PianoMan51

Look how far we've come... [video:youtube]

Go to 1:10. You'll see where the miniature Stonehenge came from!

Would it be conceivable for this forum to have a permanent topic: For Family, Significant Others, and Friends: WHAT THE OTHER GUYS HAVE! Think about it. "Honey, it's not too much to spend! Just go to KC and see WHAT THE OTHER GUYS HAVE!" (For the sake of fairness, perhaps someone can de-misogynize my wording.)

Not if you follow the rules. If the previous note was played in the upper split and the next note is played below the split point, but within the range you have defined below the split as the floating range for the upper split, then it will sound the sound assigned for the upper split. Also you can transpose each split by octave so the pitch sounded is correct irrespective of where it is played on the keybed. All described in the MainStage manual, might be worth reading it. I am a manual reader, and I do understand that there is an algorithm. And if I were playing exactly the same parts for a performance, then I could use the floating split. But when I'm making up parts as I play, which is virtually all the time, it's a whole lot easier to remember a fixed split point than to have to work within the algorithm while devising bass and piano parts. To be fair, the algorithm works 19 times out of 20 for me, but having a bass note boink out during a piano solo just throws me off my stride.

For those of you who are thinking about the wonder of MainStage floating splits, I beg to differ. I spent weeks and weeks using the floating split for bass vs. 'whatever'. It is so wonderful to be able to play bass lines up in the alto range, or take the right hand voice down into the baritone range. The problem is that on a regular basis, especially when I was getting into the music, I would break the algorithm. Hit a note that was expected to be a piano, but get a bass note. Groove=broken. Eventually just turned it off. Too bad.

My only Fatar board is an old K-2500. I've had it open so many times that I believe the Fatar keybed (not keyboard) does not do any electronic processing. There is no 'Fatar board' in the Kurzweil. So, I interpret that the difference between two and three sensors, from the Fatar point of view is that there is an additional set of sensors that the keyboard electronics can parse. This would mean that it's the keyboard manufacturer, not the keybed manufacturer, that utilizes the additional information and then decides what to do with it. This is what I think Nord is doing when they talk about the new Piano 3.

Tautology.

An old topic, but I found myself re-listening to the Progeny live Yes recordings last night. Ended up doing the following review on Amazon: Call this 'The Revealing Science of an Unproduced Yes'. Each player and vocal is positioned as it would be on stage. Jon Anderson and Alan White are center, Steve Howe guitar and vocal are left. Chris Squire bass and vocal are mid-right. Rick Wakeman is far right. Now you get to hear everything, and I mean everything that each person does. I'm a musician and I absolutely love it. It's like pulling a curtain away from the produced recordings and lets you hear what really went into it. Some new things you might hear are Jon's vocal like a tiny but profound voice in a hurricane of sound. And Steve and Rick both sound at times like two flailing maniacs at opposite ends of the stage, playing in two different universes that somehow intersect at crucial times in the most amazing ways. Chris's bass doesn't so much 'hold everything together' as create a musical space where everything else can make sense. BTW, this is all good! What you lose, and I think this will be up to individual taste, are what I think of as the 'softening' effects of the production on the studio albums. On the studio albums Jon's voice might be doubled or tripled to make it sound like it's coming from everything. And the group harmonies are always mixed to make it difficult to hear each of the three parts as distinct. They're panned together as a single 'group' sound. In the studio Steve's guitar licks between the vocals are often panned center, and Rick's harmonizing or answering keyboard lines are often panned to sit in the same place as the original guitar line. These are done, I imagine, to give Yes a more homogeneous tone, where the final work is a 'singular but complicated' musical idea. The Progeny recordings leave out this 'softening production'. Instead of hearing One United Yes we hear five crazily talented musicians producing a joyful sound that must be made understandable by the active listener. If I want to play a Yes tune for someone who doesn't know Yes or isn't a musico, I'll play them the studio recordings. But if I'm drinking some Bourbon and branch with a musician friend, I'll have them take a listen to what I think of as "the Real" Yes.

