Roland R-70

The machine on the counter

In June 1992, a British reader could meet Roland's R-70 in the pages of Sound On Sound: a grey, compact rhythm composer with sixteen oval pads on one side, a small lit display on the other, and a long pad set above the ordinary ones. The advertised question was how much human movement a programmed rhythm could hold. At £499 including VAT in Ian Steele's review, this was a serious studio purchase. It was also a machine that made its argument through its controls. A player could strike a pad for a sound, then strike a different place along the long one to alter the way a note spoke. Steele, Sound On Sound, June 1992 ↗; Roland R-70 specifications ↗.

Roland called it a Human Rhythm Composer, a name inherited from the R-8 and R-5 line. The positional pad let a player change a hit by striking a different place along its length; built-in effects supplied another way to shape the part. Steele ↗.

Sixteen pads and one strip

The ordinary pads were velocity sensitive. The long pad could change pitch, decay, nuance or pan according to where it was struck. The word nuance had a specific instrument meaning here: on sounds that supported it, the parameter could move between qualities such as an edge strike and a centre strike. It did not apply to every sound. The available variation depended on the selected instrument. Steele ↗; Roland specifications ↗.

There was a second way to use the surface. Multi-instrument mode placed one sound across all sixteen pads, each with a different pitch, decay, nuance or pan value. In step recording, those same pads could become positions in a measure. Here the R-70 changed the meaning of its controls according to the task: a sound palette, a variation map, then a time line. The interface was not simple in every respect; Steele described navigating many pages of parameters with cursor keys. Yet the pad surface offered a direct way back from the screen to the rhythm. Steele ↗.

The same panel offered two ways into a rhythm: shape the sound of a hit, or place that hit within a measure. Switching tasks changed what the pads meant.

A palette you could revise

Roland specified 210 internal instruments and 32 copy-instrument locations. Those numbers describe a fixed PCM sound library and places to reshape copies of sounds, not a bottomless store of newly recorded samples. Roland also advertised a far larger number of possible variations, but that was combinatorial marketing language. The important point is what the player could alter while composing: pitch, attack, decay, nuance, layers and assignments, then the level, pan and effect send of an instrument in the internal mix. Roland specifications ↗; Steele ↗.

The effects offered reverb or delay in one block, chorus or flanging in another. Four audio outputs included a stereo pair and two individual outputs, but the two separate outputs shared resources with the effects. The mixer’s Output/Effect mode chose between individual output 1 and reverb/delay, and between individual output 2 and chorus/flanging. Using both individual outputs disabled both effect blocks; using both blocks left only the stereo outputs available. A producer could also keep one individual output and the other effect block. Roland owner’s manual, Mixer ↗; Trask ↗.

The R-70 could store 100 patterns and 20 songs internally. MIDI and built-in FSK tape synchronisation connected it to sequencers and multitrack tape. Those facts explain why a review of a drum machine spends so much time on workflow: it was a small instrument asked to sit inside a larger room, keeping time while the rest of a recording changed around it. Roland specifications ↗; Steele ↗.

The machine offers an idea, the musician decides

Rhythm Expert could generate an editable pattern in styles including rock, jazz, samba and dance. Steele found its results usable. A model could be varied, saved into an ordinary pattern, or discarded; once saved, its parts could be edited like a rhythm entered by hand. Steele ↗.

Contemporary reviews are valuable because they still hear the instrument as new. Steele admired its sound and the time saved by Rhythm Expert; Simon Trask's July 1992 Music Technology review described strengths and limits from another player's position. The two reviews capture musicians trying to make the new box work for them, alongside Roland's own description of its capabilities. Steele ↗; Trask, Music Technology, July 1992 ↗.

A one-bar beginning in Chicago

In his 2016 Montréal lecture, RP Boo recalled seeing DJ Slugo demonstrate an R-70, then buying a store's last display unit without a manual. He dated the purchase to late 1995. At first he did not know how to extend a pattern beyond one bar, so that loop became his starting point. RP Boo, Red Bull Music Academy lecture ↗.

That account makes the R-70 more than a footnote to Chicago footwork. RP Boo says that the majority of the tracks under discussion in the lecture came from his R-70, and that he kept returning to it even after moving to MPC Studio in 2014. It does not identify the exact machine behind every individual record. RP Boo lecture ↗.

He recalled another breakthrough around 2009: bringing the R-70 and a Tascam trackboard together on “Under'D-Stat,” channelling different samples without putting them into the sampler itself. He identified the track as a Japanese-edition contribution to Legacy. RP Boo lecture ↗.

What survives in the archive

The period reviews show how the R-70 first met buyers. Its controls explain what a musician could do with it. RP Boo's oral history follows one musician's long relationship with the instrument. Together, they carry the story from the shop counter into years of work.