Question about using Xite-1 as a synth module in 96 kHz

The Sonic Core XITE hardware platform for Scope

Moderators: valis, garyb

Mikael-R
Posts: 47
Joined: Fri May 08, 2015 11:33 am

Question about using Xite-1 as a synth module in 96 kHz

Post by Mikael-R »

Hi,
I have done some tests with a 6 dsp card. Now I’m considering an Xite-1 and plan to use it as a kind of super synth module and run everything live with midi for a fast and fun workflow. Question is: Will it run well at 96 kHz in a setup like the following or will it be buggy and cause phase problems due to spread out dsp-load or something?

Example: - 4-5 synths, maybe including something demanding like minimax and something less demanding like uknow007, maybe a relatively complex modular patch and one or two more synths. Polyphony on at least 2 synths.
- Every synth should be connected to their own fx chain. Eq, reverb, delay, sometimes compression chorus, phasers and other effects.
- Effects on the master channel like reverb, eq, compression and possible others. It would be preferable to mix live without bouncing down audio tracks into the DAW. Mastering live is not necessary though.

I’m not sure how much the answer will affect the decision since there isn’t much competition anyway and it isn’t really necessary to run everything live (just desirable) but still it would be good to know if it works well in 96 kHz with that kind of setup.

Also, something a bit unrelated that I have been thinking about lately: Is midi jitter from the DAW a problem to you? Could a device like the USAMO from expert sleepers be a potential fix (audio out from the DAW which then is converted to midi in a hardware box)?

Question summary:
1. Will it work well in 96 kHz?
2. Additional question: Does long fx chains cause noticeable delay? (harder to play from midi keyboard)
3. Comments about midi jitter, USAMO?

PS. When I say “live” I don’t mean on stage, so it doesn’t need to be super stable, just stable.

/Thanks
Mike (Sweden)
User avatar
garyb
Moderator
Posts: 23248
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by garyb »

1. in general, yes. there is plenty of DSP for this. there are other factors in play besides DSP, however. it's possible that some projects like you describe amy run into SAT connection or other issues, but generally, i don't see a problem.

2. maybe, but mostly no. it's similar to a long chain of FX pedals or rack mount boxes. there is a delay, but it's pretty much unhearable. the main exception would be for parallel bass sounds. those might suffer a small amount of phase cancellation, which can be solved by delaying the dry signal a few samples. this is the same issue as with hardware. it won't create any latency that would effect playability in about 99% of situations. the Convolve EQ requires significant processing time, it's probably the main exception. basically, it would have to be a pretty freaking long chain to become an issue, in which case the resulting mud couldn't be played like a normal instrument anyway.

3. midi jitter is mainly an issue of the sequencer. it's very low into the hardware inputs, it's higher through windows.
Mikael-R
Posts: 47
Joined: Fri May 08, 2015 11:33 am

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by Mikael-R »

3. midi jitter is mainly an issue of the sequencer. it's very low into the hardware inputs, it's higher through windows.
1. Are you saying that the jitter is higher if I route midi internally to Xite-1 and lower if I route it externally via a good midi interface?

2. I guess the USAMO box could be used to send note information to Xite's external midi in while I could send midi controller information internally so I not overload the external midi stream?
User avatar
garyb
Moderator
Posts: 23248
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by garyb »

1. maybe. if windows is involved, maybe not. many interfaces are not really jitter-free for the same reason as the sequencer to Scope connection is not. the difference is obvious when a hardware sequencer is used. a hardware sequencer to Scope is typically very low jitter.

2. for music that must be jitter-free, that looks very promising. i haven't seen the external midi ports get overloaded before, but i suppose that it's very possible. i think the USAMO box is probably the ticket, if your sequences must be that tight, like when stacking kick drums.

for controller data between Scope and the sequencer, there's also Scope Sync which is free:
http://www.scopesync.co.uk/
jksuperstar
Posts: 1638
Joined: Mon Nov 15, 2010 12:57 pm

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by jksuperstar »

In my testing, the fastest MIDI I/O loopback I could get was with an iConnectivityMIDI4+ (USB2.0 -> Port -> loopback Cable -> PortIN -> USB), which was about 0.55ms. Since a Hardware MIDI cable takes 0.33ms to send one byte, that's pretty good. However, not sure how this is possible, since the USB MIDI standard groups midi messages into 32 bits...that totals 1ms for a typical midi message. Maybe the iConnectivity device cheats? Anyways...details. Message Jitter was 0.43ms.

XITE (PC out to XITE -> port -> loopback cable -> port -> XITE back in to PC) was total of 2.57ms. Message Jitter was measured at 0.41ms.

