Pianoteq iOS

[HammondDave]

Seems like our musician colleagues can’t tell the difference. Do you really think a typical listener in a club could?… especially when other instruments are in the mix?  

[Reezekeys]

Again, some interesting comments. As was mentioned previously in this thread, #1 is my iOS piano which is a "re-creation" of my Native Instrument's  "New York" piano. #2 is Pianoteq.

 

3 hours ago, drawback said:

In this comparison I’m definitely liking your #1 (at the end of the video) as opposed to the thinner, less nuanced sound of #2.

 

I note that it seems, from the comments I've read so far, that my piano has a few more votes here than Pianoteq. I don't consider this any kind of "scientific" analysis (for one, it's a pretty small sample size!), but I must admit to being surprised at this outcome since, first of all, the "NY Piano" I sampled started life as Native Instrument's "Akoustik Piano" plugin... in 2005! This sample is 18 years old! Secondly, my iPad version is significantly cut down from the original; for one thing, there are no key up samples. There are also no pedal resonance samples, and no sustain pedal noise samples. Also, my samples are stretched one semitone in each direction - I've sampled every minor third (three semitones) while I believe the real NY Piano has every semitone sampled. I have less velocity layers in my piano as well (nine, I believe the actual NY Piano has ten). On paper, my piano is way behind on its feature set compared to Pianoteq - which is why I feel that it's my piano (#1) that has less nuance and sounds "thinner" than #2, the opposite of your view! 🙂  In this solo context I think I can hear the lack of these "extra" samples plainly, but then again this is a piano I've lived with since I first started playing gigs with a laptop - we're going back about 15 or 16 years. I know exactly what to listen for. 

 

13 hours ago, ElmerJFudd said:

Can you share how many velocity layers the desktop version is and how many you resampled?  How large is the resulting library?  

 

Nine layers as I mentioned. Most samples (except some high notes) are ten seconds long, unlooped. Some of the higher notes are less than ten seconds long. There are a total of 270 samples, and they add up to about 645MB (they're .aiff). I used Logic's Autosampler to make this. It creates a Logic Sampler file (.exs) that AudioLayer can import. Thank you iCloud drive!

 

3 hours ago, U.Honey said:

It sounds like #2 has a little more reverb. But, if the reverbs are the same with both of them then I guess what I'm hearing is the modeled string resonance on #2.

 

I agree with this. I think the resonances being modeled give the sound ambience that could be thought of as reverb, especially where the note density is higher. This actually informs my personal theory as to why my drier, "less complex" NY Piano in this example might have been preferred by some - the sound seems a little clearer, the notes cut through the mix with more clarity. This at the expense of a certain richness in tone the Pianoteq piano has.

One thing I want to say is I'm very much not meaning to imply we don't need these extra sonic components to make a good piano sound - we do! My cut down version of the NI piano is for gigging only, and I made it back when my iPad was a nine-year-old Air 2 so was trying to conserve resources as much as possible. Then I brought it out on gigs and found out I didn't miss the key-ups, resonances, noises, etc. On a recording, listening critically on studio monitors, yes, for sure I do miss them. I guess my ears aren't that discerning when I'm on a gig! 🙂 

 

As far as this video: I really like Pianoteq very much. I do think there is a certain quality to the modeled sound and it is slightly jarring to me after years with sampled pianos. But, the way it reacts to velocities, like the hammer "thunk" not suddenly jumping out on the high notes, which I hear on my piano (that's velocity switching, I may need to do some adjusting!), also how smooth the timbre changes are, are very cool. I didn't really think this was going to make me a convert and I was right about that - still I'm impressed with Pianoteq, and having this on iOS is a great thing.

[AnotherScott]

I like #2 better, but it sounds less like a real piano. So I guessed that was pianoteq before I took note of your reveal.

 

I don't love either of them. They both suffer from what is, to my ears, the plunkiness of exaggerated attacks, common to many digital pianos. Though considering the customizability available in Pianoteq, maybe that can be tamed.

 

Nice playing, though, especially on an action that is not so amenable to piano playing.

[Kawai James]

45 minutes ago, AnotherScott said:

Nice playing, though, especially on an action that is not so amenable to piano playing.

I wondered if the action was the main reason for the "plunkiness of exaggerated attacks"?

But yes, some great playing!

As for identifying which is which in the latest video, I believe its:

#1 - Pianoteq
#2 - Pianoteq (but a different model/mic setting)

#1 - Sample

#2 - Pianoteq

 

Cheers,

James

x

[ElmerJFudd]

Point being that the player, foremost, has to be happy with the sound of their instrument and how it feels to play it.  It doesn’t matter if it’s acoustic, electric or digital.  After that it’s what the player has to say with it that matters.  

