Yamaha GS1 Electric Piano patch

I've never played the Yamaha GS1 but have heard songs recorded using the electric piano patch - the very same sound on the East West Colossus patch named "GS1 - EP1 ENS". I love that electric piano sound as it has "swir" and a sparkling high end that makes it very good for playing lead melody. No, this is NOT the e. piano sound that made DX7 famous.

I would love to re-create that GS1 patch in the FM7 or FM8. Anyone has this patch/algorithm?


Mukaiya

 

Hi Mukaiya,

there are a few classic DX7 patches trying to recreate the GS1 Sound avaiable, but I never had a closer look at them...

You can find the GS1 algorithm on the following page...

http://members.tripod.com/kmi9000/kmi_multim.htm

...Summa

 

I've yet to come across a DX7, or any patch for Yamaha FM synth, that comes anywhere close to the GS1 electric piano sound.

Thanks for the link to the GS1 algorithm. I'm still looking for the patch/parameters to recreate that sound in FM8.

 

Is an audio demo of that "GS1 - EP1 ENS" avaiable? Except the few times I was "forced" to built one I have not much use for that kind of sound in my music? Still, would be interesting to hear what's so special on a GS1 epiano.
Other that that I've been told that the built in chorus is partly responsible for the characteristic sound of the GS1...

 

I have audio samples that I can post. How do I do it?

Mukaiya

 

You could ZIP it and using the paperclip symbol to attach it to your post...

 

OK. Here's it is.

 

Well I rather tought about a clean example from the East West Colossus Piano, one that allows me to detect the frequency relations...

Other than that I hope you don't mind that I delete the file, since this is copyrighted material...

 

No problem with the deletion. You did the right thing.

Here it is. There are 2 mp3 in the zip file. One played using the Colossus "GS1-EP1" patch and the other the "GS1-EP1 ENS" patch. I believe GS1-EP1 is the raw sound while GS1-EP1 ENS added the GS1 onboard (chorus/tremolo/ensemble?) effect.

Thanks for looking into this.

Mukaiya

 

Sorry, can't find the attachment...

 

I tried uploading the zip file. The uploading seemed to be going ok and the progress bar was moving, but after about close to 10 minutes, the upload windows showed a "this page cannot be found" error, like what you'd see if trying to access a non-existing URL on a browser. What could be wrong?

My zip file size is only 2.37 MB.

Mukaiya

 

Never tried to upload bigger files here, since this is a relatively new feature, could be some kind of time out or high server load. You could try to split it up into multiple files, but I haven't tried that either.

...Summa

 

Finally got the attachment uploaded. Please have a listen. Thanks.

Mukaiya

 

Was a bit hard but fortunately possible to find a few single notes in your play Even so it's rather a chord instrument it's easier to detect and to rebuild when hearing single notes from different regions of the keyboard...

According to the GS-1/GS-2 advertisement I found, the GS-1 was equipped with an ensemble mode. Your demo sounds very much like one of those ensembles, since the left sound is pretty different to the one on the right side. Unfortunately those samples seem to miss all of the FM typical velocity changes in sound. Other than that I can hear quite an amount of quantisation noise in the bass region, what will be pretty hard to emulate. Everything else sounds pretty much like the FM I know...

 

I can record sonatina type songs with distinct single notes, or I can play chromatic scales through the range of the keyboard. Guess that'll help.

Not sure I understand what you are trying to do. Are you attempting to recreate the EP1 sound in FM8?

Mukaiya

 

Well, I'm pretty experienced when it comes to FM-Synthesis, but not exactly a friend of E-Piano sounds, so I persistently tried to find my ways around them :angry: Anyway, since firms think of E-Pianos when it comes to FM-Sound, this is getting harder and harder lately So I guess it's a good thing to know and being able to recreate a few famous ones, so I might give it a try... :|

 

Here it is. I played single sustain note on C and G across 5 octaves, each note is played 4 times with gradually increasing velocity from 64, 85, 106 to 127.

Thanks for taking up the challenge. I used to be pretty decent with FM programming but that was almost 20 years ago. I could handle single modulator-carrier pair sonic interaction, such as if the freq ratio is 1:1 you get a sawtooth wave (if memory serves me right) but found the interaction with additional modulator, feedback modulation, etc to be unpredictable. There was no clear way of knowing what algorithm to use, let alone what freq ratio to use, to recreate a given sound.

Anyway, hope you have success in recreating the GS1 EP1 ENS sound on the FM8. I look forward to that

 

Velocity wise I don't think the GS-1 reacted that way, early FM weren't equiped with filters and the variation of FM Amount sounds pretty much different. So this is probably a confession on sampling...

When adding a bit feedback

This is pure training, after a while you simply know that 1:1.5, 1:3, 1:6, 1:9 ... are for instance fine for bellish spectres...

When it comes to algorithms, their meaning is simply overrated. Sure some are preferable for voice type sounds, others are better when wanting a waveshaper between the Operators or identical lines to detune. But when using a good editor-software that allows to copy operator settings or having free definable algorithms (like in FM7/8), one can change the configuration while making the sounds pretty easily so there is no need to spend too much thought on them. I simply connect Operators the they way I need them...

 

Had a busy weekend tinkering with my computer, but finaly here's a version that might need a bit finetuning velocity and envelope wise. It also misses the spice of the original, but with all operators in use and only one Saturation Operator, I was running out of options, so using a bitreducer might be helpfull...

 

Thanks for spending your weekemd on this! I agree, it is missing the spice of the original sound. One suggestion - can you construct an algorithm just like the one in GS1 and go from there? That way at least you can get "getting the right algorithm" out of the way and just focus on getting the freq ratio correct.