@bvideo

Thanks for reminding me of that, I forgot that Noel jumped in at some point. Yes this bug certainly has a long history.

Even so, I guess I must have missed an important step somewhere, which would explain why CW is seemingly oblivious of the issue. If anybody knows what that step is, please let me know and I'll fix my mistake.
 
EDIT: Hope you don't mind, I added your bullet point verbatim to the list.

@brundlefly

"Oblivious" in the sense that they are patting themselves on the back for "fixing" what is known as the "MIDI crosstalk" bug, but apparently being unaware of the fact that what they fixed is a tiny subset of what is commonly referred to as the MIDI crosstalk bug, even as far back as Feb 2009.

Yes I know it is nitpicky but we should make sure CW knows there is much more to this than just not being able to record from two different external inputs. All the other related functionality that has been described in detail is still broken.

I'm not sure what to call this bug, as it manifests in different ways. Could be "MIDI crosstalk between VSTi outputs and track inputs", or "Crash when multiple VSTis send MIDI events out", or similar. Not sure.

In my view any bug that has been reported and identified needs to be fixed no matter what it is called.

+1000000000000000000000000000000000000000000000000000000000000000000000000000000000000000 and 0000000000000000000000000000000000000and0 

I had an interesting thing happen when once I tried for the heck of it, to daisy chain 7 beringer bcx2000 units via real 5 pin midi all on their own differing channels to one physical midi port.

It worked! For a moment I had beat a system

But as I played back a project, 3 bcf2000s running under MCU Emulation would have 1 of 8 of the fader twitch along about 1/4" with the music like it was synced to something. I turned off all automation I loaded a blank project and it still persisted. Turned off the metronome, no change

This was with 8.5, I have not tried that stunt again.
But as I recall Silktone's sissue from way back,
I wonder....

How about we call this this the 'Enable Midi-Out VSTi X-Talk Bug"  (EMOVXTB for Short).  (As opposed to the Physical Port Midi X Talk Bug (PPMXTB))

This was was the one thing that would have moved me to X1.

Hopefully the future is not too far away.

JR

Yes the hardware crosstalk is unrelated to this Sonar bug. MIDI hardware was never designed to merge MIDI in that way.

The original MIDI spec only allowed for a total of 16 channels. But 16 channels are woefully inadequate for a DAW. That is why the concept of a "port" was created (something that doesn't exist in the MIDI spec). A port can be something like an external MIDI keyboard's input, the output of one MIDI track, or the output of a VSTi plugin. Essentially anything you can select as the input to a MIDI track. A port is supposed to give you 16 unique MIDI channels, completely isolated from another port's 16 MIDI channels. Sonar can't get that part right, and with some MIDI routing configurations, bleeds MIDI events into unrelated ports. Any "professional" DAW should be able to get this right - folks this is MIDI Routing 101 that Sonar fails at bigtime.

The analogy I like to use is a stereo analog mixer. In this case a "port" is the same as a "track". So each stereo track has 2 channels, L and R. The L channel in track 1 is not the same as the L channel in track 2 (yes they are called the same thing, but they are supposed to be isolated from each other). Now if the same bug existed in this analog mixer, then what would happen is if you feed a signal into the L channel of track 1 while recording something else on track 2, track 1's signal would bleed into the L channel of track 2 (at full 0dB level no less) and would be recorded onto track 2's L channel on the multitrack reel-to-reel.

Some people claim you can work around this issue by only using the R channel in track 2. Well obviously that is a problem if you actually wanted to use the L channel in track 2. Such "work-arounds" also have other issues too which I have pointed out before. If your workflow happens to trigger this bug, you are out of luck. There is no effective work-around for this bug.

I believe the reason CW is incapable of fixing this is because of my theory that Sonar suffers from the "spaghetti code" problem. Basically their code is such a mess that they can no longer track down complex bugs, and fixing bugs often result in other things breaking unexpectedly. Unfortunately there is no quick fix for this problem. X1c isn't going to magically make Sonar a usable DAW again, nor is X1d or X1e. The more they fudge with the code, the bigger the mess becomes. That is why I'm moving to Studio One as may main DAW, although I will be stuck with Sonar for a while since I still have some unfinished projects. Absolutely no more new projects in Sonar, that is for sure. I have zero confidence in Sonar and it isn't going to change anytime soon.

@ Noel

Well you certainly don't give me any reason to believe anything else other than the conclusions I came to. New bugs are popping up left right and center and this points to the issues I brought up.

