Part II
[14:23] OK - I have been monitoring the discussions on the list so I have seen all you preparations - let's talk about some of the issues.
[14:24] Obviously we have NOT abandoned the Triton. We are fully behind it. It is our flagship synthesizer.
[14:24] Each PCM expansion board we create is a huge task for our voicing resources. Nearly equal to voicing a new synth.
[14:24] *** Joins: todd (todd305@ool-18bd7a21.dyn.optonline.net)
[14:25] We just finished the dual board Orchestral Collection EXB-PCM0607 - 32 MB of samples, 256 Programs and 256 Combinations.
[14:25] It has not yet shipped in the US, but will late this month, for a list price of $480.
[14:26] And we plan to make more boards. We want to do another dedicated piano board, we are thinking about a board of cut-up ethnic percussion data good for dance genres...
[14:26] We have a great B3 that we bought to develop the CX3, maybe a board of organ...
[14:26] *** Parts: ipi (eldiablo@200.64.192.88)
[14:26] *** Parts: Saturna (Saturna@pm3-164.efn.org)
[14:26] *** Joins: Saturna3 (Saturna@pm3-077.efn.org)
[14:27] But in regards to sound support we have been maxed out trying to develop enough boards to support the product.
[14:27] If you have up to 8 slots in a rack we assume you want WAY more than 8 boards to choose from.
[14:27] *** Parts: Saturna1 (Saturna@pm3-072.efn.org)
[14:27] And we're just getting started in this filed - Roland is far ahead of us. :-)
[14:27] And they're far larger than us, too!:-)
[14:28] *** Joins: tritonfunny (newbie@sdn-ap-009neomahP1040.dialsprint.net)
[14:28] So we are certainly active trying to support the product. But there are many other ways that you have discussed, right Arvon45? :-)
[14:29] i am sure he agrees...
[14:29] While we have not had enough resources to also be voicing new PCG data, I work with ANY valid 3rd party company to get a Triton at a great discount, or even a loaner.
[14:29] That's how Pro-Rec, and Kid Nepro and others have been able to afford to develop for the Triton.
[14:29] *** Joins: ipi (eldiablo@dup-200-64-192-101.prodigy.net.mx)
[14:30] I have personally converted their Trinity MOSS data, or Z1 MOSS data to Triton format to help them.
[14:30] So we try to encourage and facilitate 3rd parties to support us.
[14:31] Regarding sample support in native format, I have provided Tritons to MANY of the sample CD companies and urged them to go native, but I can't force them.
[14:31] Unless I were to pay them directly, right? That's beyond my financial scope.
[14:32] But as Paolo from Korg Italy mentioned the other day we do have the Triton Sampler Starter Pack that KorgNet developed, as well as the Trinity drum loops package that I made, which now has a PCG file for loading the loop into your Triton.
[14:33] I am continually surprised that the 3rd parties haven't jumped on the chance - there are 10s of thousands of Triton owners in the market, and it has to be finacially enticing. But the programming is rather deep, perhaps the loop-library people are dominating the filed.
[14:34] So as far as sounds go we have been supporting and will continue to support within our capabilities.
[14:34] *** Joins: delerium (big_dirty3@d150-17-239.home.cgocable.net)
[14:34] Let's move on now to design and OS issues - remember - this isn't one-sided - you'll get to ask me plenty of questions soon.
[14:35] #1 resampling, right?
[14:35] right...
[14:36] It can't be done. We tried - it won't work. The data "collides" in the bus and the timing is bad. No use trying to resample if the results are bad.
[14:36] The same thing affects the ability to sample while the sequencer is playing. unfortunately, the Triton was not originally designed to do these things, and only new hardware design can change that. Software can't.
[14:37] One thing that may answer a lot for all of you, is that the original target when designing the Triton was a mid-priced prelacement BELOW the trinity, to replace the trinity and N364 business.
[14:37] So a lot of design decision were based on keeping the price low.
[14:37] *** Joins: MrKorgTriton (vista500@rm01-24-29-194-172.ce.mediaone.net)
[14:38] That's why no digital I/O, and a number of other issues.
[14:38] *** Joins: Bubbleface (danrek_98@adsl-151-202-176-93.nyc.adsl.bellatlantic.net)
[14:38] Unfortunately (or fortunately, depending on how you look at ti), we did too good of a job. We made a product that was in most ways so superior to our goal that it became the flagship.
[14:39] And so now it looks like, why didn't we go 'all the way" with the design and feature set.
[14:39] But we can't change certain aspects.
[14:39] *** Parts: Navid (~triton@ubr-33.189.12.apopkavine.cfl.rr.com)
[14:40] This may help to understand why we did what we did - we're not "that" stupid, but certain decisions affect your ability to do other things.