[Artomas]

I could tell which was which, mostly by the way the 1st example was panned, as if you used 2 mics above a soundboard, with the left ear's mic more towards the bass end, and vice versa. It reminded me a little of how the CFX samples are panned in my Yamaha DP (EDIT: I always use the Yamaha's line outs into fx into a mixer and then headphones from there, but I think the Yamaha's headphones outs are supposed to be panned differently from the line outs?).

 

My preference is the first Pianoteq example, the one with all the fx. I know. I think that's because, even though I don't really play "ambient" music, I use fx as if I were playing it twice. (I'm an amateur, usually solo noodling on a P-115. I don't play classical, and haven't played an acoustic in a while, for your statistical reference.)

 

In the revised demo, I think it could go either way... but I'm very likely going to add delay and reverb, and sometimes even filter and distortion etc... so I'd prefer to start with a fuller and punchier piano (in this comparison, Pianoteq) before adding those fx which tend to push the instrument further back. IOW, I think the more natural, and clear but thin, nature of the samples would get smeared anyway. The reason I say it could go either way is, I think I could  use EQ to thicken up the samples or help tame the modeling, for any need I'd have. I agree with the observations of the people who chose example 1, and yet I'd slightly prefer 2, based on hearing the same things. Maybe I just like fx?

[Mighty Motif Max]

Listening to the revised video, I like them both equally I think. I actually lean toward Piano 2 more this time around - the odd metallic resonance I noticed in the first demo is gone. There are a few notes where it vaguely reminds me of a Kurzweil piano tone (mostly in the octave above middle C) as well. I think as a general purpose piano I would prefer Piano 2.

[Al Quinn]

On the revised video they both sound good to me. How wonderful that we have instruments of this quality for iPad!
 

I’ve played that song many times but not for many years. Still should have recognized it but did not. Chick Corea and Gary Burton kill it on their In Concert, Zurich album.

[Al Quinn]

Is it just me or doesn’t the Pianoteq 8 Steinway D in this video sound different than Rob’s video? It’s my understanding that the iOS version should sound the same as the Mac version but the piano in this video sounds warmer to me. I’m wondering if the iOS and Mac versions really sound the same. He starts playing at around the 2:00 mark.

 

 

[Reezekeys]

10 hours ago, AnotherScott said:

I don't love either of them. They both suffer from what is, to my ears, the plunkiness of exaggerated attacks, common to many digital pianos. Though considering the customizability available in Pianoteq, maybe that can be tamed.

 

Yes I can hear that "plunk" on my piano, less so on the Pianoteq - and I think it has to do a bit with me just hitting the keys a little harder once I start playing in time. It's an uptempo jazz piece and I got a little too excited! The more rubato section at the beginning has less of that, imo. Scaling my velocities down a bit would help - of course the proper time to do that is while I'm playing!

 

7 hours ago, Al Quinn said:

Is it just me or doesn’t the Pianoteq 8 Steinway D in this video sound different than Rob’s video?

 

They sounds pretty similar to me, though he is using a preset that includes effects & eq while I turned all those things off and did some mild EQing in a DAW to try and match the balances I'm used to hearing with my piano.

 

8 hours ago, Artomas said:

I agree with the observations of the people who chose example 1, and yet I'd slightly prefer 2, based on hearing the same things. Maybe I just like fx?

 

Yes from what you wrote it seems you like more ambience & fx. I play old-guy music (jazz, r&b) and my preference is for a more vanilla piano sound, on the bright side to cut through in a band situation with drums, bass etc.

 

10 hours ago, Kawai James said:

As for identifying which is which in the latest video, I believe its:

#1 - Pianoteq
#2 - Pianoteq (but a different model/mic setting)

#1 - Sample

#2 - Pianoteq

 

You were close! #1 is always my sample, #2 is always Pianoteq. Same exact settings on both throughout the video.

 

8 hours ago, Al Quinn said:

Chick Corea and Gary Burton kill it on their In Concert, Zurich album.

 

 Oh yea!!

[Kawai James]

4 hours ago, Reezekeys said:

You were close! #1 is always my sample, #2 is always Pianoteq. Same exact settings on both throughout the video.

Ah, okay. Perhaps I misinterpreted what is being shown on the iPad screen during the different passages.

It seems that at some points, the AUM app is connected to Pianoteq, at other points it is connected to the sampler app.

[Reezekeys]

Good catch noticing that! I zoomed in on my iPad screen and see what you mean. The first time through my intro you see AudioLayer bypassed, yet it's my New York piano playing, not Pianoteq. This was a post production move. I decided to run the rubato intro twice to compare the pianos in that setting, where I played a little softer and held some notes longer. Since I recorded midi as I filmed, I rendered the song from beginning to end with both pianos and chose appropriately while editing the final video. Starting with the second time through my rubato intro, the iPad screen is accurate to what you're hearing.

