taking the time to listen. I really appreciate your feedback!

 

I am glad you brought out this balance improvement. The purpose of our calibration procedure is to provide the recording setup with a flat frequency response. The balance is now better because it is more accurate to what the instrument actually does.

 

 

Now, if you could also achieve the same results in other instrumental settings, say a string quartet or a jazz piano trio . . .

 

We have not gotten to those yet but we will and I will let you know as soon as I have something

 

The other thing that we do have is the flamenco, drums and rock sound tests.

 

We heard version 1.0 of those sound tests on our previous thread PlayClassics master file giveaway for CA members The flamenco was good, but when we got to the drums and rock it just did not make it. Version 1.0 was not completely transparent, and it did show more on those samples.

 

But here we are again. We have a new calibration that has shown to be more transparent (I would say totally transparent) and I think that should really make a difference on how those samples will sound.

 

I would like for us to listen to those samples. If the sound is consistent with what we are hearing on this classical albums then I think we could safely say we have reached transparency.

 

Who wants to try?

[Mario Martinez]

Mario,

 

Just checking if the same layout has been used for the 2.0 recordings.

From your post in the earlier 1.0 thread:

 

We did not re-record the albums. The recording are the same (so the physical setup is the one that was used to begin with)

 

What we have done is develop a new calibration on the gear side of things (the physical setup remains the same) We have run that calibration on all our previous recordings (except the Debussy Preludes because that was recorded long ago with a different physical setup)

[Mario Martinez]

I would like to explain something about the Drums samples. This might only affect people who are using large stereo setups to reproduce these recordings (by large I mean the distance from the listening point to the speakers)

 

You might find that when you playback the Drums samples on your systems, the image of the Drums set is bigger than it should be in relation to the playback image of the piano.

 

I have taken the time to draw some more pictures see if I could explain why this is happening

 

 

Here is a sketch of how all our piano recordings are made. The picture on the left corresponds to the recording setup. The picture on the right would be the image produce on your systems:

As you can see, the recording setup is built so that we can achieve a realistic size image during playback.

 

 

Now, here is a sketch of how we should have recorded the drums:

Had we done it this way, the playback image of the drums would have had a realistic size too (at least compared to the playback image on the piano recordings)

 

 

But this is what we did:

We were lazy enough not to move the piano out of the stage (guess we should have) so we ended up recording the drums closer to the mic than they should have been. That is the reason for the playback image of the Drums samples being a little bigger than it should (specially if you compare it to the piano recordings)

 

I hope this makes sense

[Mario Martinez]

Here is some more useful information about the image of both, the piano and the drums:

 

There is nothing wrong with the images other than the fact that they are out of scale with each other. If you wanted to "see" a playback image of real size all you would have to do is move the speakers to match the scale of the recording setup.

 

Here is a sketch of the piano setup. To reproduce a real size image of the piano the speakers should be placed 11,365' apart. (If you played the drums here then the drums image would be too big)

 

 

Here is a sketch of the drums setup. To reproduce a real size image of the drums the speakers should be placed 7,456' apart. (If you played the piano here then the piano image would be too small)

 

 

We could have easily kept the same scale in both recordings by placing the drums in the same place as the piano. That way both images would have been "in sync" with each other no matter what the size of your system was.

[Mario Martinez]

This is really interesting info. My speakers are approx this distance from each other.

 

 

So does it work? Do the drums produce a real size image on your system?

 

 

Our monitors are 11.365' apart so I really have not had a chance to try it myself...

[Mario Martinez]

Hah, I forgot about commas versus decimal points momentarily and was wondering why I would need to have my speakers miles apart!

 

Mario, very nice indeed to see you back here with these files of very beautiful music. Though I'm traveling and can only listen through headphones for a while, I would very much appreciate the code for the new Albeniz Iberia and Cabrera Debussy.

 

Thanks!

 

 

Thanks Jud, its nice to see you too.

 

I sent you a PM with everything, the two albums and the experimental samples.