[14:40] *** Joins: Navid (~triton@ubr-33.189.12.apopkavine.cfl.rr.com)
[14:40] So after resampling let's look at Sys. Ex. in the sequencer mode.
[14:41] Funny story - but that's how I first met Shakil, remember S?
[14:41] *** Parts: tritonfun (newbie@sdn-ap-006neomahP0462.dialsprint.net)
[14:41] *** Parts: zaplin (~fgfg@212.72.3.107)
[14:41] Or maybe it's not right for me to say funny - I apologize if you all take that the wrong way.
[14:41] *** Joins: Saturna (Saturna@pm3-147.efn.org)
[14:41] *** Joins: steskilt (steskilt@mp-216-42-242.daxnet.no)
[14:42] basically Shakil called us up and read us the riot act for not being able to do it.
[14:42] *** Parts: stuart_pryer (triton@modem-87.elured.dialup.pol.co.uk)
[14:42] *** Joins: stuart_pryer (triton@modem-87.elured.dialup.pol.co.uk)
[14:43] I have been having discussions with Engineering for a while about this - Korg has NEVER had this features in any workstation so far, but we recognize the value in it. I am still hepoful that we can do it, but as you can expect I can't promise you today that it will happen for you in a new OS>
[14:43] yes.. i remember
[14:43] But I am fighting for it for you, for me, for all of us.
[14:44] As someone in the group guessed, Korg is usually very concerned about file compatibility issues, and adding this WOULD change the file format of sequences.
[14:44] But I am continually working with them to recognize that this should not be a reason not to do things.
[14:44] So I am not promising the features, but I am promising that I will continue to push for it.
[14:44] *** Parts: Saturna3 (Saturna@pm3-077.efn.org)
[14:45] And I DO have influence...
[14:45] lets hope for the best
[14:45] yeah
[14:45] There's a group of features that can best be understood by realizing what can and can't be done in a product like the Triton, vs. virtual synths.
[14:46] This relates to things like new filter designs, increased polyphony etc.
[14:46] The Triton is based around a custom ASIC chip which does many of the voice architecture issues.
[14:46] The filter design are hard-designed into the chip. The buss structure is as well.
[14:47] The only way to increase polyphony would be to change the host processor to a quicker one AND to make changes in the ASIC.
[14:47] So certain aspects of the product are 'fixed" and live in the chip design.
[14:47] *** Parts: id (~id@80.75.0.249)
[14:48] Virtual synths that basically use some "generic" DSP (Motorola, in the case of Virus etc.), the Sharc or even the NEC DSP we use in the MOSS and MS-2000 world allow a different type of design flexibility.
[14:48] And you all are already well aware of those advantages.
[14:49] But look at those products - none offer the high polyphony (for instance) of a ROM-based synth.
[14:49] *** Joins: Tokke (triton@212.233.44.55)
[14:49] So there are certain very real advantages to the custom ASIC approach. And some disadvantages.
[14:50] What's the best route for the future? It's unclear, but trust me that we are working in both directions - the OASYS project and the MS-2000 and even the CX-3 are proof that we are working the in the DSP-based world as well.
[14:50] So filter design changes are out of the question.
[14:51] OK, let's jump into probably the most heated/controversial area you have been discussing.
[14:51] AKAI handling.
[14:51] yes!!
[14:51]
[14:51] drum roll
[14:52] crash...
[14:52] First of all, I'm sorry for ANYONE's mis-interpetation of our spec, our ads, our brochures, our manuals, OK?
[14:53] ok...
[14:53] But we don't feel that we misled the public. I do understand how with "little" information one can assume, but we never promised that the Triton was or could be an AKAI library player.
[14:54] If it could, trust me, I would have trunpeted it to the heavens in my ads and materials.
[14:54] You would hae seen statements like "And use your Triton to play back ANY AKAI library, with no programming or fuss needed!"
[14:54] *** Joins: jnr (midihead@dialup-64.156.99.65.Dial1.Orlando1.Level3.net)
[14:55] We would have "run" with that feature.
[14:56] So first of all, we list AKAI as one of the sample foramts we can load - AIFF, WAV, AKAI S-1000/S-3000 and Trinity native (except for the compressed format).
[14:56] I think one of the least clear points was when we say that we can load from the Program.
[14:57] *** Joins: Saturna1 (Saturna@pm3-040.efn.org)
[14:57] This is hard to explain in short copy, but I think you all can understand what happens when you load in WAV and AIFF - you only get individual samples - no keymaps. No tuning. No relative volume.
[14:57] *** Parts: ipi (eldiablo@dup-200-64-192-101.prodigy.net.mx)
[14:58] We load from the AKAI program to be able to get these critical parameters so we can make a valid multisample.
[14:58] Again - I'm empathizing with your desire to get more. I want more too.
[14:59] When I was at Ensoniq we developed a versin 3 for the ASR-10 that loaded AKAI and Roland. Even with our sampler voice, which was better-suited to these formats than the Triton we found many problems playing back libraries.
