The No.1 Website for Pro Audio
New Roland Jupiter X / XM
Old 22nd February 2020
  #3931
Lives for gear
 

Quote:
Originally Posted by warg View Post
New hands-on video of the Jupiter-X posted last night - supposedly first in a series:

So they use the center 7 segment displays to show a parameter's current value, hadn't noticed before. I think there's enough place left on the left to display something else, like the stored value or maybe an abbreviated version of the parameter's name*. Just to be a little less reliant on the far left main screen.



* Or a custom message on startup like the BSII.
Old 23rd February 2020
  #3932
Gear Maniac
 

Quote:
Originally Posted by ToyBox View Post
I think there's enough place left on the left to display something else, like the stored value or maybe an abbreviated version of the parameter's name*.
I you look closely at the vids from the NAMM, that's exactly what it does.

Edit : my bad, it's bank/preset number or model only. There are parameter values going up tu 2048, but there's still enough room indeed.
Old 23rd February 2020
  #3933
Gear Maniac
 
mkastrup's Avatar
Anyone managed to get Jupiter XM Arp to MIDI Clock sync proper with external gear without feeding it notes from a sequencer ?

In SYSTEM/ARPEGGIO you can set Arp Sync to OFF-BEAT-MEASURE

With OFF the Arp will start when you hit a key.
With BEAT the Arp should start on next beat.
With MEASURE the Arp should start on next measure.

Problem with OFF is that you have to hit they key excatly ON the beat for the Arp to sync proper to the recieved MIDI Clock. This is impossible to do manually and have to be done from a sequencer.

Problem with BEAT is that it start to play at the correct BPM but at a wrong place within the beat.

Problem with MEASURE is the same as BEAT except it picks up on measure.
Old 23rd February 2020
  #3934
Lives for gear
 
xanax's Avatar
^ that issue was mentioned couple pages back. Seemingly impossible to sync correctly which is nuts (iArp is a big feature of this synth).

Please don’t hesitate to report this bug to Roland via Roland Backstage portal!
Old 4 weeks ago
  #3935
Here for the gear
 

Another Jupiter-X video has been posted by Mr. Tuna on YouTube - this one featuring the JX-8P sounds:

https://www.youtube.com/watch?v=0qN4_IL5EwU
Old 4 weeks ago
  #3936
Lives for gear
 
VennD68's Avatar
This one is IMHO the best example thus far.
It certainly captures that "something something Roland" in the sonic department incredibly well and is certainly a good sounding digital board.
You could easily gig with this and leave your OG's at home or get a sonic imprint that isn't all that dissimilar to what one would expect using retro production techniques.
Lots of EBM, Darkwave and so on to be found in there as well I'd imagine.

Do yourself a favour though and skip the first 3 minutes......
Old 4 weeks ago
  #3937
Gear Maniac
 

Quote:
Originally Posted by VennD68 View Post
This one is IMHO the best example thus far.
Just the usual factory Scene (edit : « 05-10 Upside Down »), i.e. nothing different than the Xm being played with a larger keyboard or sequenced.





Provided you tame the i-arp a bit, you sure can waste several hours noodling with these things, even better with a hold and a volume pedal.

Last edited by Poumtschak; 3 weeks ago at 09:53 PM..
Old 4 weeks ago
  #3938
Lives for gear
 
LIFEPASS4's Avatar
Quote:
Originally Posted by Poumtschak View Post
Just the usual factory Scene, i.e. nothing different than the Xm being played with a larger keyboard or sequenced.





Provided you tame the i-arp a bit, you sure can waste several hours noodling with these things, even better with a hold and a volume pedal.
I was thinking of a second XM that's if the price dropped , REV 2 beat it , have to have one , still like XM , yes hours of noodling , have a BigSky , might mess with it ,better then staring at 2 much TV
Old 4 weeks ago
  #3939
Lives for gear
 
Oscar1's Avatar
Quote:
Originally Posted by Poumtschak View Post
Just the usual factory Scene, i.e. nothing different than the Xm being played with a larger keyboard or sequenced.