 

I cannot wait until you are able to hear this on your regular setup

[Mario Martinez]

The drums imaged beautifully as per your description

Having information like this really makes listening enjoyable.

 

My speakers are set 7 ft apart.

 

After hearing the drum test tracks the Rock track was a little underwhelming.

The drums in the rock track are muted in comparison and seem somewhat restrained.

The musicians also seem to be a little tentative at the beginning before settling in later.

Rock may not be suited for this setup in my opinion.

 

 

Hi Nikhil,

 

thanks so much for the feedback

 

I understand what you are saying about the rock sample…

 

Please keep in mind that this kind of experiment has never been done before. Yes, there might be some rock recordings that have been made without the use of close miking, but even if they were recorded using only two microphones, they were mixed/mastered to try to achieve the particular rock sound that the engineer/producer had in mind. This mixing/mastering process is not a choice, it is a must. When you record any instrument on any room, hall or studio, the sound of the take is always deformed by the acoustics of that particular venue. You need to mix/master it to make it sound "good". Plus you also have to take into account the fact that, if they were recording an album, the musicians probably spent countless hours trying different settings in their instruments to get the best possible sound out of them.

 

Here we have something different. We are not mixing or mastering anything. What you are hearing is just the "true" sound of the instruments. Getting our setup to produce that "true" sound is what has taken us 8 years of research. The hall has been worked during years of trail and error to try to minimize all the acoustic problems and the calibration has been developed specifically for that hall to take care of whatever acoustic problems were left. That way, when you record any instruments in this setup, you know that what you are hearing on the recording is the "true" sound of the instrument without any coloration or deformation. This does not mean that the sound on the recording is going to be "pretty" it just means that it is the sound that the musicians produced (by themselves without the intervention of any engineer, producer…) That should justify the fact that this particular sample is never going to sound like the "rock" we all have in mind.

 

I also have to say that the musicians actually did a great job. They are not recording artists. They do live gigs. They are used to having an engineer construct their sound at the mixing table. When I told them we were going to record this way (that means that they were going to be completely exposed without the help of any touch up) they had the courage to do it and I am really grateful for that.

[Mario Martinez]

I have now A/B compared 5 tracks with 2.0 calibration to their predecessors. I would say the 2.0 process is a total success bringing improvements in varying degrees of improvement to the 5 that I have compared. I have described three in previous posts, namely, the soprano, the tenor and Iberia piano tracks. Now the remaining two.

*The drum solo, track 3. The overall sound of the 2.0 has a little more gain and is more transparent. The kick drum has more impact and the drum head sound is clearer now. The higher gain and transparency of the snare and cymbals make them sound more balanced with the kick drum (previously they were subdued and duller in comparison). I am even able to hear the mechanical noise of the high hat and occasional ringing of the cymbals. A revelation! Those who have ribbon and diamond tweeters are likely to get what I have described. This track has usurped the Sheffield drum track that has reigned for a long time in my playlist. The Sheffield one was apparently recorded with multi microphones with cymbals sounding too widely distanced. The 2.0 recording with two microphones has the drum set pieces grouped together like a real one. The only problem is that it was recorded from the back of the drummer and the cymbals still sound more subdued than if they were placed in front nearer to the microphones (and to the listener when reproduced in an audio system).

* the flamenco track 2. I have archived the previous version because in my system the singer was too soft in comparison with the guitar. In this regard, the singer is positioned aback and the guitar accompaniment sounded like he was the soloist in focus.

The 2.0 calibration has brought about a little more gain and transparency and the singer seems to be standing slightly forward now. If someone asks me to play a flamenco guitar track, this is the one I would use.

 

 

Thank you Francis for your very detailed feedback.

 

I am really happy to hear about the improvements on all five pieces.

 

This is great news for us!

[Mario Martinez]

I do not understand how and what you actually do to compare the 1.1 or 2.0 versions with the LIVE SOURCE. Anyway I do not need to know because I am not making any recording. You know how to improve step by step is suffice for users like me.

