The No.1 Website for Pro Audio
 Search This Thread  Search This Forum  Search Reviews  Search Gear Database  Search Synths for sale  Search Gearslutz Go Advanced
SSM/CEM Chips, Do i understand correctly?
Old 17th January 2013
  #1
Lives for gear
 
xanderbeanz's Avatar
 

Thread Starter
SSM/CEM Chips, Do i understand correctly?

I just wanted to check whether i've got all this info right:

From what i understand, before these chips came along, all synthesizer electronics were discrete, a bunch of capacitors and transistors and cogs and wheels (joking about the last 2 XD) then, sometime in the 80's when microchips started to become cheaper and cheaper, they miniaturized alot of these components and bunged them on a chip, so all the wiring turned into traces on the chips, and the transistors themselves became the chip's complex set of gate arrays.

So they did a VCO chip, a VCF chip, etc, (I understand my Mono/Poly has a load of them in) and then they either wired these up to pots and sliders like they always used to (like in the M/P) or, they used a CPU to digitally control them (like in Prophet 600) now, the primitive CPU's lead to a bit of stepping between values, as the analogue values were crudely converted to digital ones using 4-8 bits (i think it's 5bit on the poly 800 for example)

Of course, DCO's came in later, further integrating oscillator circuits into the digital domain, but for now, lets discuss 2 key things.

1. Did i get most of that information right? Is my basic understanding of the CEM/SSM process ok?

2. What exactly is lost in the miniaturization of the circuits? for example people rave on about the SH2 for it's big "discrete" oscillators, can CEM or SSM oscillators ever hope to compete with discrete ones if they're still voltage controlled and not sterile in terms of pitch? Does the chip making process take out a load of voodoo magic away from the oscillator circuit? Also, what about filters? I really like my M/P's SSM 2044, and i really like my C64's sid filter, but is there something missing in the design? could a chip ever really compete with a real moog ladder filter? (I'll probably be buying a moog later this year for THAT sound anyway.)

None of this really matters of course, lots of chip based instruments sound great, but the topic fascinates me and i'd like to know more

Discuss...
Old 17th January 2013
  #2
Lives for gear
 
Altitude909's Avatar
 

1. Yep, that's the gist of it
2. Super tight control of every single component. At the end of the day though, the whole is greater than the sum of the parts. There were plenty of curtis and SSM parts that sounded very good (Prophet 5?). The main thing to understand is what these chips did was to enable big polys to become affordable enough for most every body. Ever seen the inside of the Korg PS synth? Its a nightmare of huge voice boards, super expensive to make where everything has to be discrete. With filters/oscs on single chips you can have 6 voices on one board easily.

Obviously, it went too far at one point where you had the whole synth on one chip (CEM3394->barf) where too many corners were being cut for the convenience of price and usability so everything started to sound the same and analogs fell out of favor
Old 17th January 2013
  #3
Lives for acid
 

The late 70s, not the 80s, is when these chips began manufacture. But otherwise, you're spot on.
Old 17th January 2013
  #4
Jose Ramón Alvarado Villa
 
Don Solaris's Avatar
Quote:
Originally Posted by xanderbeanz View Post
2. What exactly is lost in the miniaturization of the circuits?
Each component in this miniature circuit inside the chip is made of exactly the same material (talking about material purity) and thus will have exactly the same properties. It is very unlikely that few microns away, next transistor will behave a bit different, such as you would expect when you have two "hardware" transistors made during two different runs, of materials with different purity, neither their properties perfectly match.

Further more, if heated, components inside the chip will heat up together and no difference will occur between individual parts. Their thermal properties are most likely the same down to xxyy decimal points. Going to chip design has its benefits/drawbacks. In VCO circuit design, you might want to avoid 100% identical component values. OTOH in ladder filter circuit you want transistors to be as close match as possible.

Keep in mind that BIG part of the sound is the chip's surrounding electronics, gain staging, auto tune routines (on CPU controlled synths), heaters, etc. Yes, SH-2 does indeed sound BIG. Its single oscillator in PWM eats all 4 Monopoly's. OTOH, SH-2 can not produce that characteristic (i call it hollow/fat) charming sound of SSM VCOs. So there's no real winner in here. Nothing that one should worry about. SSM LP filters sound phenomenal. Probably second best thing to Moog ladder.

Recommended to watch: http://www.youtube.com/watch?v=z47Gv2cdFtA
Old 17th January 2013
  #5
Lives for gear
 
xanderbeanz's Avatar
 

Thread Starter
Thanks guys
Old 17th January 2013
  #6
Gear Guru
 
Yoozer's Avatar
Quote:
Originally Posted by xanderbeanz View Post
Of course, DCO's came in later, further integrating oscillator circuits into the digital domain
Only the control. A DCO still spits out an analog voltage. Digital means that there's an extra step involved - D/A conversion (but that's for PPG Waves or DW8000s)

Quote:
Can CEM or SSM oscillators ever hope to compete with discrete ones if they're still voltage controlled and not sterile in terms of pitch?
The Alesis Andromeda uses Anadigm chips. How sterile are those?
Old 17th January 2013
  #7
Jose Ramón Alvarado Villa
 
Don Solaris's Avatar
Quote:
Originally Posted by Yoozer View Post
The Alesis Andromeda uses Anadigm chips. How sterile are those?
VCOs are in free flow (if you let them by disabling all the tuning) but the signal path is a bit on the clinical side, particularly the filter section. Also there is some sort of bandwidth limit - you can't get that VCO high sheen such as what Jupiter 8's VCO produce. IMO, it definitely sounds like a chip based synth. Not that it sounds bad. It's a killer synth!!! And i'm sure it will sound better as it ages.

