reliable MIDI I/O
hello ken, is there really a problem with pulsar midi? from my experience i only have problems with the sequencer when output midi notes to creamware seq dest. So i have bought midiex3 (steinberg) and it get better but always worst (in cubase 5.1) than hardware sequencer (nova arp, arpeg01,modular seq, vst instruments) (when i send midi from nova's arp to pulsar midi-in it's similar (accurate) when i route it to midiex). What i found is that when i use pulsar in xtc mode i have really tight timing.. Now i use Sx that in my ear has good midi), as u can see i don't have many external synths so maybe there is problem with pulsar midi that i have not encountered, i don't know if you talked about that..
-
Counterparts
- Posts: 1963
- Joined: Tue Aug 19, 2003 4:00 pm
- Location: Bath, England
I've never had problems with the CWA card's MIDI, but I do sometimes get some jitter from within Cubase, especially when moving into a section (beat 1 of the bar) where there's a lot of MIDI stuff going on.
What happens is that cubase stops outputting any MIDI for a second or two, then offloads the last 1,000 MIDI notes/CCs that it forgot to play all at once: =BOOM!=
Royston
What happens is that cubase stops outputting any MIDI for a second or two, then offloads the last 1,000 MIDI notes/CCs that it forgot to play all at once: =BOOM!=
Royston
Hi,
I've never had any midi issue, for me the sequencer source is rock solid too.
MIDI is something that involves very much the computer os and it's relation with the hardware. Cubase, that is not a champion of MIDI stability itself, is very much affected by the overall machine. A proof of this is how different, and this is documented, is the MIDI stability in, say, 98se and ME, just to consider old OS's, and how a tweak of XP is necessary.
Using other apps for MIDI management doesn't change what Scope MIDI ports do, just optimize what is on the OS side.
I couldn't really say why my machine has such a tight midi going on in every direction from and to Scope environment, but I noticed that the performance of my P III-s of only 1.26 GHz is for many aspects superior to some P 4 systems around. I'm sticking with this machine despite the 512Mb of ram limit of the 815e chipset, because it's stable and reliable as nothing I've seen.
I've never had any midi issue, for me the sequencer source is rock solid too.
MIDI is something that involves very much the computer os and it's relation with the hardware. Cubase, that is not a champion of MIDI stability itself, is very much affected by the overall machine. A proof of this is how different, and this is documented, is the MIDI stability in, say, 98se and ME, just to consider old OS's, and how a tweak of XP is necessary.
Using other apps for MIDI management doesn't change what Scope MIDI ports do, just optimize what is on the OS side.
I couldn't really say why my machine has such a tight midi going on in every direction from and to Scope environment, but I noticed that the performance of my P III-s of only 1.26 GHz is for many aspects superior to some P 4 systems around. I'm sticking with this machine despite the 512Mb of ram limit of the 815e chipset, because it's stable and reliable as nothing I've seen.
-
kensuguro
- Posts: 4434
- Joined: Sun Jul 08, 2001 4:00 pm
- Location: BPM 60 to somewhere around 150
- Contact:
well, the problems seems to be when I'm doing something extreme, like doing dumps, or using i-cubeX (realtime A/D converter that converts voltages to MIDI). In terms of playing and triggering events, Pulsar does just fine. But it seems to be overloaded quite easily. (aka, the "clock signal" overload causing stuck notes)
Who knows, it could just be my cables... but I remember reading somewhere on Z that Pulsar's MIDI implementation wasn't 100% compliant to the standard. Which would explain why I have to use "slow" transfer mode to update my Nord electro and i-cubeX. So, it's these occasions I'm talking about.
<font size=-1>[ This Message was edited by: kensuguro on 2004-08-17 10:34 ]</font>
Who knows, it could just be my cables... but I remember reading somewhere on Z that Pulsar's MIDI implementation wasn't 100% compliant to the standard. Which would explain why I have to use "slow" transfer mode to update my Nord electro and i-cubeX. So, it's these occasions I'm talking about.
