The (real) Jupiter 8 Clone

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Joe., lisa, luketeaford, Kent

User avatar
ACME
Learning to Wiggle
Posts: 38
Joined: Wed Jan 17, 2018 8:39 am
Location: Germany

Re: The (real) Jupiter 8 Clone

Post by ACME » Thu May 21, 2020 6:24 am

beatkamp wrote:
Tue May 19, 2020 2:41 pm
Hi Marcus,
I would suggest that you simply sell the manufactured and fully populated boards, perhaps as a set, and let the buyer wrestle with them.

If you offered a full set of CPU Board x 1, Module Controller x 2, Module Board x 4, Interface Board x1 I would buy it right now.

I see the Interface & CPU boards up on Ebay already and that made me smile. Frankly, if the Module Boards and Module Controller Boards were up there now, I would be buying a set right now.

With such an offering, the strange and wonderful projects that would be the fruit of your work should dazzle all who behold them.
I could do that. But only to people who know exactly what they are doing and have the skills to build the missing panel boards themselves. In the future I will surely be able to offer the panels, at the moment I don't need them, because the potentiometers and switches are built on breadboards. The Bender board is not absolutely necessary for the operation of JP8. You can simply bypass it.

User avatar
ACME
Learning to Wiggle
Posts: 38
Joined: Wed Jan 17, 2018 8:39 am
Location: Germany

Re: Re:

Post by ACME » Thu May 21, 2020 6:30 am

Sir Ruff wrote:
Wed May 20, 2020 11:19 pm
ACME wrote:
Thu Dec 19, 2019 3:21 am
Hello, everybody,

I am currently building and developing 2 different Module Boards at the same time. This is because I use two different ENVGens. I think that I have to decide ;-)

The digital ENV genes (Druid ENVGEN8) are stable, but have a bug at startup that is not fixed yet. The envelopes are also not as organically round as with analog ENV Gen.

The AS3310 have the nicer envelopes. They also fit better with the original ones. Their disadvantage is that they require more effort for the conversion of the control voltages. The KB-ENV tracking is also complex to control. Could you do better without this function? The disadvantage that the ENV times in the warm-up phase are not correct is probably bearable. Almost all other famous analog synths use the 3310 as well.
I bought an original CEM3310 for comparison, but unfortunately it doesn't seem to work.

What is your opinion?
Roland envelopes typically have a very specific sound due to their slopes. If the CEM3310 clone can be shaped the same way (or close) then that might work, but as Tobb says a discrete envelope might be easiest in long run.

If you do use the 3310 and have to remove the keyboard->envelope tracking, I personally don’t think that’s a huge loss. Probably one of the least used features.
What is the difference in sound?
Well, I have the IR3R01ENV from Roland. The envelopes are absolutely comparable to the 3310's. You don't hear any difference at all, and on the oscilloscope there are no significant deviations. Even in an A/B comparison in a Jupiter 8 you could not hear any differences. The envelopes do not have a special "sound". At least I did not notice any.

User avatar
clusterchord
Super Deluxe Wiggler
Posts: 1203
Joined: Tue Sep 07, 2010 8:46 pm
Location: Zagreb, Croatia
Contact:

Re: turning into a DIY JP8 - your suggestions

Post by clusterchord » Thu May 21, 2020 10:34 am

ACME wrote:
Wed Nov 27, 2019 8:30 pm
It would be helpful to find someone who would develop a new CPU for the extended desktop JP8 together with me.
We would also have to decide if this version should work with the IR3109 in the filters, or if new filters should be built with other OTA, e.g. 13700 or 3080. Then we won't talk about a Jupiter 8 any more, but that doesn't interest anyone nowadays anyway.
hi,

i use 14-bit Jup8 for past 15 yrs. i'd like to give my 2 cents, since you asked on several points thruout the thread..


great that you can limit the ammount of IR3109 to just 8 pieces in the filter. imo, filter definitely needs to stay IR3109. for three reasons:

1 accurate sound
2 best availability out of all vintage filter ICs
3 a DIL-16 clone will surely come up sooner or later, since all of them are. even if it will sound slightly different (they always do) a prospective builder will have a choice to put that in or find some originals.

thing is when recreating a behemoth from yesteryear, its a sum of its parts. everything is important. once you start thinkin in terms of what is "good enough", you might as well build a completely different synth. but then all bets are off.

same with VCAs, i hope we can choose either,, orig (expensive) BA662 or the clone (from RE303) project?

old CA3080 also sounds great, but it will change the sound. not sure about the Rochester reissue. never heard them. LM1xxxx serie i heard on many Roland clone attempts, and it always sounds off. the non linearity characteristic is different, clean, bright as betlehem. no saturation. no growl in lo mid.


that said, i'm not against well chosen subtle additions to vintage designs, but ones that can default to original behaviour if desired. a new CPU that governs midi, midi cc. possibly add a second randomized digital lfo that can sync to incoming clock, would be welcome, just that if i turn it off.. it still works exactly like Jup from 1981.
ACME wrote:
Thu Dec 19, 2019 3:21 am
The digital ENV genes (Druid ENVGEN8) are stable, but have a bug at startup that is not fixed yet. The envelopes are also not as organically round as with analog ENV Gen.

The AS3310 have the nicer envelopes. They also fit better with the original ones. Their disadvantage is that they require more effort for the conversion of the control voltages. The KB-ENV tracking is also complex to control. Could you do better without this function? The disadvantage that the ENV times in the warm-up phase are not correct is probably bearable. Almost all other famous analog synths use the 3310 as well.
I bought an original CEM3310 for comparison, but unfortunately it doesn't seem to work.