Provided you tame the i-arp a bit, you sure can waste several hours noodling with these things, even better with a hold and a volume pedal.
I do love the sound, but when you watch few videos and they all use the exact same sound and pattern, the thing became pretty sad. They should start releasing these without any presets.
Old 4 weeks ago
  #3940
Quote:
Originally Posted by Oscar1 View Post
I do love the sound, but when you watch few videos and they all use the exact same sound and pattern, the thing became pretty sad. They should start releasing these without any presets.
Yeah, I’m kind of noticing that trend too - it’s like “OK, the Stranger Things score demos are well represented - let’s hear something we *haven’t* heard before.” It’s such a capable synth with a huge sound and massive potential... it would be nice to hear a little more of that than cheesy 1-finger arp demos. If it weren’t for guys like Gattobus showing off more of its potential, I would never have had any interest in it. Thankfully I had enough time with one at NAMM to know what the X/Xm can do, so it’s a shame there aren’t better representations of it yet!
Old 4 weeks ago
  #3941
Quote:
Originally Posted by whinecellar View Post
it would be nice to hear a little more of that than cheesy 1-finger arp demos
I resemble that remark
Old 4 weeks ago
  #3942
Quote:
Originally Posted by rsaintjohn View Post
I resemble that remark
Ha - I’ve heard you do otherwise, my friend ;-)
Old 4 weeks ago
  #3943
Gear Nut
JP8000 on the XM?

Anyone tried to emulate JP8000 on the XM? The supersaw in the VA synth sounds much better than the Integra 7/ JD-XA and it seems promising. I've tried using Bitcrush to get that aliasing that I remember from the JP8080 I used to have but it's long gone and I don't have any real reference anymore. I think I can get it close, but has anyone who still has a JP8000 tried to do it?
Old 4 weeks ago
  #3944
Lives for gear
 
gentleclockdivid's Avatar
 

Afaik the supersaw in the xm is also sample based .
Old 4 weeks ago
  #3945
Gear Addict
I've attached files from when I did a quick test on the supersaw - using ableton to send a midi note to the xm and recording (both via usb) , so the timing should be identical.

xm va saw1.wav is a single supersaw instance

xm ss comb.wav is 2 separately recorded instances played back simultaneously with one phase inverted
Attached Files

xm ss comb.wav (421.3 KB, 1067 views)

xm va saw1.wav (520.6 KB, 1082 views)

Old 4 weeks ago
  #3946
Lives for gear
 
gentleclockdivid's Avatar
 

No you should record a pass , and record a second note (same note ) , and make sure there are no modulatin effects on or lfo's modulating stuff

You should line them up exactly ( sample accurate ) and invert the phase of one recording , if cancelling out the change is big that it's sample based .
Old 4 weeks ago
  #3947
Gear Addict
I'm not following, how are these 2 things different?

"2 separately recorded instances played back simultaneously with one phase inverted"

"No you should record a pass , and record a second note (same note )"


- - -

to be clear, what i did was have a midi clip in ableton sending a single note to the xm and an audio track recording the xm output.

I did that twice, then played the 2 resultant audio-clips back simultaneously in 2 audio tracks, one with its phase inverted, with both routed to a third audio track for recording the output.
Old 4 weeks ago
  #3948
Lives for gear
 
gentleclockdivid's Avatar
 

Are they lined up perfectly , meaning starting at the exact same sample position ?
Just post the two different recording without phase inversion and modulation , effects etc ..just pure dry ) , I 'll have a look
Old 4 weeks ago
  #3949
Gear Addict
as the play and record clips were both triggered at the same time via scene and it was all via usb I would expect it to be sample accurate.

Unfortunately I don't have both raw supersaw original recordings - only a single and a combined. I'll redo it.
Old 4 weeks ago
  #3950
Gear Addict
I've redone it - same as before, ableton: midi clip note trigger>audio clip record, triggered by scene all USB:

ssaw1/2.wav - the raw files
ssaw comb.wav - combined with inversion
ssaw comb 13smpl.wav - combined with inversion and 13 sample delay on non-inverted

xm setup:
zen int model
initialize tone
filter off
output thru
all effects off
analog feel 0
osc type ssaw
ssaw detune 64 (default)
Attached Files

ssaw comb 13smpl.wav (393.0 KB, 1034 views)

ssaw comb.wav (396.0 KB, 1025 views)

ssaw1.wav (421.3 KB, 1028 views)

ssaw2.wav (421.3 KB, 1019 views)

