PlanetZ forums - scopeusers.com

Hi !

2 questions on DSP assignment:

Project 16Bit 44.1K

All ASIO-2 I/Os, WAVE Source-1 and hardware I/Os on DSP#1
1x VDAT on DSP#2
DSPs #3 - #6 free for the communications w/ the 4 slots.

Slot-1 / DSP#7 / STM2448X / DSP#15 Control Room / DSP#11 B-2003
Slot-2 / DSP#8 / Minimax monophonic (1 voice)
Slot-3 / DSP#9 / Profit-5 (5 voices like the original)
Slot-4 / DSP#10 / Vectron (8 voices like the original)

Question 1:
What´s best for additional MIDI I/O source/destination modules ? - ´cause I´ll need many ports and usage of MIDI OX virtual MIDI cables.

Question 2:
Increasing polyphony for Profit-5 from 5 voices to 6 - 8 voices resulted in DSP limit warning for XITE-1
Is one of the 333MHz SHARCs (DSP#9 here) at it´s limit w/ Profit-5 and 6-8 voices polyphony and doesn´t the DSP routine spill over to the next free DSP (in the same slot),- DSP#17 p.ex ?

The DSP warning disappeard when back to 5 voices for Profit-5 though.

Is there a better solution of DSP assignment for Profit-5 and 8 voice polyphony by re-assigning Profit-5 to any other free DSP in my arrangement above ? (Slot-2 DSPs #12 / 16, Slot-3 DSPs #17 / 13 or Slot-4 DSPs #14 / 18)

Same question might come up when adding UKnow7 synths and a STS 4000 sampler,- or a Modular III.
And there aren´t any FX loaded up to now ...

Bud

I believe at this point that for any of the 'native' Scope mixers and devices, there is no more need to assign DSP's.
They've already been optimized with the V5.1 release afaik, which MAY be why you're getting the errors.
It's backwards I know, but I think that's the current status.

Otherwise I have P5 (6 voices for MIDI guitar) and several other synths, two mixers, ModIV, and various inserts all loaded in one project in Xite with DSP left over.

Some of the other guys know more of the DSP-related tricks than I do, so I expect them to chime in at any time now.

Greg

It's true, if you drop a STM2448 into a blank routing window, you'll see a jump in 7,8,9 and a little on 10. They are very efficient when left alone at that location (those DSPs seem to have more routing resources among themselves and to DSPs #1 and #2.) So inputs get nicely routed from ASIO and VDAT to those DSPs also.

I need to ask S|C directly how much routing resources are available between each DSP. A quantitative answer would probably solve a lot of guess work here.

Assigning Asio to dsp 2 has a good benefit for me with STM2448X.

For the profit5 with 6 voices, I don't know. Maybe you can try to rebuid your project by saving all presets and load Profit5 at the bigining...Just an idea.

JoPo wrote:Assigning Asio to dsp 2 has a good benefit for me with STM2448X.

For the profit5 with 6 voices, I don't know. Maybe you can try to rebuid your project by saving all presets and load Profit5 at the bigining...Just an idea.

Thx for all the answers incl. these above ...
I´ll rebuild the/ my live performance project tonite.

According to ASIO on DSP#2,- does that mean every ASIO module (ASIO-1 / ASIO-2 source and destination modules) will benefit from being assigned to DSP#2 ?

Is that also a good idea when using VDAT ?

Because I´m not thinking in direction of recording in 1st place, but building up a project for realtime playing /gigging w/ as much as possible devices available in one project (possibly as an autoloading "startup") and therefore saving as much ressources as possible,- VDAT might become handy as a player for backing tracks if these are necessary.

It would be also great to know where to place monophonic synths ideally because these probably don´t need much ressources.

I know, B-2003, even fully polyphonic, doesn´t need much DSP,- and I assume Minimax and Prodyssey monophonic too,- isn´t it ?

In fact, yesterday nite, I finished a project incl. the 2448X, Control Room, headphones dest., ASIO-2 source and destination 8 stereo channels, 2 MIDI sources (32 MIDI channels), MIDI monitor and B-2003, Minimax (1 voice), Profit-5 (5 voices), Vectron (8-voice), UKnow7 (8-voice), Lightwave v5 (8-voice),-

When I started loading a STS-4000, there weren´t enough ressources for DSP#9 anymore,- so for now I have to rebuild the project and start DSP assignment w/ the right devices in the right sequential order.

B.t.w.,- the project described above loads as a "startup" and works,- I tried that 2 times before I stopped working @ ~7:00 a.m this morning.