Either way, 2.57ms, for a ROUND TRIP midi connection, through the router and out hardware and back in, is really unnoticeable. Jitter between the two interfaces is basically the same. That's faster than nearly any audio interface out of the computer. I did all testing using a program called MidiTest.exe, and found MME was better to use than DirectX MIDI (your sequencer may have an option to select which driver type is used).
Mikael-R
Posts: 47
Joined: Fri May 08, 2015 11:33 am

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by Mikael-R »

Maybe scopesync could be expanded to send jitterfree notedata via audio, not just controller data?
I don’t know how much jitter really matters. I haven’t tested it myself, I only have read about it on forums. I’m probably a bit overly picky in one way, but in another way it make sense to aim for the best possible performance.
Anyway, thanks for the information. Now I feel 100% confident in buying that machine.
User avatar
garyb
Moderator
Posts: 23248
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by garyb »

great!

well, if jitter is extreme, it will make the music messy and bad. small amounts of jitter are unavoidable.

just for a little bit of clarity, the best human performance will have more deviation in timing of notes(jitter) than most midi systems today. :)
User avatar
tlaskows
Posts: 1512
Joined: Fri May 20, 2005 4:00 pm
Location: Ontario, Canada
Contact:

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by tlaskows »

Hmm. I never have any major MIDI timing issues...

I like to use separate MIDI port for each instrument. I like to use a lot of breath control and aftertouch data and that fills up the MIDI pretty fast :o

The only MIDI timing issues I can remember was with a SoundBlaster AWE32 on Windows 95 or 98. Windows MIDI timing used to be a disaster. Bought a cheap 486 junk laptop with windows 3.1 and a serial MIDI interface and it was SOLID! Well, probably not Atari ST solid...

-Tom
User avatar
Bud Weiser
Posts: 2684
Joined: Tue Sep 14, 2010 5:29 am
Location: nowhere land

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by Bud Weiser »

garyb wrote: ... the best human performance will have more deviation in timing of notes(jitter) than most midi systems today. :)
True !

But there´s also a difference in human timing inaccuracy or machine timing inaccuracy.
It´s about when machines were designed to eliminate human timing inaccuracy, they now introduce a different kind and feel of inaccuracy,- machine type inaccuracy so to say,- and it´s much more obvious for the human ear vs the human timing inacurracy.

B.t.w. as good as p.ex. ATARI MIDI timing is vs other computer hardware/OS, already in the 90s not only me but also other musicians and producers recognized MIDI timing issues and w/ ATARI too.
In fact, when MIDI was introduced and everyone out there set up a MIDI studio and made experiences, the timing hardliners still prefered analog clock and trigger pulses over MIDI for a very long time (in fact as long as it was possible).
So, from the very beginning of MIDI, it already was stigmatized for timing inaccuracy.

I remember using the Garfield Minidoc clock synchronizer together w/ all my former analogue synths and drum machines etc., all connected simultaneusly and using different clock rates and there were 2 cool arpeggiator circuits devided in note values and all worked well and like expected.
At this time, it was quite normal it was tight.

Bud
S|C Scope/XITE-1 & S|C A16U, Scope PCI & CW A16U
User avatar
tlaskows
Posts: 1512
Joined: Fri May 20, 2005 4:00 pm
Location: Ontario, Canada
Contact:

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by tlaskows »

Hmm,

whatever happened to that Yamaha thingy that was supposed to use Ethernet? I've only heard about it a few times, but I don't think anyone liked the idea...

-Tom
User avatar
garyb
Moderator
Posts: 23248
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by garyb »

Bud Weiser wrote:
garyb wrote: ... the best human performance will have more deviation in timing of notes(jitter) than most midi systems today. :)
True !

But there´s also a difference in human timing inaccuracy or machine timing inaccuracy.
It´s about when machines were designed to eliminate human timing inaccuracy, they now introduce a different kind and feel of inaccuracy,- machine type inaccuracy so to say,- and it´s much more obvious for the human ear vs the human timing inacurracy.

B.t.w. as good as p.ex. ATARI MIDI timing is vs other computer hardware/OS, already in the 90s not only me but also other musicians and producers recognized MIDI timing issues and w/ ATARI too.
In fact, when MIDI was introduced and everyone out there set up a MIDI studio and made experiences, the timing hardliners still prefered analog clock and trigger pulses over MIDI for a very long time (in fact as long as it was possible).
So, from the very beginning of MIDI, it already was stigmatized for timing inaccuracy.

I remember using the Garfield Minidoc clock synchronizer together w/ all my former analogue synths and drum machines etc., all connected simultaneusly and using different clock rates and there were 2 cool arpeggiator circuits devided in note values and all worked well and like expected.
At this time, it was quite normal it was tight.