Old 4 weeks ago
  #3951
Gear Addict
I've just redone it again 3 more times and the delay needed to render the weakest signal varied each time, as does the level of that weak signal - it was 10 samples (very weak combined signal), 0 samples and then 160 samples (still a stronger signal than the other tests).
Old 4 weeks ago
  #3952
Gear Nut
Quote:
Originally Posted by Paul_xyz View Post
I've just redone it again 3 more times and the delay needed to render the weakest signal varied each time, as does the level of that weak signal - it was 10 samples (very weak combined signal), 0 samples and then 160 samples (still a stronger signal than the other tests).
doesn't that mean it's not samples.

I don't know, using the Jupiter X model, the supersaws I am getting with mild eq and slight bit crushing sound a bit more convincing JP8080, at least what I remember, than the System 1/8 and slightly better than V-synth, although V-synth had the more detailed control over the oscillator. But that's my subjective impression so far.
Old 4 weeks ago
  #3953
Gear Addict
I don't think phase cancellation on its own is a good enough barometer for whether something uses samples or not.

A pure calculation algorithm with no random element that uses the same start value/position on trigger would also phase-cancel with a second inverted recording.

What's interesting is just how close to null the xm can get. One possibility is that it's using an algorithm with a variable start value/position and either the diff is coded non-linearities, or else it's caused by a difference in processing rate/resolution vs sample rate/bit depth.

There's detune at play in both supernatural and zencore too. That could be done via pure calculation algorithm, or by layering sampled saw waves and varying the playback speed/pitch.

Fully expect to have my ass handed to me on the above by someone far more knowledgable than me on all this stuff.
Old 4 weeks ago
  #3954
Lives for gear
 
gentleclockdivid's Avatar
 

Quote:
Originally Posted by Paul_xyz View Post
I've just redone it again 3 more times and the delay needed to render the weakest signal varied each time, as does the level of that weak signal - it was 10 samples (very weak combined signal), 0 samples and then 160 samples (still a stronger signal than the other tests).
O.k. here are my findings
I took both of your recordings an inverted the phase of the second one ,almost complete silence -84 db ...but I suspect this is because minor differences in recording .


I then took the left channel of your first recording and the left channel of your second recording and alligned them sample perfect , again almost complete silence , the samples look identical .
A supersaw never repeats itself identically , so I think that again these are sampled based supersaws (jup 80 , integra , fa )

It's pretty easy , if your hold a note long enough and repeat , the waveforms will look identical






I don't get your delay part 130 samples , you just prepare them afterwards ..you cut of the silence before each recordings , to do so you zoom in on sample level and cut everything before the very first sample for each recording or don't cut but make sure they are alligned sample accurately .
It's no rocket science

Last edited by gentleclockdivid; 4 weeks ago at 01:11 AM..
Old 4 weeks ago
  #3955
Gear Nut
Quote:
Originally Posted by gentleclockdivid View Post
O.k. here are my findings
I took both of your recordings an inverted the phase of the second one ,almost complete silence -84 db ...but I suspect this is because minor differences in recording .


I then took the left channel of your first recording and the left channel of your second recording and alligned them sample perfect , again almost complete silence , the samples look identical .
A supersaw never repeats itself identically , so I think that again these are sampled based supersaws (jup 80 , integra , fa )

It's pretty easy , if your hold a note long enough and repeat , the waveforms will look identical






I don't get your delay part 130 samples , you just prepare them afterwards ..you cut of the silence before each recordings , to do so you zoom in on sample level and cut everything before the very first sample for each recording or don't cut but make sure they are alligned sample accurately .
It's no rocket science
This doesn't prove they are samples. They could be mathematically generated and would still cancel out. JP8000 might not of because of the older processor, and precision of calculation back then, giving small variances. You can't prove your assertion this is a sample based on phase cancellation.
Old 4 weeks ago
  #3956
Lives for gear
 
goldphinga's Avatar
 

I put out a run through of every single preset from every engine months ago on my channel!!

