Opinions on Kurzweil K2500 and Alesis QS8 ???
david raila
I'm looking seriously at these two... although I can't get my
hands on the QS8, I have heard the QS7 and am hoping the action is good
on the QS8. Although the K2500 is a fully integrated system, I have
lots of PC hardware lying around, so the QS8 with the provided software
may be a nice way to go.
Anyway, I wonder if anyone has opinions on features of
these machines. I've said very little to avoid direvting any opinions.
Thanks
David
Petteri Ihalainen
david raila (ra...@cs.uiuc.edu) wrote:
: I'm looking seriously at these two... although I can't get my
K2500 = Lexus
QS8 = Beetle
--
Petteri Ihalainen at Elma Oy
ihal...@elma.fi || http://www.elma.fi/ || 040 555 9223
DaveBryce9
<< K2500 = Lexus
QS8 = Beetle >>
Upon what are you basing this qualitative analysis?
dB
Alesis Keyboards
Adam Levin
david raila (ra...@cs.uiuc.edu) wrote:
: I'm looking seriously at these two... although I can't get my
: hands on the QS8, I have heard the QS7 and am hoping the action is good
: on the QS8. Although the K2500 is a fully integrated system, I have
: lots of PC hardware lying around, so the QS8 with the provided software
: may be a nice way to go.
: Anyway, I wonder if anyone has opinions on features of
: these machines. I've said very little to avoid direvting any opinions.
I have no experience on the Kurz. The action on the QS8 is quite
impressive -- a little softer than I'd like, but quick and still quite
good.
The price difference is pretty big, obviously. The Kurz wasn't an
option for me because I'm not really a synth programmer, at least not
yet. I didn't want a sampling type machine, although I have the
option with the Alesis.
I think it depends what you're looking for. If you want full control
over any sound, get the Kurz. If you want some control over sounds,
but you want something that works great right out of the box and
doesn't cost as much as a Hyundai, go for the Alesis. They really are
rather different instruments.
-Adam
--
Adam Levin -- ale...@stevens-tech.edu -- http://www.stevens-tech.edu/~alevin
O- Do you ever get the feeling that the story's too damn real and in the
present tense, or that everybody's on the stage and it seems like you're
the only person sitting in the audience?
PGP fingerprint: 33 30 6F 10 3E 46 FA 20 A5 03 A8 1A FB E2 25 3A
J Durham
Adam Levin wrote:
>
> david raila (ra...@cs.uiuc.edu) wrote:
> : I'm looking seriously at these two... although I can't get my
> : hands on the QS8, I have heard the QS7 and am hoping the action is good
> : on the QS8. Although the K2500 is a fully integrated system, I have
> : lots of PC hardware lying around, so the QS8 with the provided software
> : may be a nice way to go.
> : Anyway, I wonder if anyone has opinions on features of
> : these machines. I've said very little to avoid direvting any opinions.
They're not comparable, really. However, if you want an analogy, I'd
say: one is a Hyundai and the other is a Cadillac. They'll both get you
_there_, but only one has the optional seat-warmer. You guess which.
On the other hand, the actual keyboard "action" is another issue. I'm
simply talking about the real innards.
YoungChang
>> K2500 = Lexus
>> QS8 = Beetle
> Upon what are you basing this qualitative analysis?
Ignorance. What else?
-Y(?)
Athanasios(Tom) Zougas
daveb...@aol.com (DaveBryce9) writes:
><< K2500 = Lexus
> QS8 = Beetle >>
>Upon what are you basing this qualitative analysis?
I'd say price: sell your Lexus so you can afford a K2500; sell your Beetle
so you can afford a QS8 :-)
DaveBryce9
<<They're not comparable, really. However, if you want an analogy, I'd
say: one is a Hyundai and the other is a Cadillac. They'll both get you
_there_, but only one has the optional seat-warmer. You guess which.
On the other hand, the actual keyboard "action" is another issue. I'm
simply talking about the real innards.>>
As a former Kurzweil product specialist and a current Alesis one, perhaps
I might comment.
If you listen to the raw sounds on both, they're quite comparable...our
base piano sample is much bigger (around 7 meg, which is one meg short of
K2500's entire base memory), stereo (not a mono piano sample panning from
left to right - real pianos don't do this), and sampled at a higher rate.
Ours is also not compressed. None of our samples are. K2500 is a 48
voice, QS8 is 64 voice (yes, K2500 is capable of 196 voices, but only with
about 5 DSP waveforms - square, saw, sine, etc.). K2500's onboard effects
are a Digitech DSP256XL (6 years old) 16 bits internal capable of 4 fx on
one bus, ours is basically a Q2 - 8 fx across four discrete busses, 24
bit internal resolution. Except for the drum channels, K2500 can only do
3 layers, QS8 can do four. QS8 has a computer port and a digital output
standard...K2500 may or may not have the dig out depending on whether or
not the DMTi ever comes out (last known list price of DMTi was $1999, same
as the base price of QS8), and will never have a computer port. K2500 has
8 meg base ROM, expandable to 28 with daughter board and Rom blocks A and
B (approximate cost = $1050.00), while QS8 starts with 16 meg, and can go
to 32 much less expensively (around $500.00 max). K2500 can only transmit
8 channels of MIDI from the keyboard at once, QS8 can do 16. Also, the
MIDI response time of QS8 is significantly faster then K2500, making it a
superior controller (see Keyboard magazine article on MIDI timing a few
months back - no one is faster or more accurate than we are).
Conversely, K2500 has resonant filters, an onboard sequencer (the best one
available in a workstation), and can sample...up to 128 meg (which will
cost a pretty penny in SIMMS). QS8 can only download 100 sequences from
computer to flash cards (regardless of length and resolution of the
sequences, so you can store a hundred 200,000 event sequences at 480 ppq)
and 16 meg of sample to flash ROM (approximate price = two 8 meg cards @
$179 each). Because the user samples become waveform ROM when burned to
flash cards, load time for 16 meg is under a second; and, the samples stay
in the machine when it is turned off. The sequences on the card play back
instantly.
However, for the price of a fully loaded K2500, you can get a QS8, an
ADAT, a brand new computer (which will wipe out any onboard sequencer), a
sampling program (which will outperform any dedicated sampler), a
sequencing program, a small mixing board, and take a vacation. This only
leaves the resonant filters...how much is it worth to have these?