 

i am not sure what is meant by the 'calibration' that you have posted about, but whatever it entails, you seem to have perfected it to an extent that is quite uncommonly found in commercially available piano recordings. Now, if you could also achieve the same results in other instrumental settings, say a string quartet or a jazz piano trio . . .

 

Can you explain this a little more? The reason I ask is that I am struggling to find differences between the two "Cabrera Plays Debussy" versions. I have to qualify that my setup is stand mount speakers that do not go below 47Hz. If anything I am finding the v1.0 version to be a tad more open.

 

See if this makes sense:

 

Say we built a full range instrument designed to PRODUCE the same color on everyone note. If you allowed that instrument to vibrate freely (without the constrains of any walls around it) you would be able to HEAR that same color on every note. (Let's name this sound "A")

 

Now imagine you take that instrument into a room. The walls on that room will constrain the way the instrument vibrates. Depending on where on that room you place the instrument the distances from the different walls will affect each note in a different way. So the color of the instrument will not be homogeneous any more. It will now PRODUCE a slightly different color on every note. (Let's name this sound "B").

 

If you want to listen to it you will have to place yourself somewhere within that room. Depending on where on that room you place yourself the distances from the different walls will deform the sound of each note in a different way. This color deformation will not be homogeneous. For each note, you will now HEAR a different color from the one the instrument produced. (Let's name this sound "C")

 

But it does not stop there, because you are not listening live, we are making a recording. So instead of your ears, we are using two microphones and some gear to record that sound. Depending on your choice of mics, mic arrangement and gear the recording chain will also deform the sound. This color deformation will not be homogeneous either. So for each note, you will RECORD a different color from the one your ears would have heard. (Let's call this sound "D")

 

So, what are we doing? Our project is about getting as close to "A" as possible. We worked the hall to minimize the differences between "A", "B" and "C". Then we developed a calibration to took care of the rest.

 

How do we know if we actually got to "A"? We are using our setup to record voice, piano, guitar and drums. These instruments are extremely different from each other. Yet, we are recording them all the same way. They could have all been recorded on the same take. The fact that the sound quality is transparent across all those instruments means we have been successful on achieving our goal of actually recording that "A" sound (call it "true" sound, "absolute" sound… you name it)

 

So if you are comparing versions, you should focus your attention on the color of the instruments. I do not know if it helps, but I like the way Guido explained it:

 

It's not only the clarity and detail, but also the balances across the frequency spectrum from bass to treble that seem more convincing than in v 1.0. When 1.0 was indeed excellent, 2.0 is simply stunning.

[Mario Martinez]

[ATTACH=CONFIG]28686[/ATTACH][ATTACH=CONFIG]28687[/ATTACH]

Now having listened to the Iberia 1.0 and 2.0 I can say:

- you notice the volume increase by 2 dB

- I have attached screenshots of the first track of the Iberia analysed by JRiver

In practical terms you sense a bigger soundstage with v2.0 compared to v1.0 while retaining the level of detail and ease of listening. The performance itself is excellent too. After several switching between the two options it was easy to stay with the new v2.0 as my preferred choice. Therefore the new calibration is definitely successful .

 

Hi Wouter,

 

thanks so much for listening

 

Yes, that is very interesting about the soundstage. You can hear that here in the studio too. The previous version had some issues on some frequencies. It is very interesting to see that just fixing those issues affects that sense of soundstage.

[Mario Martinez]

We have had some more reports of people having problems with the downloads.

 

I have reported this issues to our hosting service to see if they could fix it.

 

As a backup plan, I have uploaded the Albeniz Album to Dropbox.

 

If anybody else is having trouble with the download please tell me and I will send you a direct link to download the Dropbox file.

 

I am really sorry for the inconvenience this is causing.

[Mario Martinez]

Thanks so much Mario!

Great to hear the new versions.