<font size=-1>[ This Message was edited by: kensuguro on 2004-08-17 10:34 ]</font>
-
BingoTheClowno
- Posts: 1722
- Joined: Wed Nov 12, 2003 4:00 pm
- Location: Chicago
- Contact:
-
Basic Pitch
- Posts: 627
- Joined: Fri Aug 15, 2003 4:00 pm
Hi Ken,
I know this is not exactly what you were looking for, but I have been using a Steinberg Midex8 for ages now, it has been nothing but rock solid and if you using Cubase or Nuendo it takes advantage of their supposid LTB technology, either way, its a great device even though it connects USB.
Cheers!
I know this is not exactly what you were looking for, but I have been using a Steinberg Midex8 for ages now, it has been nothing but rock solid and if you using Cubase or Nuendo it takes advantage of their supposid LTB technology, either way, its a great device even though it connects USB.
Cheers!
-
Liquid Len
- Posts: 652
- Joined: Tue Dec 09, 2003 4:00 pm
- Location: Home By The Sea
Here is my two cents, I realize not everyone has the same problem with the cards, and some of the issues raised may not be significant to all users.
MIDI is the part of the Pulsar cards that I have the most problems with by far. I am unable to synchronize the MIDI clock with Cubase's output, which prevents me from using that aspect of synth plug-ins (not to mention the modular). Sometimes MIDI controllers assigned to plugin parameters stop working until I delete and add them again. Or else, once I assign them, they change some aspect of the instrument's sound. These two things alone are annoying - I have outboard gear and it responds pretty reliably to MIDI messages, but something that should be much more powerful and configurable, is in fact severely compromised (at least in those regards). I know I can reliably sync Pulsar synths with an external clock source, but this wouldn't be very convenient to say the least. I find the bugs more or less force you to cut everything to audio as soon as you can, 'while it still sounds right' - which is much less powerful than being able to tweak individual parameters right to the end.
But the most disappointing thing is the problem with stuck notes. I would use my system live in a second except for this. I have tried a half-dozen keyboards and dozens of MIDI cables and I keep getting stuck notes - which often do not even respond to MIDI resets but do get fixed if I change the polyphony. For recording, its not a big problem, but it kind of kills the idea of live performance.
I'm not trying to slam the Creamware cards here, I realize they are working within the constraints of (1) PC architecture and (2) reliably written plugins. But the plugins I have the most problem with generally aren't 3rd party, of which I have few, but ones that Creamware sells themselves. I guess you can't have everything, and I will admit Creamware redefined what I thought a PC softsynth was capable of (or what a soundcard is capable of, for that matter).
MIDI is the part of the Pulsar cards that I have the most problems with by far. I am unable to synchronize the MIDI clock with Cubase's output, which prevents me from using that aspect of synth plug-ins (not to mention the modular). Sometimes MIDI controllers assigned to plugin parameters stop working until I delete and add them again. Or else, once I assign them, they change some aspect of the instrument's sound. These two things alone are annoying - I have outboard gear and it responds pretty reliably to MIDI messages, but something that should be much more powerful and configurable, is in fact severely compromised (at least in those regards). I know I can reliably sync Pulsar synths with an external clock source, but this wouldn't be very convenient to say the least. I find the bugs more or less force you to cut everything to audio as soon as you can, 'while it still sounds right' - which is much less powerful than being able to tweak individual parameters right to the end.
But the most disappointing thing is the problem with stuck notes. I would use my system live in a second except for this. I have tried a half-dozen keyboards and dozens of MIDI cables and I keep getting stuck notes - which often do not even respond to MIDI resets but do get fixed if I change the polyphony. For recording, its not a big problem, but it kind of kills the idea of live performance.
I'm not trying to slam the Creamware cards here, I realize they are working within the constraints of (1) PC architecture and (2) reliably written plugins. But the plugins I have the most problem with generally aren't 3rd party, of which I have few, but ones that Creamware sells themselves. I guess you can't have everything, and I will admit Creamware redefined what I thought a PC softsynth was capable of (or what a soundcard is capable of, for that matter).
-
blazesboylan
- Posts: 777
- Joined: Sat May 25, 2002 4:00 pm
- Location: The Great White North
- Contact:
Interesting.
