I have found a major bug in Logic Platinum 5.5.0 Windows. Testing it requires only one click. Change track 2 output to a bus between 1 and 9. Mac users should test it also. http://groups.yahoo.com/group/logic-users/files/LPW%20550%20Aux%20Bug.zip 5.5.0 the last Windows version? I don't think so. :D

Hans Sjöblad wrote: > I have found a major bug in Logic Platinum 5.5.0 Windows. > Testing it requires only one click. Change track 2 output to a bus > between 1 and 9. Mac users should test it also. Reproduced and reported. Thanks, Hans. Sascha

>I have found a major bug in Logic Platinum 5.5.0 Windows. >Testing it requires only one click. Change track 2 output to a bus >between 1 and 9. Mac users should test it also. >http://groups.yahoo.com/group/logic-users/files/LPW%20550%20Aux%20Bug.zi p >5.5.0 the last Windows version? I don't think so. :D Tried it on OS X...interesting....I got a spinning beach ball and I had to Force Quit Logic. After that none of the keyboard shortcuts seemed to work in any of the programs I had running. So I tested this on my own autoload...no problems whatsover. May I suggest that you trash this "song" and build a new one form scratch and see if the problems still appears. Cheers -- Check out the web page for my band updated 11.11.2002 . New home address : http://koti.welho.com/jpennane

> I have found a major bug in Logic Platinum 5.5.0 Windows. > Testing it requires only one click. Change track 2 output to a bus > between 1 and 9. Mac users should test it also. > What are u encountering? I don´t see any bug. (LAWP 5.5)

> > I have found a major bug in Logic Platinum 5.5.0 Windows. > > Testing it requires only one click. Change track 2 output to a bus > > between 1 and 9. Mac users should test it also. --- In logic-users@yahoogroups.com, "Newtown Music" <muhammed@n...> wrote: > What are u encountering? I don´t see any bug. > (LAWP 5.5) Instant and permanent freezing of the application.

> So I tested this on my own autoload...no problems whatsover. May I > suggest that you trash this "song" > and build a new one form scratch and see if the problems still appears. This song file was built from scratch. The steps involved were: Re-install Logic. Choose audio driver. Create 64 Aux and 64 Bus objects. Adjust audio object settings.

> > So I tested this on my own autoload...no problems whatsover. May I >> suggest that you trash this "song" >> and build a new one form scratch and see if the problems still appears. > >This song file was built from scratch. The steps involved were: >Re-install Logic. >Choose audio driver. >Create 64 Aux and 64 Bus objects. >Adjust audio object settings. OK....I certainly haven't used that many objects. Do you really need them all ? Have you tried this same with 32 Aux and Bus Objects ? Cheers -- Check out the web page for my band updated 11.11.2002 . New home address : http://koti.welho.com/jpennane

"Hans Sjöblad" <w92nqlq6r001@...@sneakemail.com> wrote: > I have found a major bug in Logic Platinum 5.5.0 Windows. > Testing it requires only one click. Change track 2 output to a bus > between 1 and 9. Mac users should test it also. > > http://groups.yahoo.com/group/logic-users/files/LPW%20550%20Aux%20Bug.zip Confirmed here too. Freezes Logic. What are the parameters for this? I've assigned tracks to busses loads of times but have not got this freeze. Is it to do with the max number of busses or something? Ben Hall www.benhall.co.uk

On a fine day, 03-02-2003, Hans SjXblad <w92nqlq6r001@...> wrote: >I have found a major bug in Logic Platinum 5.5.0 Windows. >Testing it requires only one click. Change track 2 output to a bus >between 1 and 9. Mac users should test it also. > >http://groups.yahoo.com/group/logic-users/files/LPW%20550%20Aux%20Bug.zi p No problem over here. I used my own autoload, assigned track 2 to various busses, put some audio on the track, hit play: plays back fine. LAMP 5.5, OS 9.2.2 on a G4/350. -- Hendrik Jan Veenstra <h@...> Omega Art: http://www.omega-art.com

