Classic D-50 Sounds
At what is now a long distant point in the past I acquired a Roland D-110 (me too! - Ed) multi-timbral sound module based on the Linear Arithmetic S+S (sampling plus synth) synthesis method made famous by the Roland D-50. At the time I got it because it was cheap and it allowed me to purchase not only that unit but a couple of others such as a Yamaha TX81z and a Roland U-220 (for more traditional sounds) primarily to be driven by my 8-bit computer at the time. This was a common configuration back in the day for those old enough to remember.
It was also one of the first affordable multi-timbral units with multiple audio outputs - making it eminently mixable (back when we all had mixing desks) - Ed.
So accustomed and attached to the D-110 I became that I purchased a second unit in case anything happened to the first and I regularly use both in different configurations. I did acquire a D-50 later which was interesting because it allowed for a personal perspective on the debate about the differences between the two machines. Likely you have read similar opinions focused around the idea that the D-110 was considered to be the D-50's 'little brother' thereby implying that somehow it was less capable. I would contentiously suggest (howls of rebuttal no doubt to follow) that when one examines both architectures that you will find, and hopefully I will demonstrate, that the differences are in fact not as pronounced as you might think. Such examination of any synths architecture you may own is crucial to extracting the maximum capabilities from it and, dare I say it, you might actually be encouraged to program the thing based on that knowledge. This is an astonishing concept to many when dealing with synths with 2-line displays and hierarchical button driven menu systems.
But you now have the ability to edit your machines with the right librarians on your computer, so why wouldn't you? Further I have also developed a Max4Live prototype that allows me to graphically edit a D-110 within Ableton Live. A D-50 editor is also in the works.
So let's examine the two machines. Firstly they both share the same Linear Arithmetic (LA) synthesis method, period. Roland engineers leveraged the idea that sound can have two distinct 'parts'. The first relates to that magic first second or so in the timeline that we know as the attack transient. For many sounds this contains a good deal of the tonal and harmonic richness that characterises the sound. Further as the note is sustained over time it develops into a more uniform waveform that analog synth programmers would be familiar with. Roland however gave us the best of both worlds and implemented it in both machines. LA provided the ability to use carefully crafted PCM waveforms that generated the rich attack transient and also digital oscillators that could generate the familiar waveforms we associate with analog machines. Further you didn't need to use the PCM waveforms at all, the digital oscillators could be used by themselves to produce pretty damn good analog representations of their own accord. Both the D-110 and D-50 can do all of the above. The tone shaping and dynamics of the timeline are controlled by envelopes that again, are implemented in both machines. Both have the same digital filters. At the very broadest level where the two machines part ways is that the D-50 is not multi-timbral, which ironically enough when you examine the architectures in detail gives the D-110 a significant advantage.
In LA Synthesis the smallest unit of sound that Roland implemented is called a partial. A partial is in effect a container that binds an oscillator that generates a standard waveform or a short PCM sample stored in ROM, a digital filter with cut-off, resonance and an envelope, and a digital amplifier with its own envelope also. Further synth generated waveforms in the traditional sense also have pulse width. The same partial configuration is implemented in both the D-110 and D-50. Immediately outside the partial boundary however we identify the first difference between the two machines. The D-50 implements no less than three LFO's. Each can be routed to pitch, filter cut-off and amplifier level. The D-110 only has one LFO which is routed to pitch only.
Additionally only pulse width modulation is possible on the D-50. LA also recognises that one partial by itself makes for a pretty thin sound so they decided to implement four partials that can be mixed or combined using differing configurations, which Roland refered to as Structures. Interestingly enough both the D-110 and D-50 have four partials at their disposal but they are combined slightly differently. In the D-50 the partials are put into pairs and encapsulated into a 'tone'. The D-50 therefore had two tones each containing a pair of partials, which they called an upper tone and a lower tone. The D-110 also paired off its partials but contained both pairs within the same tone. This was largely to facilitate the multi-timbral features of the D-110. The D-50 approach can in fact be replicated by the D-110 simply by using just two partials in each tone and combining those tones (timbres) into a D-110 patch with each triggering on the same Midi channel.
The only fundamental difference as stated above being the lack of an LFO on the filter and amplifier. While there is a slightly different 'mixing' approach for partials and tones in both machines the fundamentals remain the same with respect to HOW the partials are combined to begin with. As stated just now this was done using a structure which is really just a fancy term for differences in partial routing within the tone itself. In particular was the fact that LA also implements a ring modulator that can be included in the selected structure or routing. As a point of fact the D-110 and D-50 offer the same structures with the D-110 offering six additional structures than the D-50.
So we can see that apart from some minor routing differences there isn't that much different between the two machines particular when focused on the key sound generators and shapers and so on. But it is worth noting some other departures. In a D-50 patches are treated slightly differently but this is largely terminology to make the idea of the D-110 being multi-timbral a little easier to follow. A 'patch' in a D-50 can be generally compared to a 'timbre' in a D-110 in that both contain the tone(s) and their partials and some measure of control or performance parameters. A D-50 implements an equaliser and chorus within the tone block where the D-110 does not.
Both machines implement reverb, directly in the D-50 patch and globally on the D-110's multi-timbral output. In the context of a single timbre application, where the reverb is applied in either machine is of little consequence. With chorus things are a little different. On-board chorus has been a synth staple for many years and the D-110 doesn't have it. However, whether the scenario of two tones containing two partials, OR one tone containing four partials with the former having chorus in the tone block and the latter relying on chorus after the fact can still produce very close approximations between the two machines, again it comes down to your understanding of the routing.
The D-50 also featured a number of key modes that utilised the upper and lower tones and how they are voiced across the keyboard such as splits, duals and other single or multi-voiced options. This can easily be achieved on a D-110 as each of the eight timbres (one tone, four partials) can have a key range set for each, the rest in terms of multi-voicing is down to how you manage the assignment of Midi channels.
Can a D-110 do a D-50 sound?
After a number of years trading parameters between the two machines I can honestly say that the answer to question is yes, BUT not all the time. There are a great many sounds that can be replicated quite happily in both machines even with the differences in tone mixing. These differences can be ironed our fairly quickly when you understand how each machine mixes and combines each pair of partials and account for them accordingly. Where things get a little contentious is in the area of modulations.
The D-50 is famous for lush, evolving sounds that in effect draw heavily from subtle (and sometimes not so subtle) use of LFO's. Remember the D-110 only has one LFO whereas the D-50 has three that can be routed to each section within a partial. This makes the D-50 are far more flexible machine in the minds of many and I can't disagree with that. That said, any sound utilising just one LFO can be replicated on either machine. Quite ironically when I acquired my D-50 I found myself spending more time with D-110 simply because I was curious to see whether anything I did on the D-50 could be done on the D-110.
The reality is that in 75% of cases it could which over the years has led me to the view that in some quarters the D-110's reputation is thoroughly undeserved. It is, in reality, an extremely capable machine and will continue to sit proudly in my rack for as long as it lives.
But perhaps you can tell the difference from this audio example:
Jason Durbin (aka Lagrange Audio) has been a synth and music tech enthusiast for 30 years since getting his hands on his first synth in 1983 at the tender age of 16. He hasn't earned a single Aussie dollar from music but the journey has been nothing short of incredible and he has met and interacted with some amazing people along the way. Jason is a true enthusiast doing it for nothing more than the pure love of it.
A sneak preview of the soon to be released Zebra 2 update
Dan takes us through the new module and the updates