I wanted to reply to this because I think this is a particularly notable feature of the Synchron Steinway.  On most VSTs, repeated notes and fast trills are possible but often hard work and never terribly crisp sounding.  That's true even of the Synchron CFX (an instrument I overall prefer, but not in this respect).  The Synchron Steinway, makes them a lot easier and a lot crisper; it is undoubtedly the best piano VST I've yet tested in this respect.

I wonder if you have "Repetition Smear" on in the Edit tab?  If so, try turning that off and testing again.

Originally Posted by: fatis12_24918

there is something so perfect and well balanced in the Yamaha CFX Synchron Piano, that still surprises and captures me every time I play it... deep bass, vibrating but never metallic... sweet treble, sharp but never nasty... superb uniformity across the whole range.  So finally I loved the Steinway D last update, but Yamaha CFX is my favorite one. Does anybody agree?

Yes actually, I do.  They're both very fine instruments and I will quite happily play either, but if you told me I could only keep one, I'd marginally choose to keep the CFX.

In my case that's a reflection of my acoustic prejudice as well, though; I actually prefer the sound and feel of an acoustic CFX to a Hamburg D, which I realise is sacrilege to many people.  The acoustic CFX is a fine instrument; completely different from other Yamaha offerings.  I think the VSL instruments quite accurately capture the acoustic characteristics of both instruments, to the point of replicating my acoustic preference.

Originally Posted by: stephen limbaugh

Originally Posted by: William

Somehow I never hear any music by them.

Allow me to flex. 

https://www.youtube.com/watch?v=5VdWcNWQQlI

https://www.youtube.com/watch?v=jz2Q3kNsIIk

https://www.youtube.com/watch?v=tBn5yVYG96k

https://www.youtube.com/watch?v=QisezFa_oUU

 

If the suggestions in this were implemented, I would say they would have the following direct musical benefits:

• Better "out of the box" sound requiring less EQ to fix the noise.
• Better playability requiring fewer adjustments to the MIDI after a performance.
• For film scoring, piano parts tend to air on the soft side (the opposite of the current curve)... could be more inspirational for composers!

 

Granted, this is what *I* would do.  If none of these are implemented, then I certainly have workarounds and handy "save presets" options once I get it dialed in how I like it.   I have a feeling though VSL will introduce a curve... they're really good about listening to their customers when they have reasonable requests.

Fantastic stuff, Stephen!!  Dazzling performance of the Rachmaninoff, and your Millennial Suite is superb and with such passionate performance.  Deeply impressive.

Originally Posted by: opus64

Hi Karvala,

Interesting point, but i'm not sure I understand something if you don't mind clarifying:

My understanding of how the piano was sampled is that multiple mics are placed in the room with fixed settings(preamp gain, etc).

Considering only a single pedal/etc condition layer, the robot plays a key at multiple velocities and the microphones capture the resulting sound.

Unless VSL is changing the mic settings for velocity(perhaps to accomodate dynamic range?) or post-adjusting the amplitude of the samples during playback, it seems to me the amplitude and timbre are linked at the recording point and if the microphones are linear, should represent what the instrument sounds like.

If these two are truly linked at sampling time, then it seems the midi->velocity map should allow to adjust for the problem, which in my view is that for a relatively soft keypress velocity the resulting sound is more ff or even sfz.

I'm probably missing something....

Thanks. 

That's a fair question.  If there were no post-processing, you would be absolutely right, but in practice there is a *lot* of processing of the samples after recording, and amplitude adjustment is certainly one of them.

Have you ever heard the Salamander piano - a Yamaha C5 VST?  It was essentially a fun project that someone did on their own, and it's probably the closest you'll come to a piano VST which has simply been sampled and  assembled with minimal post-processing (but even in that, they did some).  It's worth having a listen if you haven't , and you'll see just how raw and 'recorded' the sound is, and it will give you some idea of the amount of work which is needed after recording to turn it into a usable VST rather than a series of triggered individual piano recordings.  Indeed, one might even say that the quality of the post-processing is as important as the recording setup in determining the final quality of the VST.

Originally Posted by: opus64

Hi all,

Thanks for the replies.  In the past few days I did a direct comparison with the VSL Vinenna Imperial and it does not have the same problem.  The Imperial runs fine with a linear curve off the Kawai NV10, much like Garritan and my other VSTs.  There is something definitely different with the VSL Steinway D.

