Piano World Home Page Forums Our Most Popular Forums Digital Pianos - Electronic Pianos - Synths & Keyboards The DPBSD Project!

snazzy, modern KAWAI DPs (all models released since the ES6, with the exception of the EP3) feature 88-key sampling.

As do most, if not all, Roland DPs.

However, I'm afraid I honestly don't know the average length of the samples before looping - I suspect it varies depending on the region of the keyboard being sampled/played back.

Cheers,
James
x

madshi, yes, in an ideal world, we would have full 88-key sampling and long, long samples.

However, for one reason or another this is not economically viable, hence the need for a compromise between the two.

Cheers,
James
x

That's what I figured you would say.

Thanks,

Snazzy

snazzy, I can fully understand why you'd be cynical.

However believe it or not, in raising this question, I'm not terribly concerned by the particular sampling method employed by any specific DP manufacturer. Instead, I'm just trying to stimulate a debate among other forum members examining how limited sample memory resources can be allocated.

Cheers,
James
x

I'm glad you understand, James.

Thanks once again.

Snazzy

Ha Ha! Hey Snazzy, I thought you were finished with this thread?


Best regards,

Colleen

Well I don't have a massive amount of experience with DPs but I can tell you that I hear stretching and looping on my SP200, the looping really doesn't bother me in fact it reminds me of the beating you'd hear on a not perfectly tuned acoustic.

On the other hand the stretching does annoy me, you get a block of notes all sounding the same then a sudden change in timbre on the next block.

So my gut feeling is that looping is better than stretching but I'm sure it depends on how well it's implemented.

If this was indeed the only option, I'd definitely take the longer sample. Sample stretching is not nearly as destructive to the basic sting tone as looping (provided stretching isn't overly abused).

But these either/or scenarios strike me as missing the whole point of Flash being very inexpensive. Five years ago you could float a better small ROM defense, but these days it should be laughed out of the room.

Lets take this a bit farther. For the price of the glowing CP1 logo you could probably quadruple the sample memory, which would eliminate stretching and perhaps double all of the sample lengths. You can guess which one I would prefer.

dewster, wasn't there some discussion in a previous thread regarding the suitability for standard 'flash' memory to store samples? I forget the exact argument - perhaps something along the lines of latency or access time/transfer speed required for real time playback.

However, regardless of the type of memory used, I agree that increasing the amount of memory capacity is a rather more sensible use of resources than a sexy light-up logo.

Cheers,
James
x

Hi Madshi,

Don't misunderstand me, I think the test is great; it also keeps Dewster busy, for I have a feeling he is more of a spec man than an avid player.

The impression I got from Snazzy was that he felt there were too many opportunities for the test to be rigged or flawed for anyone to really base any actual faith in it; I can see where he is coming from, but I also see the value in some sort of analysis.

Snazzy tends to be very passionate about his beliefs, and whilst I do not share them all, he certainly has both the playing credentials and real world experience to back up what he says.

My thoughts? Unless Dewster can do the tests hands on with each piano, they will always be at best "mostly" reliable; I just feel like several others, that the more "spec-minded" individuals would take away more than they should from the reviews, and not even try a particular instrument that would, if they actually sat down and played it, be chosen over a more highly spec'd and perhaps more expensive instrument.

One thing I do before I buy, is simply go to the local shop and have a conversation with the repair department guy, and find out the frequency of repairs on certain brands and models; the guy I deal with is readily forthcoming with what instruments are problematic, as long as the information remains between us.

I think Dewster would be the ideal person to develop a mobile tester; he's got the know how and the passion for details.

Whether he feels it's worth his time or not is another matter.

Best regards,

Colleen

Ha!

Snazzy's had more comebacks than Frank Sinatra.

Steve

You guys are missing the point. Style is equally as important for stage image as having a great sound.

I doubt if the home pianos will carry the cool logo, but you can bet they will have the same internal workings(with the exception of the un-graded keyboard).

In another thread there was a discussion on the look (or lack of look) of the CP-300.

Companies who realize their pianos will be played in front of an audience that may contain budding keyboardists, realize the "look of cool" is important, and an integral part of promotion and ultimately, more sales.

You should be very much aware of that, James, considering your position with Kawai.

Perhaps the very conservative company prefers the Kawai-it look. Ha Ha.

