Our Nemesis

[Giorgio Tommasini]

Of interest, Scarbee\'s report \"No problem. modulation and pitchbend (on track 1)does not affect track 2 at all\", seems to indicate that a previously reported bug, i.e. propagation of vibrato and tremolo to all loaded instruments, rather than intrinsic to Gigastudio, could be related to some particular settings. Given the practical relevance of the problem, I feel that Nemesys should give an authoritative answer in this Forum.
Otherwise, we could start to meditate on the meaning of Nemesis in ancient Greek: \"a deserved, endless punishment coming from above\"

Giorgio

[KingIdiot]

hehheh good one Giorgio...

I fooled around with this problem, and with Cubase, I get the problem,...but its hard to define exactly what happens. It doesn\'t occur as soon as I impliment Vibrato on a track, the track to which I\'m not applying vibrato has to be playing back atleaast one sample.....AND THEN WHILE vibrato is applied to the first track THEN the track I don\'t want vibrato applied to has to playback a second sample. This is when the track I don\'t want vibrato on gets vibrato. Its failry odd.

------------------
Really...I am an Idiot

[toby]

>>,...but its hard to define exactly what happens.<<

nevertheless, could you try again and rephrase your findings. I didn\'t get it.

[jbibbo]

Test to see if my password works. My account seems to have gone away. If this works I will respond on NemeSys\'s behalf to the question.

[jbibbo]

Giorgio,

This message is from NemeSys.

I have followed some of your threads regarding tremolo and vibrato being incorrectly propogated to multiple instruments.

Our tests have shown that tremolo and vibrato are correctly applied to individual instruments on different midi channels. Here are some of the ways we verified the results.

First we took two instruments from the GM 500
set, let say A.gig and B.gig. We then took all the modulation out of both instruments, all layers, and all regions. IMPORTANT: you need to make sure that you select ALL the regions and ALL the layers when you want a global setting to effect everything. Then we set up both instruments using LFO1 frequency of 2, and depth of 1200. We tried both cases of controller - Ext Mod Wheel and Ext/Int Mod Wheel. We then loaded a.gig to channel 1 and b.gig to channel two. Driving the sampler directly from the MIDI controller and driving the sampler via a sequencer (Cakewalk Pro 9) yielded the same results: Using the mod wheel on channel 1 only effects the LF01 on A.gig and using the mod wheel on Channel only effects the LFO1 on B.gig.

We repeated the test using LFO3 - same results.

In setting up this test, there is only one thing that tripped me up. Because I was dealing with a multilayered instruments, when changing the parameters I forgot to select all the layers. Consequently, my edits were only applied to \'half\' the instrument. Also if you are using \"Ext/Int Mod Wheel\" setting then if the \'Internal depth\' is nonzero then you will hear some modulation, regardless of where you external mod wheel is set.

We feel strongly that sampler is behaving properly.
The fact that Scarbee, a very knowledable giga user, is reporting similar results should also give you confidence.

I hope this helps in some way.

Regards,
Joe (NemeSys)

[Chadwick]

Hey JBibbo,

I\'ve never run into this problem, but I\'d be interested if someone could theorise what factors outside of Gigastudio might cause this phenomenon to appear, as there have been more than a couple of users who report vibrato appearing unintentionally on channels where it wasn\'t (apparently) routed.

There are so many options, maybe making suggestions will help via a process of elimination.

We should also be mindful of the fact that certain bugs can be \'conditional\' on certain factors which are not obviously related to the fault (the moon being in the 7th house, your mouse pointing to 3 O\'clock, two identical gigs being loaded in different slots in the same performance)

One example I can think of is where you try and assign a sound to a channel and the wrong sound is assigned.

Say you have ten sounds loaded (but not necessarily assigned), one of which is \'bell voice\' and another which is \'funk drums\'. Even though bell voice is loaded, you forget and double click on your destination channel, find its directory and double click to load \'bell voice\'. Immediately \'funk drums\' appears in the channel slot to which you were assigning \'bell voice\'.

Wierd, but you then remember that you already had \'bell voice\' loaded, so you detach the \'funk drum\' patch from the channel, grab the \'bell voice\' gig from the loaded instruments pane and drag and drop it onto its channel slot. Immediately the \'funk drums\' gig appears in that channel.

Wierd, and at this stage you\'re hoping that you didn\'t somehow save the \'funk drums\' instrument as the first instrument in the \'bell voice\' gig. So you unload \'bell voice\' from memory, double click on the channel slot, navigate to the directory with \'bell voice\' in it and double click on \'Bell voice\'. Voila! Bell voice is correctly assigned.

