51

Re: A new Italian temperament: CHas

Hm...I'll have to test that out some more. I didn't try going back to paste into text. I do remember making a change of about a half cent and seeing the yellow bar leap up a tab on the interface, but as you say, it may just be for the eyes.

But I think I hear a difference, even if just moving the sliders physically up or down to say .5 cents. Many of those Victorian well tunings that I worked on earlier involved pitch changes that were less than a cent--I just estimated the amount visually.

We can try it out by taking two notes, say an E3 and an E4 and tuning\typing one up .9 cents and the other down .9 cents, and then pasting that tuning in. Surely we'll be able to hear the difference on an octave if it changes. In the middle of something right now. If you get the chance to try it out, let us know...

52

Re: A new Italian temperament: CHas

You know, I always relate that to my tuner's day. To tune I listen to an interval and regulate its beat rate, comparing it with the beat rate of the neighbor (now that I adopted ALfredo's method, but before I was comparing intervals at different ranges, looking for acceleration or identical beat rates - between intervals that where different, and only have a similar partial match - that is way harder to be precise than comparing intervals in sequence)

Then I may tune with a slider in Pianoteq, (or I may use a formula and hope for the best)

The parameters that are not taken in account in Chas are the ear behavior (asking for higher pitches in the treble, hearing better in the medium than there)
and the room acoustic, (that delay the pitch and make the tuner add stretch,  for instance in a church or very reverberant place, by listening to the sound coming back after reflection and regulating the pitch on that tone reflexion ).

But I feel that the high treble can be stretched a little more if wanted (at the expense of resonance) .

Or may be the whole treble tuning can be done in that hearing mode, simply that may be tiring.

Last edited by OLEK (01-03-2010 19:49)

53

Re: A new Italian temperament: CHas

Jake Johnson wrote:

We can try it out by taking two notes, say an E3 and an E4 and tuning\typing one up .9 cents and the other down .9 cents, and then pasting that tuning in. Surely we'll be able to hear the difference on an octave if it changes. In the middle of something right now. If you get the chance to try it out, let us know...

You will better hear it if you listen to fast beating intervals, or even to the 5ths, the octaves are highly misleading because of many levels of coincident partials (2,4, 6, 8, 10,12)

look there for beat rates samples :
http://home.broadpark.no/~rbrekne/beats.html

Do you know how to listen ONLY to the beat rate, by "ghosting" ? works perfectly in Pianoteq, you even have the tonal pedal if the interval is large. that is a way to listen only at the partial match level. (I say, good for study !)

THAT goodie is also a perfect learning tool to understand where are the beats : http://www.colemantools.com/www.coleman … ndex.shtml only 2 paper strips with the partial serie located at the back of the keys, one on the lower note, one on the top note, the partial matches are seen immediately.

(sold 8 USD, and old )

Last edited by OLEK (01-03-2010 19:59)

54

Re: A new Italian temperament: CHas

OLEK wrote:

Do you know how to listen ONLY to the beat rate, by "ghosting" ? works perfectly in Pianoteq, you even have the tonal pedal if the interval is large. that is a way to listen only at the partial match level. (I say, good for study !)

I don't.  I've seen a little information about it, but assumed that it was over my head, since I can't really hear exact beat rates when listening to notes. I hear beats, but counting the rates appears to beyond me. Any suggestions or links to information about ghosting and listening for beat rates? (I DO understand what ghosting is. I just don't understand its relation to hearing beats.)

Last edited by Jake Johnson (01-03-2010 20:46)

55

Re: A new Italian temperament: CHas

Jake Johnson wrote:
OLEK wrote:

Do you know how to listen ONLY to the beat rate, by "ghosting" ? works perfectly in Pianoteq, you even have the tonal pedal if the interval is large. that is a way to listen only at the partial match level. (I say, good for study !)

I don't.  I've seen a little information about it, but assumed that it was over my head, since I can't really hear exact beat rates when listening to notes. I hear beats, but counting the rates appears to beyond me. Any suggestions or links to information about ghosting and listening for beat rates? (I DO understand what ghosting is. I just don't understand its relation to hearing beats.)

To count beat rates they are related to a second, so the tuners learn to have that second time interval in mind.
Then it is not more difficult than counting the measure in music, you of course have to evaluate the beat speed, then they are compared with similar beat speed a step above (same interval)  or with beats speed produced at the same level but form another  interval.
This may be fun for you as you work from a tuned piano.

For instance : to listen to the M3
beat speed, wee look at the partial serie of the 2 notes and we see that the first coincident partial is at  5:4 (you get it ?) and a second at 10:8 - an octave above. if you release the "wire" from the dampers on the M3 (simply pushing the keys without sounding the notes,, or wait for the tone to stop) then you struck the note at the partial level, you excite that partial match for the 2 notes and the result is (very clearly in pianoteq) the beat only, not mixed with the fundamentals. It helps to now where are the beats, to understand where to listen for, and to hear them more clearly.

The 3d beats are not the cleaner ones usually.

If you want to appreciate if a 5th is tempered and how much (kind of half a beat second for instance) you can listen only to the beat , striking a note one octave above the top note (3:2 relation).

Or you can compare the beat speed obtained when opening a note a tenth under the top note, and a Msixth under the lower note : those 2 intervals have a beat (fast) produced at the same level than the 3:2 5Ths, it is said to be easier to compare the beat speed of the 6th-10th to ascertain the one of the 5th, than to listen directly to the slow beat of the 5th (it is faster done also, no need to wait a few seconds for the beat to develop) in the end most tuners forget how to listen to the slow beats and they are then fighting with iH all along the tuning, the 5th speed progression being only ascertain with the comparison of fast intervals, which is in fact easily misleading , because comparing 2 different intervals with a different harmonic content.

I'll surely find links that explain that better than me ...





Edited ; sorry for the false partial match on the M3

Last edited by OLEK (03-03-2010 23:10)

56

Re: A new Italian temperament: CHas

Olek: Thanks. You've just opened up a lot for me. (And given me several hours of homework...)

Gilles: You're right. It seems to round the numbers off when you copy it back out of the Detuning pane. If I detune one note in the text file to -7.5, and another to -7.0 and copy the tuning to the Detune pane, the difference between the two is reflected in the yellow bars, but going the other direction, both notes are tuned to -7 when I paste back into the text file. A bug?

In any event, after going back and listening some more, I think I do hear the difference with copied changes under a cent. Play C2 and C3 together for a few seconds. Now play the G between them and let them all decay. Try flatting C2 less than a cent and sharping C3 under a cent. Now play the G. Reverse the pitching of the C2 and C3, so that it's the C2 that is sharper, etc. Something going on there with the beats. (I think I hear more of a difference between the default pitches and either variation than I hear between the two variations. In other words, regardless of which C is raised and which is lowered, the effect on the G is audible.)

Of course the difference in additional pitch change between C2 and C3 may total more than a cent (raising one .5 cents and lowering the other .7 cents, for example), but that's why I hope the information to the right of the decimal places is being copied--it's the difference between the width of the intervals, and how that lines up partials\beats, not just the cent change per note, that makes a difference. A change of about .8 changes the 4th partial by a herz, and some partials above it more. Bad that it rounds off. (Don't beat me up too bad, Philippe.)