[drawback]

54 minutes ago, Kawai James said:

Ah, okay. Perhaps I misinterpreted what is being shown on the iPad screen during the different passages.

It seems that at some points, the AUM app is connected to Pianoteq, at other points it is connected to the sampler app.

That was my clue as well. They both sound good, but which Pianoteq pianos did you use? I’m finding Steinway B gives me the best root sound to edit around. Next would be the Bechstein. In all the pianos though, there’s a certain tone in the lower octaves I can’t seem to fix - some call it metallic, I suppose.

[Reezekeys]

38 minutes ago, drawback said:

They both sound good, but which Pianoteq pianos did you use?

 

I did use the Steinway, I figured that would be a better choice to compare to my NI New York piano since that's also a Steiny.

[Art-0252]

Incredible playing as always @Reezekeys!
In both examples I preferred Pianoteq before I knew which it was, and I think the reason for that is the same reason that I have gravitated towards Pianoteq when practicing in the studio over the last 6 months or so: The dynamics of playing are going to be naturally more expressive with modeling because you essentially have 127 velocity layers ( limited only by the midi resolution of your controller). I love the C7 in Keyscape, and I think to my ears it sounds slightly richer than the Yc5 model in Pianoteq, but when I play PT it feels alive under my fingers in a way that no digital piano ever has to me. The ability to dig in and hear the nuance of dynamics just makes it so enjoyable to play imo, but of course YMMV.

Modartt hit it out of the park by bringing this to iOS. I hope their investment pays off for them because I know it is costly to enter the iOS ecosystem, and the pricing paradigm of iOS leads to challenges others on this thread have already talked about.

 

Having this on my iPhone for no additional cost to the desktop license of PT8 I already had blows my mind and gets me thinking about the exciting new possibilities for a lightweight jam rig when i don't want to haul my heavier more expensive keyboards to something like a patio blues Jam or basement rehearsal.

 

PT + B3x + a synth app like Tal-uno-lx + a lightweight keyboard with a usb audio interface (ie mx61 or modx6) would be a pretty sweet lightweight setup!

 

 

[octa]

#2 in second batch is my fave. #1 in first very good too. Can tell you’re more used to dynamics of #1. Cool playing & tune! May have to steal some of yer licks. 
 

When at home with studio monitors, I like running garritan, modern u & keyscape at 96 kHz as it adds a little presence & oddly enough garritan seems to run better at higher sample rate. Wondering if sample rate would make a noticeable difference w/ modeled piano. (higher sample rates are only available in Pro)
 

If there’s a sale, I may spring for Pro. Paid for Stage and really enjoying Steinway D and bluthner.  Would like to have full keyboard available on Bechstein & Clav too. 

[Nievski]

2 hours ago, Jazz+ said:

I’m wondering how Pianoteq iOS will respond dynamically when playing from the USB port of my little 10 pound Casio CT-S1. 

 

So far so good on my CT-S1, but still tweaking the velocity curve. It's a nice battery operated setup! 

[Kawai James]

11 hours ago, Reezekeys said:

Good catch noticing that! I zoomed in on my iPad screen and see what you mean. The first time through my intro you see AudioLayer bypassed, yet it's my New York piano playing, not Pianoteq. This was a post production move. I decided to run the rubato intro twice to compare the pianos in that setting, where I played a little softer and held some notes longer. Since I recorded midi as I filmed, I rendered the song from beginning to end with both pianos and chose appropriately while editing the final video. Starting with the second time through my rubato intro, the iPad screen is accurate to what you're hearing.

 

Got it, thanks for the explanation.

 

So am I correct in thinking that (for some passages) we're essentially hearing a (re)rendering of a MIDI file initially recorded while playing a different sound, rather than playing and recording both sounds "live"?

 

This is probably a minor point, but it could be said that pianists respond to the sound they're hearing and adapt their playing in real-time.  So a MIDI file recorded while playing a sampled VI and then played back using Pianoteq, may not have the same "feeling" as if it was recorded while playing Pianoteq directly.

 

Cheers,

James

x

[Reezekeys]

12 hours ago, Kawai James said:

So am I correct in thinking that (for some passages) we're essentially hearing a (re)rendering of a MIDI file initially recorded while playing a different sound, rather than playing and recording both sounds "live"?