Which is the Cadillac, and which is the Hyundai? Which serves most
people's needs better? Admittedly, there are people for whom each of
these instruments is a fine choice...
Also ***in my opinion***, the QS8's pianos, organs (we sampled Emerson's
C3!), clavs, Rhodes and drums are much more to my liking...our synth
samples come from Emerson's modular moog, Oberheim Matrix 12 and OB-8 and
a host of others, and knock my socks off (Keith likes it, too...he used
two on this last tour - one on the Steinway MIDI grand, and one inside it
(the Steinway's MIDI response wasn't fast enough for him). BTW, he's not
an endorsee...) - but then again, I work for Alesis, so who'll believe me?
Bet this post elicits a response or two...let the flame wars begin!
dB
Alesis Keyboards
Frank Boon
DaveBryce9 wrote:
> Also ***in my opinion***, the QS8's pianos, organs (we sampled Emerson's
> C3!), clavs, Rhodes and drums are much more to my liking...our synth
> samples come from Emerson's modular moog, Oberheim Matrix 12 and OB-8 and
> a host of others, and knock my socks off (Keith likes it, too...he used
> two on this last tour - one on the Steinway MIDI grand, and one inside it
> (the Steinway's MIDI response wasn't fast enough for him). BTW, he's not
> an endorsee...) - but then again, I work for Alesis, so who'll believe me?
> Bet this post elicits a response or two...let the flame wars begin!
> dB
> Alesis Keyboards
Dave,
I like it when you do this! First stating all the dry facts (like
allways)... but this time you couldn't hold you [personal] horses:
hehehe, good work.
BTW: I have a friend who I jam with sometimes: he owns a K2500 and I
have a (humble) QS6. And he is really amazed of the sound quality,
fatness, and sound sources come out off my little box. And he paid 2,5
times it's price.
Cheers,
Frank
--
Frank Boon Delft Institute for Earth-Oriented Space Research
phone : +31 15 278 6221 mailto:frank...@lr.tudelft.nl
fax : +31 15 278 5322 http://dutlru8.lr.tudelft.nl/~frank
Blake Stone
DaveBryce9 (daveb...@aol.com) wrote:
: As a former Kurzweil product specialist and a current Alesis one, perhaps
: I might comment.
: If you listen to the raw sounds on both, they're quite comparable...our
: base piano sample is much bigger (around 7 meg, which is one meg short of
: K2500's entire base memory), stereo (not a mono piano sample panning from
: left to right - real pianos don't do this), and sampled at a
: higher rate.
Anyone serious about solo piano will surely pick up the Piano ROM
for the Kurzweil, which is stereo, and sampled at a much better
rate than the on-board one. The size of a sample doesn't impress
me nearly as much as the sound, and I've never heard a more
realistic piano than my K2500X. Beyond being a beautiful sample,
the ability to shape the sample with the Kurzweil's filters makes
it an expressive instrument.
: Ours is also not compressed. None of our samples are.
Lets be clear on the compression issue ... the Kurzweil ROM
compression is only amplitude compression. It allows the decay
of a sound to utilize the full 16-bit dynamic range, and is not a
"lossy" compression technique like Roland's.
: K2500 is a 48 voice, QS8 is 64 voice (yes, K2500 is capable of
: 196 voices, but only with about 5 DSP waveforms - square, saw,
: sine, etc.).
The K2500 can also accomplish a great deal more with fewer
voices. VAST processing allows complex tricks using a single
voice where traditional architectures would need several layered
voices.
: K2500's onboard effects are a Digitech DSP256XL (6 years old)
: 16 bits internal capable of 4 fx on one bus, ours is basically
: a Q2 - 8 fx across four discrete busses, 24 bit internal
: resolution.
Too true. The effects on the Alesis offer great value for your
money. Until the KDFX ships later this year, the K2500's effects
are its weakest link.
: Except for the drum channels, K2500 can only do 3 layers, QS8
: can do four.
Try turning that around. On 8 of the 16 MIDI channels, the K2500
can do 32 layers, while the QS8 can do only 4.
: QS8 has a computer port and a digital output standard...K2500
: may or may not have the dig out depending on whether or not the
: DMTi ever comes out (last known list price of DMTi was $1999,
: same as the base price of QS8), and will never have a computer
: port.
Excuse me? The DMTi is a format and sample rate conversion tool,
in a completely different league than the QS8's digital out. Any
K2500 with the KDFX or sampling option has a digital output. As
for a "computer port" it's hard to ignore the standard SCSI port
with full SMIDI implementation on the K2500.
: K2500 has 8 meg base ROM, expandable to 28 with daughter board
: and Rom blocks A and B (approximate cost = $1050.00), while QS8
: starts with 16 meg, and can go to 32 much less expensively
: (around $500.00 max).
Of course, you could add 32MB of RAM to the K2500 for around
$250, and access any sample library you are interested in.
: K2500 can only transmit 8 channels of MIDI from the keyboard at
: once, QS8 can do 16.
I have NEVER needed a 16-way split or layered keyboard, but I
make daily use of the ribbon controllers and flexible assignment
of other master keyboard options that I've never seen on any
instrument but the K2500.
: Also, the MIDI response time of QS8 is significantly faster
: then K2500, making it a superior controller (see Keyboard
: magazine article on MIDI timing a few months back - no one is
: faster or more accurate than we are).
The timing question is an issue of how fast an instrument
responds to MIDI, not how fast it sends it. This has nothing to
do with its suitability as a controller.
: Conversely, K2500 has resonant filters ...
To dismiss VAST as "resonant filters" is doing the Kurzweil a
huge disservice. I hope you were a little more forthcoming when
you were a Kurzweil product specialist. VAST offers a huge array
of real-time sound shaping tools which no other instrument can
come close to touching.
: ..., an onboard sequencer (the best one available in a
: workstation), and can sample...up to 128 meg (which will cost
: a pretty penny in SIMMS).
About $1000, which is peanuts for 12 MINUTES of stereo CD-quality
sampling RAM.
: However, for the price of a fully loaded K2500, you can get a
: QS8, an ADAT, a brand new computer (which will wipe out any
: onboard sequencer), a sampling program (which will outperform
: any dedicated sampler), a sequencing program, a small mixing
: board, and take a vacation.