I need to reread Olek's post about ghosting note and listening to beats. It's going to take me some time to understand.

Last edited by Jake Johnson (02-03-2010 01:22)

57

Re: A new Italian temperament: CHas

OLEK wrote:

If something misses it may be at the "phase" level, which may be certainly very difficult to modeling , even to analyze, and may be is even partially lost in the sampling rate in any case (be it on samples or on models, as to me it involve instant changes in the direction the sound radiates (?) ). I heard that Yamaha installed vibration goodies to provide tactile sensations within the keys, on their last model, may be they give a more realistic reproduction of the phase effects (?).

I'm not sure what you mean by the phase level. You might want to try moving the mics around a little to hear the effect. But you mean the phasing of the string motion? (Not sure I understand what that has to do with the tactile sensations in the keys, though...???)

58

Re: A new Italian temperament: CHas

Jake Johnson wrote:
OLEK wrote:

If something misses it may be at the "phase" level, which may be certainly very difficult to modeling , even to analyze, and may be is even partially lost in the sampling rate in any case (be it on samples or on models, as to me it involve instant changes in the direction the sound radiates (?) ). I heard that Yamaha installed vibration goodies to provide tactile sensations within the keys, on their last model, may be they give a more realistic reproduction of the phase effects (?).

I'm not sure what you mean by the phase level. You might want to try moving the mics around a little to hear the effect. But you mean the phasing of the string motion? (Not sure I understand what that has to do with the tactile sensations in the keys, though...???)


I really dont know myself, I just like to understand why the attack and the life of the tone lack that crispness and air that exists on the real piano.
I think of the phase because the way the strings couple is evolving they change orientation during the sustain, they also have different time between the initial movement and a first stabilization of the waves.

All those waves are reflected in the case and felt within the keys.
There is also standing waves that are filtered by the action and hammer/shank resonant frequencies, and that are delivered to the string at the moment of the attack.

As it is now possible to reproduce stereo effects with only one speaker, I believe that it is with a phase treatment that the tone can be made more realistic.

Depending of the voicing, of the tuning (unison particularly) a piano tone can go toward the ceiling of fulfill the room with lot of projection.

A tuner can even feel the tone traveling along the wire, even if it is a somehow psychoacoustical effect, it is real, I guess it depends of the way the energy provided by the hammer/action is used.

Thoses different effects may well have their source somewhere. I only see the coupling (phase) variations (orientation and timing) between the different elements to be the cause .

We all now those days that a pianos is not just a soundboard, a plate and strings, but that all the elements are concurring to the tone. Factories make test to listen to the tone radiated by the keybed, the strenght of tone and the voicing quality is felt under the finger in the keys.
When modeling, the appreciation of what makes the tone, may leave some parameters aside, I suppose, because they can't be accessed with simple enough rules, or they cant be recorded efficiently.

Last edited by OLEK (02-03-2010 08:47)

59

Re: A new Italian temperament: CHas

Hello all,

may be I have bad news for you.

Seem to me that the scl chas file have only one effect :
the frequency correction due to iH is not taken in account , so A= 440, the A5 = 880 and so on.

I have seen that by experimenting with Tunelab : indeed the higher notes as A
are alittle higher with the scl file that with "all notes flat" file, but not by much (may be it is normal, in fact)

But the A3  is at 220 hz, A5 880 Hz may be the display is not precise enough, I will check again, indeed it only may be 1661 Hz at A6 level so it is possible that it is not seen so easely, but I doubt)

Any deviation you see may be due to the iH of the notes themselves, (that one is raising a lot the partials)  the more just sensation due to that : more justness, more lining of the fundamental frequencies with the perfect theoretical spectra.

Anyone can check that better than me please ?

I suggest, as Philippe told us yet, that a 88 notes scl file may be neccessary to be sure.

the 0.49 cts deviation at A5 may show. it does not.

Thanks

Thanks Jake for noticing !
OLek

Last edited by OLEK (03-03-2010 16:21)

60

Re: A new Italian temperament: CHas

(All: remember that C3 = middle C for Pianoeq. Olek is writing in terms of C4 = middle C. Shame on you Olek.)

Last edited by Jake Johnson (03-03-2010 18:59)

61

Re: A new Italian temperament: CHas

I would like to summarize my current thoughts regarding CHAS tuning.

First 12, 24 or 88 notes scala files give the same result. I thought there was a difference because varying keyboard velocities give different audible results. Using a fixed velocity now shows no difference. Displayed frequencies in the Spectrum Profile are also the same.

Second, these displayed frequencies are very deceiving since they are limited to integer values and are off by several hertz as we go up the scale as compared to the theorical flat tuning. On top of that, the high displayed frequencies (see for example A6)  are different for different presets using the flat scale...(Maybe there is some included stretch in the presets' analyzed samples that prevents a perfect theorically flat scale).

It's difficult to know if this is just to display in these Note Edit panes values that are cleaner to look at, or if this really limits the feasible changes to things that are audible, meaning that no unnecessary precision may be given by the user. This is perfectly correct with the assumption that integer cents or integer dB increases are enough for audible changes of single notes.

But the CHAS tuning works because very subtle changes are made that impact the combination of notes in chords by tempering intervals differently so partials interact better (see Alfredo Capurso's papers for details).

A 12 note scala file approximates well this effect to the ear in my opinion, but the exact theorical values given in the paper cannot be implemented since scala files are based on C instead of A and can't be stretched downward by subtracting because Scala doesn't allow negative values before the implied 0.0 which is allocated to C. If there is a way to change the reference from C to A, I didn't find it.

I thought maybe aural tuning using also a standard stretch and the Detune pane to correct the differences might be useful but now in light of all these previous findings I doubt it.

So the only way to try out the theorical CHAS tuning correctly would be to be able to give the 88 exact floating point note frequencies as input to pianoteq instead of a scala file. These frequencies are easily obtained with the published semitone value of 100.038318440222 cents.

But then maybe such a fixed table of frequencies doesn't exist in pianoteq and everything is the result of other computations so this would be impossible. In that case I guess the current approximation with a scala file, with or without further stretching is the best that can be done.

Last edited by Gilles (03-03-2010 21:08)

62

Re: A new Italian temperament: CHas

Gilles, I agree with you, a good way to implement a very precise temperament is to specify the 88 notes frequency values in a scala file. I paste here a recent post from the piano world forum (piano tuner-technicians section). Remember that if the last note in the SCALA file has a value different from 1200.0 (or 2/1), then the inharmonicity will be by-passed and the frequencies will be exactly those specified in the SCALA file.