> > So I tested this on my own autoload...no problems whatsover. May I >> suggest that you trash this "song" >> and build a new one form scratch and see if the problems still appears. > >This song file was built from scratch. The steps involved were: >Re-install Logic. >Choose audio driver. >Create 64 Aux and 64 Bus objects. >Adjust audio object settings. I have now created 64 Bus and Aux objects and I can't reproduce the bug. What about creating (or moving) those auxes and busses to another enviroment layer ? Cheers -- Check out the web page for my band updated 11.11.2002 . New home address : http://koti.welho.com/jpennane

I've now tested my new Logic song with 64 Auxes and Buses and it won't freeze. Here are a few ideas that occured me during the making of it: 1) I'm sure you have abled 64 buses in preferences ? 2) When I converted your Windows song from Asio to Core Audio, I noticed that while the Tracks were changed, the buses and auxes still said ASIO ? 3) Too many objects in one enviroment layer ? My Auxes and buses are on their own enviromenr layer Hope any of this has some help.. Cheers -- Check out the web page for my band updated 11.11.2002 . New home address : http://koti.welho.com/jpennane

> > > So I tested this on my own autoload...no problems whatsover. May I > > > suggest that you trash this "song" > > > and build a new one form scratch and see if the problems still appears. > > > >This song file was built from scratch. The steps involved were: > >Re-install Logic. > >Choose audio driver. > >Create 64 Aux and 64 Bus objects. > >Adjust audio object settings. Jorma Pennanen <jorma.pennanen@p...> wrote: > OK....I certainly haven't used that many objects. Do you really > need them all ? I guess my requirements aren't relevant to this issue. > Have you tried this same with 32 Aux and Bus Objects ? No, I haven't. You can't find this bug with a slightly modified setup. That's often the case with bugs; they require special circumstances to be seen. (A language question: what do bugs do when they become apparent? What's the word? Do bugs "happen" or "occur"?) If Logic had a maximum track number of one million, they ought to test it with a million tracks. It's quite possible that Emagic or any of the beta testers have not used even a relatively small number of aux sends the way I have. I believe that methodical and exhaustive software quality assurance would have eliminated a bug like this. This bug can be discovered easily by simply using the features of the program. In fact, none of the bugs I've found, require more effort to notice than using the features in normal circumstances. Neither Emagic nor the beta testers have used a number of features once or the bugs were left intentionally. To date, I have reported several bugs to Emagic without corrective action taken on their part. Why? The word used by Emagic was "priority". There are bugs since 4.x. Their persistence cannot be explained, but only excused by priority. Take a look at my bug report to Emagic and see how many remain unfixed: http://groups.yahoo.com/group/logic-users/message/116615 If you think some of the "bad features/bugs" don't qualify as bugs, you don't have to tell me that.

--- In logic-users@yahoogroups.com, Jorma Pennanen <jorma.pennanen@p...> wrote: > I have now created 64 Bus and Aux objects and I can't reproduce the > bug. What about creating (or moving) those > auxes and busses to another enviroment layer ? I created a layer per object type and moved them. No change.

> No problem over here. I used my own autoload, assigned track 2 to > various busses, put some audio on the track, hit play: plays back > fine. LAMP 5.5, OS 9.2.2 on a G4/350. Please try my file instead of your autoload.

--- In logic-users@yahoogroups.com, Jorma Pennanen <jorma.pennanen@p...> wrote: > I've now tested my new Logic song with 64 Auxes and Buses and it > won't freeze. > Here are a few ideas that occured me during the making of it: > 1) I'm sure you have abled 64 buses in preferences ? Sure. > 2) When I converted your Windows song from Asio to Core Audio, I > noticed that while the Tracks were changed, the buses and auxes still > said ASIO ? That's strange. My song was built step by step manually from scratch to the point you see. I can't see how it might have gotten corrupted. > 3) Too many objects in one enviroment layer ? My Auxes and buses are > on their own enviromenr layer There's no limit mentioned in the manual.

Hans, After reading your "manifesto" all I can say is I hope you are joking. Your argument that beta testers should load the program to unrealistic levels is ridiculous. In essence you are arguing that if the software can present 64 busses that the program had better work that way, and well. This equates nicely to the concept of: "I still have checks, so there must be money to spend in my bank account." Nowhere in the world of host based processing DAW's would anyone with your expectations find satisfaction. You are looking for problems, not trying to use the application, and if 64 aux busses is a necessity for you to create music, then you have a problem no DAW can solve.