Best music store purchase of my life.
Old 7th January 2017
  #8
Gear Head
Does anyone know if you can swap out a Curtis CEM3340 with a SSM2030? Not that I can find an SSM2030 anywhere....
Old 7th January 2017
  #9
Lives for gear
 

ok, while on the topic of SH-2 sounding huge (and it does), I was wondering, which sounds huge-er, an SH-2, or a roland jupiter 8 in single voice mode? both are discrete oscillators, does the chip filter on the jp-8 make much of a diff (and one of the best filters on a chip out there)? not sure why that'd be. but I've heard the jupiter isnt great for bass, and the SH-2 I know from personal experience is huge-er than huge.
Old 7th January 2017
  #10
Lives for gear
 

Quote:
Originally Posted by alwaysuptil1 View Post
Does anyone know if you can swap out a Curtis CEM3340 with a SSM2030? Not that I can find an SSM2030 anywhere....
the answer is no and yes.

no you can't pull the original chip out and do a direct swap. the outlying support circuit and traces lead to different legs on the chips and have different ranges as far as the resistor's and capacitor's ratings, values and tolerances. those differences alone if you re-wired the legs to the corresponding traces would cause mayhem and unstability and undesirable results.

yes in that if you redesign the support circuits and rewire it you could get it to function as a replacement but by the time you did you'd have saved time just making the boards and populate them from scratch
Old 7th January 2017
  #11
Lives for gear
 

Quote:
Originally Posted by fromthepuggle View Post
ok, while on the topic of SH-2 sounding huge (and it does), I was wondering, which sounds huge-er, an SH-2, or a roland jupiter 8 in single voice mode? both are discrete oscillators, does the chip filter on the jp-8 make much of a diff (and one of the best filters on a chip out there)? not sure why that'd be. but I've heard the jupiter isnt great for bass, and the SH-2 I know from personal experience is huge-er than huge.
when is it colder? in the winter, or when you're in the mountains?

discrete vs. integrated
analog vs. digital
dco vs. vco
mac vs p.c.

you've just re-asked the same old tired subjective question.

you can make a discrete design sound sterile and thin, and you can make a dco chip based synth sound huge.

it isn't the underlying parts but the way they are used.
Old 7th January 2017
  #12
Lives for gear
 
Mefistophelees's Avatar
Quote:
Originally Posted by xanderbeanz View Post
I just wanted to check whether i've got all this info right:

From what i understand, before these chips came along, all synthesizer electronics were discrete, a bunch of capacitors and transistors and cogs and wheels (joking about the last 2 XD) then, sometime in the 80's when microchips started to become cheaper and cheaper, they miniaturized alot of these components and bunged them on a chip, so all the wiring turned into traces on the chips, and the transistors themselves became the chip's complex set of gate arrays.

So they did a VCO chip, a VCF chip, etc, (I understand my Mono/Poly has a load of them in) and then they either wired these up to pots and sliders like they always used to (like in the M/P) or, they used a CPU to digitally control them (like in Prophet 600) now, the primitive CPU's lead to a bit of stepping between values, as the analogue values were crudely converted to digital ones using 4-8 bits (i think it's 5bit on the poly 800 for example)

Of course, DCO's came in later, further integrating oscillator circuits into the digital domain, but for now, lets discuss 2 key things.

1. Did i get most of that information right? Is my basic understanding of the CEM/SSM process ok?

2. What exactly is lost in the miniaturization of the circuits? for example people rave on about the SH2 for it's big "discrete" oscillators, can CEM or SSM oscillators ever hope to compete with discrete ones if they're still voltage controlled and not sterile in terms of pitch? Does the chip making process take out a load of voodoo magic away from the oscillator circuit? Also, what about filters? I really like my M/P's SSM 2044, and i really like my C64's sid filter, but is there something missing in the design? could a chip ever really compete with a real moog ladder filter? (I'll probably be buying a moog later this year for THAT sound anyway.)

Discrete transistors are made in the same process as chips so they wont be any different to chips.

There's probably better control over the materials on the chips so the capacitor and resistor tolerances will be tighter.

The parts on chips are not still perfect though, the parts are all still slightly different, but probably to a lesser degree than discrete parts.

The fact the whole circuit is on the same bit of silicon means the temperature will be a lot more stable because heat can move around, although individual components can still get cooler and hotter.

All this means is chips should be more consistent and stable but this does not guarantee pitch stability. The prophet 5 uses chips and it has a tune button - so does the Andromeda.

There's a lot of classic synths use chips and they're *all* digitally controlled so I don't think there's any valid argument that chips sound worse, maybe a little different.

Quote:
could a chip ever really compete with a real moog ladder filter?
Listen to an SSM and you'll find out!


BTW the Modal 002 has digital oscillators and it sounds absolutely huge.
Old 7th January 2017
  #13
Lives for gear
 
Westlaker's Avatar
Quote:
Originally Posted by fromthepuggle View Post
but I've heard the jupiter isnt great for bass
You've heard wrong.
Old 7th January 2017
  #14
With these old chips a major difference in sound between two synths using the same chips is primarily the gain staging. A good example is the Polaris vs the Xpander. Both have almost identical circuitry disregarding the extra filter circuitry in the Xpander but there's quite a bit of difference in the gain staging throughout and thus the two synths sound night and day different from each other. This even holds true for CEM3394 based instruments to some degree.

I'd love to have modern recreations of the SSM2044 and SSM2040 filter chips. Both are such gorgeous sounding examples of synth IC's.
Topic:
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Similar Threads
Thread
Thread Starter / Forum
Replies
corganb / Electronic Music Instruments & Electronic Music Production
22
Megacurve / Electronic Music Instruments & Electronic Music Production
0
Acid Hazard / Electronic Music Instruments & Electronic Music Production
0
demonfuzz / Electronic Music Instruments & Electronic Music Production
4

Forum Jump