I do not beleive there is anything wrong with the sample set, the problem would be remedied by adding a 1-D velocity curve editor.  The only case in which a 1-D velocity curve editor could not address this is if there was (a) a sparse sample set and (b) strong non-linear characteristic to how the samples were originally captured.  For example if the sample set had 8 layers and most layers were recorded near ff-sfz..not much can be done there.

However with so many sample points on the velocity axis, just fixing the curve would fine even if the 'robot' velocity curve for layers was non-linear.

Someone commented on the VSL video and how the pianist did not notice this, this is a good observation. I would note that that this was a specific weighted MIDI controller which has some implemented velocity mapping curve.  It could be that the VST was designed to work very well with that in-house controller, but that doesn't mean it would work with others, which is why having a 1-D velocity curve editor is standard on all other VSTs.

I agree that if the velocty curve issue is fixed as far as I can tell this could be the definitive sampled Steinway D.  What great sound, and ambience, absolutely mesmerizing.

Unfortunately it's not as simple as that; if it were there would be no problem because there are countless free velocity curve editors already out there which can easily be used with Synchron Pianos already.

The problem is that the perceived loudness of the piano comes from two things: objective amplitude (sound pressure level) and timbre.  If I record a piano really loud, and then greatly reduce the volume on playback, it doesn't sound like a soft piano, it sounds like a loud piano being played back quietly.

A velocity curve editor enables the user to change the mapping between the input and the amplitude-timbre combination (because the two are fixed together in the sample).  However, if the amplitude-timbre mapping is wrong, which in this it clearly is, that cannot be fixed by changing the velocity curve.  I could, for example, edit my velocity curve to require me to play much harder in order to achieve higher velocity layers, which would help to fix the input-timbre mapping.  However, that would have the effect of making it much harder to achieve higher amplitudes, i.e. it would distort the input-velocity mapping.  Or instead, I could edit the velocity curve to ensure that the input-velocity mapping is okay (which it more or less already is in this case), but that would then create (or leave/not fix) the problem with the input-timbre mapping.

In other words, you cannot, by adjusting the mapping of the input to different samples, fix a problem which is caused by internal inconsistency in the samples in terms of the amplitude and the timbre.  The amplitude-timbre mapping of the samples needs to be adjusted and that can only be done by VSL.

Originally Posted by: William

Is this another dynamic range issue?  In other words, perhaps the timbral velocity is mapped properly but the amplitude differences triggered by velocity are simply too much. So dialing down the dynamic range could fix it.  But that is only a speculation as I don't have this instrument.  

No, if anything the other way around.  The amplitude mapping is more or less okay; it's the timbral mapping that's heavily skewed in favour of fff.  If the amplitude were similarly skewed that would actually be better because then it would be quite straightforward to fix with a velocity curve.  Unfortunately, that's not the case, so we await some magic from VSL.

Yes, I must say I notice this too, and several users over at PianoWorld have also commented on it.  One of them suggested that the likely cause is the very different sampling techniques used, with the CFX sampled using the built-in facility, while the Steinway was sampled using the robot.

Regardless, it is slightly curious that it was not normalised in some way prior to release.  It suggests that VSL regard the velocity curves/midi sensitivity of the two as broadly equivalent, but it's hard to see how.  The Steinway sounds almost brutally harsh on a linear velocity curve and zero midi sensitivity.

Definitely an SSD is the way to go, and SSD prices have dropped dramatically in the last six months, so this is a good time to buy.

In regard to experiencing the problem on the Steinway but not the CFX: it's apparent that the Steinway is significantly more demanding of the CPU.  With the same number of mics (and both sample sets on the same SSD drive), I can run the CFX at a lower latency without artefacts than the Steinway.  I'm not sure why that should be;  perhaps Martin can comment?

Depending on how you host Synchron Piano, there are easily ways around this.  For example, the free SAVIHost can host Synchron Pianos and thus the Yamaha CFX, and includes a builtin velocity curve editor which can with a single click-and-drag map all midi values below whatever value you choose to be remapped to 'off' and therefore not sounding.  There, you now have silent keys; job done. 

You're very welcome; glad you like it. 

I've attached another that I use here as well.  Quite similar to the previous one in terms of overall feel, but using some different mics.  The main difference is a bit less sparkle from the ambient mics, but a bit more clarity overall.  I call this my Everyday one, and use it for longer playing periods when the other one can become a bit tiring.

Here's mine; uses six microphones so a bit resource-heavy.