Aye, there's the rub. The K2500 isn't for everyone. It isn't
inexpensive, but it is an incredible instrument and continues to
evolve and expand as new hardware options and operating system
versions are released. The QS8 is a marvel of affordability, and
has a pretty reasonable array of controller features centered
around a decent feel. It cheapens your image to try to pretend
that it is also a match for a much more capable instrument.
--
-------------------------------------------------------------------------
Blake W. Stone bst...@dkw.com
Technical Director - DKW Systems "Art may imitate life, but life
http://www.dkw.com/bstone imitates TV" - Ani Difranco
mid...@erols.com
LLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL
Congratulations on a VERY coherent response to some
rather audacious comparisons/remarks.
May I add?...
Blake sez... "It cheapens your image to try to pretend
that it is also a match for a much more capable instrument."
Hey. Way too late for alesis to worry about a cheap image! Some
GREAT alesis gear here and there but the cheap image is in place. Would
that "small mixing board" be the one alesis put out a while back?
Also I am weary of defending the kurz effects. THEY SUCK. However
the alesis effects are not the answer.
alesis sez..."However, for the price of a fully loaded K2500, you can
get a QS8, an ADAT, a brand new computer (which will wipe out any
onboard sequencer), a sampling program (which will outperform
any dedicated sampler), a sequencing program, a small mixing
board, and take a vacation."
Personally, I would rather take my sig other to the local
motel 6 ($62) for that vacation and return home to my K2500RS/mac/DA88s.
I skipped SEVERAL vacations to buy this stuff.
Anne
mid...@erols.com
LLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLLL
Congratulations on a VERY coherent response to some
rather audacious comparisons/remarks.
May I add?...
Blake sez... "It cheapens your image to try to pretend
that it is also a match for a much more capable instrument."
Hey. Way too late for alesis to worry about a cheap image! Some
GREAT alesis gear here and there but the cheap image is in place. Would
that "small mixing board" be the one alesis put out a while back?
Also I am weary of defending the kurz effects. THEY SUCK. However
the alesis effects are not the answer.
alesis sez..."However, for the price of a fully loaded K2500, you can
get a QS8, an ADAT, a brand new computer (which will wipe out any
onboard sequencer), a sampling program (which will outperform
any dedicated sampler), a sequencing program, a small mixing
board, and take a vacation."
Personally, I would rather take my sig other to the local
KURZWEIL
As to comparing the sound quality of the instruments, I leave it up to the
end use r to decide - that is a subjective opion. Go play both instruments
and see what you think.
However, I do have a few factual comments regarding Dave Bryce's
statements.
It is true that the stock piano sample on the K2500 is mono. However, we
do have a 4 meg stereo piano sample which comes on the daughterboard
option. Admittedly, this is something that adds to the cost of the
instrument. Dave also mentions the fact that our ROM samples our
compressed, seeming to imply that this is somethig negative. In fact,
there is nothing negative about this at all. Kurzweil has a proprietary
method of storing samples on ROM which allows us to cram much more
information onto the chip. This allows us to give you more sample roots
and longer samples for the same amount of room in ROM. So just because
Alesis uses 7 megs for their piano, it doesn't mean that we can't use 4
megs and have samples that compare in quality
Again, I suggest you compare the two instruments, listening to the stereo
piano in the 700s bank. Your ears can decide which you like best.
Yes, the onboard effects processor is a Digitech 256. With the release of
the KDFX option sometime early next year, you will be able to add our own
very high end effects chip. I don't have details on it now, but I am
certain people wil be suitably impressed by its quality. It iwill offer
VAST like realtime control capabilities. Again, admittedly, this ads to
the cost of the unit.
>>Except for the drum channels, K2500 can only do 3 layers, QS8 can do
four.<< This one is really a hoot. 8 of the 16 channels can be designated
as "drum channels". those channels can contain programs which are up to 32
layers, not 4!. the other 8 channels are limited to 3layer programs.
needless to say, the Kurzwiel is more flexible in this issue.
The K2500 does indeed have digital out. With the sampling option
installed, yo have two channels of digital out. With the KDS out option
connected to the DMTi, which is a digital audio sync box, you have 8
channels of digital out. The DMTi and KDS out should start shipping next
month. List price on the DMTi is $1310. I don't have a price yet on the
KDS Out.
Dave's comments on the cost of the ROM block upgraes are correct. I would
only comment that what you are paying for is the Kurzweil quality samples
as well as the incredibly detailed programming for the presets.
Dave's comments about response time making for a superior controller are
curious. The response time has to do with how the instrument responds to
MIDI messages, whether from an external source or from itself. I will not
dispute the fact that Alesis has excellent response time. But controller
capability has to do with the sending of information, not receiving. I
don't know enough about the QS8 to comment on its controller capabilities,
but the K2500 is truly outstanding in this area. You have 8 sliders, 2
ribbon controllers, pitch and mod wheels, mono pressure, 4 footswitch
jacks, 2 continuous contoller pedal jacks, and a breath controller input
(which can be used in place of the second CC pedal). Each of these
physical controllers can send a different MIDI message on each of eight
MIDI channels - in other words, moving a slider could send 8 different
messages if you want. You also have a very sophisticated and programmable
arpeggiator, which is part of the controller capabilities.
Dave also mentions the resonsnant filters, but that really only is the tip
of VAST There are 60 different DSP functions capable of modifying sound,
including the resonant filters. There are 17 different types of filters
alone. From any objective viewpoint, VAST is still the most powerful
synthesis engine available in any commercially available product out
there.
Finally, I doubt Dave's claimn that you can get a QS8, computer, sampling
hardware and software, sequencing software, and a mixing board for the
price of the K2500 - but I don't want to waste time arguing numbers.
The bottom line is that the K2500 is clearly a much more expensive
instrument than the QS8, but we believe that its capabilities are clearly
worth that price difference. Many people seem to agree. You will have to
decide yourself.
David Fox
Senior Product Specialist
Kurzweil Music Systems
Cliff Chang
As an Alesis user, I have a few things to say about Alesis product in
general:
1) They try to cut the cost too much here and there beyond reasonable
limit, especially in quality control. Some people are so grateful how
"helpful" Alesis was when they had a problem with Alesis gear, and
(almost) forget that the whole problem originated from Alesis. Of course
it should be taken for granted that the manufacturer be responsible for
their product for a reasonable period of time ("We guarantee this will
work for 90 days or probably longer" does not impress me.) Why can't
Alesis spend little more on quality control than just sticking one of
those "Q.C" stickers outside?