Finally. Now it makes sense. Thank you. After years of reading your posts, with many thoughtful and interesting observations, you are welcome, and I'm pleased to have helped.

Cool! You've described the end product. I've described how you do it. Two sides to the same coin. Done?

Let me try this one more time. 'Unique' is undefined. 'Not Identical' is undefined. Fugetaboutit . These terms make understanding about this stuff impossible. 1). L is what comes from the Left output on your keyboard. 2). R is what comes from the Right output on your keyboard. 3). L+R is the strict summation of the two signals. (This is the same as using the 'Mono' Left only jack on most boards.) 4). L-R is strictly the sum of the L signal plus the phase inverted R signal. (You might get this if your mixer has a phase button. Or as Zaphod B. has pointed out, if you apply the L and R signals to the two input pins of a balanced XLR input, the input amplifier will output the difference: L-R) This was all done back in 30s or 40s when M/S was invented and used. There's no DSPs, no algorithms, no secret sauce. They didn't have any then, and didn't need it. Just '+' and '-'. This is easy to do in software, or in a well-featured digital mixer, or with a few OpAmps. What's missing from my perspective is a piece of commercially available hardware, most likely a mixer, that can output both L+R and L-R.

Oh man, the last thing I wanted to do was to ruin anyone's fun. Yours included. Former EE here. For years I've bit my tongue when reading some of the posts in this thread. All kinds of magic and voodoo must be involved! And bless him, Aspen was the cause of much of that. But I didn't want to effect his business, given his very long passion for bringing us these products. And they are ground breaking for keys. But over the past year though I've seen more and more people try to jury rig their SS3 into a full range flat response system. Adding a sub. Then adding a quality powered speaker on top. And then you came along as a brother in arms ready to roll your own too, and I started chiming in. And finally one day I had too much time on my hands, put on my professor's hat, and wrote down IMNSHO what's what. If this seems targeted at you (although it did start as a short response to you) I apologize. Wasn't the case. I look forward to reading how your experiments come along.

Hey guys. Actually I was only trying to help by adding some clarity to the conversation. Really, just start out with an engineer's description of M/S recording. The term 'unique' is, intentionally or not, misleading, since it has no strict definition. Go back to the beginning, then you will understand what left and right mean. Aspen's words describing this can be great for a non-technical understanding, but if you really want to get it: figure out M/S. It's gloriously simple once you wrap your brain around it. Did you do your homework? If yes, skip to next step. If not, go back to the beginning. Then when you try to understand translating an M/S recording into L and R stereo and then back to M/S speakers, it will make sense and you won't get lost in translation. Hmm, wasn't that a movie? I loved Scarlett.

For most of us the L+R signal is easy. But creating the L-R requires a phase inversion of the R signal. When I did this I was using all VSTs running under Plogue Bidule. This gave me the ability to create two separate output channels, L+R and L-R. The L+R signal when directly into a 10" self-powered Turbosound speaker. The L-R went into a power amp and then out to an 8" speaker. Not a speaker cabinet, a speaker. Criterion for the speaker was high power handling, high quality mid-range, no coax or whizzer cone for highs, and an open speaker frame to allow the air to move from both sides of the cone. I didn't worry about lows because the open speaker frame naturally attenuates lows no matter what you do, so I didn't even try. In one of Aspen's posts he had mentioned that his favorite spot for locating the side speaker was directly behind the front speaker. So I bought a "NOS Atlas Soundolier 510-8W Cylindrical Bidirectional Baffle - White Finish" on eBay to hold and protect the speaker, and attached it with a small hook to the back of front speaker. B3, Rhodes, Wurly all sounded great through this. Suddenly the whole room felt alive with the sound. Piano not so much. It just seemed to difuse the sound. When I play LH bass I will also put a tiny bit of delay on the signal and route a high-pass version of just the delayed signal to the side speaker. Hope this helps.