Bud
yep, true!
Mikael-R
Posts: 47
Joined: Fri May 08, 2015 11:33 am

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by Mikael-R »

Another question:
I have read that there is a bug regarding midi cc assignment recall.
I’m thinking about completely remote controlling synths like minimax from the DAW by creating a VST clone that sends out cc messages, even storing and recalling presets and easily automate any parameter via that clone. Or will this not work because of the bug?
User avatar
tlaskows
Posts: 1512
Joined: Fri May 20, 2005 4:00 pm
Location: Ontario, Canada
Contact:

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by tlaskows »

I think that won't work, because by default no CCs are assigned as far as I know...

-Tom
User avatar
Bud Weiser
Posts: 2684
Joined: Tue Sep 14, 2010 5:29 am
Location: nowhere land

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by Bud Weiser »

tlaskows wrote:I think that won't work, because by default no CCs are assigned as far as I know...

-Tom
Depends on device, some CCs might be pre-assigned,- pitch wheel, sustain pedal etc., others aren´t.

You CAN assign any MIDI CCs to device´s GUI knob/button parameters and store these assignments in MIDI presets for each device.
But when you close the project and recall it later, assignments might be gone, sometimes all, sometimes partially.
You cannot make it total recall and working flawlessly.
Instead and when the project is up again, you load MIDI presets for each device manually again and hope all the assignments from yesterday will be there and if not you start re-assigning.
That´s unfortunately not a reliable solution for live usage and it depends on for you available setup times, your personal presence and willingness to work around this all the time.
When you have roadies setting up your gear at noon while you´re still traveling in the bus, that might be not possible at all, especially when there´s not enough time before soundcheck or no soundcheck at all or you and the band arrive late and so on.

In the home studio, it´s not the dealbreaker per sé because this bug exists since the PCI cards and isn´t XITE related at all.
You see, most use the system w/ PCI cards and/or XITE in the studio being satisfied.

Also for me, the original intention was using SCOPE/XITE as a super synth module.
I can do in the studio, but I gave up for live because I´d need it total recall and that is what I hope for in SCOPE 6.

For live, I´d need a fully loaded autostart project and when it´s up, all should work as it should, so any roadie after some instructions can power up XITE and the computer and after a few minutes (project dependent) load time can press some keys and do a line check.
And when you come to the venue, you press a button for the 1st MIDI Prg.-Change number which recalls overall state of your rig and play the 1st tune in the setlist.
That´s how it always worked for me w/ hardware and should work w/ any virtual devices too.

Bud
S|C Scope/XITE-1 & S|C A16U, Scope PCI & CW A16U
JoPo
Posts: 2306
Joined: Wed Jun 05, 2002 4:00 pm
Location: VRRAAaaooOôOooommmh
Contact:

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by JoPo »

Oh ! USAMO box looks very interesting !

Is there anybody around here who would have test it ? With Xite-1 ?

Receiving midi data through ScopeSync would be awesome !
> > > > > > > > > > > > --- Musica --> here ! ---< < < < < < < < < < < <
User avatar
tlaskows
Posts: 1512
Joined: Fri May 20, 2005 4:00 pm
Location: Ontario, Canada
Contact:

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by tlaskows »

Yes Bud. I know you can assign CC # to any knob or button. But what's the point if they can't be saved as referred to many posts on here :(

-Tom
User avatar
garyb
Moderator
Posts: 23248
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by garyb »

it depends on the device and how the developer made it.

for example, the STM mixers recall 100% just by saving the project.
User avatar
Bud Weiser
Posts: 2684
Joined: Tue Sep 14, 2010 5:29 am
Location: nowhere land

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by Bud Weiser »

tlaskows wrote:Yes Bud. I know you can assign CC # to any knob or button. But what's the point if they can't be saved as referred to many posts on here :(

-Tom


Well, Dawman gigged w/ it and as a synth module for years in Vegas and made good cash ...

Gary´s friend, a keyboardplayer, uses XITE for synths live too ...

It seemed they found workarounds to use it.
Me personally, I never got it running the way they did and failed already when storing MIDI presets.

Bud
S|C Scope/XITE-1 & S|C A16U, Scope PCI & CW A16U
User avatar
tlaskows
Posts: 1512
Joined: Fri May 20, 2005 4:00 pm
Location: Ontario, Canada
Contact:

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by tlaskows »

Honestly, I've never even saved my mappings. So maybe I should try it out first. I just read the horror stories, but others don't seem to complain.

-Tom
User avatar
garyb
Moderator
Posts: 23248
Joined: Sun Apr 15, 2001 4:00 pm
Location: ghetto by the sea

Re: Question about using Xite-1 as a synth module in 96 kHz

Post by garyb »

it depends on the device.
Post Reply