Here is how to proceed after creating your own SCALA file (below an example of an 88 notes SCALA file content, remember that the first note is always omitted):
- tuning section, click on the "mu" microtuning button
- import the SCALA file via the "scale" menu
- choose "88 notes scale" in the keymap menu

Example of flat tuning 88 SCALA file, copy paste the content between the dotted lines and name the file allnotesflat.scl:
-------------------------------------------------------------
! allnotesflat.scl
!
all notes flat temperament
87
!
100.0
200.0
300.0
400.0
500.0
600.0
700.0
800.0
900.0
1000.0
1100.0
1200.0
1300.0
1400.0
1500.0
1600.0
1700.0
1800.0
1900.0
2000.0
2100.0
2200.0
2300.0
2400.0
2500.0
2600.0
2700.0
2800.0
2900.0
3000.0
3100.0
3200.0
3300.0
3400.0
3500.0
3600.0
3700.0
3800.0
3900.0
4000.0
4100.0
4200.0
4300.0
4400.0
4500.0
4600.0
4700.0
4800.0
4900.0
5000.0
5100.0
5200.0
5300.0
5400.0
5500.0
5600.0
5700.0
5800.0
5900.0
6000.0
6100.0
6200.0
6300.0
6400.0
6500.0
6600.0
6700.0
6800.0
6900.0
7000.0
7100.0
7200.0
7300.0
7400.0
7500.0
7600.0
7700.0
7800.0
7900.0
8000.0
8100.0
8200.0
8300.0
8400.0
8500.0
8600.0
8700.0

-------------------------------------------------------------

63

Re: A new Italian temperament: CHas

But Philippe, what we may enter here are are well  deviations in cts from the theoretical values ? or a frequency ?

If a frequency wher is A 440 ?

If a cts value departing from 100-200-300 etc, the bottom notes may be less than 100 and the top ones more, to represent the stretch of Chas (that stretch may add, by evidence.

I've find that the central octave used in Chas is  beating less fast than with a standard tuning, may be that allow for very little stretch in the medium range, and the inversion of fifths (the top one being larger than the bottom one in the "temperament octave") protect the harmony (fifths that go beating more and more while raising in the treble provide a very different kind of harmony, and this is fight by the octave stretch, while in Chas the ratio being coherent the stretch is installed naturally.

I will try to get to something with Tunelab. A good tool to check or tune Pianotq by the way.

In Tunelab, iH constants can be entered manually (Thanks Philippe for the values).
And curves showing the partial relation are easy to see. the stretch can be modified and frequency values for all notes seen at the same time on the screen

Only problem is still that a temperament concept remain computed within a 2:1 octave in some zone, as it may be a necessity for computations. (the starting note can be changed in case of an uneven temperament, but using an identical ratio on each note dont provide nothing,

May be a progressive ratio would but then the octave above the temperament may be tweaked so the progression follows well

That is something to experiment, as we can consider Chas as an uneven temperament, if compared to a 2:1 octave the last semi tone is too short. I am unsure of the numbers to use (cts addition most probably)

Last edited by OLEK (04-03-2010 00:38)

64

Re: A new Italian temperament: CHas

I forget to say, for practical purpose, you can decide that you want only the fundamental to drive the display in Tunelab ; it works relatively low in the scale, after that too much for the sound card. The reading is less precise when in the bass, but it is better as when you see the hz value in the screen that is the one of the fundamental. (if not partials are displayed, as they are more prominent in tone, and they are more used to tune octaves when descending under a3 for instance)

If you want to see the iH level you can simply look at the level of a partial of the note played.

And if a basic stretch is possible to describe Chas , then recording or entering the iH constants for as many notes you want will correct the iH in the same way that Pianoteq is doing, making a stretch curve that only allow at last the octave to be really 2:1 (I dont know if other partial relations are used in the iH correction for Tunelab, but there is not real reason, as the tuning stretch (artificial stretch) is added on the top of that.
(but may be there is a somewhat "standard tuning model that implies more than one octave, I'll ask about that)

Sorry to be out of topic, while it is not really ...

PS Jake, I apologize the basic partial relation for M3 is 5:4 (not 8:6 !) a 17th above the bottom note. another beat occurs one octave above (and in Pianoteq I also hear a beat at the 10 th level which I dont know what is the real cause may be a substraction from the 2 others ? (?)

Last edited by OLEK (04-03-2010 01:04)

65

Re: A new Italian temperament: CHas

Olek,

In Philippe's template, you would overwrite each of the sample freqs that Philippe entered with those read by Tunelab (where the fundamental was read). Scala lets you enter either herz or cent values or cent values that are deviations from ET. Scala also lets you enter the note numbers, to keep track of which freq is on which note. (I think you just enter the text to the left of the freqs and then tell Scala to "Hide the first column." There's a button for that.)

The advantage of this approach is that Scala lets one enter very precise freqs.
The disadvantage, of course, is that Tunelab can't read all of the notes on the fundamental.

About Tunelab and other etd's using upper partials:

Since we can roughly calculate the iH of the strings, would it be possible to work backwards from the upper partial that is read? If we know the iH of a string and the freq of its 6th partial, can we come close to finding the fundamental? Or would the uncertainty about the exact iH of each string make the margin of error too wide?

Last edited by Jake Johnson (04-03-2010 02:10)

66

Re: A new Italian temperament: CHas

Jake Johnson wrote:

About Tunelab and other EDT's using upper partials:

Since we can roughly calculate the iH of the strings, would it be possible to work backwards from the upper partial that is read? If we know the iH of a string and the freq of its 6th partial, can we come close to finding the fundamental? Or would the uncertainty about the exact iH of each string make the margin of error too wide?

Jake you are ready to make a tuning software !

THis is what is computed by all EDT's, and in a sense, by Moddart  when building and Instrument.

Look at the "edit iH constants" window in Tunelab, you will have those readings, if some samples have been recorded, or none if you did not. Then you can enter 0.66 and , for instance, and look at the difference of pitch provided higher and lower of A4 (note49)

In the usual mode of display, under A4,the partial used is the 2nd, then the Third, then in the low bass it may be anyone from the 3 to the 12eme, as long as it relate to the 2 notes '(top note and bottom note of the interval tuned, an octave generally)
This is to allow more precision and to focus on the same kind of octaves (same size) that the ones traditionally used in tuning . ( octaves and intevals are named by their partial relation to have an idea of their "size in regard of the ih)

The higher partial relations are used to provide an "opening" of the octave.  Then the fast beating progression of intervals is slowed, and it goes in the direction of what the human ear is asking, a higher treble (for the bass the lowering is due to the iH, so the partials of the bass notes relate well to the medium of the keyboard).
Many tuners tend to focus on one sort of octave by region of the keyboard, " listening at different partial levels and changing that partial match formula when descending/raising thru the scale. This is a well know method to have a little control on the stretch. Is it really accurate ? I would not say so, as what is the sign of a good progression is the quiet evolving of the speed of the RBI (Rapid beating intervals) such as the 3ds the 6Ths and all them with octaves added inside (17th 24 th..)


The  sampling of iH is done in any case prior to a tuning, (to use an EDT) I did not mean that Tunelab cannot be used in the Bass in fact it may work fine with another partial displayed, one which is in the audible range of the sound card.

I've find it useful to have only fundamentals, essentially to have a fast view of what is heard or what is tuned, as it allow to see the real frequencies of the note at the screen, computed or tuned.

On my Pocket PC the fundamental may be heard down to D2 73 Hz (while singing I cant sing lower ! on a piano I recall it goes a little lower )

I f I where to tune a piano with Tunelab I would use the proposed partial set, for a higher precision in principle .

Last edited by OLEK (04-03-2010 11:22)

67

Re: A new Italian temperament: CHas

After sampling notes in the M3 instrument, A49 (440)  find 0.66 iH constant, then

the A5 (with the scl file loaded hence no IH gestion) is 881.5 Hz (before the iH reading it was 880 Hz of course).