On a fine day, 04-02-2003, mbinvest <mbinvest@...> wrote: >After reading your "manifesto" all I can say is I hope you are joking. Your >argument that beta testers should load the program to unrealistic levels is >ridiculous. In essence you are arguing that if the software can present 64 >busses that the program had better work that way, and well. This equates >nicely to the concept of: "I still have checks, so there must be money to >spend in my bank account." Nowhere in the world of host based processing >DAW's would anyone with your expectations find satisfaction. You are looking >for problems, not trying to use the application, and if 64 aux busses is a >necessity for you to create music, then you have a problem no DAW can solve. Although I understand your argument, and up to a point sympathize with it, he _does_ have a point (if the bug is systematically reproducible). If Logic is supposed to have 64 auxes, then all 64 auxes should work, right? And if every song with 64 auxes freezes the computer, then they'd better include only 48 or 32 or whatever works. I just wonder how on earth one ever discovers a bug like this? I never would use 64 auxes, and surely wouldn't go through the effort of creating them all, one by one, just to see if it can be done. -- Hendrik Jan Veenstra <h@...> Omega Art: http://www.omega-art.com

--- In logic-users@yahoogroups.com, Hendrik Jan Veenstra <h@k...> wrote: > I just wonder how on earth one ever discovers a bug like this? I > never would use 64 auxes, and surely wouldn't go through the effort > of creating them all, one by one, just to see if it can be done. Well ... I use 16 auxes as input monitors (so the sends work all the time). I use another 6 for multiple outputs for a multitimbral VSamp VST instance loaded with drum samples (could do with more from that instance), and have another 6 available for a multitimbral Phatmatik (ditto, I want 16 outputs for that). I can certainly see a situation where I might have another 3 or so VSamp VST instances loaded, with 6 auxes available for each instance's multi-outs, so now we're up to 46 auxes. Add a few more so I can run some sends on bussed submixes, and maybe I'll want to try some softsynth demos ... get the picture? I'm not saying all those will actually have signal present at any one time, just that I might want to setup a song with it all ready to go. My current Autoload has 56 auxes instantiated, with Fadermapper-assisted control access to them. Once you start using multitimbral multi-out softsynths, you chew up auxes really fast. If 64 auxes are available, I want them all to work... John Pitcairn ------------------------------------------------------------------ Midi controller learn in Logic. Soft takeover. Lots of memories. Tracks remember control assignments. Write track automation. Fadermapper demo: http://www.revolver.co.nz/fadermapper/ ------------------------------------------------------------------

--- In logic-users@yahoogroups.com, "mbinvest <mbinvest@y...>" <mbinvest@y...> wrote: > After reading your "manifesto" all I can say is I hope you are joking. Your > argument that beta testers should load the program to unrealistic levels is > ridiculous. In essence you are arguing that if the software can present 64 > busses that the program had better work that way, and well. This equates > nicely to the concept of: "I still have checks, so there must be money to > spend in my bank account." Nowhere in the world of host based processing > DAW's would anyone with your expectations find satisfaction. You are looking > for problems, not trying to use the application, and if 64 aux busses is a > necessity for you to create music, then you have a problem no DAW can solve. I don't think my argument is "beta testers should load the program to unrealistic levels". That kind of a statement would, indeed, be ridiculous. I fail to see how that "I still have checks, so there's gotta be money to spend in the bank." is analogous to this situation. In my view, it is, however, analogous to a situation where you are trying to exceed the limits of your hardware, eg. trying to play lots of tracks when your hard drive can handle less. What I mean is that before you claim your software has a certain feature, I think it goes without saying that you verify it. I'm not talking about beta testers doing this kind of exhaustive testing. It's the job of the company. I don't see anything unrealistic about using all routings. It is uncommon, but not unrealistic. I suppose you are aware of big consoles with hundreds of channels and buses. Ever wondered why they build so many channels? Does using all those buses and auxes constitute "unrealistic" use in your view? Where do you draw the line of "realistic" use? FYI, I was in the midst of creating a system enabling flexible routing, not bug hunting, when this bug happened. Thanks for the feedback.

