Piano World Home Page Forums Our Most Popular Forums Piano Tuner-Technicians Forum TuneLab - Sampling

The TuneLab manual recommends sampling 4 to 6 notes before beginning. (It seems that octaves are recommended.) When I tune with the VT, if the piano is flat, I usually sample everything from A4 to A0, then A5 and A6, recalc, then begin tuning/pitch-raising with overpulls from A0 to C8 with inharmonicity reading left on. Unless the piano is way off, one pass with some minor tweaking later is often sufficient.

I need to ask about two things:

1.) If I follow a similar procedure with TuneLab and take many samples, will I confuse the split-scale mode and mess up the tuning calculations?

2.) If I choose to sample 50, 70, or 88 notes, will TuneLab "consider" all this data in its calculations re creating the curve, etc.? (Input into the black box, not tweaking the output?) Can it process all this extra data? Or, would it be overloaded?

Sorry to ask about all this, but just testing TuneLab on one or two pianos doesn't mean I'm ready to take it out in the field. I was hoping for some experienced users views:)

Thanks,
-Joe

1. Tunelab "knows" you want split scale by sampling two notes next to each other.

2. Like a graph of scattered points over which you project a smooth curve, tunelab doesn't adjust to non-linear data points.

No, you won't overload it, but you will get better results by:

Careful measuring, checking number is repeated with multiple measuring.

Pre-tune/check some octaves to make sure that the curve generated matches the piano. (playing with the stretch arrows and partial selection)

Ron Koval

I understand.

I see. It won't adjust its curve? Will it take into account notes/unisons when measured (after calculation) that deviate slightly from the curve and give me good octaves, double octaves, triple octaves, etc.?

It doesn't seem so difficult to try various types of octaves on pianos that are close to pitch.

Thank you Ron,
-Joe

Hi Joe,

I know this isn't really about tuning with VT, but I noticed something you said:

Hope you don't think I'm getting hung up on details, but: If you sample in Coarse Mode, I don't think it's necessary to recalc before you begin tuning, since in Coarse Mode VT constantly updates everything anyway.

Maybe Ron or David WB will correct me if I'm wrong ...

I think I used to do that too.

If I understand you correctly, it is possible to do something along those lines. You can insert red-marked set points on the graph for notes you've tuned aurally and, to a certain extent anyway, adjust the curve so it passes through them.


Jeff

Hi Jeff,

I thought the notes locked in coarse mode as well as in fine mode. I'll remember your advice. Thanks.
Btw, I'm using one of the earlier versions of the VT software on the VT100. I really like the line graph interface for overpulling. I don't know if changes in the latest version are only to the GUI, or, if Dave C. also changed what's under the hood.

This is exactly what I'm interested in. The VT performs almost flawlessly on 95% on the pianos I tune. It's amazing! But, there are still some pianos that I end up tuning mostly or fully by ear. (The very inferior ones.) I'm not sure I fully understand what the VT is doing when I adjust stretch. Tweaking stretch seems like it could be more obvious and maybe a little faster in TuneLab. So basically, I could set all the A's to each other by ear, tweak TL's curve to align with the A's, and then I'm good to go?

Thanks,
-Joe

Joe,

The method you describe with the A's will work on TuneLab, at least if your preferences fall within certain parameters -- with any shifts between sections of the scale not being too drastic. There are limits, because TuneLab's curve responds very much like a unified whole.

I also wonder about first calculating a tuning with octave types in the bass and treble that sound like a good start, then tuning some sample octave and fifth multiples over the scale. One could then go into fully-manual adjustment mode (all four arrows) and tweak the calculated curve according to markers like described above. I don't know if the result would be much different than just starting with aurally-tuned A's, but it might be.

Jeff

Of course when I was discussing tempering the A's, I was considering using more than just octave checks.

Thank you for the additional advice re TL. Either this coming weekend or the following one (depending on how much free time I get), I'll strip mute a grand piano and really tear into tweaking TL.


Thanks,
-Joe

Jeff, I think we were both correct, but I may have misunderstood your point.