This is a very wide opening, to me more than the basic iH correction (as the second partial of A4 is 1.65 cts high, at A4 one beat second is almost 4 cts, at A5 it may be around 2 cts if I am not wrong.



AS Gilles said, may be there is some stretch parameter that is included in some instruments, due to the computation of the scale, or whatever.

68

Re: A new Italian temperament: CHas

Philippe Guillaume wrote:

Gilles, I agree with you, a good way to implement a very precise temperament is to specify the 88 notes frequency values in a scala file. I paste here a recent post from the piano world forum (piano tuner-technicians section). Remember that if the last note in the SCALA file has a value different from 1200.0 (or 2/1), then the inharmonicity will be by-passed and the frequencies will be exactly those specified in the SCALA file.

Here is how to proceed after creating your own SCALA file (below an example of an 88 notes SCALA file content, remember that the first note is always omitted):
- tuning section, click on the "mu" microtuning button
- import the SCALA file via the "scale" menu
- choose "88 notes scale" in the keymap menu

This is exactly what I used previously to get an 88 notes scala file. I used the following "piano.kbm" file from the Scala program catalog which should be the same as the embedded "88 notes scale" in pianoteq:
---
! Key-for-key mapping for 88-key piano with A4=440
! Size:
0
! First MIDI note number to retune:
21
! Last MIDI note number to retune:
108
! Middle note where scale degree 0 is mapped to:
21
! Reference note for which frequency is given:
69
! Frequency to tune the above note to (floating point e.g. 440.0):
440.0
! Scale degree to consider as formal octave:
88
---

It seems this is the only way to map starting from A instead of C, but still the mapping is always upward if I understand well, not upward after A3=440 and downward below as is specified for CHAS.

This kbm file says A3 (midi 69) = 440.0 but it is incoherent with the fact that scale degree 0 is mapped to key 21 (A-1). the only way this can work is with flat temperament with semitones of 100.0 cents...

If I suppose A-1 gets the exact theorical value of 27.5 hertz, then A3 diverges a bit. To illustrate, here are the exact frequencies published for CHAS in the left column followed by the frequencies we obtain with an 88 notes scala file on the right:

27.47079916    27.5
29.10494208    29.13587997
30.8362945     30.86907279
32.6706391     32.70536725
34.61410253    34.65089653
36.67317587    36.71215863
38.85473637    38.89603808
41.16607037    41.20982897
43.61489764    43.66125929
46.20939719    46.25851673
48.95823454    49.01027604
51.87059073    51.92572799
54.95619293    55.01461012
58.22534694    58.28723916
61.68897162    61.7545456
65.35863537    65.42811011
69.24659473    69.32020229
73.36583536    73.44382157
77.73011537    77.81274072
82.35401133    82.44155178
87.25296688    87.3457148
92.44334436    92.54160954
97.94247947    98.0465901
103.7687391    103.8790429
109.9415828    110.0584482
116.4816277    116.605445
123.4107172    123.5419
130.7519943    130.8909807
138.5299786    138.6772328
146.7706483    146.9266622
155.501527     155.6668216
164.7517754    164.9269028
174.5522892    174.7378344
184.9358017    185.1323843
195.9369935    196.1452701
207.5926082    207.8132744
219.9415751    220.175368
233.0251393    233.2728398
246.8869996    247.1494349
261.5734541    261.8515007
277.1335549    277.4281416
293.6192724    293.9313831
311.0856683    311.4163453
329.5910797    329.9414276
349.1973141    349.5685029
369.9698556    370.3631252
391.9780838    392.3947476
415.2955055    415.7369552
440                440.46771
466.1740795    466.669612
493.9051646    494.4301746
523.285876      523.8421171
554.4143445    555.0036744
587.394538     588.0189251
622.3366093    622.9981389
659.3572636    660.0581454
698.5801487    699.3227235
740.1362677    740.9230157
784.1644166    784.9979655
830.8116479    831.6947818
880.2337618    881.1694304
932.5958266    933.5871548
988.0727296    989.1230285
1046.849762    1047.962539
1109.123237    1110.302209
1175.101145    1176.350251
1245.003852    1246.327262
1319.064828    1320.466964
1397.531437    1399.016981
1480.665753    1482.239667
1568.745442    1570.412982
1662.064687    1663.831424
1760.935172    1762.807006
1865.68712      1867.670303
1976.670399    1978.771555
2094.255691    2096.481837
2218.835725    2221.194297
2350.826595    2353.32547
2490.669145    2493.31667
2638.830446    2641.635462
2795.80535      2798.777228
2962.118149    2965.266813
3138.324321    3141.660289
3325.01239      3328.546803
3522.805887    3526.55055
3732.365436    3736.332856
3954.390958    3958.594386
4189.624011    4194.077486

It's the left column, the theorical published frequencies, that I would have liked to input in pianoteq. As I said before, the integer values in the Spectrum Profile don't help much since they are off by quite a bit.

69

Re: A new Italian temperament: CHas

Hello Thanks Gilles for your work.

I also worked so I could try a "historical temperament" with Tunelab, all intervals being a little larger than usual. (0.49/12)

It only works for 12 notes, (C to B)  after that one have to find the correct octave to pursue in the temperament.

But while I was testing (on a real piano, trying to follow those 12 notes up and down)  that allowed me to discover the presence of a huge resonance as soon as one can obtain an even beating between 12 and 17 th. In fact the pitch of the note tuned is somehow "self locking" because of that strong resonance (seem to me)


As I have yet given false wishes, I have to make more tests before explanations, but it is (and it may explain why the Chas form is relatively easy to tune, because as it ask for a huge precision, it sound difficult to attain, while it is not, really)

It is a kind of "gold number" that helps to raise resonance of more notes than in the usual way we tune.
Then indeed the sensation is that we don't fight the iH, as things tend to settle naturally, and the opening of the tone is relatively easy to perceive and to recognize

Is what is perceived at the piano exactly what the formula describe ?
That I am unable to say.

I'll post the link to the recording (tests double octave and 12 ths and their beat tests

Last edited by OLEK (04-03-2010 22:52)

70

Re: A new Italian temperament: CHas

Good news! Philippe confirmed that using an 88 notes scala file along with Pianoteq's keymap "88 notes scale" will correctly produce the left column frequencies I just posted i.e. the exact frequencies for the CHAS tuning without further stretching for inharmonicity, as published in Alfredo Capurso's paper.