Wierd.

My point is that if I told you that loading an already loaded gig into an extra channel makes the wrong gig load, you could test for days and not find it. It\'s only happened to me twice in two months. And it\'s happened to more than one other user.

Maybe we can make suggestions to the guys with the vibrato problems to help eliminate all possible sources outside Giga.


[This message has been edited by Chadwick (edited 02-02-2001).]

[This message has been edited by Chadwick (edited 02-02-2001).]

[KingIdiot]

ok here goes.

I open Giga and Cubase


I load French Horn Sustain LOOP from QLB Brass into Port one Channel one. It has vibrato attached to the Mod Wheel.


I load Turbo Analog Sawtooth into Port 1 channel 2 (which also has Vibrato attached to the Mod Wheel).


I draw in a not at C4 for 4 bars on CHANNEL 2(sawtooth)


I draw in a note at E4 for 2 bars on CHANNEL 1 (Come on...the French horn...pay attention http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif)


Both notes set to start at the same time


Set the sequencer to start playing a bar before the notes are to start


Select Channel 1 on Cubase


Hit Play


When the notes start I kick the Mod Wheel up to full vibrato


when the French Horn Stops playing, the Turbo Analog Sawtooth sound keeps playing and you can tell it has Full vibrato

HOWEVER If I do all the above but set the Sawtooth sound to begin playing BEFORE the French horn, Vibrato isn\'t applied to it.


It happens without the Sequencer as well (I need to use the hold/sustain controller)


Of course there are alot more variations of this bug, but this is one simple way it can be reproduced


I\'ll write it in a bug writing form if ya guys want, I used to be a software tester for about 10 years. http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif

BTW I\'m using a Roland XP-30 as my controller. I\'m pretty possitive it has nothing to do with this tho, since it doesn\'t record Mod wheel info to any cahnnel but the one that is selected.

------------------
Really...I am an Idiot

[This message has been edited by KingIdiot (edited 02-02-2001).]

[Giorgio Tommasini]

Hi Jbibbo,
What is not entirely clear in your post is whether your experiments were performed by simultaneously playing back a series of notes on both channels. As pointed out by others, if you have vibrato/tremolo applied to some notes on channel 1 and simultaneously play a series of notes on channel 2 with vibrato/tremolo set to zero, vibrato/tremolo will be added to some notes on channel 2, depending on the time relationship to note on/off on channel 1.
Please try this and let us know about the results.
Thank you,
Giorgio

[arcusa]

I repeat this report problem I posted on the 12-26-2000 10:31 AM related to the \"tremolo\" madness. Apparently most customers has notice that problem, one way or the other. I hate tremolo since that is happening...
--------------------------------------------------------------------------------
I love GStudio, and now, I could not live without it, but...I read some complaints about the unwanted problem in GStudio, that happens to me also. Need help, or is that a bug?
Suddenly, with no reason, ALL MIDI CHANNELS at GS start an unwanted & nasty vibrato, sometimes a light one, and other -to many- times, a heavy one. I tried GS in a \"fresh\" computer just for playback, with nothing else installed on it but GS, and the vibrato-trembling comes up very often. Obviously I tried that, after I was using GS with Logic as a sequencer and thought it might be something to do with it. I disabled all midi messages except notes, and the same thing happened. So I tried GSs in new, \"fresh\" computer. No sequencer. No Mod messages from my keyboard, but tremolo, yes.

I discover something though: in the midi control page on the GStudio, the first fader is assigned by default to \"Mod\". This fader, \"by default\" -nobody knows why- is leveled up a bit in every midi channel with no reason. Even if I, patiently, brought down every fader of every channel of every port (64 total) down, saved as default start page, and worked from then on, the same annoying unwanted tremolo-vibrato-mod comes up whenever you are in the middle of your inspiration, with no reason. ¡Who likes that!

Any ideas?

Thanks for your help,

arcusa

Note: I have downloaded the last version available of GStudio and I own a Dell Pentium III 1GH with 526 RAM; SoundScape Mixtreme card and Logic Audio.

[jbibbo]

Chadwick,

I am not quite sure what suggestions to make past the ones I already posted. However after reading Giorgio and KingIdiot\'s responses, I am thinking that maybe our tests are not identical, so I will give another shot.

BTW, if you can find a deterministic case on your instrument loading example, I would be very interested. I am all to familiar with \'conditional problems\', but sometimes it is possible to understand the \'conditions\' required, such that we can easily reproduce the problems. Any help you can give me would be appreciated. Thanks.