Yes, that's exactly what the Fender versions of Aspen's design do. You input a mono guitar, add stereo effects, and the effects sounds are broadcast into the room. Let's start at the beginning. (You're asking why should I listen to this guy? In a previous life I was a telecoms engineer and I re-created the SS3 but with higher quality components two years ago. This is no diss on Aspen. It's a brilliant invention and I wish him all the best) 1). Damn it, Google "M/S recording technology" and really learn it. If you open your mind and do some simple arithmetic you can do this in one hour. 2). Don't go on to the next step before you did step 1. Really. 3). Aspen's genius is to invert the M/S recording process and use it for reproduction. Since an M/S recording is done from a single room position this means that the inverse reproduction system can be done from a single room position. This eliminates so many problems that exist in trying to present a stereo image to an audience. 4). The center channel in an M/S recording is done with an omni mic. This means it captures everything. This is the same as taking a stereo recording and summing the left and right channels into L+R. This is everything. The phrase "common to both L and R signals " is deliberately confusing. It's L+R. And this is what is presented via the front-facing speaker. Technically, L and R signals that are identical and in-phase are boosted 6dB. 5). In M/S recording the second mic is a Figure-8 configuration positioned at 90degrees as close to the Omni mic as possible. To invert this for reproduction, we're going to use a Figure-8 speaker: which is exactly what every standard speaker is when it's not placed in a box. When a Fig-8 mic records, it recognizes sound from both ends of the diaphragm, but they are 180 degrees out of phase. Go back to step 1. If you haven't internalized it stop here. Back to our discussion, the Fig-8 mic has processed left and right directed signals, but in opposite phases. The way to state this exactly is that the Fig-8 mic has recorded L-R. 6). So, to invert and reproduce the original signal we are going to take the two stereo input signals (L and R) and subtract one from the other: L-R. 7). We then amplify L-R and drive the second speaker, positioned at 90 degrees w.r.t. the front speaker. Just like the Fig-8 mic that was used in recording. From one side it sends L-R and from the other side it sends that same signal but 180 degrees out of phase which is -(L-R) which is R-L. 8). That's it. This is not an analogy. This is the way it works. L+R to the front, L-R to the left, and R-L to the right. With two speakers and two amps. 9). My personal experience with this is that acoustic piano benefits slightly from this technology. It's like salt. Use to taste. 10). Keyboard sounds with lots of L-R components do really well with this technology. The result is not the same as a Leslie or as Fender Rhodes Stage, but gives a very sensual feeling of the sound filling the room. Pop quiz on Friday.

I'm not sure I understand the question... It's not a case of the top speaker being mono, or performing as such. The whole tower is an integrated M/S stereo system. I blended the volumes pretty carefully. The top speaker is essentially just acting as an extention of the SS front array. Truth be told, the sound is much bigger than it was with just the SS - the bass boost on the K8 is way helping with that. The stereo bloom is especially fun, too. Can't thank Al and Tim enough for the tip. dB Dave... what does M/S stand for? Mono/Stereo? You also mentioned M/S signal in one of your posts? hatricklov, everything about the SS3 comes from understanding M/S recording theory and then recognizing the contribution that Aspen has made by commercializing an M/S speaker. Google it. Really. He crossed the diamond with the pearl.

I auditioned a Chroma 2-3 months before the DX-7 came out. Bought the DX, and it did me great service, but I really regret not having spent the $5k or so to get the Chroma.

There's an old one from Roger Miller: "You know one time I actually got all my shit together. It was too heavy to lift."

This one. The first record I borrowed in high school that sounded at first, really strange, with a sense of (maybe transformative) menace to, not the body, but to the spirit. This was definitely the gateway drug that set me up really listen to ELP, Yes, KC, etc. when they came out. "gather all my friends unto me, and I'll strangle them with words."