I've never had a problem with MIDI controller messages (Liquid Len mentioned removing then re-loading a module in order to control the parameters). Just about everything I do right now is remote-controlled via MIDI controller messages. However through experimentation I found that an 80 ms delay is needed between MIDI controller messages.
I came to this conclusion controlling the PEQ4 -- I first send a "band chooser" message to control the minus sign; then I send a frequency, Q, or gain message. If the messages are too closer together (< 80 ms) then the first one doesn't take effect before the 2nd one comes in. Therefore you get a frequency, Q or gain change on the wrong band in PEQ4.
Nevertheless I've sent thousands of controller messages via MIDI over the past 2 months or so, and never lost one except due to this timing issue.
YMMV of course, but I'm mostly happy with SCOPE MIDI handling. Especially the relative ease of setting up parameters to control -- contrast with Nuendo. Uy!
I would still agree, though, that the MIDI implementation is one of the weakest parts of SFP/SCOPE. For starters, I'd like to get rid of that 80 ms delay requirement. Resetting 1000 MIDI parameters takes quite a while... Secondly, having multiple MIDI ins, to control more than 120 parameters per module, would be nice.
(Even though my requirements are oddball, I suspect these 2 improvements would benefit a lot of people.)
Cheers,
Johann
I've never had a problem with MIDI controller messages (Liquid Len mentioned removing then re-loading a module in order to control the parameters). Just about everything I do right now is remote-controlled via MIDI controller messages. However through experimentation I found that an 80 ms delay is needed between MIDI controller messages.
I came to this conclusion controlling the PEQ4 -- I first send a "band chooser" message to control the minus sign; then I send a frequency, Q, or gain message. If the messages are too closer together (< 80 ms) then the first one doesn't take effect before the 2nd one comes in. Therefore you get a frequency, Q or gain change on the wrong band in PEQ4.
Nevertheless I've sent thousands of controller messages via MIDI over the past 2 months or so, and never lost one except due to this timing issue.
YMMV of course, but I'm mostly happy with SCOPE MIDI handling. Especially the relative ease of setting up parameters to control -- contrast with Nuendo. Uy!
I would still agree, though, that the MIDI implementation is one of the weakest parts of SFP/SCOPE. For starters, I'd like to get rid of that 80 ms delay requirement. Resetting 1000 MIDI parameters takes quite a while... Secondly, having multiple MIDI ins, to control more than 120 parameters per module, would be nice.
(Even though my requirements are oddball, I suspect these 2 improvements would benefit a lot of people.)
Cheers,
Johann
-
Liquid Len
- Posts: 652
- Joined: Tue Dec 09, 2003 4:00 pm
- Location: Home By The Sea
Just to clear things up. To reset a MIDI controller on a softsynth I just have to delete the setting that's there (and isn't working) and add it again. I don't have to unload or reload the actual module, ever.On 2004-08-17 22:04, blazesboylan wrote:
controller messages (Liquid Len mentioned removing then re-loading a module in order to control the parameters). Just about
-
blazesboylan
- Posts: 777
- Joined: Sat May 25, 2002 4:00 pm
- Location: The Great White North
- Contact:
Oops, OK.On 2004-08-18 01:08, Liquid Len wrote:
Just to clear things up. To reset a MIDI controller on a softsynth I just have to delete the setting that's there (and isn't working) and add it again. I don't have to unload or reload the actual module, ever.
Still, I've never had to do reset MIDI cc settings, either.
Are you sure your controller isn't sending a burst of messages? Otherwise I see no reason for SFP needing to be "reset".
Cheers,
Johann
P.S. If you assign the same cc on the same MIDI channel to 2 different SFP parameters, the first one will still display that cc # setting, but it will not respond to it.
-
Liquid Len
- Posts: 652
- Joined: Tue Dec 09, 2003 4:00 pm
- Location: Home By The Sea
My system doesn't appear to work in exactly the same way. Are you using 3.1 instead of 4.0 by chance?