 

Sadly, I have limited equipment on hand here, but could nonetheless hear an improvement. Couldn't say much beyond that timbres seemed better. The few samples of Iberia and Debussy I played were all more enjoyable with the new calibrations (albeit playing in crappy old Sennheisers).

I will be happy to post more impressions once I get my decent headphones in a couple weeks.

 

Thanks Luke,

 

well, that is good news so far!

 

I cannot wait until you get the good ones

[Mario Martinez]

Hi Mario,

 

Downloading for me was a total no brainer...

 

I've listened to your Albeniz recording and samples and did found them very enjoyable.

My office set up is not adjusted as per your suggestion. The sound was crystal clear, smooth.

Should I say very precise, detailed ? Yes.

Too much? not exactly, but I've felt like some life was missing. Probably due to focus and lack of distance between my B&W speakers.

 

I will try on my other setup asap to check difference with a more usual and valid configuration for a more pragmatic assessment.

 

Thank you very much

 

We use an ORTF mic arrangement. That means the image that you see on playback is formed by volume differences and timing differences.

 

If the arrangement of the stereo playback system is very precise (the distances between the speakers and the listening point) then the ORTF offers a perfect image, but if the distances are not correct, then the timing differences will not make sense any more and that will not just affect the image but it will also affect other aspects of the sound too.

 

If you do have a chance to play them on a "regular" stereo configuration please do try. This recordings are very sensitive to the stereo arrangement of the playback system.

[Mario Martinez]

Here is how we are doing so far:

 

We have given out 54 albums. 25 for Albeniz Iberia, 23 for Cabrera plays Debussy, 3 for Chopin Polish Songs and 3 for Songs of Paolo Tosti.

 

In addition to the Gift Codes we have also received some donations; I want to tell you I am really grateful for that too.

 

Please feel free to post your impressions. Should you have any questions about the recordings, the web page, ...anything, please do ask, I will be happy to answer all your questions.

[Mario Martinez]

Some reference recordings try to sound overtly ‘hi-fi’. This one does not. The piano sounds superb, but the more one listens, the less one focuses on the instrument. One appreciates the insight ...into the music and the musicianship. Well done.

 

Thank you, ...what else could we hope for

 

this is from the about on the PlayClassics page:

Music is communication.

It is communication between the artist and the listener. The more obstacles there are between the artist and the listener the more the message will deform and the worse the communication will be.[/Quote]

 

All we want to do is provide that direct communication with a natural sound, so you can enjoy the artists!

[Mario Martinez]

Here is an SPL file you can use to calibrate the volume of your system to the volume of the actual performance.

 

You can download it on any of our regular formats (Master, DVD, CD, MP3)

 

While playing this file in stereo, adjust the volume of your system until your SPL meter reads 75dB at your listening point.

 

All our recordings (except for the drums and rock) are recorded with the same level, so once you calibrate your playback system you will be hearing all the recordings at real live level.

 

The drums and rock are 24dB louder than this. If you wanted to get real live levels of the drums and rock you would have to turn the volume up by 24dB. But I would not recommend trying to achieve that volume on your systems for the safety of your ears.

 

Here are the links to the files:

 

TRTv2.0.master_SPL75dB

TRTv2.0.dvd_SPL75dB

TRTv2.0.cd_SPL75dB

TRTv2.0.mp3_SPL75dB

 

Have fun

[Mario Martinez]

We are happy to announce we have a new release on our catalogue:

 

"Conversations" by Enrique Bernaldo de Quirós, piano.

 

This album contains works by Beethoven, Schubert and Liszt.

 

I hope you do find it enjoyable.

[Mario Martinez]

We have upgraded our server. Downloads are now faster than before. The performance of the previous server was very much location dependent. Some people reported really long download times. That should not be a problem any more.

 

Thank you all for your support, it is really gratifying to know you are enjoying the music

[Mario Martinez]

Hi everyone,

 

yes, we are still giving out two albums for each Computer Audiophile member.

 

All you have to do is ask