Jump to content


Please note: You can easily log in to MPN using your Facebook account!

Q's for those who've worked with Synth Mfg


Recommended Posts

It often strikes me (most recently in some Behringer discussions) when people discuss synth designs and desires, that they don't tend to understand real world manufacturing considerations and OPEX (operational expenditure) vs CAPEX (capital expenditure) and how decisions have to be made based on projected run rates and risk mitigation. A number of folks in our community have worked with some of the manufacturers, and to the extent that they can, it would be very interesting to me, at least, to hear some inside insight into some of the decision making process.

 

Sorry to put dB on the line, but I think the Alesis Andromeda is an excellent case study. The model was technically A6 which was code for ASICs - Application Specific Integrates Circuit. When making an analog synth, you can go completely discrete components - transistors (Moog Ladder) - or use things like OTA's (Operational Transconductance Amplifiers)...(MS20, etc), or costly chips like the Curtis chips that went into synths like the Prophet 5. These are widely available but relatively expensive per unit. It may not seem like it until you add up the component list and look at the circuit board printing, stuffing, soldering, etc.

 

The concept of the ASIC is to have a customized chip designed in house to do exactly what's you want. It takes a lot of per unit cost away...OPEX...but has a huge initial investment....CAPEX. So the gamble was that the initial investment would payoff by selling enough units that the CAPEX would be paid of and the units would become profitable. My guess is that I think didnt happen.

 

This becomes even bigger in modern keyboards with things Theo likes to talk about .... DSP vs FPGA vs some sort of RISC processor or whatever.

 

Insights?

Dan

 

Acoustic/Electric stringed instruments ranging from 4 to 230 strings, hammered, picked, fingered, slapped, and plucked. Analog and Digital Electronic instruments, reeds, and throat/mouth.

Link to comment
Share on other sites



  • Replies 9
  • Created
  • Last Reply

Top Posters In This Topic

Top Posters In This Topic

In my experience, features & specs we request are very much put to the number crunching filter of net unit profitability. The actualy math of that number crunching is very dependent on the resources of the specific company, especially as it comes to the CAPEX, as you can start/stop/start production runs based on demand and control the OPEX

 

A large company like Yamaha had enormous resources at their disposal, not just $$, but also ability to design and fab their own components within their various divisions. I don't believe any other company had the ability to bring something like the DX7 to market with what was needed to develop Chowning's research to a mass (musical) market product. During my hardcore time with them from DX7II through the SY's, VL's, CS/AN/EX & FS1r, I would say most of their CAPEX was 'internal' amongst their various divisions and subsidaries, which IMHO allowed them to take risks others couldn't in developing & bringing unique synth technologies to market.

 

When you think about something like the VP1, even at $30K that price point, the hardware inside was only possible as they did their own VLSI's -- think of the power of a 2008 era 2.5 GHz Xeon CPU in 1993 dollars, and it had the equivalent of 8 of them. The EX5, the cool yet glitchy Kronos of the time -- is actually a good example where they missed the mark in specs vs costs as they invested the CAPEX & kept the features but didn't commit enough OPEX resources to the hardware as by then it had to cost-compete in the market ruled by straightforward ROMplers.

 

OPEX will limit things like number of switches, knobs, buttons, display size, etc but not necessarily the overall features or capabilites of modern "synth on a chip" gear, which as you note is dependent on the willingness to invest in the CAPEX. Fortunately, because of the modern powerful / cheaper off the shelf options of chips like general purpose DSP's & FPGA's the CAPEX is in the coders salaries not the chip dev as in my past days with Yamaha. Which is related to why there's so many softsynths out there compared to hardware...

 

Just my 2 cents...

 

Manny

People assume timbre is a strict progression of input to harmonics, but actually, from a non-linear, non-subjective viewpoint, it's more like a big ball of wibbly-wobbly, timbrally-wimbrally... stuff

 

Link to comment
Share on other sites

Sorry to put dB on the line, but I think the Alesis Andromeda is an excellent case study. The model was technically A6 which was code for ASICs - Application Specific Integrates Circuit.

That's actually not accurate.

 

Alesis had internal codes for every product that were a letter followed by a number. QS8 was Q8 (QS series, 88 key), QS7 was Q7, QS6 was Q6. Andromeda was A6 because it was Analog synth, 61 keys. The ASICs thing was just a bonus...for some.

 

As the marketing manager for the division at the time, I disliked the A6 name becoming public - I still do. A) It's an Audi, and B) it sounds like it's a 6 voice.