There´s one thing driving me nuts w/ Vectron,- it doesn´t show the "sequencer" indication and the recalled preset name in LIVE BAR and it also doesn´t show a recalled preset name it it´s GUI.
The only control is yourear or opening the preset bank, remind the MIDI Prg.-Change No. being sent and look for this number in the list to verify if you setup works perfect. Showstopper when in a hurry.

I´ll try Vectron Player as an alternative tonite.

For the next try,- I think I should begin w/ a simpler mixer like STM 1632X (auto assignment) and ASIO-2 32Bit float sources and destinations (DSP#2),- need these for some VST and Reason,- then put STS-4000 on DSP#9 (16-voice likea AKAI), continue w/ the polyphonic devices (orig. polyphony like the hardware instruments) and throw in Minimax monophonic and B-2003 in the end.
Up to now I don´t have a key for Prodyssey, so I´m unable to test it w/ XITE-1 because it only works on my PCI card.

What´s up w/ insert FX like (mono) overdrive, flanger, phasor and aux send FX like stereo-chorus, mono delay and room ?
Do these need much DSP ressources on XITE-1 ?

Which DSPs in the 4 slots are the less effective ones but probably powerfull enough to assign Minimax monophonic, B-2003 and simple live-FX to ?

thx

Bud

In fact, yesterday nite, I finished a project incl. the 2448X, Control Room, headphones dest., ASIO-2 source and destination 8 stereo channels, 2 MIDI sources (32 MIDI channels), MIDI monitor and B-2003, Minimax (1 voice), Profit-5 (5 voices), Vectron (8-voice), UKnow7 (8-voice), Lightwave v5 (8-voice),-

Well ! It's an allready good synth project, is n't it !

In the little jpg attached, You'll find a summary of the dsp assignment..

I guess everyone's waiting for Scope 6 with a good manual (for all the dsp assignment gas plant) and a more 'automatic' dsp management to use all the power of xite...

JoPo wrote: Well ! It's an allready good synth project, is n't it !

In the little jpg attached, You'll find a summary of the dsp assignment..

thank you !

JoPo wrote: I guess everyone's waiting for Scope 6 with a good manual (for all the dsp assignment gas plant) and a more 'automatic' dsp management to use all the power of xite...

confirmed !



Bud

Just in case you want to get experimental.
I always load ASIOs, MIDI and the STM2448 first.
Then make sure I have all of the effects in the mixer that I want.
This also helps show me where the DSP power is getting used.
Then I leave 7/8/9 and 10 as is.
Start loading synths on 11/12/13/14/15/16/17/18.
But after each synth is assigned, immediately got to the bottom of the drop down addigment Menu and clear it.
Then add the polyphony you want. Repeat this on every chip that appears unused, ( usually 256 -400 ) and if you have Modular patches you want add them last as they will retain the DSP assignment.

This works best for me....

Cheers.

XITE-1/4LIVE wrote:Just in case you want to get experimental.

Yep,- I want because I think we can squeeze out a lot out of XITE-1.

XITE-1/4LIVE wrote: I always load ASIOs, MIDI and the STM2448 first.

O.k. and thx,- cool !

What´s w/ a WAVE source and VDAT ?
Would it be a idea to load these to DSP#2 + ...

I´d like to have a WAVE source in the project just to play some music coming from any media player app like VLC player or such,-pure entertainment, you know.

VDAT, as a backing track player, is eventually not necessary but I want to check it out.
Can be,- I can go w/o VST in some cases because I´m also using hardware instruments.

XITE-1/4LIVE wrote: Then make sure I have all of the effects in the mixer that I want.
This also helps show me where the DSP power is getting used.
Then I leave 7/8/9 and 10 as is.

Thx for that advice,- but does that mean, once you´ve loaded mixer and FX, these all go automatic to the SLOT-1 (DSP#7/8/9/10) ?

What´s up if I want a STS4000 for my AKAI sample library then ?
I recognized it wants ressources from DSP#9.

Now, if I think about using a smaller mixer like STM1632X for the live keyboard rig,- is it worth a try loading STS-4000 on DSP#9 in the configuration you described,- or is a STS-4000 a no go for any realtime players project ?

XITE-1/4LIVE wrote: Start loading synths on 11/12/13/14/15/16/17/18.

Thx for that, good idea because I wasn´t sure and setting up trial ´n error projects might become a endless task.

XITE-1/4LIVE wrote: But after each synth is assigned, immediately got to the bottom of the drop down addigment Menu and clear it.
Then add the polyphony you want.