Here is the scala file to use :

---

! chas.scl
!
chas temperament without inharmonicity
87
!
100.038318882533
200.076636882091
300.114955302867
400.153274154802
500.191592345311
600.229910658558
700.268229170886
800.306547579976
900.344866290436
1000.38318454203
1100.42150297336
1200.45982154193
1300.49814004033
1400.53645839507
1500.57477693634
1600.61309533657
1700.6514138122
1800.68973205365
1900.72805059276
2000.76636908475
2100.80468745035
2200.84300599411
2300.8813241215
2400.91964255186
2500.95796186672
2600.99628013653
2701.03459855033
2801.07291689319
2901.11123514696
3001.14955381662
3101.18787139104
3201.22618988988
3301.2645083965
3401.30282710402
3501.34114593067
3601.37946441464
3701.41778239709
3801.45610095191
3901.49441971533
4001.53273766749
4101.57105608141
4201.60937469044
4301.64769310961
4401.68601164913
4501.72433003252
4601.76264836962
4701.80096695059
4801.83928536594
4901.877603754
5001.91592235181
5101.95424053159
5201.99255906407
5302.0308775266
5402.06919614558
5502.10751456003
5602.14583299108
5702.18415143658
5802.22246984666
5902.26078819135
6002.29910655411
6102.33742516183
6202.37574360248
6302.41406228998
6402.45238105003
6502.49069856574
6602.52901787102
6702.56733518509
6802.60565437913
6902.64397304413
7002.68229150255
7102.72060962034
7202.75892833981
7302.79724658811
7402.83556449645
7502.87388346934
7602.91220187592
7702.9505204923
7802.98883858298
7903.02715731485
8003.06547553316
8103.10379400434
8203.14211209111
8303.18043058019
8403.21874901926
8503.25706762933
8603.2953860523
8703.33370445092

---

It should then be used instead of the 12 or 24 values file.

71

Re: A new Italian temperament: CHas

Gilles and Olek:

EDIT: Oops. Cross post. Didn't see that this was a new list of freqs, Gilles. (In any case, do read below. Useful for the future, perhaps.)

This may not solve the problem, but it's worth knowing. It's Jim Aikin's overview of Scala, in which he says that you write "set middle 60" to set  the temperament to begin on note 60. (See the example near the bottom of the page.)

http://emusician.com/tutorials/square-one-scaling-up/

So is it possible to create the 88 note file, and then type "set middle 1" to make sure that the freqs start on the correct note? I don't know. Just ran across this a few minutes ago.

Last edited by Jake Johnson (05-03-2010 09:57)

72

Re: A new Italian temperament: CHas

For those interested, here is the scala file to use for the CHAS temperament with inharmonicity, namely the inharmonicity of the default C3. It is based on adding the CHAS variation to a pure 2:1 tuning (2:1 tuning does not mean that octaves are tuned in a ratio 2 but that the second partial of the lowest note of an octave matches exactly the fundamental of the upper note).

Don't forget to select "88 notes scale" in the keyboard mapping menu.

Copy paste the text between the dotted lines in a file called chasinhC3.scl.

---
! chasinhC3.scl
!
chas temp with C3 inharmonicity
87
!
100.074705253158
200.14613576181
300.212866494716
400.273049633295
500.328981194239
600.382531455092
700.435162454973
800.487941942146
900.541554156055
1000.59630766325
1100.65214021992
1200.70862044661
1300.76494596415
1400.81993739487
1500.87202779321
1600.91924718556
1700.96378827263
1801.0074360002
1901.05158412415
2001.09724804251
2101.14507406712
2201.19534526769
2301.24798386254
2401.30255014528
2501.35823801535
2601.41386709479
2701.46787143493
2801.51828478912
2901.56730818402
3001.61674210052
3101.6680057926
3201.72215258737
3301.77988114709
3401.84154282013
3501.90714516003
3601.97635177826
3702.04847872765
3802.1224875411
3902.19697499258
4002.27015961618
4102.34445079462
4202.42188267471
4302.50413555747
4402.59255353657
4502.68815845697
4602.79166027288
4702.90346398538
4803.0236733016
4903.15209126004
5003.28821797853
5103.43124567908
5203.58005098966
5303.73777050565
5403.90724180153
5504.09103273055
5604.29146776336
5704.51065161344
5804.7504902365
5905.01270955093
6005.29887213832
6105.6103922397
6205.94854939667
6306.3145010633
6406.7092943115
6507.13846283025
6607.60749163997
6708.12187256452
6808.68715951338
6909.30902403074
7009.99331163254
7110.74609950662
7211.57375633972
7312.48300490077
7413.48098832065
7514.57534080752
7615.7742635778
7717.09119282957
7818.54034284013
7920.13684805722
8021.89691411044
8123.83797909889
8225.97888676967
8328.34007333519
8430.94376994184
8533.81422288265
8636.97793403537
8740.46392401326
---

73

Re: A new Italian temperament: CHas

Thanks.  cool

(But Alfredo will say that your treble 12th's are  w--i--d--e.  Says they should be narrow even after the unisons are raised...)

Last edited by Jake Johnson (05-03-2010 10:23)

74

Re: A new Italian temperament: CHas

Jake, it's not my fault if treble strings have a strong inharmonicity wink

75

Re: A new Italian temperament: CHas

Philippe Guillaume wrote:

For those interested, here is the scala file to use for the CHAS temperament with inharmonicity, namely the inharmonicity of the default C3. It is based on adding the CHAS variation to a pure 2:1 tuning (2:1 tuning does not mean that octaves are tuned in a ratio 2 but that the second partial of the lowest note of an octave matches exactly the fundamental of the upper note).

Don't forget to select "88 notes scale" in the keyboard mapping menu.

Copy paste the text between the dotted lines in a file called chasinhC3.scl.

---
! chasinhC3.scl
!
chas temp with C3 inharmonicity
87
!
100.074705253158
200.14613576181
300.212866494716
400.273049633295
500.328981194239
600.382531455092
700.435162454973
800.487941942146
900.541554156055
1000.59630766325
1100.65214021992
1200.70862044661
1300.76494596415
1400.81993739487
1500.87202779321
1600.91924718556
1700.96378827263
1801.0074360002
1901.05158412415
2001.09724804251
2101.14507406712
2201.19534526769
2301.24798386254
2401.30255014528
2501.35823801535
2601.41386709479
2701.46787143493
2801.51828478912
2901.56730818402
3001.61674210052
3101.6680057926
3201.72215258737
3301.77988114709
3401.84154282013
3501.90714516003
3601.97635177826
3702.04847872765
3802.1224875411
3902.19697499258
4002.27015961618
4102.34445079462
4202.42188267471
4302.50413555747
4402.59255353657
4502.68815845697
4602.79166027288
4702.90346398538
4803.0236733016
4903.15209126004
5003.28821797853
5103.43124567908
5203.58005098966
5303.73777050565
5403.90724180153
5504.09103273055
5604.29146776336
5704.51065161344
5804.7504902365
5905.01270955093
6005.29887213832
6105.6103922397
6205.94854939667
6306.3145010633
6406.7092943115
6507.13846283025
6607.60749163997
6708.12187256452
6808.68715951338
6909.30902403074
7009.99331163254
7110.74609950662
7211.57375633972
7312.48300490077
7413.48098832065
7514.57534080752
7615.7742635778
7717.09119282957
7818.54034284013
7920.13684805722
8021.89691411044
8123.83797909889
8225.97888676967
8328.34007333519
8430.94376994184
8533.81422288265
8636.97793403537
8740.46392401326
---

Thank you Phillippe

The secretions of crustaceans are these sea creatures' special features!
A.k.a. I like shells!