On a fine day, 03-02-2003, Hans SjXblad <w92nqlq6r001@...> wrote: >I have found a major bug in Logic Platinum 5.5.0 Windows. >Testing it requires only one click. Change track 2 output to a bus >between 1 and 9. Mac users should test it also. > >http://groups.yahoo.com/group/logic-users/files/LPW%20550%20Aux%20Bug.zi p Okay, confirmed on Mac OS 9.2.2, LAMP 5.5 as well. However, I would be curious to know under which _exact_ circumstances the bug occurs. E.g. I removed aux 64 and all references to it, and the song still crashed. Since personally I don't find this particular bug very interesting (won't ever use 64 auxes :), I don't feel like pursuing it any further, but I think that whoever files a bug-report should test stuff like this -- i.e. how many busses/auxes are needed, etc? And, since in the testsong a few of the busses are actually assigned to some output or send: does that influence the occurence of the bug? Are you completely sure that assigning track 2 to bus 1-9 doesn't create some audio-loop? (I've had Logic crash when doing things like setting Aux1 output to bus 2, and then sending bus 2 to bus 1 which again was input to aux 2, or some setup like that). With this many objects, I can't be bothered to check the exact routing of them all, or manually switch them all to 'no output' and re-assiging outputs one by one -- but that _is_ the stuff you should test imo when filing this as a bug. Esp. the audio-loop part would be user-error instead of a genuine bug... BTW, you don't actually use a song with a setup like this as your autoload, do you? Would drive me completely crazy... -- Hendrik Jan Veenstra <h@...> Omega Art: http://www.omega-art.com

Hans wrote: >If Logic had a maximum track number of one million, they ought to test >it with a million tracks. It's quite possible that Emagic or any of >the beta testers have not used even a relatively small number of aux >sends the way I have. I believe that methodical and exhaustive >software quality assurance would have eliminated a bug like this. This >bug can be discovered easily by simply using the features of the >program. In fact, none of the bugs I've found, require more effort to >notice than using the features in normal circumstances. Neither Emagic >nor the beta testers have used a number of features once or the bugs >were left intentionally. I did build yesterday an enviroment with 64 Auxes and 64 Buses..it does not crash (well not at least by changing the output to any of the 64 buses). Maybe there's something else wrong with your song (like what Hendrik suggested). Hope you find a solution to this. Cheers -- Check out the web page for my band updated 11.11.2002 . New home address : http://koti.welho.com/jpennane

--- In logic-users@yahoogroups.com, Hendrik Jan Veenstra <h@k...> wrote: > Okay, confirmed on Mac OS 9.2.2, LAMP 5.5 as well. > > However, I would be curious to know under which _exact_ circumstances > the bug occurs. E.g. I removed aux 64 and all references to it, and > the song still crashed. Since personally I don't find this > particular bug very interesting (won't ever use 64 auxes :), I don't > feel like pursuing it any further, but I think that whoever files a > bug-report should test stuff like this -- i.e. how many busses/auxes > are needed, etc? I don't agree with you. I don't think users have the obligation to determine the cause of bugs, in order for their bug-report to be considered. > And, since in the testsong a few of the busses are actually assigned > to some output or send: does that influence the occurence of the bug? > Are you completely sure that assigning track 2 to bus 1-9 doesn't > create some audio-loop? All sends are bypassed, thus no possibility for feedback. > (I've had Logic crash when doing things like > setting Aux1 output to bus 2, and then sending bus 2 to bus 1 which > again was input to aux 2, or some setup like that). This should be now ok according to documentation. > With this many > objects, I can't be bothered to check the exact routing of them all, Neither can I be bothered to do that. > or manually switch them all to 'no output' and re-assiging outputs > one by one -- but that _is_ the stuff you should test imo when filing > this as a bug. Esp. the audio-loop part would be user-error instead > of a genuine bug... I'm used to software that can't be crashed by user error. :D If there is a paragraph against this kind of routing in the manual, I will accept my responsibility for the problem. The manual should state explicitly what kind of routing is prohibited. > BTW, you don't actually use a song with a setup like this as your > autoload, do you? Would drive me completely crazy... Of course. And I also use 8 sends on every track. Not. :)

--- In logic-users@yahoogroups.com, Jorma Pennanen <jorma.pennanen@p...> wrote: > I did build yesterday an enviroment with 64 Auxes and 64 Buses..it > does not crash (well not at least by changing the output to any of > the 64 buses). Maybe there's something else wrong with your song > (like what Hendrik suggested). Hope you find a solution to this. I said in my earlier post, that this does not occur in a slightly different environment. Is your audio object routing exactly similar to mine? If it is, and it doesn't crash, I'd like to have it, because I wanna get on with this aux send routing thing I was building. I need it to be exactly like mine. There is nothing arbitrary in my setup. It is basically a real environment, of course not a complete one.

