Piano World Home Page Forums Our Most Popular Forums Piano Forum Steinway in Wired

Everything beneath the first two rows makes me want to gag.

I don't remember the exact % I was told, but I think even Steinway(perhaps the brand bought more by non pianists than any other high end piano)sells less than 50% of their pianos to non pianists. As far as maintenance is concerned, my guess is that only a small percent of piano owners take good care of their pianos.

I had some shares in Steinway, and followed their financials closely. They were not on the brink of bankruptcy. IIRC, they were in a pretty good situation with their debt.

When Paulson purchased them, I got double what I paid for my shares.

I'd rather have the shares, though. I could never own a Steinway, but I could say I OWNED Steinway, even if it was just a pile of spruce shavings in the corner. :-)

Maybe Steinway should offer a year's worth of free lessons with the purchase of a piano.

With what those pianos cost it could be a lifetime of lessons and still be very profitable.

The article mentions neither the Yamaha Disklavier, nor the Bösendorfer CEUS. Neither were designed for "dusty old saloons".

Expand the comments on the bottom of the article. More entertaining than the article, as usual. Most [of the commenters] didn't seem to take the article too well also saying things like how Yamaha has been doing this for decades.

One comment caught my attention: The article was published on April 1st.
Did Steinway confirm it?

Definitely not an April Fool's joke. Has anyone seen one yet?

I wonder how different it is from the LX retrofit system.

indeed, yamaha first brought the tech to the US in '87 -- 28 years ago. the burgett brothers funded their purchase of M&H on the back of their (at the time) PianoDisc success.

yamaha continues to expand disklavier, also allowing remote playing -- a couple years ago the local yammy dealer had a live concert with sarah mclaughlin 'playing' the piano in the room (and many other rooms simultaneously). it was pretty cool, but not sure about its commercial reach.

the dealer also told me that without disklavier he'd probably go out of business -- it helps sell that many more pianos.

although the steinway system may have some higher resolutions i doubt it justifies the cost difference, especially to the normal ear. but people are paying a premium anyway for a steinway, if only for the nameplate in the parlor.

to create a real differentiator from disklavier, which on a comparison matrix would tromp the spirio, steinway should get its stable of pianists to record pieces for it. then, you could have some real bragging rights about who's playing the mozart sonata in your living room. that might come at an additional price of course.

as to the article's headline "Steinway’s New Piano Can Play a Perfect Concerto by Itself" -- unless i'm missing something (like its ability to reproduce an orchestra), no it can't.

Yamaha wasn't the first, either. When they introduced the Disklavier in the U.S. in 1987, they bought out and discontinued the cassette-driven Superscope/Marantz Pianocorder system that had been on the market since 1978. Yamaha's discontinuing the Pianocorder is what spurred the Burgett Brothers (piano dealers at the time) to build their own replacement, founding PianoDisc.

Prior to the Pianocorder, there were a few other commercially-sold solenoid-driven systems in the U.S., including the Teledyne "Piano Player" and the Orrtronic "Piano Play-orr", but the Pianocorder was really the first system to achieve much success, selling around 16,000 units during the decade the product was on the market.

From the mid-1970s until the Disklavier was introduced in the U.S., Wayne Stahnke (designer of the LX / Steinway Spirio system) was doing his early work with high-resolution, closed-loop, solenoid-based reproducing pianos, including the SE system licensed to Boesendorfer. The LX/Spirio system maintains a lot of the standards introduced for the SE and is compatible with recordings made on it. Playback is open loop, but the system still manages to achieve performance that is nearly SE-level in quality.

What does all that mean: high-resolution, closed loop?

Is high-resolution related to some type of digital sampling bit-rate? What's closed loop?

Solenoid-based player piano systems are generally based on the MIDI spec developed in the mid-1980s that permits 127 levels of expression for notes (7-bit resolution). In practice, player systems use only a subset of this range, typically around 24-95, a convention established by Yamaha allowing some headroom at each end. High-res is anything above that, but typically 10-bits of resolution (1024 levels), but as before, there are many unused levels at the top and bottom of the range. In the case of the Disklavier Pro, the same 24-95 range gains three additional bits of resolution, giving a usable range of about 576 expression levels. The LX and SE systems are 10-bit as well but similarly use only a subset of the available range. On the low end, the streaming format used by QRS Pianomation systems supports only 15 expression levels.

There are different ways of getting a system to play notes at varying levels reliably at the correct moments in time. Many player systems (including Steinway's) are open loop, meaning that they fire note solenoids blindly, factoring in calibration data configured by a technician, and use software tricks to try and achieve the intended results. This can work surprisingly well, even though the system is only guessing at what is physically happening. For greater accuracy, a closed-loop control system adds position and velocity sensors to each note so that the system KNOWS what's happening and can control the solenoid drive mid-strike based on the measurements, thus closing the "loop". The extra sensors add considerably to the cost of the system, and it's only feasible to include them in high-end, factory-installed systems like the Disklavier Pro, not retrofit kits like PianoDisc. It will be interesting to see whether Steinway eventually goes closed-loop with the Spirio system.

great info, thanks mark

Lots of thoughts as I read through this...