In update:
100-5
5-Nov-01

are what were added in this update. On page 3 of the downloadable .pdf file for update 100-5:
I think I misunderstood your point. The reason I was recalc'ing was so that the originally measured pitches would alter with more data. But, the VT WAS designed to be "one-touch," so recalc'ing isn't absolutely necessary.

-Joe G.

Not sure I'm totally following you now either, Joe, but in hopes of clarifying:

What I originally referred to was how the entire tuning VT is calculating as it goes constantly updates in Coarse Mode, whereas in Fine Tune Mode targets typically lock as their notes are tuned and all the targets that follow are calculated with the previous locked targets in mind.

I didn't think you might be talking about the overpull markers in Coarse Mode, that they stay the same unless an "Erase" command is pushed. Even with the overpull markers for previously-tuned notes staying the same, the tuning is still being updated as you go -- at least that's my understanding. In other words, the markers for notes you've tuned don't change, but future markers are set according to updated info, regardless of the previous markers. By the end of a Coarse Tune pass with markers, the earlier set markers may be quite far from indicating the updated internal targets for those notes.

So, when you switch to Fine Tune, it isn't necessary to push "Recalc," even though for sometime I used to do that anyway. (At least that's how the current version is.)

You were originally talking about sampling before pitch raising/tuning, then recalcing. I just related that to my own experience, in which I usually sample in Coarse Mode without overpull, unless I'm doing a full pitch raise anyway, in which case the sampling just goes along with that. I'm assuming in your VT version overpull (and markers) can be switched on or off in Coarse Mode, but maybe I'm wrong.

I hope that all didn't just muddy the waters further. I apologize if I misunderstood something you said. If so, please point it out.

I'm curious about the earlier VT version you have, whether the line/graph interface in overpull you mentioned is something different than the current set-up.

Jeff

Somehow got double-posted

I think that's how it works in Coarse mode as well. This seems to have been programmed in update 10-5:
(From the downloadable .pdf file for update 100-5.)

It also works like this in my version.

Of course, maybe the latest version of VT'as firmware is different. I haven't been able to find the downloadable .pdf manual for 100-7. I berlieve that I'm using the next to the last version of the VT firmware: 100-6b. If you know of a command I can enter to verify the firmware version, I'll be happy to enter it and report back.

-Joe

Hi Jeff,

It does display in the lower right of the screen when starting. I'm using 100-6b.

Ah-ha!!!

I think I see where we are getting mixed up!!!

I will sample the notes in Coarse Mode WITHOUT tuning them. I sample everything from A4 down to A0, then A5 and A6. Then I will go to Fine Mode and recalc. Yes, it is automatically recalc-ed when you go to Fine Mode. Then I go back to Coarse Mode to begin the pitch raise.

The reason I use this pattern is because I've found (like many others) that pitch raising works very well if it is done by simply going from left to right. So, I don't tune as I sample if there is a pitch raise to be done. I've found that by following this procedure, i can "sometimes" get away with making even large pitch raises and have the piano be in good tune after only one pass... and stand in good tune later.


This misunderstanding is MY fault.

Sorry about that.
-Joe

PS. I didn't understand that you meant going to Fine and then the recalc-ing being automatic.

Also, even if the piano is close, I'll often still sample A4 to A0 +A5 +A6 in Coarse and then go to Fine so I can tune left to right to get the most stable tuning possible.

Joe,

Your pitch raising sequence and explanation make sense. When you're just sampling, do you have overpull and its markers turned off? (Just curious.)

I hate to approach the beating of a deceased four-legged, and it may not really matter, but I think I still haven't found the words to explain exactly what I mean when talking about how, in Coarse Mode, the entire tuning is still potentially updating as one proceeds.

You posted:
Although, according to the current manual, pg. 29:

"After an overpulled string begins to settle, the markers and what they represent are no longer valid. Therefore both markers are automatically erased and the overpull target is cancelled two minutes after moving to the next note."


Maybe my use of the term "targets," in previous posts, was problematic. I agree the overpull markers stay where they are (at least for two minutes), once VT has read those notes, unless an "Erase" command is pushed. But in talking about "targets," I was trying to refer to something else besides the display and overpull markers -- really something else entirely.