I completely agree with the sentiment of the opening post (and great playing!), but I would make a plea that demos should cover a range of styles and capabilities, not just focus on virtuoso music from the 19th and early 20th centuries.

As many professional pianists know, to make a Mozart sonata sound convincing on a modern concert grand is a lot harder than a Liszt Etude.  Both have their place in demos: I want to hear the full dynamic and tonal range in a great romantic work, but I also want to see how the instrument handles something much more exposed.

I also want to see some purely technical demos.  Playing a simple scale is hugely informative for both acoustic pianos and VSTs.  If there are flaws in an instrument, they are much more likely to be apparent within a single demo that is simple and exposed rather than complex and sonorous.

So let's have great romantic works by Chopin and Liszt, let's have some complex early 20th century works by Ravel, but let's also have Bach fugues, Mozart sonatas and simple scales.  Let's also ban all post-processing (no compression/expansion, no EQ), and then we would have a set of demos that could really show the capabilities and limitations of the instrument.

I'm sure we'll see them.  VSL will have learned a lot in the production of the CFX and have a good idea how to do it more efficiently (and cost effectively) for the next instrument; I'm sure they will want to make good use of that knowledge.

There are a lot of Steinway D VSTis out there, but none of them are entirely convincing for one reason or another, so there is definitely an opportunity there.  There are many fewer Bösendorfers, and they are even less convincing, so that's wide open.  Would be nice to see VSL versions of both. 

Hi David,

Good choice; I find the extra mic positions very much worth having to give some extra options for the sound, and I always find myself using them.

Just to briefly describe the process of adding the full library in my experience:-

1.  Once you've bought the piano, you'll get an activation code for the Full Library Upgrade.

2.  You enter this activation code in eLicencer in exactly the same way you did for the first one, and it will create what it calls a Collection licence, which will include the original one ("SY Yamaha CFX") and the upgrade ("SY Yamaha CFX L2").  That's all you need for the licencing.

3. In MyVSL you'll find a new download file ("Yamaha CFX - Multi Library Download").  Downloading that file and opening it in the same Vienna Download Manager you already have installed will give you a list of the 5 new mics (and a couple of information files), and you can uncheck any you don't want to download from there.  There original 5 mics will not be in this list.

4. You can download the mic files to wherever you want.  Once downloaded, you run the installed in exactly the same way as you did previously for the standard libraries.  You don't need to keep the download files afterwards, but if you want to keep them in case you need to reinstall at some point, then you can move them to wherever you want.

5. The installed additional microphone files will now be wherever you told the installed to put them.  Again, these can be moved around to wherever you want, as long as you remember to add whatever folder(s) they're in to the Settings of Synchron Pianos when you next run it.  Do that before clicking any preset.

6. Now you can run any preset to load the samples, and all 10 microphones will automatically be available for you.

That's it!  It's very straightforward; much more so than most other piano VSTs.

Ah, if only it were that easy.  Player adjustment can get you so far, but this is hard-limited by the sound in this instance.  Even playing that fugue subject (which is just one example; I could give you dozens) staccato, the room resonance overlap is strong enough to smudge it.  There is no way to maintain the tempo and avoid the smudging; it's not a playing issue.

I fully appreciate that some people will prefer the new sample length; I didn't feel the previous samples were clipped myself but opinions will differ on this.  That's exactly why an adjustment control would be ideal - then each player can customise it to their own taste, and indeed to the requirements of different repertoire even. 

This is a very well sampled and nice-sounding instrument you have here.  The various updates to the software to address particular issues have also been very welcome, and it's great to a see a developer engaged with the community like this.

The updated have, perhaps inadvertently, created one issue for me, which is a problem with the room resonance in the release samples.  The latest update - 1.0.582, which solves some important issues - and the previous update - 1.0.530 - have seen an adjustment of the release samples, in which the room resonance has become more prominent.  I can understand the desire to make the piano as resonant as possible on a single-note or single-chord basis - it sounds luxurious - but it is a bit of a problem for daily playing where notes are being smudged into each other much more than they used to.

To give a very simple example of the issue, I recorded just a Bach fugue subject and two chords in two versions - 1.0.464 (which comes first in each comparison in the example) and 1.0.582 (which comes second).  The first fugue subject is relatively clean, while in the second the notes are slightly smudged.  The chords show the slightly increased resonance in the 1.0.582 version.  Just to be clear: these were created from the *same midi file* and used the *same custom VSL profile*, so this is not a case of playing differently in the two versions or using different mic settings; this is entirely down to different release sample behaviour.  The example is at: https://tinyurl.com/ybqdgtw4"> https://tinyurl.com/ybqdgtw4.