Now I´m stuck w/ my limited english.
What´s a addigment Menu ?
I asked LEO, but there´s no translation available unfortunately.
Can you please post a shot which menu to clear after assigment of synths to DSPs and how ?

XITE-1/4LIVE wrote: Repeat this on every chip that appears unused, ( usually 256 -400 ) and if you have Modular patches you want add them last as they will retain the DSP assignment.

Numbers shown as "256-400" means DSP unused ?
Where do I see this ?
In that mysterious "addigment" menue ?

XITE-1/4LIVE wrote: This works best for me....

Cheers.

Thx Jimmy,- lots of great info for me in this post.
Seems, you´re the only one using XITE-1 w/SCOPE 5.1 live,- so, what works best for you will work for me too,- I´m sure.

best

Bud

Bud, check the post above about dsps assignments. 7,8,9,10 are the first dsps of each slot. Slot #1 is then 7, 11, 15. Slot #2 is 8, 12, 16. etc. So the mixers span all slots. the assignment "window" is just a notepad with information Ralf from S|C sent out a long time ago. So you won't find it in the menus

You find the DSP usage meter in Set -> DSP Load menu. By default, it shows the number of calculation cycles, or click the % to see in percent (unused is then 3% instead of 256-400, for slot dsps 7-18, and 38-40% for the older dsps 3-6 used for communication).

I am beginning to use the Xite live, as a live mixer for the whole band, plus effects, and my whole bass sound chain. It is difficult to get that much working stabile, since we use lots of midi to control the mice (more than scope allows, so we need multiple mixers to pull it off).

jksuperstar wrote:Bud, check the post above about dsps assignments.

Thx, J.K. ...

I understand better now,- made a PDF from all the info here I´ll use later when working w/ my DAW.

For live, my target is shrinking rack size(s).
I already sold several 19" synth rack modules which will be replaced by XITE-1 and I´ll ditch rackmount FX boxes too.
The idea is also replacing the keyboard mixer and 19" rack line mixers w/ XITE-1 and connecting all the hardware keyboards directly to a AD/DA.

Well, the MIDI ...

IMO,-we´d need more hardware MIDI connections for XITE-1(D).

I use a parallel port 8x8 MIDI interface as well as USB MIDI from Kurzweil PC3 and Edirol PCR300, but meanwhile and after recognizing my computer´s CPU load (up to 40% !!!) when playing glisses across a keyboard being connected via USB, I suspect MIDI OX causing this because it happens even there isn´t any VST stuff involved.
I´m using MIDI OX to connect a recognized USB port to XITE-1 MIDI In source #1 & #2 .

Today, I´ll check that w/ the LTP port MIDI interface too and see if this makes a difference.

Since I plan integrating Reason 6 and a VST host for some additional VSTis, a up to 40% CPU (not DSP !!!) load, caused by virtual MIDI cables, is totally unacceptable.

It´s only dreaming,- but XITE XTDM port is unused,- so that would be a nice connector for a dedicated S|C /Ferrofish hardware MIDI interface offering more physical MIDI I/O ports directly connected to XITE.

Or,- SCOPE itself should be able to recognize USB MIDI as well as any connected hardware MIDI interface, serial, parallel and USB 1.1. / 2.

In fact,- the one and only physical MIDI In of XITE is already needed for a fader controller box (JLCooper Masterfader Pro) here.

Bud

I am doing the exact thing as you..pairing equipment down, especially allows me more control when jams or practices happen outside my studio.

for midi, I use a MOTU MIDI EXPRESS (micro) to handle my midi. It is the only MIDI interface I know of that can do routing internally at the same time providing I/O to the computer. So I don't need midi ox to do that routing for me. Midi is then routed into the XITE midi port, not through the computer. But the Daw can still have access to the same ports, and send seq midi to xite all the same. Otherwise I agree, more midi to the xite would be best. now I just have to be smart how I use the 16 channels in/out of the xite.

jksuperstar wrote:I am doing the exact thing as you..pairing equipment down, especially allows me more control when jams or practices happen outside my studio.

If all works like I imagined and expected, time will tell ..., studio gear and live gear will be the same for me.
I maintained 3 rigs in the past decades, studio recording gear (analog & digital), hardware keyboard rig w/ racks in my home and empty racks prepared to fill w/ additional gear for gigging/touring.
There were times I had almost everything twice and partially I´m suffering from that up today because of the effort on service, the costs and time investment searching parts for vintage gear.