Player pianos have been around a long time. Like some folks have posted, I'd much rather hear someone play than a robotic/mechanical player.

The first time we had a party after getting our piano I hired someone to play. My main purpose was to keep people from putting drinks on it! But I was surprised how much it added to the atmosphere and was enjoyed by many people! It was such a hit that I've hired the same guy multiple times now, for two holiday parties and Spring soirée we just held. Friends of ours liked it so much that they bought a piano and plan to hire the same guy to play at their next party. In retrospect, it shouldn't be a big surprise, because--as we all know--pianos are awesome.

The main thing I think of when I read this thread, though, is so tangential that I think I'll start a new thread.

Mark - I've hesitated to write you before, due to your great background and knowledge of solenoid-based player systems, the use of MIDI and/or ESEQ specs and the many very useful conversion tools you have developed over the years for old and new users of this technology. Your explanation of the 127 level MIDI-spec (7-bit range) and it's 72 level useful range (24-95), and 1024 level HD range (10-bit range) and it's 576 level useful range is the best explanation I've seen... Thanks! My question is, with the newer LX/Spirio 10-bit playback systems and even new PianoDisc HD / ProRecord III 10-bit playback/record systems, how is this additional 3-bits of fidelity being stored? The MIDI standard only accounts for 127 levels... is there an enhanced standard in use, when Live Performance, Zenph Re-Performance or Spirio produce an 1024 level HD 'historic' (e.g. Glenn Gould) performance or HD emulation (e.g. Rachmaninoff) of a scanned old roll, or when ProRecord uses their HD TFT system to "record" a performance"? Are these in some proprietary formats, wherein any conversion to the old MIDI standard would drop fidelity? Conversely, what do these systems do when they are reading original MIDI files only encoded as 7-bit, 127 level performances? Do they convert to 10-bit 1024 level playback via interpolation or is this resolution just wasted?

In high-res MIDI streams, both Yamaha and PianoDisc pass the high 7 bits (of 10) as the MIDI velocity in the note-on commands. They use adjacent non-note MIDI commands to provide the three extra LSBs (least-significant bits) for note velocities. Yamaha uses a combination of key aftertouch and control change messages (documented in their "XP" spec), and as far as I know, PianoDisc uses key aftertouch only (no spec is available). The extensions are transparent by design; playing an enhanced stream on a system that doesn't know how to interpret the extra MIDI commands will just result in the coarse 7-bit velocities being used. Likewise, if the stream doesn't include the extra commands, then an HD playback system will just assume the lowest three LSBs to be 0.

Incidentally, Yamaha doesn't offer any commercial high-resolution recordings, and as far as I know, PianoDisc does not either. Both companies claim that HD playback enhances the existing standard-resolution music libraries via more accurate control during playback. Yamaha does provide some free HD tracks recorded at the Piano E-Competition events that it hosts.

As for the LX/Spirio stream, it isn't based on MIDI, and no backwards compatibility is needed, so it always passes 10-bit velocity values. Steinway has stated that all music for Spirio will be recorded, edited and released in high-definition and will be provided at no additional charge with purchase of the piano.

So, do you happen to know what file format that LX/Spirio is based on? Stahnke said that the LX will be resistant to technology's changes over time, but I really doubt that is true. Any design and file format will become obsolete sooner or later.

I have one (two) customers that do make their own solenoid midi driven systems, added on player piano, accordion, (quite a challenge), an external one to be installed on any keyboard, next will be a percussion set.

Material obtained from old hard disks, minimum investment, mostly midi card, and transceiver for remote control.

They have a blog, if you are interested.
http://www.dailymotion.com/video/xlo7eb_duo-accordeon-et-piano-midi_music
http://www.dailymotion.com/video/x1kxyh0_bobinage_music
http://www.dailymotion.com/video/x17l3pv_solenoide-sa-carcasse_music

I think the point is that the core LX/Spirio system doesn't integrate any rapidly-changing technology like floppy disks, optical disks, USB, Wi-Fi, Bluetooth, web interface, etc. as past player systems have done. It just accepts an analog audio signal containing a modulated digital stream. Any future device with an analog audio output should be able to drive it, even 50 or 100 years from now. A device that can generate the stream dynamically can add control over tempo, volume, and transposition, sync with accompaniment or video, etc. This is the direction Steinway has been going with Spirio (currently via an iPad app). It seems to be a trend: PianoDisc iQ similarly isolates the core player system from the user interface. QRS PNOmation and Yamaha Disklavier haven't fully accomplished the separation yet but are moving in that direction.