Giorgio/KingIdiot,

I am out of the office today, but I will rerun the tests trying to match KingIdiot case Monday morning. I don\'t have the QL brass, so hopefully it is not instrument specific. I will also use CWP9. The easiest way to make sure I am hearing what you are hearing, and to make sure it is not a system specific problem is for someone who is having the problem to create two simple .gig (maybe with a single sample), and a .mid file. You could then zip them up and send us your test case. KingIdiot, if you do file a bug report with regards to this issue please send it care of myself, Joe Bibbo. Thank you

I will report back later tomorrow.

Regards,
Joe (NemeSys)

PS Arcusa, the midi control surface is set to values that by definition of the MMA MIDI 1.0 specification are the reset state. If you would like the default state to be something else, all you have to do is set the control surface to the desired value and choose the \'Save as default performance\' under the File menu.

[KingIdiot]

I\'ve posted a ZIP file with two simple instruments and a MIDI file that will show this bug.

It occurs even with playback of the MIDI file in Media Player, and the MS MIDI router set to playback to Memesys Port 1
http://home.earthlink.net/~ohthat/stuff/testvibrato.zip (\"http://home.earthlink.net/~ohthat/stuff/testvibrato.zip\")
is the direct link to the file


Load test gig one to channel 1

and test gig 2 to channel two

playback the midi File, it will first reset the Mod Wheel to zero, then at about 10 seconds the sounds will trigger. I kick up the Mod wheel only on Channel one, and both channels get vibrato. This can be heard clearly after channel one\'s note stops playing, and Channel two\'s note plays for two more measures

Also, after Channel one\'s note stops playing, and Channel two\'s note plays for about another measure, I brought the Mod wheel back down in channel one, and Channel two\'s note still had Vibrato applied to it.


------------------
Really...I am an Idiot

[Simon Ravn]

I really like that you people are using time on fixing this bug, KingIdiot and JBibbo. It would be nice to get it removed already after the program has been out for 2-3 years..... http://www.northernsounds.com/ubb/NonCGI/images/icons/smile.gif)))))))))

[KingIdiot]

have you guys found if the bug is Giga Studio specific? I mean, does GigaSAMPLER work fine?

I wish I had the time to really get into testing this bug and any others, but I\'m jsut swamped during hte day, and I get drained so much by the end of it, I\'m not really into fooling with audio much (let alone looking at puter screens...but I\'m addicted to the internet)


------------------
Really...I am an Idiot

[Simon Ravn]

KingIdiot, I remember the bug being in Gigasampler too, so it´s been there all along...

[Bill]

I was able to reproduce the vibrato bug in King’s sample.

Another quirk is that even though channel 2 is getting a mod value of 127, it is not shown on the MIDI control surface, which remains at 0.

[KingIdiot]

yeah, it looks to be a bug in the playback routines, not directly the Midi Control routines.

Can\'t believe I never saw it until lately, and I had to TRY and get it to reproduce. I can see how it would happen tho. I jsut reccently got into programming Vibrato and/or tremelo into some of my GIGs, so I\'m sure I would have run into it.

------------------
Really...I am an Idiot

[jbibbo]

KingIdiot,

Thanks for your example. Yes, I too hear the problem. I played with it a bit, and it seems that the bug has a condition that probably explains why I missed it in my first tests.

The time relationship between the MIDI note on on channel 2 to the MIDI controller mod wheel on channel 1 determines whether or not you will experience the bug. If you move the note on to measure four or eight the problem goes away.

In any case, it is definitely a bug. I will make sure the appropriate developer is notified, and that it gets fixed for v2.2.

Thanks again for your help - KingIdiot, Giorgio, Chadwick ...

Regards,
Joe (NemeSys)

[KingIdiot]

yeah its definately a tricky bug. Its god to know that you guys know about it now.



------------------
Really...I am an Idiot

[Rackthemout]

Good work everybody ! Thank God Nemesys recognise this vibrato problem as a bug.
For a while I thought I was losing my mind!
When will we see V2.2 ?????????

[KingIdiot]

BTW Joe.

If there are any bugs reported to you guys, that you haven\'t been able to reproduce, or are looking for conditional problems, some of us might be willing to try them and give you our results, in hope of it being fixed.

I mean if we have the time, and its not much effort, I don\'t see why I wouldn\'t do it.....well besides the fact that I said I wouldn\'t ever go back into testing software again..... http://www.northernsounds.com/ubb/NonCGI/images/icons/wink.gif but its in my blood...sigh

------------------
Really...I am an Idiot