As there are already many good ways increase to increase the resonance of the sound in the instrument (8 out of the 10 mics in the full version, or 4 out of 5 mics in the standard version, do this, as does the very good builtin reverb), I would suggest one of two solutions to this issue:-

1. There should be an adjustment control for the release sample resonance; this would be the ideal solution, as it would allow players to customise it exactly as they wanted.  Most piano VSTs have something like this, and this instrument clearly has separate release samples and they are clearly adjustable across the different releases, so I'm hoping this will be possible, but only the developers will know how difficult this is.

2. If 1. is really not possible, then release an updated version incorporating all the latest fixed, but with the samples at least as dry as they were in 1.0.464, which otherwise remains the version that I will have to use because of the playability issues of the longer/louder room resonance.

Thanks!

Hi Paul, thanks for this.  I'm happy to confirm that this new release appears to have resolved the chopping problem as far as I can tell.

I've been enjoying giving this a test drive, and it's a very nice instrument that can sound very beautiful.  I like the CFX as an acoustic instrument anyway, and this captures it very well.  Excellent clean samples.

It's still early days for this and I know a few minor issuse are being ironed out at the moment.  Can I draw your attention to one general issue which you might want to look at to give the instrument that final polish that will really make it shine?

It's the issue of matching tonal characteristics and spatial characteristics across the keyboard in a seamless manner.  At the moment there are a few notes which pop out in a rather distracting manner when playing, and this doesn't need to be the case.  They pop out either because their tonal characteristics are notable different from surrounding tones, or because their spatial positioning is different from the surrounding tones.

There are a few of these; some examples from the Close 1 mic, where they are particularly prominent, are:-

1. Tonal: B4.  Play a scale from a few notes below and above including this with just the Close 1 mic, and you'll hear how different this is to A4 and C4; while the surrounding notes are quite similar to each other.

2. Spatial:  C5 is the most egregious example - if you play a surrounding scale you'll hear the rest of the notes as relatively centred but this one pops out as being across to the left somewhat.  Some lower tones (G2,G#2,A3) are also quite prominent in this regard.

I'm sure these can be easily fixed and would make the ,playing experience that bit more enyoyable.  Once the more pressing issues have been addressed, it might be worth having a close listen across all the tones and mics and ironing out other similar examples, just to give it that final polish. 

Has this been broken again in the most recent release?  I started playing from the previous version (so not the first one with the problem), and I didn't notice any problem, but I've installed the latest version I definitely notice this happening now.  Essentially, sample release sounds as though it's just completely cut off if the sustain pedal had been down.  Doesn't happen if the sustain pedal was not used; then the normal room release sound occurs.

EDIT: Yes, confirmed; I just reinstalled the previous (1.0.464) version and there is no chopping in that, but installing the latest version (1.0.530) brings it back unfortunately, so that's another for re-patching.

You're very welcome, and just for extra info, I have since downloaded and installed the latest version, and although the samples still don't load with junction links used, it no longer causes a crash (the samples simply remain unloaded), so I think you've somewhat fixed that one already. 

"Fixed: Default frequencies for keys off by one note in EDIT view"

Sorry, but that's not fixed.  When you first fire it up, indeed the default frequencies are now okay.  But as soon as any samples are loaded, the default frequencies in the EDIT view are once again off by one note.

Just to give both of you some extra info here:-

1. No, this definitely does not require the April update.  I have not installed that because I'm not part of the early adoption scheme, and given the reported problems with it I wouldn't install it anyway.  I can run the VSL CFX fine, so that update is not required.

2. This may or may not be relevant to your problem Craig, but is something that you should be aware of, Paul.  On Windows filesystems, hard/junction links, which were pointers to other filesystem locations but which are treated by the OS as extensions of the existing filesystem, cause the VSL CFX to crash.  They work fine with literally every other piece of software I've ever tried, including other VSTs, so there is something amiss here.  It only occurs when there is an attempt to load the actual samples.  That is, if the hardlink points from C: to E:, and the software the files in C: and they are actually in E:, which should be treated as C: because of the hardlink, in practice they' are not treated the same.  don't know if it's in the VSL loading mechanism, or related to elicencer which is probably quite strict about locations, but it's worth keeping in mind.  Certainly a crash should never be the intended behaviour, so this amounts to a bug, and is useful to know in case anyone wants to move the software and was hoping to use junction links.