On a fine day, 04-02-2003, Hans Sjöblad <w92nqlq6r001@...> wrote: >--- In logic-users@yahoogroups.com, Hendrik Jan Veenstra <h@k...> wrote: >> Okay, confirmed on Mac OS 9.2.2, LAMP 5.5 as well. >> >> However, I would be curious to know under which _exact_ circumstances >> the bug occurs. E.g. I removed aux 64 and all references to it, and >> the song still crashed. Since personally I don't find this >> particular bug very interesting (won't ever use 64 auxes :), I don't >> feel like pursuing it any further, but I think that whoever files a >> bug-report should test stuff like this -- i.e. how many busses/auxes >> are needed, etc? > >I don't agree with you. I don't think users have the obligation to >determine the cause of bugs, in order for their bug-report to be >considered. Well, you're free to disagree obviously :). Maybe I meant more something like: if I run into a bug, I'd like to know exactly "how it works" myself, and therefore tend to figure it out completely before sending in a bug report. To satisfy my own couriosity *and* in the hopes that the more precise I am, the bigger the chances are that the bug actually gets fixed and doesn't hang around until version 13.7. >All sends are bypassed, thus no possibility for feedback. Okay, I'll take your word for it. > > (I've had Logic crash when doing things like >> setting Aux1 output to bus 2, and then sending bus 2 to bus 1 which >> again was input to aux 2, or some setup like that). > >This should be now ok according to documentation. This was in 5.5 as far as I remember. So not okay... >I'm used to software that can't be crashed by user error. :D I suppose the smiley indicates a certain amount of cynism? If not, I'd be curious to know which software can _not_ be crashed by user error... Well, Photoshop maybe -- probably one of the most stable applications on earth. >If there is a paragraph against this kind of routing in the manual, I >will accept my responsibility for the problem. The manual should state >explicitly what kind of routing is prohibited. Yeah, yeah, but that kind of semi-legalese won't fix the bug for you, will it? > > BTW, you don't actually use a song with a setup like this as your >> autoload, do you? Would drive me completely crazy... > >Of course. And I also use 8 sends on every track. Not. :) Well, I don't use 64 auxes, but I _do_ use 8 sends on every track -- religiously. Reverbs mainly. Really makes the music better :-). -- Hendrik Jan Veenstra <h@...> Omega Art: http://www.omega-art.com

--- In logic-users@yahoogroups.com, Hendrik Jan Veenstra <h@k...> wrote: > Well, you're free to disagree obviously :). Maybe I meant more > something like: if I run into a bug, I'd like to know exactly "how it > works" myself, and therefore tend to figure it out completely before > sending in a bug report. To satisfy my own couriosity *and* in the > hopes that the more precise I am, the bigger the chances are that the > bug actually gets fixed and doesn't hang around until version 13.7. Oh, I agree with you. It's just that running "Crashing, starting Logic, tweaking" loop gets a bit tedious after a while. Would you care to find the exact cause of this bug? > >All sends are bypassed, thus no possibility for feedback. > > Okay, I'll take your word for it. All sends are bypassed, thus no possibility for AUDIO feedback, was what I meant. This bug seems to have something to do with feedback. > > > (I've had Logic crash when doing things like > > > setting Aux1 output to bus 2, and then sending bus 2 to bus 1 which > > > again was input to aux 2, or some setup like that). > > > >This should be now ok according to documentation. > > This was in 5.5 as far as I remember. So not okay... Yep. A function of the mal-type. > >I'm used to software that can't be crashed by user error. :D > > I suppose the smiley indicates a certain amount of cynism? If not, > I'd be curious to know which software can _not_ be crashed by user > error... Well, Photoshop maybe -- probably one of the most stable > applications on earth. Hint well received. None of the large software I've used is crashproof. Actually, I was thinking of TWO things when I wrote that line. The other was that usually software design restricts user actions so that invalid actions can't be performed. > >If there is a paragraph against this kind of routing in the manual, I > >will accept my responsibility for the problem. The manual should state > >explicitly what kind of routing is prohibited. > > Yeah, yeah, but that kind of semi-legalese won't fix the bug for you, will it? No. It won't. Just my view on what the user's responsibility is in case of software misbehavior. > > > BTW, you don't actually use a song with a setup like this as your > > > autoload, do you? Would drive me completely crazy... > > > >Of course. And I also use 8 sends on every track. Not. :) > > Well, I don't use 64 auxes, but I _do_ use 8 sends on every track -- > religiously. Reverbs mainly. Really makes the music better :-). You're not sending to 8 different reverbs are you? What do you use on your send destination buses? My use: max 3 verbs per track simultaneously, stereo enh, pseudo stereo, chorus, flange, ensemble, special fx.