76

Re: A new Italian temperament: CHas

here is a demo using the above tuning

http://www.forum-pianoteq.com/uploads.p … 20demo.mp3

Played on my CP33 using the following settings:

keyboard: 'Hard Touch' setting
                 'Two' instrument presets enabled (any two)

Pianoteq: 'C3 Solo Recording' preset
                 Velocity = [10, 95, 105, 116, 127; 0, 105, 116, 123, 127]
                'Concert Audience' reverb

Last edited by sigasa (05-03-2010 21:26)
The secretions of crustaceans are these sea creatures' special features!
A.k.a. I like shells!

77

Re: A new Italian temperament: CHas

OLEK wrote:
sigasa wrote:

here is a demo using the above tuning

http://www.forum-pianoteq.com/uploads.p … 20demo.mp3

Played on my CP33 using the following settings:

keyboard: 'Hard Touch' setting
                 'Two' instrument presets enabled (any two)

Pianoteq: 'C3 Solo Recording' preset
                 Velocity = [10, 95, 105, 116, 127; 0, 105, 116, 123, 127]
                'Concert Audience' reverb


Hello ALl,

I had little time to check those days !

You all made amazing things !. thanks for that !

The Chas tuning file provided by Philippe seem to be exactly what we obtain with Chas tuning.

Anyway , it respects the basicl ratio.

If there are differences thy may be that Alfredo allow the 5th to get purer in the treble and basses, when in there (with the scl file) , they look similar in any region (bottom m3R faster than topM3 in the 5th)

The 12ths are tempered all along, the octaves doubles etc are good, and no artificial stretch is needed as when tuning with Chas

What I understood is that with Chas the 12ths stay similar, the 5ths evolve, but I cant have the correct understanding at the moment if that is to be with iH correction or no- probably with, and probably it is more aural than anything else - but in that case quid of the formula ?)

I thought that the "inversion of 5ths rate" was the correct consequence of the Chas ratio applied to the piano iH.

I hope ALfredo will be back (in a few days) to answer those points.


(the 12ths are not over sized, if the Chas formula is applied - the ratio express the final result, the large fifths are tuned for practical purpose when making ( on a real piano) a one note pre tuning, (3.5 octaves) that is ready to settle down when the unisons are tuned.

I will check the justness in the Chas, but suffice to listen to Sigasa record to know this is Chas tuning (the basses particularly have that special resonance)


Thanks for all that !

The example in standard tuning is excellent also for comparability.


Isaac

Last edited by OLEK (06-03-2010 15:39)

78

Re: A new Italian temperament: CHas

Here is a link to the scl file that Philippe prepared the progression :

http://docs.google.com/leaf?id=0B6GjQDk … &hl=en

please feel free to download and try.


in keymap choose 88 note scale (important)

Last edited by OLEK (07-03-2010 14:10)

79

Re: A new Italian temperament: CHas

I have listened to some midi files with the Chas 88 notes and the standard setup.

I am under the impression, that, due to more resonance, the use of the sustain pedal may be lighter with the Chas.

the timbre behavior differ, also, the low medium and basses get more prominent .

I wonder what others have to say.

80

Re: A new Italian temperament: CHas

OLEK wrote:

I have listened to some midi files with the Chas 88 notes and the standard setup.

I am under the impression, that, due to more resonance, the use of the sustain pedal may be lighter with the Chas.

the timbre behavior differ, also, the low medium and basses get more prominent .

I wonder what others have to say.

I do like Chas IH but find it a little flat in the treble. However the bass and mid ranges are beautiful. Philippe sharpems the treble. Philippe, do you have a scala file for the Chas IH with sharpened treble? big_smile

Chris

Last edited by sigasa (07-03-2010 20:57)
The secretions of crustaceans are these sea creatures' special features!
A.k.a. I like shells!

81

Re: A new Italian temperament: CHas

Why not use the octave stretching? That's the purpose of this parameter. It won't change much the middle range but principally the treble.

82

Re: A new Italian temperament: CHas

Philippe Guillaume wrote:

Why not use the octave stretching? That's the purpose of this parameter. It won't change much the middle range but principally the treble.

Thank you, will do

Chris

The secretions of crustaceans are these sea creatures' special features!
A.k.a. I like shells!

83

Re: A new Italian temperament: CHas

Philippe Guillaume wrote:

Why not use the octave stretching? That's the purpose of this parameter. It won't change much the middle range but principally the treble.

Thank you again Philippe. here is the same demo midi file with Chas IH Octave Stretching set to max. (3.00)

http://www.forum-pianoteq.com/uploads.p … max%29.mp3

Some may think this colourless! I think it is somewhat colourless regarding lack of harmonics, but has a charm of it's own? Sounds very warm! I like it anyway! smile

The secretions of crustaceans are these sea creatures' special features!
A.k.a. I like shells!

84

Re: A new Italian temperament: CHas

The truth is,.. in Chas tuning , on a piano, the fifths get "pure" in the high treble (or eventually a little bit more but not in a really noticeable way - say Alfredo.
The same in the basses. That point I did not check yet in my tunings, I only begin with the method, but I feel that it happens "naturally" when tuning in the treble , coming from a medium with Chas.

In The Chas88 with ih, these intervals keep the same (very good) ratio all along the keyboard. That make that raised harmony yet, but indeed I like more crispness in the treble.

The suggestion to use stretch parameter is good, but I wonder what is the cause.
I thought that, due to the piano iH, that U shaped curve for the 5 ths may fall in place naturally (while the 12th keep a consistent beat rate)

Any reason ? Alfredo will be back in a few... I hope for his input.

May be also the way the unisons are tuned (with the outer strings tending to be high) "push" the pitch high, more in the treble than in the medium, and that this is not strictly reproduced in Pianoteq.

As the externlt strings are tuned to couple with the center string at the 2nd partial level, it put the spectra in the iH more, and raise the pitch (probably) that more in the treble due to more iH.

or the 0.49 number only reproduce partially the formula basics.

Last edited by OLEK (07-03-2010 23:01)

85

Re: A new Italian temperament: CHas

sigasa wrote:
Philippe Guillaume wrote:

Why not use the octave stretching? That's the purpose of this parameter. It won't change much the middle range but principally the treble.

Thank you again Philippe. here is the same demo midi file with Chas IH Octave Stretching set to max. (3.00)

http://www.forum-pianoteq.com/uploads.p … max%29.mp3

Some may think this colourless! I think it is somewhat colourless regarding lack of harmonics, but has a charm of it's own? Sounds very warm! I like it anyway! smile

Thanks Sigasa, It is better for clarity of the treble,

But  probably the stretch would be better if less "straight" (it is not, really but a hair )  , I mean less stretch in the begin of the treble, the same amount in the high treble, a little less for the bass , where it can stay linear)

Would you post the same tuning with limited reverberation ? It helps a lot, any tuning or piano will sound better with a little reverb. less easy to compare.

If a good tuning is done the piano may have its own "respiration", then the sustain pedal may suffice to open the tone when necessary.

When I say "U shaped curve for the 5th,  I talk of beats in fact it is same usual "S" shape we see in the Railsback curve (and which is present yet here).