XITE-1 seems to be the ideal solution, replacing many synths and other hardware devices and offering total recall of routing projects and device configurations for the gigging and recording tasks.
I´ll keep a few vintage instruments,- Minimoog D, Oberheim Xpander and one dedicated FM machine.
I´d wish there would be a FM synth in SCOPE reading Yammi DX7mkII and or TG77 sysex because I don´t have the time to program all the patches new from scratch.
Maybe Open SCOPE opens the doors for some future surprises.
If not, there´s Solaris, but that´s another big investment too.

jksuperstar wrote: for midi, I use a MOTU MIDI EXPRESS (micro) to handle my midi. It is the only MIDI interface I know of that can do routing internally at the same time providing I/O to the computer. So I don't need midi ox to do that routing for me.

I also have these options later, once a new rack will be realized.
Normally, I use a MIDITEMP PMM88E as the MIDI frontend for my controller keyboards, but actually I use XITE-1 striktly in my home studio rig/environment because it was the easiest way to set it up here after the purchase.
Actually I mimik w/ this setup what I plan to do later in the very compact new rig,- but I have to buy some more stuff to realize that.
I also have a assortment of 8x8 computer MIDI interfaces.
These do routing in hardware too, but only w/ their original crappy drivers.
I replaced the driver for my Opcode/MusicQuest 8PortSE by a Earthvegaconnections Direct Music driver for Windows (32&64Bit available) and it´s different worlds.
I own 2 of these, so I have one in spare which is great always, but there are also MOTU MIDI Timpiece II, Emagic and a Steinberg MIDEX-8 USB.
The 8PortSE is the best by far !

jksuperstar wrote:
Midi is then routed into the XITE midi port, not through the computer. But the Daw can still have access to the same ports, and send seq midi to xite all the same.

Yep,- but 1 physical port is only 16 channels of MIDI and that´s not enough.
I´d like to see at least one port more or ideally 4 or these for 64 MIDI channels.

jksuperstar wrote: Otherwise I agree, more midi to the xite would be best. now I just have to be smart how I use the 16 channels in/out of the xite.

Exactly !
Squeezing lots of continuous controller data together w/ all the notes created by realtime playing thru 1 MIDI port isn´t the optimal solution.

I have a shitload of physical I/O ports available for the MIDI outboard, owning Sycologic MIDI matrix switchers in additon the the stuff mentioned above,- but the bottleneck is and keeps the one and only MIDI IN port of XITE-1(D).

I wonder what CopperLAN will offer for XITE-1(D) because we´ve seen that on a pic of Musikmesse 2012.
http://www.copperlan.org/index.php/copperlan

Bud

Well we already know the internals of Open Scope will allow more accurate / wider bandwidth control signals. Hopefully MIDI can map easily to the new protocol for allowing interface to legacy devices, as well as integrating newer Copperlan generation devices.

All is speculation at the moment but some good will come.

dante wrote:Well we already know the internals of Open Scope will allow more accurate / wider bandwidth control signals. Hopefully MIDI can map easily to the new protocol for allowing interface to legacy devices, as well as integrating newer Copperlan generation devices.

All is speculation at the moment but some good will come.

Because I´m technically interested and like these ideas of MIDI over LAN like rtp-MIDI and KISSBOX etc.,- I took the time and read some more about CopperLan during last nite.

This is very exciting if it will be integrated in SCOPE 6.
If it works as described, all the MIDI limitations will be gone because it works cross platform and shall be able to recognize any 3d party USB virtuial port as well as any ext. hardware interface being connected to whatever available computer connection, once the OS detects & accepts the device as a MIDI device/port.

There will be much more possibilities then.

B.t.w.,- following advice from Jimmy, JKsuperstar and others here,- I managed to create a new large project for potential live gigging and/or usage of XITE-1 as a "MIDI tonegenerator /pre-mixer & FX" machine in the home studio.

I threw in by automaticDSP assignment:

XITE-1 hardware analogue audio I/Os
AES/EBU source/destination
XITE-1 phones destination
ADAT A source and destination
XITE-1 MIDI source/destination

ASIO2 float32 source/destination modules (4-ch. In for a VST host and Reason6 stereo outs and 2 ch. back to VST/Reason)

2 sequencer MIDI source/ destination modules conneced to MIDI OX virtual MIDI cables/ports

STM1632X (13 stereo channels, 3 mono channels, 4 aux sends, 4 stereo aux returns)
AUX-1: Mastererb Classic
AUX-2: SC Plate Reverb
AUX-3: SC Chorus/Delay
AUX-4: Mono Ducking Delay

I missed a mono>stereo signal splitter in SCOPE 5.1 software to use the mono AUX sends w/ stereo FX stereo inputs,- so,-
I duplicated mono AUX sends and connected both to L+R inputs of AUX 1 - 3 FX.