It also seems to me that CW decided that the "MIDI crosstalk" bug is purely related to recording from multiple external inputs at the same time. As I have already pointed out, the "MIDI crosstalk" bug was discussed in detail in a thread in Feb 2009, and specifically points out the crosstalk from VSTi outputs into other MIDI tracks during recording. That is clearly what is referred to as the MIDI crosstalk bug. If CW decides to fix a different bug and then proclaim that the MIDI crosstalk bug has been fixed, then of course it is going to be pointed out that clearly it hasn't. And you mention the crash resulting from using multiple VSTi MIDI outputs, hence ignore the fact that even with a single VSTi output you will get crosstalk into unrelated tracks.  My fruitless attempts through the years to get CW to fix this issue is directly related to VSTi outputs bleeding into other MIDI tracks while recording. Not the issue with recording from multiple inputs. Not the crash resulting from multiple VSTis sending MIDI data out. Why are you completely ignoring this particular issue?

Now you also say the multiple VSTi crash is an unimportant bug and hence doesn't need fixing (triaged into oblivion), but did you happen to notice that the OP started this thread specifically due to this exact problem?  Sonar's inability to support multiple instances of plugins like Catanya, Thesys ets has been brought up numerous times through the years. Another example of a breakdown in bug triaging since apparently fixing the splash screen (that nobody cares about) is more important than Sonar being unable to get basic MIDI routing right.

Sorry, I'm not withdrawing anything I said, as I firmly believe that Sonar suffers from the issues I pointed out. Any developer looking at Sonar from the outside will come to the same conclusions based on the painfully obvious symptoms.

Brandon Ryan [Cakewalk
]

SilkTone


@ Noel

New bugs are popping up left right and center and this points to the issues I brought up.

They are? X1b introduced new bugs popping up left right and center? I understand you are frustrated but it's just not a factual comment. The only thing I know that X1b introduced was a preview bug in the Browser for non-acidized loops at host tempo.

Really I think you guys should look at the forum postings. New bugs are discovered daily. Do you really want me to enumerate them? They are all right there in plain sight. Things that worked before but are now broken.

Now you also say the multiple VSTi crash is an unimportant bug and hence doesn't need fixing (triaged into oblivion), but did you happen to notice that the OP started this thread specifically due to this exact problem?  Sonar's inability to support multiple instances of plugins like Catanya, Thesys ets has been brought up numerous times through the years. Another example of a breakdown in bug triaging since apparently fixing the splash screen (that nobody cares about) is more important than Sonar being unable to get basic MIDI routing right.

I'll let Noel respond to this but two comments I have: 1) I didn't read anywhere that the bug was "unimportant" or "doesn't need fixing". Those are purely editorial comments. I think what Noel said was that it wasn't the highest priority and explained why. There's a pretty big difference. Having said that I understand your frustration with this particular issue and why you would perhaps make these inferences, but it wasn't what Noel said as far as I could see. 2) What is this splash screen fixing you speak of that you feel has taken priority as a result of triage breakdown?

1) Well, to me "triaged into oblivion" means "not important to fix anytime soon". What other conclusion can you possibly come to? And we are not talking one or two releases here, we are talking 7 or 8 major + patch releases later since CWBRM-1336 was logged and Sonar still can't get basic MIDI routing right.
 
2) From your very own list of bugs fixed in X1b: "Fixed an issue causing the splash screen to appear too briefly". That is just one example. There are dozens of fluffy, low-hanging fruit bugs that have been fixed. If those same developers spent their time fixing bugs like, say, making Sonar get basic MIDI routing right, Sonar would be in a much better shape right now. Nobody cares whether the splash screen is shown too briefly. Bad triaging, period (or a complete lack of it).
 
The bottom line here is that if developers are left to their own devices (cowboy coding), they will always choose the low-hanging fruit. Without proper management from above, the harder-to-fix bugs will always be skipped over in favor of the fluffy, easy to fix bugs ("Hey look, I fixed 7 bugs today!"). As a developer with my own list of bugs, I know how tempting it is to get into such a bad groove. In those cases, bugs need to be triaged and prioritized properly.
 
And sorry but I don't think you have a clue how frustrating this has been. Read comment #29 in this thread again. I never wanted this to get to this point, and I have been a proponent of CW since the early days when "Twelve Tone Systems" had a product called "Cakewalk Pro" that didn't do anything other than MIDI sequencing. But really, I mean come on, the current state of Sonar is just horrible by any standards (accept those that apparently follow the one and only workflow for which CW does have a test case). Can you really say you are proud of Sonar in its current state?

Problem reporter works for me:
http://www.cakewalk.com/s...act/problemreport.aspx

SilkTone,

Without getting into too much detail about our triage or management practices, let me inform you that your assumptions are based on minimal solid information.

For example, fixing the splash screen was a 5-minute fix and it's important to some users.
If it's not important to you, I understand.  But fixing it doesn't imply that another issue is any less important.

You've thus far managed to elicit a response from Product Management, Product Marketing, and the CTO of the company - all of whom have acknowledged your issue and the fact that it must be fixed.

At this point, I think the thread has inarguably been solved:
"MIDI Cross Talk fixed?"

If by MIDI Crosstalk you mean the ability to record from multiple hardware devices, yes.
If you mean the bug causing a crash with multiple VSTs sending MIDI out - no, but it will be.

I sincerely hope this helps.