Regardless of the company, the main focus is sales and profit; if they can get away with making something that sounds better than the competition with the least amount investment in R&D and completely new (and expensive) technology, they will make every effort to do so, and thus please the shareholders and remain in business.

Regards,

Colleen

At the risk of stepping on another landmine, I would like to say that I was actually pulling my punches somewhat in the CP1 review, and for the following reasons:

1. Believe it or not, I don't generally like being the bearer of bad news. Lawrence was kind enough to provide the sample, and I didn't want him to think for one second that I was ripping his new DP to shreds just for fun.

2. I have a history of disliking the flurry of hype and fog of confusing technical terms surrounding the CP1 launch, and I didn't want my review to appear as something of a revenge tactic or sour grapes.

3. While the looping is probably the best I've heard, I was quite shocked at the short sample lengths, the stretching of the lower octaves, the relative shortness of the note decays, and the typically poor sympathetic resonance. So I erred on the side of caution in an attempt to offset any bias introduced by my shock.


Would I buy one of the newest CPs? Probably not for our home studio. Everyone here knows I'm rather picky when it comes to sound, and the CFIII sample doesn't meet my minimal requirements. IMHO, YMMV, the opinions of this poster may not reflect the general consensus view, etc. <insert beating a dead horse icon here>

Would I recommend others buy them? Preliminary indications suggest the CP50 will be sufficiently better than the P-155 to warrant steering people who are in the market for a ~$1500 USD DP in that direction. But I guess we'll see, and I look forward to trying them out when they hit the local stores.

This is why we can't have nice things.

Dewster, dear, you poor man; you are so picky you quite probably won't buy a new piano until Snazzy's best friend Nyiregynazi puts his Blather up for sale, and buys himself an Avant Grand.

Having said that, I'm aware that Lawrence is happily enjoying his new CP-1, unlike many of us who only wish we could afford one.

Personally, I have to say I base my purchase on the actual playing experience, with some knowledge of specs, but not so much that it influences me to distraction.

I really can't imagine you being satisfied with anything being made today, and I am really glad I don't have that problem; no, my final choice may not have mind-blowing specifications and totally perfect replication, but at least I will totally enjoy the digital piano for what it is; a genuinely versatile musical instrument in it's own right.

Perfect? No. Perfect for Colleen? You bet.

By the way, you have my thanks for doing such a bang up job on the testing, even with all it's potential flaws, it is the best we've had here so far.

Best regards,

Colleen

Indeed.

Hahah! Exactly.

I don't know modern Rolands, but older ones had problems with short sample length (and fast decay), so 88 key sampling doesn't help that much.
In fact, my big question is: are "they" try to make us all fools and purposely make the technical development slower that it can be? I think, that in some aspects, yes. It's easier to make great sounding 50 sample piano than 88; many don't hear difference, so producers save their money.

But anyway, recordings of CP-1 sound great maybe worth about half it's price, but still impressing.

I.......[sings in his best Sinatra impersonation] did it myyyyy waaaaaay.

Scoo-be-doo-bee-doo!

I just uploaded a recording of my trusty old P-80:

http://docs.google.com/leaf?id=0BwW...TI3LWI0NTMtMDAwYjA2NWUwMzM3&hl=en_GB

Hope I did it right!

I've been thinking about replacing it for a while, but not found anything I like much better yet, which probably says more about my hearing than anything.

Cheers, Steve

Yamaha CP1 CFIII Analysis Pix

Just thought people might like to see this, so we can all hallucinate together.

Open them in a new window to see them full size, or check them out at MediaFire.


The above is a spectral phase view of the note C4. Actual sample ~2 seconds, obvious visual looping after that.


The above is a spectral pan view of the note C5. Actual sample ~1.8 seconds, obvious visual looping after that.


The above is a spectral pan view of the lowest notes. Obvious note stretching, both visually and audibly.


The above is a spectral phase view of the highest notes. Obvious visual note stretching.


The above is a spectral phase view of the layer test. There appears to be a layer switch at the cursor, and another 4 steps in from the right side of the screen (I believe I can hear the highest one).


The above is a spectral pan view of the layer test. There appears to be a layer switch at the cursor.

So three layer switches implies a 4 layer sample. The layer blend is very smooth so the layer count is a bit of an educated guess.