Last edited by OLEK (07-03-2010 22:46)

86

Re: A new Italian temperament: CHas

For Isaac OLEK as requested

http://www.forum-pianoteq.com/uploads.p … Reverb.mp3

Played on my CP33 using the following settings:

keyboard: 'Hard Touch' setting
                 'Two' instrument presets enabled (any two)

Pianoteq: 'C3 Solo Recording' preset
                 Velocity = [10, 95, 105, 116, 127; 0, 105, 116, 123, 127]
                 Reverb Off
                 Level Compensation ebabled
                 5th  mic enabled

Chris

Last edited by sigasa (08-03-2010 10:55)
The secretions of crustaceans are these sea creatures' special features!
A.k.a. I like shells!

87

Re: A new Italian temperament: CHas

Sounds very good, Sigasa. (The solo recording preset has a lot of room in the sound, however...Hard to hear what's happening with beats, etc. Any change of doing a Close mic version that's similar?)

88

Re: A new Italian temperament: CHas

sigasa wrote:

For Isaac OLEK as requested

http://www.forum-pianoteq.com/uploads.p … Reverb.mp3



Chris

Thank you Chris, it is cleaner, easier to understand the output. indeed some resonance (due to the reverb) is lowered. I feel you did not use any stretch there. is it the case ?

89

Re: A new Italian temperament: CHas

OLEK wrote:
sigasa wrote:

For Isaac OLEK as requested

http://www.forum-pianoteq.com/uploads.p … Reverb.mp3



Chris

Thank you Chris, it is cleaner, easier to understand the output. indeed some resonance (due to the reverb) is lowered. I feel you did not use any stretch there. is it the case ?

no, I used maximum stretch again (3.00). There were a few other alterations though OLEK

here are the settings:

First load default C3 Solo Recording preset
Turn Reverb Off
Adjust Octave Stretch = 3.00 (max)
Load up CHas IH (88 note scale enabled!)
Enable Level Compensation
Enable 5th mic

Hope this helps

Chris

Last edited by sigasa (09-03-2010 23:38)
The secretions of crustaceans are these sea creatures' special features!
A.k.a. I like shells!

90

Re: A new Italian temperament: CHas

New recording (Schumann Romance (No. 2 of three))

http://www.forum-pianoteq.com/uploads.p … nt%201.mp3

Settings:

Default C3 Solo Rec Variant 1
Reverb Off
CHas IH Tuning
Octave Stretch 3.00 (max)
Level Compensation enabled

Chris

Last edited by sigasa (10-03-2010 00:22)
The secretions of crustaceans are these sea creatures' special features!
A.k.a. I like shells!

91

Re: A new Italian temperament: CHas

Thank you all for the relevant progress. I like these recordings very much.

I have some questions: what can be done to make sure we are listening to Chas? Is there a way to aurally check the 12ths progression? For instance, by slowly playing choromatic 12ths, say 3/4 seconds for each 12th? (Maybe alternating bottom 4ths and upper 5ths too).

Philippe, would you be equipped (at your best time available) for measuring the iH deviation curve for Chas frequencies, meaning the equivalent of the Railsback curve?

Olek, you write:

..."I thought that the "inversion of 5ths rate" was the correct consequence of the Chas ratio applied to the piano iH.
I hope ALfredo will be back (in a few days) to answer those points."...

What I could aurally notice (tuning centre strings) is: from low notes, up to C3 (on a real piano) 4ths and 5ths converge, i.e. 4ths get less and less wide, 5ths get more and more narrow. At C3 they seem to have the same beat rate. From C3 up, 4ths and 5ths diverge: 4ths get wider and wider, 5ths go more and more narrow. From A3-E4 (little approx) 5ths invert and start going progressive less and less narrow, to the point that, in high trebles they can sound pure. This can gain progressive octaves and constant opposite equal beating 12ths (delta-narrow) and 15ths (delta-wide).

I'm not sure about how much iH is influencing this, but I have been able to draw the Chas beat form on all harpsychords tunings (40+) EDIT: "and any size pianos, listening to beats", so I could only "freeze" the iH understanding and wait for deeper investigations.

Regards, a.c.

Last edited by alfredo capurso (10-03-2010 10:29)

92

Re: A new Italian temperament: CHas

Below is a copy of a list of essentials (by no means a complete list) for a CHas tuning. I posted it on PW and Alfredo made the included corrections:

1. The bearing is set over THREE OR MORE octaves (FROM C6 DOWN TO STRINGS CROSSING, ON CENTRE STRINGS).
2. The intervals OF THESE octaves are divided up equally, as much as possible, although of course iH on some pianos may require repitching some notes slightly.
3. The lowest possible partials are listened to when checking beat rates:
octaves at 2:1 , doubles at 4:1, 12ths at 3:1 ; 5ths at 3:2 not using the checks that compare 2 fast beating intervals, as the M6 M17th to check the 12th size.
4. Double octaves and 12ths beat equally.
5. Double octaves are very slightly wide, CONSTANT ALL ALONG THE KEYBOARD.
6. 12ths are very slightly narrow, CONSTANT ALL ALONG THE KEYBOARD .
7. Octaves beat very slightly (PROGRESSIVE) wider as they move to each end of the keyboard.
8. But the 12th's remain slightly narrow, AND CONSTANT ALL ALONG THE KEYBOARD
9. M5's become PROGRESSIVELY just or nearly just in the upper regions.
10. 4ths become LESS WIDE as they approach C3.
11. Unisons are PREFERABLY very slightly wide to reinforce the 2nd partial.
12.(SUSTAIN) Resonance is to be preferred to (ATTACK) power: Although the bearing\temperament octave (THREE+ octaves in CHas) is slightly wide of theoretical ET, the upper keyboard is often "milder," than many contemporary ET's, since lower partials are listened to for beats." (Milder relative to more extreme ET's with lots of stretch and\or pure 12ths.)

I really like the recordings using PianoTeq, but I must say that I worry about the use of stretch in some of the fxp's and mp3's. Of course, one should get whatever sound one wants. It's impossible to argue with beauty. So I'll ask Alfredo to do that: Is the stretch, or are other things, moving the tuning too far from the narrow 12ths, etc of CHas?

(I have to admit that one problem I'm having is that the recordings of CHas that I've heard were done with a close-up mic that may not be omnidirectional--the intent was to capture the tuning instead of to capture a performance. But those recordings are also often very good--they have a close, intimate feel, and bring out the resonances of the chords well, since the mics don't pick up much of the resonances from the room.)

93

Re: A new Italian temperament: CHas

Jake Johnson wrote:

Below is a copy of a list of essentials (by no means a complete list) for a CHas tuning. I posted it on PW and Alfredo made the included corrections:

1. The bearing is set over THREE OR MORE octaves (FROM C6 DOWN TO STRINGS CROSSING, ON CENTRE STRINGS).
2. The intervals OF THESE octaves are divided up equally, as much as possible, although of course iH on some pianos may require repitching some notes slightly.
3. The lowest possible partials are listened to when checking beat rates:
octaves at 2:1 , doubles at 4:1, 12ths at 3:1 ; 5ths at 3:2 not using the checks that compare 2 fast beating intervals, as the M6 M17th to check the 12th size.
4. Double octaves and 12ths beat equally.
5. Double octaves are very slightly wide, CONSTANT ALL ALONG THE KEYBOARD.
6. 12ths are very slightly narrow, CONSTANT ALL ALONG THE KEYBOARD .
7. Octaves beat very slightly (PROGRESSIVE) wider as they move to each end of the keyboard.
8. But the 12th's remain slightly narrow, AND CONSTANT ALL ALONG THE KEYBOARD
9. M5's become PROGRESSIVELY just or nearly just in the upper regions.
10. 4ths become LESS WIDE as they approach C3.
11. Unisons are PREFERABLY very slightly wide to reinforce the 2nd partial.
12.(SUSTAIN) Resonance is to be preferred to (ATTACK) power: Although the bearing\temperament octave (THREE+ octaves in CHas) is slightly wide of theoretical ET, the upper keyboard is often "milder," than many contemporary ET's, since lower partials are listened to for beats." (Milder relative to more extreme ET's with lots of stretch and\or pure 12ths.)