Thanks for the head's up. Read your post and just bought one. I love the L'Otary. It's on a totally different level than the built-in FX in VB3 1.4.

Good work for you!

This is either going to make M-S seem simpler or harder depending on whether you understand it as an engineer or as fluffy marketing-speak. (I've been both.) On the recording end: 1). The center mike (M) is omnidirectional. It picks up sound from all directions in mono. Since our interest is in Left versus Right, let's call this signal L+R. (Rather than Front/Back or Top/Bottom). 2). The side mic (S) is figure-eight and captures sound coming from the Left direction 'in' phase and sound coming from the Right direction '180 deg out of phase'. The recorded S signal is L-R. 3). We do this because when de-coded from M-S to Left/Right stereo the L and R stereo signals can be collapsed to mono without phase distortion. Given the lack of mono playback systems, this method hasn't been used as much lately, because (at least to my ears), the results from separated L and R mics can sound better on stereo playback systems. Check out the mic positions available from Ravenscroft or other piano sample VSTs to hear the difference. To decode the two channel M-S to L and R stereo speakers we: 4). Add M+S: ((L+R) + (L-R)) = 2L which becomes the Left stereo signal. 5). Subtract M-S: ((L+R) - (L-R)) = 2R which becomes the Right stereo signal. What (I believe) the SpaceStation does very cooly is this: 6). Take the two input signals, L and R, add them to become L+R which becomes the original M signal which is amplified as a mono signal and directed towards the front. 7). Take the two input signals, L and R, subtract them to become L-R which becomes the original S signal which is amplified as a mono signal but is transduced by the open-backed speaker into an in-phase sound pointed Left, and a 180 deg out-of-phase sound pointed Right. 8). The audience's ears become the de-coders. Do the arithmetic yourself to get a feel and for extra-credit. Re: the Casio Privia, I've owned one for almost 15 years, a PX-100. The stereo piano sound collapses perfectly into mono. This leads me to think that Casio (back then) started out with a mono piano sound and doctored it to simulate stereo. Or perhaps they used an M-S system for sampling. This works great with the SpaceStation because the front facing speaker is sending a nice mono signal that sounds good by itself. We'll have a pop quiz on Monday!

Mr. Reeze, As a Bidule convert I tried the same thing last fall. It was fun. Clone VB3 through L'Otary sounds dynamite, acoustic piano not so much. Here's what I did: 1). Recognize that Aspen's model is the inverse of M-S recording, so the front speaker is L+R and the side speaker is L-R. (those are mathematical adding and subtracting in Bidule) 2). For the side speaker I just used an 8" speaker, freestanding, and placed it directly behind my 10" powered speaker box. When you set up a Bidule mixer for L+R into one channel and L-R into a second channel, you can immediately how much real difference is in the original L and R signals. For instance, the internal Leslie sim in VB3 has very little difference between the L and R. L'Otary has much much more, and therefore sounds really great with the pseudo SpaceStation. Here's what I hypothesize about acoustic piano and the SpaceStation concept. If the piano was originally recorded in M-S or in phase-aligned stereo, all is good. The front facing speaker plays a signal that collapses into mono nicely, and the side speaker plays the ambient parts. But if the piano was originally recorded with a widely spaced set of microphones (to sound optimal in a widely spaced pair of speakers) when the left and right are collapsed into the (L+R) front, we hear a comb-filtered piano from the main front speaker and it sounds bad, no matter what comes from the side speaker. Mr. Man

No.

Nice post Dave. I notice your line: "There were a whole bunch of posters there who had an entirely different vibe". It's the vibe here that makes this forum so attractive. And it's just like making music, you have to listen to everyone else, care about the result, and add in what's missing. There's an old Dale Carnegie quote: "People don't care what you know until they know that you care". Thanks to you, our mostly hidden conductor, and all rest of you consistently caring KC members!