2) I think the samples of acoustic instruments on the SoundCanvas
are more realistic than those in QS-series.
Cliff Chang
--
*-------------------------------*
* *
* Cliff zenta Chang *
* *
Ben Cox
DaveBryce9 <daveb...@aol.com> wrote:
>However, for the price of a fully loaded K2500, you can get a QS8, an
>ADAT, a brand new computer (which will wipe out any onboard sequencer), a
>sampling program (which will outperform any dedicated sampler), a
>leaves the resonant filters...how much is it worth to have these?
Heck: leave off the vacation and pick up an analog synth instead, and
now you've got the resonant filters as well.
__
Ben Cox c...@transarc.com
DaveBryce9
<<As to comparing the sound quality of the instruments, I leave it up to
the
end use r to decide - that is a subjective opion. Go play both instruments
and see what you think.>>
I couldn't agree more...
<<Dave also mentions the fact that our ROM samples our
compressed, seeming to imply that this is somethig negative. In fact,
there is nothing negative about this at all.>>
Nor did I say there was...however, it is my opinion that some longer
samples suffer when compressed, as does all digital audio.
<<Again, I suggest you compare the two instruments, listening to the
stereo
piano in the 700s bank. Your ears can decide which you like best.>>
Yes, Dave...but would you put your stock piano against my stock piano?
Ours isn't an option
<<With the release of
the KDFX option sometime early next year, you will be able to add our own
very high end effects chip.>>
Another expensive option...the Q2 chip in our unit is, once again, stock.
<<This one is really a hoot. 8 of the 16 channels can be designated
as "drum channels". those channels can contain programs which are up to 32
layers, not 4!.>>
Yeah, but how far does that go with 48 voices?
<<The K2500 does indeed have digital out. With the sampling option
installed, yo have two channels of digital out.>>
Digital out is standard on QS8
<< With the KDS out option
connected to the DMTi, which is a digital audio sync box, you have 8
channels of digital out. The DMTi and KDS out should start shipping next
month. List price on the DMTi is $1310. I don't have a price yet on the
KDS Out.>>
Ours has it right now. List price is $0.
<<The response time has to do with how the instrument responds to
MIDI messages, whether from an external source or from itself. I will not
dispute the fact that Alesis has excellent response time. But controller
capability has to do with the sending of information, not receiving.>>
Perhaps I can explain what I mean. MIDI spec sez that an event @ 31,250
kbaud (MIDI's data rate) will take about 1.3 milliseconds (source: Craig
Anderton), so a note played (three events - note on, note # and note off)
will take about 4 ms. The QS8's synth engine responds to it's keyboard
being played in 6 ms. This is lightning fast. I believe K2500 was over
25 ms., although I don't have the article with me, and I could be wrong...
<<There are 60 different DSP functions capable of modifying sound,
including the resonant filters. There are 17 different types of filters
alone. From any objective viewpoint, VAST is still the most powerful
synthesis engine available in any commercially available product out
there.>>
Fully agreed.
<<Finally, I doubt Dave's claimn that you can get a QS8, computer,
sampling
hardware and software, sequencing software, and a mixing board for the
price of the K2500 - but I don't want to waste time arguing numbers.>>
I'm sure...because you're wrong:
QS8 = $1999
ADAT XT = $3499
Mackie 1604 = $1099
Performer = $495
Alchemy: $369
PC = can be had for around $1000
Total = $8461 (all list prices)
Please, Dave...total up a K2500X, with sampling option, KDS option, DMTi
option, daughterboard option, ROM blocks A and B options, PRAM option, and
16 meg of SIMMs..I'll bet the difference is enough to take a vacation on.
I firmly believe the K2500 is a more powerful synth as far as
programmability. I just as firmly believe that the averae user doesn't
edit, based on the fact that most synths that come in for repair or sell
used have 100% of the factory patches intact. All I'm saying is I think
we've taken the average musician's need into account much better than
we're given credit for. Remember, my last post was in response to two
straight comparative slams that I felt were unwarranted... I wasn't
picking on yer stuff...I own and use yer stuff, and you know it...I was
just defending my stuff from what I perceived to be unprovoked and
uninformed attacks. I know your instrument real well...come see me at
AES, and I'll show you around an 8...you may be surprised...I know I was.
dB
Alesis Keyboards
Carl Christensen
daveb...@aol.com (DaveBryce9) wrote:
>Also ***in my opinion***, the QS8's pianos, organs (we sampled Emerson's
>C3!), clavs, Rhodes and drums are much more to my liking...our synth
>samples come from Emerson's modular moog, Oberheim Matrix 12 and OB-8 and
>a host of others, and knock my socks off (Keith likes it, too...he used
>two on this last tour - one on the Steinway MIDI grand, and one inside it
>(the Steinway's MIDI response wasn't fast enough for him). BTW, he's not
Jeez, I just picked up a QS8 no wonder I liked it (plus the dealer gave me
"an offer I couldn't refuse" ;-). The first thing I did when I bought it
was go over to my band's drummers house and jam with him, work through
Tarkus, etc! ;-)
I was a bit disappointed not to get the CD inside, but rather a coupon for a
backordered one. Other than that it's nice to have 88 keys to noodle around
on as well as some great sounds (IMHO) to bang around on. I admit it
though, I'm the type that doesn't want to program sounds on a synth anymore,
I'd rather write a computer program or practice!
Sebastian Nyberg
Well Jesus Christ on a motorbike, this is just what we need: Korg and
Alesis at each other's throat. I've got a good suggestion: why don't
both companies refrain from posting to this newsgroup. We have both
Korg and Alesis users on here who are prepared to debate differences
between the two synths to death. It's a sad day when r.m.m.s. will turn
into a battleground for synth companies. Leave it to the users,
please.
Base
nwb...@ix.netcom.com
daveb...@aol.com (DaveBryce9) wrote:
>Because the user samples become waveform ROM when burned to
>flash cards, load time for 16 meg is under a second; and, the samples stay
>in the machine when it is turned off. The sequences on the card play back
>instantly.
What is the process for doing this? Does a computer based program take
standard files (eg .WAV, etc) and convert them for download? If so,
does it convert all sub-formats (8 bit, 16 bit, mono, stereo)? Is
there any compression involved in the manipulation?
I am considering a QS8, but have not been able to find one yet - maybe
this weekend!