I'm open to being corrected if I'm wrong, but my understanding and everything I've read -- like on the VT Users Forum -- leads me to believe VT in Coarse Mode continues to update all notes in (How do I term this, to avoid confusion?) its "internal theoretical tuning," if it gathers new information telling it previously-read notes can be better placed. I believe notes can be internally-updated before the display markers corresponding to them pass through the two-minute window, but I could be wrong about that. It'd be a good question for Dave Carpenter.

If I'm right, that's different from what happens in Fine Tune Mode and it could theoretically happen even as the original overpull markers remain in place. In other words, the overpull markers were originally placed according to notes in a "theoretical tuning" that may be -- as the tuning proceeds -- updated to reflect new and more complete information. The original markers would then be obsolete, according to the updated "theoretical tuning" VT is currently working with. The markers could still be there, but they'd reflect note placements VT has since discarded.

I appreciate your efforts to decipher what i'm saying, Joe, but if you follow what I just posted you now see I'm saying VT doesn't just recalc the tuning when a switch is made to Fine Tune from Coarse, but this recalculation is going on all along when the unit's in Coarse Mode.

Let me see if I can give a concrete example:

A4 and A3 are tuned in Coarse Mode, then the notes in between. If we're using overpull the markers will display and stay locked in place, unless two minutes passes or we erase them. Now, suppose we tune the rest of the piano and decide to again go through the notes from A3 to A4, staying in Coarse Mode. The pitches of those notes will be better refined from their original placement, reflecting all the info VT has since gathered. It's not necessary to push Recalc or Erase, or do anything except just make another pass through that section in Coarse Mode.

As I say, I'm open to correction on this, but I think this is the way I've always heard it discussed, and how it makes sense to me.

It would be worth asking Dave C. if notes in Coarse Tune can be internally updated all the time, or if they have to wait until the same two-minute period expires that determines the lifespan of their corresponding external markers. I admit I'm not sure about that.

I guess, if this still seems obtuse, I'd better just let it go. I think my explanatory powers are about exhausted.

Thanks, Joe, for patiently considering all this.

Jeff

Joe,

I have to admit I just rediscovered the two-minute thing while re-reading the manual. I know I was once aware of it, but I forgot about it.

After considering this (really for the first time), I have no objective reason to believe VT in Coarse Mode is capable of holding an internal theoretical pitch for a note different than what is implied by its markers. It may in fact be that notes can be internally-updated only after the two-minute period expires, same as when their external markers disappear.

Maybe, when overpull and markers are used, notes can only be updated after they pass beyond the two-minute window; but, if overpull is switched off in Coarse Mode, they can be updated constantly.

Don't know. Interesting question though, and I'm open to input and opinions from others.

Jeff

Hi Jeff,

Just out of curiousity, which version of the VT's firmware are you using? I think the latest is 100-7-19. Again, I'm using 100-6b.


Also, to answer a question I forgot to answer before; I cannot turn off the overpull markers in Coarse Mode. I'm also limited to setting three overpulls. (If I remember correctly, the latest version lets you set 4.) Also, there are no overpull needle markers in Fine Mode.

We're both using the same HW, but different FW. There are going to be some discrepencies between your version and my version. These variations probably extend well beyond the GUI,so there may be significant differences "under the hood."

You are probably using the latest version of the FW. I stayed with an older version because I like the interface better; It gets me where I want to go; I'm well practiced with the older interface, so I know some little tricks; and I think that two historical temperaments (that I personally use) have been replaced in the newer version.

This is one really great advantage of the VT; There are multiple versions of FW that give very good results. One size does not have to fit all.

Joe

Yep, I have 100-7-19.

However, I don't know of a way to set four different overpull markers with it -- there are just three as far as I can see, nor have I heard of anyone doing that. There is a "0," alongside the other three numbers, but it's just what one goes to when disabling overpull altogether. Some people use all three numbers at the same time.

I know nothing about the older interface, but I'd be curious to see it.

I guess I'm really sort of waiting for the next version.

Jeff

I guess I misunderstood. I thought the latest version had 4 overpulls.

Aren't there overpull needle markers available in Fine Mode in your version? 100-7


-Joe

[Edit]
Never mind. I found a photo and I see the needle. I didn't realize that the manual for the latest version 100-7 needed to be purchased.
Thanks,
-Joe
[/Edit]