Linn LM-1 HiHat

This video documents the sample output of the Linn LM-1 as compared to the Oberheim DXa. The Linn LM-1 has been well noted for its characteristic 'groove' which is felt to be unique among EPROM-based digital drum machines.

Many have suggested that the slightly imperfect timing of the Linn LM-1 is responsible for the feel and I think that's at least partially true. But the LM-1 also has a unique way of creating its HiHat sound, which I wanted to show.

Most EPROM-based drum machines (re)trigger their Hat sample from the beginning of the sound in the stored EPROM. This gives a consistent but static delivery of the sound sample. The LM-1, however, has its EPROM clocked continuously, which means the sample is always being looped. To trigger the Hat sound, the LM-1 merely applies a volume envelope to the DAC. This means that the LM-1's HiHat output is a little different each time it is triggered. My guess is that Roger Linn did this either because memory was very expensive in the late 1970's or he's brilliant in realizing that each strike of a real cymbal is unlikely to be exactly the same ... or both!

The video shows the EXT CLOCK OUT of the LM-1 triggering the DXa via the Oberheim's EXT CLOCK IN (not MIDI). The top trace is the Hat sample of the Oberheim DXa and the bottom trace is the Hat sample output of the LM-1. You can clearly see the difference in the sampled waveform between the two machines.

I tried hacking my DXa to do this and it sounded kind of cool, but you really have to have the right HiHat sample for it.

Category:

Music

Tags:

• drum
• machine
• Linn
• LM-1
• LM1
• LM-2
• LM2
• Oberheim
• DX
• DXa
• DMX
• eprom
• sample
• digital
• groove
• Prince
• funk

[Synthasy2000]

Another "groove" element of the LM-1 is in it's late sounding snare.

It is about 10ms after the trigger. This is due to the snare sample not being trimmed right to the main snare transient.

All LM-1s will sound laid back!

[rolandsh1000]

I haven't measured that, but I believe it!

And there's another thing I found:

I was trying to build a MIDI interface for the LM-1 using a 3rd party MIDI-to-trigger board. The idea was to use 4066's to electrically close the front panel switch buttons in response to a MIDI trigger.

But it didn't work - I kept getting voices dropping out. I discovered the problem was you can't press two drum buttons at the exact same time! The LM-1 will only play one of the sounds!

[NEILVANCE]

Thank you so much for your demonstration!

Proves that you can't really get away with it by sampling the lm-1, for that 100% linn lm-1 sound, you need a real lm-1!

I always loved the hat's on this machine, on records the hat sounds slightly different on each trig, your demo proves the reason for why! - thank you. (proud lm-1 owner & lover!)

[hydrogxn]

this is such a great video. thank you.

[VacTrooper]

All the more reason why I need an LM-1. One of the. 500. :(

[ctspf]

A very interesting video. Nowadays you could achieve something like that programming at random (within a lilmited range) several parameters of the ADSR envelope of a sound and you could get pretty much the same result, but hey, this machine was released in 1980.

[VisceralVoids]

I have always found this video fascinating - the magic of the LM1. Every time the sample is triggered it is different. Amazing that such an (early) piece of digital gear like the Linn LM1 can have such a "human" element.

[alphabeets]

Excellent video. This will indeed make a huge difference in the overall sound. Slick design!

[Synthasy2000]

Quite possibly because the keys are scanned sequentially by the CPU.

Just put a small delay (1ms) between "presses".