Using a keyboard with several dials (evolution mk-225c), I'm assigning those different dials to control parameters in the prisma and inferno synths. I can assign the same dial to control a parameter in both softsynths simultaneously. If I assign the same dial to another control parameter IN THE SAME softsynth, it erases the first one I had set in that particular softsynth. I know it erases it because (1) the control parameter no longer changes when I move the dial and (2) when I right-click and check, the setting has been removed. Apparently on your system (or maybe because it's related to which particular softsynth you're using) it doesn't clean up properly after erasing the link.
Using a keyboard with several dials (evolution mk-225c), I'm assigning those different dials to control parameters in the prisma and inferno synths. I can assign the same dial to control a parameter in both softsynths simultaneously. If I assign the same dial to another control parameter IN THE SAME softsynth, it erases the first one I had set in that particular softsynth. I know it erases it because (1) the control parameter no longer changes when I move the dial and (2) when I right-click and check, the setting has been removed. Apparently on your system (or maybe because it's related to which particular softsynth you're using) it doesn't clean up properly after erasing the link.
-
blazesboylan
- Posts: 777
- Joined: Sat May 25, 2002 4:00 pm
- Location: The Great White North
- Contact:
Hi Liquid Len.
I'm using SFP 4.
As for which devices -- I'm only controlling mixers, EQs, compressors and so on. No soft synths.
However I only send one message at a time (from another computer), so it's not the same thing as using an external controller -- which sends a burst of messages as you turn each knob. I suspect you're hitting the timing / overload problem with SFP.
Wish I had some suggestions for how to get around it!
Has anyone spoken to CWA support about the MIDI overload problem? (Sorry if that question has already been answered somewhere. There are too many MIDI-related threads for m to keep track of at this point...)
Cheers,
Johann
I'm using SFP 4.
As for which devices -- I'm only controlling mixers, EQs, compressors and so on. No soft synths.
However I only send one message at a time (from another computer), so it's not the same thing as using an external controller -- which sends a burst of messages as you turn each knob. I suspect you're hitting the timing / overload problem with SFP.
Wish I had some suggestions for how to get around it!
Has anyone spoken to CWA support about the MIDI overload problem? (Sorry if that question has already been answered somewhere. There are too many MIDI-related threads for m to keep track of at this point...)
Cheers,
Johann
very strange (regarding hanging notes)...
it looks like the midi filter module is the culprit
I can reproduce note hangs reliably if the filter is in the path and no hangs if it's disconnected.
this will lead to a catch 22 situation if someone includes that module 'just to be safe'
I have a DX200 groovebox driving Scope synths and used the filter to avoid program changes on the synth if the sequencer changes patterns.
Btw that box produces a flood of midi data, but obviously SFP gets along with it.
Since this seems to be a more or less recent occurance it could as well be related to other midi software I have installed in the last couple of weeks.
One is Bome's Midi Translator, the other is called SQ Midi Tools.
None of this is running, but one never knows if something lurks in the background.
Nevertheless, if an external 'midi tool' has the ability to mess the filter module that's something to watch out for.
confused, Tom
it looks like the midi filter module is the culprit
I can reproduce note hangs reliably if the filter is in the path and no hangs if it's disconnected.
this will lead to a catch 22 situation if someone includes that module 'just to be safe'
I have a DX200 groovebox driving Scope synths and used the filter to avoid program changes on the synth if the sequencer changes patterns.
Btw that box produces a flood of midi data, but obviously SFP gets along with it.
Since this seems to be a more or less recent occurance it could as well be related to other midi software I have installed in the last couple of weeks.
One is Bome's Midi Translator, the other is called SQ Midi Tools.
None of this is running, but one never knows if something lurks in the background.
Nevertheless, if an external 'midi tool' has the ability to mess the filter module that's something to watch out for.
confused, Tom
I removed the 'external' midi stuff and now the filter module is working again as supposed to 
Imho this leads to the conclusion that the risk of hanging notes increases if several apps access the driver - like the sequencer and midi routing software or possibly even some OS related stuff.
cheers, Tom
Imho this leads to the conclusion that the risk of hanging notes increases if several apps access the driver - like the sequencer and midi routing software or possibly even some OS related stuff.
cheers, Tom