 

I came up with calling it Andromeda to try and make the engineers who just looooooved the wordplay of the ASICs thing to bits. The reason they're both there is after a pretty decent battle over it we compromised. Just like JP8 and Jupiter 8 are both on that synth.

 

...and that's the truth. :puff:

 

dB

 

:snax:

 

:keys:==> David Bryce Music • Funky Young Monks <==:rawk:

 

Link to comment
Share on other sites

Holy bleep, J Dan and DrSynth, thanks for the concise lesson in reality! The OPEX/CAPEX aspect solidified a few things I basically knew, but which needed a bit of focus. More than a few businesses or otherwise great ideas collapsed because they couldn't keep all fifteen plates spinning in the air for just those two more critical months. I can understand why Tsutomo Katoh of Korg slashed the sustain pedal from the original Polysix's feature set, but also why a third-party retrofit kit that provided it was welcomed by many owners. Part of synth mystique is the happy accidents and strange left turns.

 "I want to be an intellectual, but I don't have the brainpower.
  The absent-mindedness, I've got that licked."
        ~ John Cleese

Link to comment
Share on other sites

The lessons on basic economics are spot on. I think the other side of the coin are all the boutique hardware projects we hear about -- and wonder about.

 

How can a small team scrape together enough resources to design, manufacture and assemble a compelling product? Will they be able to enhance and support it down the road? Distribution, warranty, etc.

 

Keyboard toys seem to come and go these days. The bigger companies have deep pockets and durability. Yes, there are a few interesting smaller companies who have found their niche. Nord, for example, is rumored to have something like 30-40 employees. Not part of a bigger conglomerate. No way are these guys going to design and manufacture their own keybed, for example.

 

My thought is the future of keyboard hardware innovation are small, artist-focused companies where it's not all about the $$$. Yes, they need to make $$$, but that's not the goal. I would put in that category GSi, Nord and Sequential for starters. And there are others. I wish them all the best of luck.

Want to make your band better?  Check out "A Guide To Starting (Or Improving!) Your Own Local Band"

 

Link to comment
Share on other sites

BTW, there's new frontier in compute (vs. the familiar ASIC, FPGA, custom chip, standard processor, etc.).

 

Saw an article (wish I had captured the link) of some researchers using massively parallel graphic processors (think Nvidia and AI processors) to do super sophisticated multi-reflectional processing. No links were given with the audio results, but -- as a geek -- I find this completely fascinating.

 

Imagine being able to do the equivalent of ray tracing (graphics) for audio processing. Ray tracing is simulating the path of every photon in a scene: reflections, refractions, diffusion, etc. Extremely compute intensive. There goal was every note, processed through it's audio journey from source to ear. Here's what happens inside the instrument. Here's what happens outside the instrument. In the graphics world, ray tracing is the de-facto standard for realistic illumination.

 

I can't wait.

Want to make your band better?  Check out "A Guide To Starting (Or Improving!) Your Own Local Band"

 

Link to comment
Share on other sites

cphollis, the problem with GPU for audio is that it audio doesn't work well in GPU pipelines. Video parallelizes wonderfully well - that's why an current 3D gaming card has several thousand processors. You can literally compute each pixel/shape/etc at the same time. Audio doesn't work this way. It is real-time, and things like effects have to be processed in a fixed order. This is why DSP is lovely - each block can run fast, and then if you can re-order the blocks a little, you can have very fast audio processing, all in real time.

 

This is also why music PC's are best with very high core speed.... an entire channel strip in Mainstage or Cubase has to run on a single CPU so that it stays together.

 

What would be really great is a real-time OS for music products to run on... That's a pipe dream, but sure would make a difference to get rid of Windows/OSX. No one would care about DPC latency anymore....

 

The UAD accelerators are the DSP you are thinking of, as is Avid's HDX version of ProTools. But there is a reason that no one has GPU accelerated DAWs.... the technology has been available for years - its just the wrong architecture.

Link to comment
Share on other sites

Brings up an interesting question though - namely software development as a capital expense. I'm sure the demographic is changing regarding programming resources on different platforms,

Dan

 

Acoustic/Electric stringed instruments ranging from 4 to 230 strings, hammered, picked, fingered, slapped, and plucked. Analog and Digital Electronic instruments, reeds, and throat/mouth.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Unfortunately, your content contains terms that we do not allow. Please edit your content to remove the highlighted words below.
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...