[15:00] Especially the better-programmed ones which more fully used the voice architecture of the AKAI.
[15:00] So we redesigned our voice in the MR-Rack to accomodate this much desired spec.
[15:00] When I came to korg it was one of the first big topics we spent time discussing. How to someday get the Korg voice to that place...
[15:01] Today we are not there. We need a voice that allows for individual layers which can have their own filter and EG settings, and may more of them than we allow.
[15:02] It's REALLY difficult to try to approximate that with some automated routine that creates maltiple Programs and automatically place them into Combis to try to get there.
[15:02] This is one area where I agree with the caution of korg Inc.
[15:02] I've seen many of you talk about just getting what we can. But that leads to a more overt deception of you, our users.
[15:03] It means that we say we CAN load AKAI Program/parameter data, but only sometimes. And not always with a consistent success rate.
[15:03] We don't want to promise something that works "shakily" at best.
[15:04] u mean like me?.... jk
[15:04] So we do what we set out to do - get either the samples, or a multisample from the format.
[15:05] And we offered this right from the beginning, not with Version 2. Version 2 added better handling of the S-3000 format, and of stereo multisamples, but we offered this from the start.
[15:05] :-) Just saw your joke Shakil!
[15:05] u on your roll Jerry..
[15:05] One more short topic then we'll open it up, I've been hogging the bandwidth for a while now.
[15:06] We HAVE been offering OS updates. Version 2 was MAJOR, and it was not "done" when we launched. It was planned, but not executed. It took a lot of work form a few dedicated programmer to realize, and it added a lot to the sampler.
[15:06] Remember, before the Triton the last sampler Korg has made was the DSS-1!!!
[15:07] Since Version 2 we have responded to user bug reports with a number of smaller OS issues, and we of course will continue to do so.
[15:07] If you can report a repeatable problem we'll do our best to fix it.
[15:08] And version 2.1 added export of Triton samples to WAV or AIFF, which opens up the world of even more powerful editors for you to use.
[15:08] *** Quits: Aryan (aryan@bb-62-5-49-100.bb.tninet.se) (Quit: What a night ........ I have dream that one day Iranian stand by each others side)
[15:08] *** Parts: Aryan (aryan@bb-62-5-49-100.bb.tninet.se)
[15:08] So we are still "behind" the product and supporting it. Not in all the ways you would like, but we're trying. And we've been monitoring this group and others to understand what our users want and need.
[15:09] *** Parts: PALL (triton@210.56.19.127)
[15:09] Ethnic Exp
[15:09]
[15:09] So before this get's boring, let's open it up. try to ask me first about topics I've started to discuss, then we could move on to other issues, OK?
[15:09] ok guys..
[15:10] Jerry.. that was good.. thanks,
[15:10] Ethnic Expansion? I already mentioned the dance-oriented ideas. An Ethnic board also makes good sense.
[15:10] yeah
[15:10] Tapesh.. can you describe to the Group how can they get in row to ask questions
[15:11] ok guys now if you have questions to ask jerry plz msg me and just say question in your msg and I'll let you guys to talk to jerry 2 at a time
[15:11] *** Tapesh- sets mode: +v arvon45
[15:11] *** Parts: Saturna (Saturna@pm3-147.efn.org)
[15:11] Hi guys
[15:11] Hi arvon
[15:11] Id like to say thanks to jerry for coming out here today
[15:11] *** Tapesh- sets mode: +v vanderrohe
[15:11] its nice of korg to come out of hiding
[15:11] Hi Arvon, I';ve been axpecting you :-)
[15:12] concerning the resampling, you say that it is possible to read parameters
[15:12] but it's "shaky" at best
[15:12] you mean the AKAI samples..?
[15:12] bmy point is isnt that better than the akai reading the way it is currently?
[15:12] Well, this is about AKAI reading, not resampling.
[15:12] er I mean akai reading
[15:13] right on
[15:13] Hi Jerry. Thanks so much for this opportunity.
[15:13] you ther ejerry?
[15:13] I did not say that it's possible to read all the parameters. basically in the AKAI voice each Layer has parameters associated with it, and then the total program has more. We don't have that voice scenario to begin with.
[15:14] what would you call current akai reading in triton?
[15:14] certainly its more shaky than an inconsistent akai reading ability?
[15:14] OK guys.. let's try to make is 'productive'.
[15:14] Id rather get the reading right 50 percent of the time than not at all
[15:15] So we could get the total program Filter and envelope, and perhaps some aspect of the LFO, but we would not get the layer offsets and parameters. So it would not likely give you the result you want.
[15:15] I'd call our reading the ability to import in multisamples from Akai, no more.
[15:15] Can we expect to import single layer AKAI programs?
_________________________
Vic:)