 

Except for the rubato intro I was actually switching between the pianos during the filming, so was reacting to what I heard. The intro was filmed with me playing my piano, not PTeq. Before I filmed I adjusted Pianoteq's velocity response according to the wizard in the app and it felt fine to me. Listening to the intro rendered through PTeq, I hear nothing out of place wrt dynamics. Ironically I forgot that I usually play my New York piano via Midiflow, which has a velocity scaling preset with my tweaks - I didn't have it while recording, so my piano might have actually suffered a bit in this comparison - on repeated listenings I notice the velocities hitting it seem a little on the high side. I'm not re-doing this video again though! 🙂 

[Reezekeys]

On 5/9/2023 at 8:46 PM, Reezekeys said:

On 5/9/2023 at 7:05 AM, ElmerJFudd said:

Can you share how many velocity layers the desktop version is and how many you resampled?  How large is the resulting library?  

 

I misread this, you were asking about the desktop version, not my cut-down resample. You probably deduced what I meant but to avoid any possible confusion: I believe the desktop version has ten layers. 634 samples adding up to 2.9GB. Again, my resample is 9 layers, 270 samples, 644MB.

 

I was contemplating adding some of the "extras" to my piano, like release samples and pedal/sympathetic resonances, but I have to be honest and say I don't really miss them on my gigs, which are almost 100% with full bands. If I know in advance I'm doing some solo or more exposed piano, I'll probably bring the laptop.

[ElmerJFudd]

5 minutes ago, Reezekeys said:

 

I misread this, you were asking about the desktop version, not my cut-down resample. You probably deduced what I meant but to avoid any possible confusion: I believe the desktop version has ten layers. 634 samples adding up to 2.9GB. Again, my resample is 9 layers, 270 samples, 644MB.

 

I was contemplating adding some of the "extras" to my piano, like release samples and pedal/sympathetic resonances, but I have to be honest and say I don't really miss them on my gigs, which are almost 100% with full bands. If I know in advance I'm doing some solo or more exposed piano, I'll probably bring the laptop.

Thank you, I was asking about both.  And the size of your resulting library.  In that regard, Pianoteq 8 on iOS, I agree, is quite impressive in what it can do in comparison to sampling methods.  Including the minute details and the tweakability to taste in the mid and top tier versions. But these things are only noticeable to us, in solo and duo settings.

[ElmerJFudd]

8 hours ago, Jazz+ said:

Will the lag time be the same with Pianoteq iOS as it is with Korg Module American D Ivory?

As in latency?  Time it takes for midi in to become audio out?   There’s latency of course, but you can lower the buffer in settings as long as your iOS device is fast enough to keep up.  Otherwise pops and clicks.  Which iOS device do you have? 
 

But… with Pianoteq you can maybe go lower in buffer if you reduce polyphony.  

[ErrkaPetti]

8 hours ago, Jazz+ said:

Will the lag time be the same with Pianoteq iOS as it is with Korg Module American D Ivory?

On my iPad Pro 12.9” M1 there’s no lagging - everything spot on in response…

Modartt have made an incredible Pianoteq for iOS/iPadOS - runs smooth as hell…

[ElmerJFudd]

29 minutes ago, ErrkaPetti said:

On my iPad Pro 12.9” M1 there’s no lagging - everything spot on in response…

Modartt have made an incredible Pianoteq for iOS/iPadOS - runs smooth as hell…

I have the 11” variant and am able to go max poly and lowest buffer as well.  But I also quit other apps and put it in airplane mode. 

[Reezekeys]

I would expect excellent performance on any iPad Pro! It’s the less powerful iPads I’d be concerned with. I was happy playing PTeq on my 9G at a 128 buffer. Polyphony was 48. Actually I never even looked at these settings, just downloaded the app & started playing. It’s possible I could get more notes or use a lower buffer, haven’t tried yet though.

[Jose EB5AGV]

19 minutes ago, Reezekeys said:

I would expect excellent performance on any iPad Pro! It’s the less powerful iPads I’d be concerned with. I was happy playing PTeq on my 9G at a 128 buffer. Polyphony was 48. Actually I never even looked at these settings, just downloaded the app & started playing. It’s possible I could get more notes or use a lower buffer, haven’t tried yet though.

 

For sure, as those settings work on my old iPad Air 2 

[Reezekeys]

3 minutes ago, Jose EB5AGV said:

For sure, as those settings work on my old iPad Air 2 

 

I forgot to add... 48K sample rate. Is that what you're doing on the Air 2? That's impressive. PT should run on my original iPhone SE (2016) then.

[Jose EB5AGV]

58 minutes ago, Reezekeys said:

 

I forgot to add... 48K sample rate. Is that what you're doing on the Air 2? That's impressive. PT should run on my original iPhone SE (2016) then.

 

Yes, it works well, except if I artificially pound the keyboard and keep the sustain on 😅

[Doerfler]

 

[ElmerJFudd]

Some nice partners for Pianoteq on iOS. 
 

https://audiomodeling.com/iosproducts/

 

 

They have a few instruments running on iPhone as well.