Also, are the 64 voices on the QS8 reduced as layers are utilized?
Different architectures define this differently - enough so the
question may be have little meaning without a large expanation.
Thanks for the info.
Nate
Blake Stone
DaveBryce9 (daveb...@aol.com) wrote:
> > Dave also mentions the fact that our ROM samples our
> > compressed, seeming to imply that this is somethig
> > negative. In fact, there is nothing negative about this at
> > all.
> Nor did I say there was...however, it is my opinion that some
> longer samples suffer when compressed, as does all digital
> audio.
Once again, I'd like to point out that the compression is
strictly amplitude compression. When reconstructed using the
"natural" envelope of the sound you get a 100% faithful
reconstruction of the sound. No "suffering", no deviation from
the original. The technique is completely lossless.
So why bother? As a sound decays, it normally uses a limited
subset of the 16-bit dynamic range that is available. The
amplitude compression technique ensures that the full dynamic
range is expressed even in softer portions of the sample
... something no other manufacturer bothers with.
> > This one is really a hoot. 8 of the 16 channels can be
> > designated as "drum channels". those channels can contain
> > programs which are up to 32 layers, not 4!
> Yeah, but how far does that go with 48 voices?
Oh come on. You aren't really foolish enough to believe that
stacking sounds is the only possible use for multiple layers, are
you? Delayed layers work wonders for "wavesequencing" effects,
key release layers come in handy for a variety of uses, subtle or
drastic changes in modulation characterics for different zones of
the keyboard can eat up layers, etc. More layers does not equal
less polyphony, just more flexibility.
> > With the KDS out option connected to the DMTi, which is a
> > digital audio sync box, you have 8 channels of digital
> > out. The DMTi and KDS out should start shipping next
> > month. List price on the DMTi is $1310. I don't have a price
> > yet on the KDS Out.
> Ours has it right now. List price is $0.
Once again ... the KDS / DMTi combination is a very different
beast than the QS8 digital out. These offer conversion between a
variety of formats, and include asynchronous sample rate
conversion. In addition to 8 channels of K2500 sound, they
include inputs for two additional AES/EBU inputs (one of which
can be switched to optical input) for the same kinds of
conversion. Four KDS-format outputs, plus multitrack interface
outputs plus four indepedant AES/EBU stereo ouputs and an optical
output put this WAY beyond the simple digital ouput league. Very
few people will ever need to buy these upgrades.
> Perhaps I can explain what I mean. MIDI spec sez that an event
> @ 31,250 kbaud (MIDI's data rate) will take about 1.3
> milliseconds (source: Craig Anderton), so a note played (three
> events - note on, note # and note off) will take about 4 ms.
Firstly, you should learn a little more about the subject before
pretending to be able to make technical judgements. A note on
event consists of three bytes: note on & channel + note # +
velocity, and is a single event that takes ~1ms to transmit.
A note off arrives only when the key is released and is unrelated
to the latency of the initial sound.
> The QS8's synth engine responds to it's keyboard being played
> in 6 ms. This is lightning fast.
If this were true, it would be pathetically slow. In any case,
Keyboard magazine's measurements (which you are misrepresenting)
were only regarding response to MIDI in from a sequencer, having
nothing to do with the local keyboard.
> I believe K2500 was over 25 ms., although I don't have the
> article with me, and I could be wrong...
I have the article in front of me. The K2500's average latency
was actually 45 sample words, or 1.02ms. You should get these
things correct to avoid embarassment if you're lucky, or a
lawsuit if you aren't.
> I firmly believe the K2500 is a more powerful synth as far as
> programmability. I just as firmly believe that the averae user
> doesn't edit, based on the fact that most synths that come in
> for repair or sell used have 100% of the factory patches
> intact.
Of course, since the factory programs take advantage of this
programability, the user who never edits a single program takes
advantage of the synthesis engine, too.
(By the way, the statistic given above is an oft-used quote from
the era when users reset their synthesizers to factory defaults
to keep repair shops and subsequent buyers from stealing their
sounds.)
> All I'm saying is I think we've taken the average musician's
> need into account much better than we're given credit for.
Great. The Alesis QS8 ... a product for the average musician in
all of us.
Don't get me wrong. I understand that the QS8 is a great
instrument, and probably has a broader market appeal than the
K2500. I just fail to understand why you need to resort to
misinformation to represent your product, and I will continue to
correct you until you get your facts straight.
DaveBryce9
<<What is the process for doing this? Does a computer based program take
standard files (eg .WAV, etc) and convert them for download?>>
<< If so,
does it convert all sub-formats (8 bit, 16 bit, mono, stereo)?>>
Yes.
<< Is
there any compression involved in the manipulation?>>
No.
I am considering a QS8, but have not been able to find one yet - maybe
this weekend!
<< Also, are the 64 voices on the QS8 reduced as layers are utilized?>>
If by this do you mean "Does one note played on a four layer patch result
in four voices being used, yes. We have a variable retrigger voice
stealing algorithm called Sound Overlap that's among the most intuitive
with which I've had the pleasure to work. Basically, set to 0, it
retriggers the same oscillator over and over (1, 1, 1...etc.), while a
setting of 99 goes (1, 2, 3...64) before retriggering, with 98 settings in
between. This is available on a per layer basis.
Hope that answers your questions.
dB
Alesis Keyboards
Revad
>daveb...@aol.com (DaveBryce9) wrote:
>>Because the user samples become waveform ROM when burned to
>>flash cards, load time for 16 meg is under a second; and, the samples stay
>>in the machine when it is turned off. The sequences on the card play back
>>instantly.
Are the flash cards re-writeable?
Jonathan Elliott
nwb...@ix.netcom.com wrote:
> I am considering a QS8, but have not been able to find one yet - maybe
> this weekend!
>
> Also, are the 64 voices on the QS8 reduced as layers are utilized?
> question may be have little meaning without a large expanation.
An oscillator is an oscillator is an oscillator. If you've got 64 of 'em
then each layer uses one oscillator and a four layer patch, for example,
uses four oscillators per note for a total of 16 note polyphony.
The only real exception to this is the idea of a sub oscillator, as found
on a heap of old Roland synths (open wide and say "Juno") as well as
(albeit in a somewhat more sophisticated form) on the Kurzweil K2k/K2500.