I someone is aware of any complications using this technique,- please leave a comment because below, there will be a "bug" anounced later.

MicroMixerX, 4 channels stereo for synths not requireing AUX-FX and WAVE Player,- and 8 mono Ins for VDAT outputs


manual DSP assignment:

24Bit WAVE source /DSP #2
VDAT 8-channel out /DSP #2
STS4000 stereo out & 3 mono outs, 16 voices / DSP #9 / MIDI In = seq. MIDI src. 2 / MIDI Ch. 2 - X

B-2003 / DSP #11
Minimax monophonic / DSP #11
Profit-5 poly, 6-voice / DSP #12
Vectron Player poly, 8-voice / DSP #13

DSP#14 reserved for Prodyssey (until I have my key),- mono- or duophonic like the original synth

Lightwave v5 poly, 8-voice / DSP #15
UKnow7 poly, 8-voice / DSP #16

The overall DSP load is ~48 or 49% !

I stopped here and saved project w/ "save as" to a separate location, closed MIDI OX and quit SCOPE 5.1
I unchecked "Save as Startup" before I saved.

Bug ?

When reloading the project,- AUX-1 send and AUX Return-1 both were disconnected !!!

Conclusion:

Maybe it´s better to use only ONE larger Mixer instead the 1632X and MicromixerX.
I´ll investigate in a mono>stereo signal splitter,- maybe 3rd party and free,- if available.

For the guys reading this post,- I´ll post sceenshots later and once my DAW is up and running today.
What´s the best size in pixels for easy viewing and pics not being too large, but also not too small here ?

I have to convert bitmap>jpg and resize before posting attachements.

Any advice avoiding disconnected AUX-1 connections welcome !

best

Bud

There is a device called 4M8S to convert mono to stereo. I am about to put it to use for the same reason, I'll let you know how it goes.

Here's a simple M2S:
http://forums.planetz.com/viewtopic.php ... hilit=4M8S

I can't seem to find the source of the 4M8S though. I'll take another look later.

I also think there has been some random issues with the 1632X. You might be better using the 2448 (it's very smart about not using DSP if no input is connected, although it doesn't have the same singular interface that the 1632 has. Try the Cwm16-16 here for a simple, 1 screen mixer with 4 Aux sends & 4 Buses:
http://forums.planetz.com/viewtopic.php?f=16&t=30818

... here´s the screenshot of the reloaded project ...



AUX-1 send/returns are disconnected after loading the project.

Loading this project needed 6 - 7min !
Wonder whether double aux-send cables from mono aux-sends 1 -3 (STM1632X) to stereo-inputs of stereo aux-FX devices causes this behaviour or not.

Bud

I have issues with VDAT creating a tape each time it loads. Check the SCOPE window to see what's loading during start-up, it displays what's being loaded at the moment, and a progress bar too. Creating 8 tracks of so many minutes takes a fair amount of time.

ps - I'm considering moving the devices, samples, presets and projects to my SSD for faster loads. But I think the biggest issues isn't the disk, but compiling & routing....

jksuperstar wrote:I have issues with VDAT creating a tape each time it loads.

No dedicated tape loaded,- only the empty device is integrated into the project.
So,- only the "default" test-tape ...

jksuperstar wrote:
Check the SCOPE window to see what's loading during start-up, it displays what's being loaded at the moment, and a progress bar too.

I did, that´s why I know it was 6 -7 minutes.

It hung at STM1623X as well as SC Chorus/Delay devices.

jksuperstar wrote: Creating 8 tracks of so many minutes takes a fair amount of time.

See above ...
I just only loaded the output-connected VDAT itself,- quasi waiting for any upcoming tape to load.
I´d handle this that way just for the occasion I´d need something playing back.
I doubt I have to use that often though.

jksuperstar wrote: ps - I'm considering moving the devices, samples, presets and projects to my SSD for faster loads. But I think the biggest issues isn't the disk, but compiling & routing....

Yep, after reading your comments, I also now suspect STM1632X being the cause.

I have to mention,- last nite I did the re-load test too and it was under 1 minute,- but today at noon it was 6 - 7 minutes.
That´s weird.

Thx for your input !

Bud

I've also noticed that the devices get loaded in the order you added them to the project...more or less. So I always start my projects with the I/O, then mixer, then everything else. it seems to help the loading, routing, and DSP placement. Or maybe I'm crazy, YMMV....