I had problems with Sonar X3 constantly hanging, just about every single project, and even when doing mundane things.  Eventually, with help from users on this forum, the culprit turned out to be Native Instruments Replika plug-in.  Fixed that and now everything is running smooth.
 
This was after I reinstalled X3 twice, updated video drivers, reinstalled my audio interface, just about everything short of reinstalling Windows, and it turns out it was one little plug-in.

Anderton
Folks, it's only going to get worse as systems, hardware, and half-a**ed "standards" proliferate while software becomes more complex and fickle consumers demand more and more, regardless of the consequences.

I am sad to say that you are completely correct, Craig.

microapp
+1 to quarterly updates.
Cubase just released 8.5 and their forum is full of the same complaints about unfixed bugs even though they release only 3 or 4 updates per year. A couple of posts there even suggested a non-subscription subscription model would improve things.
I do not think in either CW's or SB's case the release model is the issue. I think it relates more to marketing philosophy and the need to attract as many new users as possible.

-1000 to quarterly updates!
 
+1000 to continued monthly releases!
 
The best thing about Plat is the monthly releases, I love 'em. I get that some folks have problems with such frequent updating, and for them I say don't update monthly.
 
Look, it's already your choice to do or not do as you see fit. Why take away my choice to update monthly just so you don't have to struggle with the temptation to install the latest monthly when your situation and your common sense dictates that you that you shouldn't? 
 
 

microapp
This may have been mentioned elsewhere but how about this...
Quarterly new features updates. Monthly bug fix updates.

-2000 (how many points are left?)
 
Prior to Platinum I would have jumped at this, but having enjoyed the way things have played out, I now find myself violently opposed to any changes to the current model.

Its a big misconception is that new features are the source of instability. In general this is not necessarily a fact. 
Very often new features are isolated from other features or build upon existing infrastructure. As such any bugs arising manifest in the new feature itself and NOT in pre-existing features.
Even in the case of a complex integrated feature like patch points there were very few bugs that affected other areas of the application - and this was during its development prior to release.
 
On the other hand I find bug fixes to existing features relatively more likely to introduce unwanted side effects. We regression test the heck out of these bugs when necessary (unless its a trivial fix) but despite our best intentions sometimes problems do indeed crop up and its unavoidable. This is not specific to Cakewalk by any means - all software companies face the same issues.
 
Relatively speaking in our new model the bug to fix ratio is really small compared to before when we released much bigger updates. Whenever we release a new update all eyes are on it and people report any problems found even if they are not specific to that update, so it may appear more problematic than it actually is...

As best I can tell, 99% of bugs are with backwards compatibility with saved projects.

The moral of the story, "work faster" so you don't run into this issue.

Sorry, smartass is just in my blood...

Good luck, I have only had problems with 3rd party plugs and Sound Center since upgrading to Kingston but I feel your pain man. 

Whilst I agree that the OP is experiencing issues probably unrelated to Sonar here (I strongly urge him to start a thread or something), I'm pretty gobsmacked at the number of people here who keep up with the roll back arguement as though it somehow eradicates all issues in a moment, and that somehow this is one size fits all solution suitable for all. The more I hear it the more I feel it's being used as an excuse, it's just a workaround facility after all. The out of the frying pan into the fire and beta tester arguement still exists and is not a figment of peoples imaginations.

A scheduled stability release (on some sort of cycle with post regression) will simply extinguish all those arguements and will send a pretty good message to people that they don't have to upgrade every month.

Anyway let's see what happens in December. Well I at least know what will happen now...So I am wearing a 64 bit precision bullet proof vest and a helmet.

Doktor Avalanche
Whilst I agree that the OP is experiencing issues probably unrelated to Sonar here (I strongly urge him to start a thread or something), I'm pretty gobsmacked at the number of people here who keep up with the roll back arguement as though it somehow eradicates all issues in a moment, and that somehow this is one size fits all solution suitable for all. The more I hear it the more I feel it's being used as an excuse, it's just a workaround facility after all. The out of the frying pan into the fire and beta tester arguement still exists and is not a figment of peoples imaginations.

A scheduled stability release (on some sort of cycle with post regression) will simply extinguish all those arguements and will send a pretty good message to people that they don't have to upgrade every month.

Anyway let's what happens in December. Well I at least know what will happen now...So I am wearing a bullet proof vest and a helmet.

Neil is saying that a scheduled stability release will not extinguish the arguments, and I agree with him. Shipping software (as opposed to writing it or fixing it) is its own discipline and, like anything else, the more you do it, the better you get at it. By this time next year, Cakewalk will be best in class in that regard.
 
Anyway, I don't want to be sent a message, I want my monthly updates. (Yeah, I know, I went for the cool sound bite there. Just happens to be true.)

Doktor Avalanche
Whilst I agree that the OP is experiencing issues probably unrelated to Sonar here (I strongly urge him to start a thread or something), I'm pretty gobsmacked at the number of people here who keep up with the roll back arguement as though it somehow eradicates all issues in a moment, and that somehow this is one size fits all solution suitable for all. The more I hear it the more I feel it's being used as an excuse, it's just a workaround facility after all. The out of the frying pan into the fire and beta tester arguement still exists and is not a figment of peoples imaginations.

A scheduled stability release (on some sort of cycle with post regression) will simply extinguish all those arguements and will send a pretty good message to people that they don't have to upgrade every month.

Anyway let's what happens in December. Well I at least know what will happen now...So I am wearing a bullet proof vest and a helmet.

kevinwal
Neil is saying that a scheduled stability release will not extinguish the arguments, and I agree with him.

I stated out of the frying pan into the fire and beta tester arguement still exists and is not a figment of peoples imaginations. This is a customer perception outside of software development.

Where is the quote from this 'Neil' person please....

'Neil is saying that a scheduled stability release will not extinguish the arguments,'

...This is a paraphrase. I'm not sure that statement exists please prove me wrong.

I see discussions around it from Noel but nothing really specifically related to this point IMHO. Considering I wrote my point afterwards hardly surprising either.

My wife often prepares new dishes for me... always they don't taste as they to be.. BUGS ??
I put my comments.. gradually she rectifies.