Sub oscillators have less flexibility than the real McCoy. In the former
case, the beasties are square wave only and have a fixed pitch
relationship (usually -1 or -2 octaves) to the root pitch (i.e. the pitch
_before_ modulation is accounted for) of the actual oscillator. In the
latter case, although I'm not completely sure, I think that the major
limitation is that the subs are square waves only but can be pitch
modulated either with or independantly from the actual oscillators.
--
Jonathan.
---
elli...@dirvet.tas.gov.au
#INCLUDE Standard_Disclaimer
-----BEGIN GEEK CODE BLOCK-----
GCS d+ s+;- a- C+ U P+ L W N+ o? K?
w---$ O- M++ V PS+++ PE-- Y PGP t+@
5 X R !tv b+++ DI? D+ G e++ h r--
y--**
-----END GEEK CODE BLOCK-----
Adam Levin
Sebastian Nyberg (bas...@beta.hut.fi) wrote:
: Well Jesus Christ on a motorbike, this is just what we need: Korg and
Erm, this would be Alesis and Kurzweil, not Alesis and Korg. Same
difference, I suppose...
-Adam
Ps: If I own a Korg and an Alesis, do I have to argue with myself?
Bill Gates
Sebastian Nyberg (bas...@beta.hut.fi) wrote:
: I've got a good suggestion: why don't
: both companies refrain from posting to this newsgroup.
I firmly DISagree. Dave has been an excellent source of Alesis info for
me, and having reps from other vendors here can only help the signal/noise
ratio.
Bill
Lawrence Lougheed
>I firmly believe the K2500 is a more powerful synth as far as
>programmability. I just as firmly believe that the averae user doesn't
>edit, based on the fact that most synths that come in for repair or sell
>used have 100% of the factory patches intact.
So? This doesn't prove anything. I have traded several synths and have
always restored factory defaults before taking the synth to the store. It's
not that I minded sharing my programming, but more a matter of common
courtesy. Kind of like pressing reset on a photocopier at the office, so that
the next user doesn't get any surprises.
=========================================================
Lawrence Lougheed
loug...@wat.hookup.net
Original Compositions for Roland JV-1080 and General Midi
***Coming Soon*** Original Compositions for Kurzweil K2500
http://www.hookup.net/~lougheed/
=========================================================
Chan Benson
: So why bother? As a sound decays, it normally uses a limited
: subset of the 16-bit dynamic range that is available. The
: amplitude compression technique ensures that the full dynamic
: range is expressed even in softer portions of the sample
: ... something no other manufacturer bothers with.
I find the phrase "the full dynamic range is expressed even in softer
portions of the sample" to be semi-nonsensical. Could you please
clarify? I understand how one might scale the mapping range of the
samples, but then it would seem that you need to store information
about how this mapping changes over the life of the sample. Where is
that information kept? In addition, it's unclear to me how this
technique could boost dynamic range while remaining lossless. You
either divide by the same factor that you multiplied by, or you risk
losing precision somewhere along the line.
Please enlighten me.
-- Chan
Blake Stone
Chan Benson <ch...@fc.hp.com> wrote in article
<553gvv$9...@fcnews.fc.hp.com>...
> : So why bother? As a sound decays, it normally uses a limited
> : subset of the 16-bit dynamic range that is available. The
> : amplitude compression technique ensures that the full dynamic
> : range is expressed even in softer portions of the sample
> : ... something no other manufacturer bothers with.
>
> portions of the sample" to be semi-nonsensical. Could you please
> clarify? I understand how one might scale the mapping range of the
> samples, but then it would seem that you need to store information
> about how this mapping changes over the life of the sample. Where is
> that information kept?
It is stored with the "natural" envelope for the sound, which is an
available
option when describing the amplitude envelope.
> In addition, it's unclear to me how this technique could boost dynamic
> range while remaining lossless. You either divide by the same factor
> that you multiplied by, or you risk losing precision somewhere along the
> line.
When playing back a sample at its normal volume without modulation, sure,
you recreate the exact same sound you'd get without this compression. But
what happens if you are playing back a sound with a variety of modulation
types ...
A) Looping a sound
Looping a sound that has been adjusted to a common amplitude is much
easier, and much more natural sounding. You can avoid some of the
"seasick"
feeling all too many loops induce.
B) Playing a sound at high volumes
Since most instruments have 20-bit D/A converters, if you are playing
back
individual samples at high volumes, you may actually be able to hear the
extra
dynamic range during the quiet portion of the sound.
C) Non-instrumental uses
If you're using the character of the original sample and altering it
using sound
shaping tools, you are no longer restricted to using the "natural"
envelope. Now
the entire duration of the sample provides useful timbral material for
synthesis.
--
Note that all of this only applies to ROM block samples. User samples are
stored
as the typical 16-bit linear sample we all know and love, since
representing the
extra information would require >16-bit A/D conversion and sample
processing
tools beyond the ken of a "normal" sampling repitoire.
- Blake Stone
Martin Cracauer
bas...@beta.hut.fi (Sebastian Nyberg) writes:
>Well Jesus Christ on a motorbike, this is just what we need: Korg and
>Alesis at each other's throat. I've got a good suggestion: why don't
>both companies refrain from posting to this newsgroup. We have both
>Korg and Alesis users on here who are prepared to debate differences
>between the two synths to death. It's a sad day when r.m.m.s. will turn
>into a battleground for synth companies. Leave it to the users,
>please.
Well, how many actually *useful* articles would this threads had
without those vendor opinions? I didn't see a posting from Korg, BTW.
I strongly disagree with you and like to encourage the posters of this
thread to continue.
Martin
--
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
Martin Cracauer <crac...@wavehh.hanse.de> http://cracauer.cons.org
DaveBryce9
First of all, let me say that upon rereading my post that I was definitely
ruder than I intended to be. Sincere apologies to the folks who make
Kurzweil keyboards and the folks who use them (myself included) for the
attitude...I gotta stop answering email late at night.
However, I don't think that I was entirely off base on a couple of these
points, just unclear. Perhaps I might try again, this time more civilly:
<<Once again, I'd like to point out that the compression is
strictly amplitude compression. When reconstructed using the
"natural" envelope of the sound you get a 100% faithful
reconstruction of the sound. No "suffering", no deviation from
the original. The technique is completely lossless.