Quote:
Originally Posted by whinecellar View Post
Yeah, I’m kind of noticing that trend too - it’s like “OK, the Stranger Things score demos are well represented - let’s hear something we *haven’t* heard before.” It’s such a capable synth with a huge sound and massive potential... it would be nice to hear a little more of that than cheesy 1-finger arp demos. If it weren’t for guys like Gattobus showing off more of its potential, I would never have had any interest in it. Thankfully I had enough time with one at NAMM to know what the X/Xm can do, so it’s a shame there aren’t better representations of it yet!
Old 4 weeks ago
  #3957
Gear Nut
I wanted to know if anyone who has a JP8000/8080 and an XM has managed to emulate/simulate the sounds closely, because this supersaw oscillator, however it's made, sounds really good, and there is also the feedback sync oscillator in this Zen Core model, derived from the JP8080 feedback oscillator, so I have been trying to emulate the sound of the JP8080 I used to have. I no longer have it so it's hard to compare directly.
Old 4 weeks ago
  #3958
Gear Addict
Quote:
Originally Posted by gentleclockdivid View Post
I don't get your delay part 130 samples , you just prepare them afterwards ..you cut of the silence before each recordings , to do so you zoom in on sample level and cut everything before the very first sample for each recording or don't cut but make sure they are alligned sample accurately .
It's no rocket science
I'm sweeping the delay to line the samples up at the point of peak cancellation - same end result as you cutting silence to line them up. I've done it both ways but find the delay method faster and easier.

What I've found is there's a different offset needed each run.

I've attached 2 more test runs - this time with much longer notes. This has the strongest cancelation when I delay "ssaw2 long.wav" by 21 samples (I've attached a file of this too).

I take back my talk about random starting values/positions - the only random thing seems to be number of samples it takes before it sounds. The supersaw itself looks pretty much identical each run (I've attached pics of the initial samples from the short and long runs - only useful for rough visual comparison as they were taken at slightly different zooms/level scales).

Quote:
A supersaw never repeats itself identically
There's no reason why roland couldn't generate something they call a supersaw with a pure calculation algorithm that does repeat.

The smoking gun would be an obvious sample loop-point artefact. If it is a sample then it seems to loop smoothly, which means that a supersaw can loop smoothly, which means an acceptable supersaw can loop smoothly via calculated algorithm. That said, I've not noticed any obvious looping in the samples I've taken, and a pure calculation algorithm could generate the exact same output each time it is triggered, but never repeat whilst it's sounding.

added: not sure if this helps in any way, but if I set up 2 partials identically with the supersaw, they are clearly sample tight.

I've attached a file of this "ssaws 2 partials.wav" - it plays partial 1 detune 64, partial 2 detune 64, partials 1 and 2 simultaneously detune 64, and finally partial 1 detune 64 whilst partial 2 is manually swept from detune 64 to 127.
Attached Thumbnails
New Roland Jupiter X / XM-ssaw-cmp.png  
Attached Files

ssaw comb long 21smpls.wav (3.71 MB, 549 views)

ssaw1 long.wav (3.53 MB, 543 views)

ssaw2 long.wav (3.53 MB, 549 views)

ssaws 2 partials.wav (4.28 MB, 554 views)

Old 4 weeks ago
  #3959
Quote:
Originally Posted by goldphinga View Post
I put out a run through of every single preset from every engine months ago on my channel!!
Found them - thanks! Very helpful ;-)
Old 4 weeks ago
  #3960
Lives for gear
 
LIFEPASS4's Avatar
The X/XM Facebook page , the Goldboys Club , nonpros excluded
Post Reply

Welcome to the Gearslutz Pro Audio Community!

Registration benefits include:
  • The ability to reply to and create new discussions
  • Access to members-only giveaways & competitions
  • Interact with VIP industry experts in our guest Q&As
  • Access to members-only sub forum discussions
  • Access to members-only Chat Room
  • Get INSTANT ACCESS to the world's best private pro audio Classifieds for only USD $20/year
  • Promote your eBay auctions and Reverb.com listings for free
  • Remove this message!
You need an account to post a reply. Create a username and password below and an account will be created and your post entered.


 
 
Slide to join now Processing…
Thread Tools
Search this Thread
Search this Thread:

Advanced Search
Forum Jump
Forum Jump