I really like the recordings using PianoTeq, but I must say that I worry about the use of stretch in some of the fxp's and mp3's. Of course, one should get whatever sound one wants. It's impossible to argue with beauty. So I'll ask Alfredo to do that: Is the stretch, or are other things, moving the tuning too far from the narrow 12ths, etc of CHas?

(I have to admit that one problem I'm having is that the recordings of CHas that I've heard were done with a close-up mic that may not be omnidirectional--the intent was to capture the tuning instead of to capture a performance. But those recordings are also often very good--they have a close, intimate feel, and bring out the resonances of the chords well, since the mics don't pick up much of the resonances from the room.)

Forgive my ignorance, but has anyone yet tuned pianoteq perfectly to CHas? I'm not quite sure how to tune CHas in pianoteq?

Last edited by sigasa (10-03-2010 19:11)
The secretions of crustaceans are these sea creatures' special features!
A.k.a. I like shells!

94

Re: A new Italian temperament: CHas

Sigasa--

So far, Philippe's Scala file and your playing are our closet encounters with CHas. And they sound very good.

My understanding is that Alfredo has been otherwise occupied, but will soon review the intervals.

95

Re: A new Italian temperament: CHas

Yes, my family and I moved last week, more boxes to manage.

..."My understanding is that Alfredo... will soon review the intervals."

Jake, do you mean aurally and taste wise? I did listen to the recordings and found them very nice and clean, but checking Chas form rigorously is a different thing.

I need to be able to listen to some intervals (two notes together), played slowly.

Chas 12ths and 15ths (foundamentals) could be checked random, since they are constantly equal beating. Progressive intervals need to be checked chromatically. Mainly 5ths, 6ths, octaves, 10ths and 17ths. Then I could be sure it is Chas. 

The range to be checked (on a piano) could be from C2 to A7, so to avoid (in this case) acrobatics.

Yet, If you (all) have any other idea or question I'll be happy to add about.

Sigasa, thanks for your recordings. Let me know what you need to know for tuning Chas in Pianoteq, maybe I can help.

Regards, a.c.

Last edited by alfredo capurso (10-03-2010 23:53)

96

Re: A new Italian temperament: CHas

Hm...After remembering something from another thread, I just downloaded the most recent trial version of PianoTeq, and I see that it lets you tune note by note, and of course you can open Scala files, record midi files, etc.

Some limitations: Seven or eight keys are disabled and the program stops playing after 20 minutes. You have to reload it. But you can save a preset as you go along, and then open it again.

Alfredo and Kamin--do you have midi controllers now? It's not absolutely necessary, since there's an on-screen keyboard.

There would of course be a short learning curve, but this procedure would be more straightforward than all of us posting files for you to listen to.

Last edited by Jake Johnson (11-03-2010 01:33)

97

Re: A new Italian temperament: CHas

Hello Jake,

In fact I did not find any retuning feature on my demo  3.5 version.

I could analyze th e beta rates of the Chas 88 scl, with or without stretch. in fact I did (but with STretch = 1.0). The only thing that differs is that the 5 ths stay the same ,as octaves probably).

When I record real pianos I use a very cheap goodie  Zoom H2 , it is just on the top of the piano for a vertical , and on the plate in a grand (because I dont carry a mike stand with me, i had to find a position.

So yes the miking is near  !!

The probem I have to check tunings on a digital is that my midi keyboard is only 2 octaves span, I have to use the midi output regulation so to play other octaves. I own a Roland A80 keyboard, but I will not have it back since summer.

I know that little midi keyboards used for writing scores are not expensive, but I will have no real use of them, I guess. I will see how much they cost...

The idea about Chas may be that it provide more proximity for more partials, in mor eintervals than when tuning plain ET, on pianos.

Till next.  Best regards

98

Re: A new Italian temperament: CHas

sigasa wrote:

New recording (Schumann Romance (No. 2 of three))

http://www.forum-pianoteq.com/uploads.p … nt%201.mp3

Settings:

Default C3 Solo Rec Variant 1
Reverb Off
CHas IH Tuning
Octave Stretch 3.00 (max)
Level Compensation enabled

Chris

Thanks for the recording , nice intention, it is appreciated.

Good feed backs from other people too. AT some point i stretch, the tone/harmony  begin to loose its natural condition., may be what I heard as absence of stretch initially. if stretch is added even more the impression can be of a toy piano. I believe we may keep that tone in mind while regulating the stretch, so to stay in a more close harmony.

The reverb is adding much in life quality of the tone, indeed. we do that with the unison tuning and the match to room acoustic when tuning acoustic pianos, I guess.

Best regards

99

Re: A new Italian temperament: CHas

Olek:

It's there:

1. Double-click on the Volume slider to bring up the Volume note edit pane.
2. In the upper left corner of this Volume ne pane, click on one of the wedges\arrowheads beside the word Volume. The Detune ne pane opens. (Or click on the word "Volume" and select Detune from the list.)
3. You'll want to remove the blue squares that let the user control ranges of notes. To do this, right-click on one of them and click on "Remove all control notes." Click and drag to repitch notes.
4.  You may want to detach this pane, to have more room to work in: move the mouse pointer to the left or right of the yellow box that has the demo info. When your mouse pointer becomes a hand, click and drag the Detune ne pane away from the main interface. Playing\clicking a note on the keyboard tells you where you are.
5. On the lower right corner of this detached Detune pane, there are some diagonal stripes. Click and drag on this corner to expand the Detune NE pane.
6. To give yourself more control over the rate at which dragging the mouse changes the pitch, hold the Shift key while dragging the yellow note pitch bar for each note.

Of course, there are limitations--you won't be able to use the similar pane to review the partial structure and listen to specific partials, or pitch the unisons note by note, etc, and some notes are locked down.

By the way, you can also drag other areas of the interface off to one side using the same method: In either the sound stage\mics area or the Presets list, move the mouse pointer to near the top of its window, and when the mouse pointer turns into a hand, you can drag the pane off to one side. Lets you see the mics and presets and tuning and pedals, etc all at the same time, with a large enough screen.

Last edited by Jake Johnson (11-03-2010 15:47)

100

Re: A new Italian temperament: CHas

(And you're using the Zoom H2? That mic has come up before, in another context. I thought that you might be using one. Strange--four tiny mics inside it and it's very clear. Not a big concert sound, obviously, or a big commercial production sound, but it puts the listener in a small room with the piano. Gets much better reviews than the previous version--for some reason called the H4? Thanks for the info.)