So why bother? As a sound decays, it normally uses a limited
subset of the 16-bit dynamic range that is available. The
amplitude compression technique ensures that the full dynamic
range is expressed even in softer portions of the sample
... something no other manufacturer bothers with.>>
Unless I'm entirely incorrect, I believe that there are a couple of
kinds of data compression, some lossless, some not. Things like sample
skipping, delta modulation, and amplitude scaling (which I believe
Kurzweil uses) can and frequently do result in a loss. Kurzweil has
created a compression scheme that is very musical, but I do not believe it
is as authentic as a linear sample. The natural sound of the envelope is
stripped away, and then regenerated by software. Ingenious? Certainly.
Same as linear? I don't believe so...I feel that the only products that
ours can be compared to are hardware samplers, and we would lose that
comparison because they have higher order interpolation.
<<Oh come on. You aren't really foolish enough to believe that
stacking sounds is the only possible use for multiple layers, are
you? Delayed layers work wonders for "wavesequencing" effects,
key release layers come in handy for a variety of uses, subtle or
drastic changes in modulation characterics for different zones of
the keyboard can eat up layers, etc. More layers does not equal
less polyphony, just more flexibility.>>
In some cases (e.g. drum programs, where each drum can be given it's own
parametric EQ) I agree fully. However, I have found that many of the drum
channel programs which I have tried to use utilize sound stacking pretty
heavily...not 32 voices worth, necessarily, but if you've got a program
that's stacked 6 layers deep, and you play four notes simultaneously while
holding the sustain pedal, then play four more (or have a couple of the
normal sounds playing at the same time), your polyphony is gone. This is
a source of frustration with the VAST synths for, me, and it is why I own
more than one of them.
<<Once again ... the KDS / DMTi combination is a very different
beast than the QS8 digital out. These offer conversion between a
variety of formats, and include asynchronous sample rate
conversion. In addition to 8 channels of K2500 sound, they
include inputs for two additional AES/EBU inputs (one of which
can be switched to optical input) for the same kinds of
conversion. Four KDS-format outputs, plus multitrack interface
outputs plus four indepedant AES/EBU stereo ouputs and an optical
output put this WAY beyond the simple digital ouput league. Very
few people will ever need to buy these upgrades.>>
My point, Blake, on which I was admittedly unclear, is the same as your
last line...Very few people will ever need to buy these upgrades.
However, there are those who are interested in the simple multichannel
digital out in order to communicate with the plethora of manufacturers who
also utilize it (over 100), and with the K2500, you have no choice but to
get the KDS and DMTi to achieve this, whether or not you need all of it's
functions. Also admittedly, to get sample rate conversion, and AES/EBU
outs with our stuff, you need an AI-1, which *still* doesn't put us in the
same league as the DMTi, but it certainly improves the "simple" digital
out.
(Continued...)
DaveBryce9
(continued from previous post)
<<Firstly, you should learn a little more about the subject before
pretending to be able to make technical judgements. A note on
event consists of three bytes: note on & channel + note # +
velocity, and is a single event that takes ~1ms to transmit.
A note off arrives only when the key is released and is unrelated
to the latency of the initial sound.>>
Once again, my apologies for being unclear. I was not speaking of the
unfettered MIDI spec - no one has an advantage there. But it takes time
from the point a key is depressed to when that MIDI event is transmitted.
This is dependant on microprocessor speed and elegance of coding. Same
thing on the receive side...it was this that I was commenting on. Marcus
Ryle assures me that we do these much faster than anyone. I have
forwarded this part of the previous posts to him, so I can be sure to
print exactly correct specs (sample words vs. milliseconds)...I was under
the impression tha the ones I was quoting *were* correct...I'm sure he'll
do better.
<<Of course, since the factory programs take advantage of this
programability, the user who never edits a single program takes
advantage of the synthesis engine, too.>>
Fully agreed.
<<(By the way, the statistic given above is an oft-used quote from
the era when users reset their synthesizers to factory defaults
to keep repair shops and subsequent buyers from stealing their
sounds.)>>
Partially agreed...the majority of people that I dealt with in music
stores, when working for Kurzweil, and when working for Alesis don't seem
to care much about programming themselves; and, many of the synths that
I've seen come in used or for repair had one or two sounds changed, so
maybe we're both right on this point.
<<Great. The Alesis QS8 ... a product for the average musician in
all of us.>>
I suppose I deserved that slam...I would hope that you know exactly what I
was saying, though...the average musician who buys keyboards wants an
extensive sonic palette of great-sounding, reliable sounds, and don't seem
to care much how they get them. Sure, some of us program sounds (I've
been doing it since the mid-70's), but I believe that more people don't
than do by a wide margin...most folks just want to be able to call up a
bunch of patches that they can gig with, or record with. This doesn't
make them "average musicians", just members of the majority.
Once again, I apologize for my condescending tone in the last post.
Hopefully, I have expressed myself a bit more clearly in this one...also,
as I said a while ago, I didn't intend to atack Kurzweil, just defend what
I perceived to be unwarranted, uninformed slams against our stuff (that
whole Cadillac/Beetle, Lexus/Hyundai thing).
dB
Alesis Keyboards
Sebastian Nyberg
ga...@fc.hp.com (Bill Gates) writes:
>I firmly DISagree. Dave has been an excellent source of Alesis info for
>me, and having reps from other vendors here can only help the signal/noise
>ratio.
That is correct, but when employees of different companies get into the
"our synth is better than yours!"-thing... ouch. Especially when it's
about two synths aimed at entirely different users AND in different price
categories.
Base
Btw. Sorry Korg for the Kurzweil/Korg mixup! First rule of Usenet:
Never post drunk. ;)
Malte Rogacki
In article <5546qu$1...@newsbf02.news.aol.com>, DaveBryce9 writes:
>
> (continued from previous post)
>
> <<Firstly, you should learn a little more about the subject before
> pretending to be able to make technical judgements. A note on
> event consists of three bytes: note on & channel + note # +
> velocity, and is a single event that takes ~1ms to transmit.
> A note off arrives only when the key is released and is unrelated
> to the latency of the initial sound.>>
>
> Once again, my apologies for being unclear. I was not speaking of the
> unfettered MIDI spec - no one has an advantage there. But it takes time
> from the point a key is depressed to when that MIDI event is transmitted.
> This is dependant on microprocessor speed and elegance of coding. Same
> thing on the receive side...it was this that I was commenting on.