On a fine day, 04-02-2003, Hans SjXblad <w92nqlq6r001@...> wrote: >--- In logic-users@yahoogroups.com, Hendrik Jan Veenstra <h@k...> wrote: >> Well, you're free to disagree obviously :). Maybe I meant more >> something like: if I run into a bug, I'd like to know exactly "how it >> works" myself, and therefore tend to figure it out completely before >> sending in a bug report. To satisfy my own couriosity *and* in the >> hopes that the more precise I am, the bigger the chances are that the >> bug actually gets fixed and doesn't hang around until version 13.7. > >Oh, I agree with you. It's just that running "Crashing, starting >Logic, tweaking" loop gets a bit tedious after a while. Would you care >to find the exact cause of this bug? No, but that's this particular bug doesn't interest me that much, since I can't see myself running into it in any real-life scenario (uhm, that's _my_ real-life, for good order :-). > > >All sends are bypassed, thus no possibility for feedback. >> >> Okay, I'll take your word for it. > >All sends are bypassed, thus no possibility for AUDIO feedback, was >what I meant. This bug seems to have something to do with feedback. But if you set bus-1 output to bus-2, which you use as input for aux-1, and then set aux-1 output to output on bus-1 again, or something like that, you already have a feedback loop -- no sends involved, so bypassing sends is irrelevant. And this could even be a longer loop of course, like bus1 > bus 2 > bus 3 > ... > bus 1. Quite hard to notice, maybe, esp. with such a complex setup. I think that just creating such a loop, without even having any real audio present, might crash Logic. Just try: aux 1 has bus 1 as input and bus 2 as output, aux 2 has bus 2 as input and bus 1 as output. What happens? > > >I'm used to software that can't be crashed by user error. :D >> >> I suppose the smiley indicates a certain amount of cynism? If not, >> I'd be curious to know which software can _not_ be crashed by user >> error... Well, Photoshop maybe -- probably one of the most stable >> applications on earth. > >Hint well received. None of the large software I've used is >crashproof. Actually, I was thinking of TWO things when I wrote that >line. The other was that usually software design restricts user >actions so that invalid actions can't be performed. Unfortunately users are always smarter than software designers & writers :-). > >> >If there is a paragraph against this kind of routing in the manual, I >> >will accept my responsibility for the problem. The manual should state >> >explicitly what kind of routing is prohibited. >> >> Yeah, yeah, but that kind of semi-legalese won't fix the bug for >>you, will >it? > >No. It won't. Just my view on what the user's responsibility is in >case of software misbehavior. Sure. Still I think your chances of getting the bug actually *fixed* (which is not always obvious with Emagic :) is to provide an as detailed description as possible, including exact circumstances, etc. > > > > BTW, you don't actually use a song with a setup like this as your >> > > autoload, do you? Would drive me completely crazy... >> > >> >Of course. And I also use 8 sends on every track. Not. :) >> >> Well, I don't use 64 auxes, but I _do_ use 8 sends on every track -- >> religiously. Reverbs mainly. Really makes the music better :-). > >You're not sending to 8 different reverbs are you? What do you use on >your send destination buses? Uhm... there was a smiley, yes? :-) Hell, 8 reverbs... no, never -- 12 at least :-). Seriously, I try to get away with 1 or 2 reverbs for an entire mix. Lots of (almost) reverb-less tracks, as I always feel reverb tends to muddy things. -- Hendrik Jan Veenstra <h@...> Omega Art: http://www.omega-art.com