What is your opinion?
i agree with others that a transistor based envelope should be the choice. CEM have a disticntly different shape, and sound bouncey great on P5 rev3, but don't have the acoustical twang or snap that Jup8 have. closest to it is transisotr envelope on original SH101. on polyanalog i tend to dislike the lifelesness of digital envelopes. if they are on complex deep digital synth, then it sort of goes with the territory and fits well.

i have to tell you the envelope on few demos you posted sounded off. very un snappy and un sexy which is one of the main traits of Jup8.

you are correct in thinking that decay times btwn voices on Jup8 do vary , they always do, warm or cold. however, offsets arent huge, like in musically unusable. its what gives it life, and when you run a random arpeggiator it sound almost as it are running another row of step seq varying the decay times. lively. for example: this is an ancient demo of mine, which shows the decay difference btwn voices, as well as how snappy that mother is:



Sir Ruff wrote:
Wed May 20, 2020 11:19 pm
If you do use the 3310 and have to remove the keyboard->envelope tracking, I personally don’t think that’s a huge loss. Probably one of the least used features.
i find it unavoidable and cruical part of the behaviour of this instrument. with certain sounds its just sounds natural. and it goes so well with already "acoustical" twang of IR3R01. for plinky arpeggios, to certain synth basses etc.

imo it is one of the things that makes it different from other peers like prophet5, obxa and all the monos from the golden era.

ACME wrote:
Mon Oct 07, 2019 11:57 am
From a purely technical point of view, it is possible to control Modules and ModCon by a microcontroller via Midi only. It's up to you to decide whether you need the autotune or not.
I can adjust the VCOs so exactly that I even could play without AT. Sounds a bit more "analog".
i prefer the "classic polysynth" era implementation - when you choose to, if the vcos are slightly all over the place, you press AT, it does it thing and stops.

however, i'd vote against any continous CPU control of the pitch. results are terrible and lifeless. worse than a DCO actually. some bigger manufacturers have opted for this, and entirely destroyed their analog sound.

calculating temperature offsets in slow unonbstrusive manner, might be as additional option but it has to be defeatable. Andromeda has this solved nicely: it has the momentary autotune function, it has the dreaded background cpu tune which is deteatable, and it has temperature offset, which is also defeatable.
Image

FS EU: Rene 1, Synthtech E370, MXR MF403 (pm me)
WTB EU: Jupiter 4 BA662, Octave Cat SRM (2040)

KSS
Super Deluxe Wiggler
Posts: 1931
Joined: Mon Jan 25, 2016 7:28 am

Re: The (real) Jupiter 8 Clone

Post by KSS » Thu May 21, 2020 9:28 pm

:agree: With all of this, and especially these below. But really all of it is spot on! Well done, clusterchord!
great that you can limit the ammount of IR3109 to just 8 pieces in the filter. imo, filter definitely needs to stay IR3109. for three reasons:
thing is when recreating a behemoth from yesteryear, its a sum of its parts. everything is important. once you start thinkin in terms of what is "good enough", you might as well build a completely different synth. but then all bets are off.
i agree with others that a transistor based envelope should be the choice. CEM have a disticntly different shape, and sound bouncey great on P5 rev3, but don't have the acoustical twang or snap that Jup8 have. closest to it is transisotr envelope on original SH101. on polyanalog i tend to dislike the lifelesness of digital envelopes. if they are on complex deep digital synth, then it sort of goes with the territory and fits well.
i have to tell you the envelope on few demos you posted sounded off. very un snappy and un sexy which is one of the main traits of Jup8.
clusterchord wrote:
Sir Ruff wrote: If you do use the 3310 and have to remove the keyboard->envelope tracking, I personally don’t think that’s a huge loss. Probably one of the least used features.
i find it unavoidable and crucial part of the behaviour of this instrument. with certain sounds its just sounds natural. and it goes so well with already "acoustical" twang of IR3R01. for plinky arpeggios, to certain synth basses etc.

imo it is one of the things that makes it different from other peers like prophet5, obxa and all the monos from the golden era.

User avatar
ACME
Learning to Wiggle
Posts: 38
Joined: Wed Jan 17, 2018 8:39 am
Location: Germany

Re: The (real) Jupiter 8 Clone

Post by ACME » Fri May 22, 2020 6:48 pm

I already find the subject of envelopes interesting. Currently my shortest A / D time is 0.6ms. Even shorter times don't make sense somehow. I didn't choose the R and C values of the AS3310 according to the data sheet, but found values where the ENV is about the same as the IR3R01. Indeed, I can't see any difference to R01. What is the advantage of the discrete transistor Env? What makes them more similar to the IR3R01 Env? Here is an example of the arpeggiator. There are so many ways to manipulate the sound. The ENVs are only one aspect of this.

You do not have the required permissions to view the files attached to this post.

User avatar
Randy
Super Deluxe Wiggler
Posts: 1846
Joined: Sat Sep 22, 2012 9:50 pm
Location: Mississauga, Ontario, Canada

Re: The (real) Jupiter 8 Clone

Post by Randy » Fri May 22, 2020 10:48 pm

Not sure if this is useful but the original SEM EGs are a discrete transistor design, and not very complicated at all. They may sound close to the Jupiter's, or maybe not. Just wanted to mention it.

Post Reply

Return to “Music Tech DIY”