Dave, while I agree with most of your points let me make two things clear:
1. Speed of sending notes from the keyboard and speed of reacting to
incoming notes are *not* the same thing. It is fairly simple and
strightforward to write good code that sends out MIDI data according to
pressed notes. There isn't much involved, right? I doubt that many
instruments fail in the send-timing category (on that *did* fail was the
Yamaha DX-1).
On the other hand, how fast a synth reacts to an incoming MIDI command
depends on a lot of factors; like notes already sounding, multitimbrality,
complexity of sound (number of parameters) and so on. It's that part that
the Quadrasynth apparently shines at.
2. But this has absolutely nothing to do with the suitability as a
controller keyboard as you originally implied. What "keyboard mag" did
measure was the *response* time, not the send time. As far as I know there
are synths out there that have a quite good send timing (which makes them
at least usuable as a controller) and a *horrible* receive timing.
Regards,
--
Malte Rogacki ga...@sax.sax.de
-------------------------------------------------------------
"Don't forget to TURN ON THE SYNTHESIZER. Often this is the reason why
you get no sound out of it." (ARP 2600 Owner's Manual)
-------------------------------------------------------------
Ben Cox
Chan Benson <ch...@fc.hp.com> writes:
> : So why bother? As a sound decays, it normally uses a limited
> : subset of the 16-bit dynamic range that is available. The
> : amplitude compression technique ensures that the full dynamic
> : range is expressed even in softer portions of the sample
> : ... something no other manufacturer bothers with.
>
> portions of the sample" to be semi-nonsensical. Could you please
> clarify? I understand how one might scale the mapping range of the
He should have said:
[...] As a sound decays, it normally uses a limited subset of the
16-bit dynamic range that is available. The amplitude compression
technique ensures that the full SIGNAL-TO-NOISE RATIO is expressed
even in softer portions of the sample [...]
This is something of a variation on floating-point D/A conversion.
__
Ben Cox c...@transarc.com
nwb...@ix.netcom.com
Jonathan Elliott <elli...@sun01.dirvet.tas.gov.au> wrote:
>> Also, are the 64 voices on the QS8 reduced as layers are utilized?
>> Different architectures define this differently - enough so the
>> question may be have little meaning without a large expanation.
>An oscillator is an oscillator is an oscillator. If you've got 64 of 'em
>then each layer uses one oscillator and a four layer patch, for example,
>uses four oscillators per note for a total of 16 note polyphony.
>The only real exception to this is the idea of a sub oscillator, as found
>on a heap of old Roland synths (open wide and say "Juno") as well as
>(albeit in a somewhat more sophisticated form) on the Kurzweil K2k/K2500.
My synths include Ensoniq synths - which dedicate multiple
oscillators to a "voice". Hence my question. I found that in the past,
the marketed voice count was not a particularly good indicator of
usefull polyphony - unless I was looking for a panpipe symphony :)
I did find a QS7 this weekend - no headphones and the store closed an
hour earlier than I thought :(. I shall check again in a few weeks.
Thanks for the explanation.
Nate
charles shriner
I'm sure the Alesis QS8 will be a big seller on the Holiday Inn Circuit.
DaveBryce9
<<I'm sure the Alesis QS8 will be a big seller on the Holiday Inn
Circuit.>>
Nope. No auto-accompaniment. Thanks for the kind thought, though...
dB
Alesis Keyboards
Benito Horta
Frank Boon (frank...@lr.tudelft.nl) wrote:
: DaveBryce9 wrote:
: > Also ***in my opinion***, the QS8's pianos, organs (we sampled Emerson's
: > C3!), clavs, Rhodes and drums are much more to my liking...our synth
: > samples come from Emerson's modular moog, Oberheim Matrix 12 and OB-8 and
: > a host of others, and knock my socks off (Keith likes it, too...he used
: > two on this last tour - one on the Steinway MIDI grand, and one inside it
: > (the Steinway's MIDI response wasn't fast enough for him). BTW, he's not
: > an endorsee...) - but then again, I work for Alesis, so who'll believe me?
: > Bet this post elicits a response or two...let the flame wars begin!
: > dB
: > Alesis Keyboards
: Dave,
: I like it when you do this! First stating all the dry facts (like
: allways)... but this time you couldn't hold you [personal] horses:
: hehehe, good work.
: BTW: I have a friend who I jam with sometimes: he owns a K2500 and I
: have a (humble) QS6. And he is really amazed of the sound quality,
: fatness, and sound sources come out off my little box. And he paid 2,5
: times it's price.
: Cheers,
: Frank
: --
: Frank Boon Delft Institute for Earth-Oriented Space Research
: phone : +31 15 278 6221 mailto:frank...@lr.tudelft.nl
: fax : +31 15 278 5322 http://dutlru8.lr.tudelft.nl/~frank
yea the Qs6 has some fat sounds, I made a couple of fat patches for it. my
favorite patch of all that comes with the QS6 is the '74 Square. I use
that one 90 percent of the time.
mid...@erols.com
Or use the Alesis DataDisk SQ to play back MIDI sequences.
dek...@aol.com
yes for the price of a Kuz you can get
a Qs8, an ADAT, computer, a small mixer, and who knows what else.
And you can use it all to make humdrum stuff. will it make
the music you want?
DaveBryce9
<<And you can use it all to make humdrum stuff. will it make
the music you want? >>
I respectfully submit that good gear doesn't make a mediocre artist any
better...some of the best demos I've heard were done with a Rhodes, a mic,
and a four track...Keith Emerson playing an old out of tune upright piano
would still be Keith Emerson.
dB
Alesis Keyboards
shawne...@gmail.com
> : I'm looking seriously at these two... although I can't get my
> : hands on the QS8, I have heard the QS7 and am hoping the action is good
> : on the QS8. Although the K2500 is a fully integrated system, I have
> : lots of PC hardware lying around, so the QS8 with the provided software
> : may be a nice way to go.
>
> : Anyway, I wonder if anyone has opinions on features of
> : these machines. I've said very little to avoid direvting any opinions.
>
> K2500 = Lexus
> QS8 = Beetle
>
>
this is only accurate if your speaking about the "brain" in the keyboard because if you want to talk straight up playability as in how your driving that car the K2500 rides like a tank and the alesis drives like a porsche. I have both and wouldnt be without. the alesis happens to also be one of the best stage pianos/midi controllers for its time too. 64 note polyphony etc.