Har-Bal suspends VST development
Posted on Mar 20, 2005 02:25 pm
Administrator Since: Apr 03, 2002
From Har-Bal.com (I know cuz I put it there)
"It is with regret that we are announcing the suspension of development on the Har-Bal VST plugin. We are acutely aware that this news will undoubtedly disappoint many of you though we feel that this is a necessary decision taken in the best interests of the future development of Har-Bal. What follows is the explanation of our reasoning.
When the proposal for a VST plugin was first mooted it was from a position of naivety on a number of levels: a limited understanding of the plugin architecture and the limits it would impose on harmonic balancing; the belief that VST is a single standard; and the belief that translation to a plugin would be a straightforward process not requiring a great deal of time and effort; belief that we could make a VST plugin of comparable quality to the standalone version. >From the experiences we have had in the past few months we have gained a better understanding of the nature of plugins , if we were to release one, and the ways in which they would hamper the future development of Har-Bal and the harmonic balancing approach.
VST, DirectX and no doubt pretty much all the plugin standards, have a very clear cut separation between a process and the source. Essentially a plugin has no knowledge of the source material structure and generally cannot query it (offline processing accepted). But this aspect is crucial for the future development of Harmonic Balancing. Essentially what it boils down to is that we would be investing a great deal of time and effort into creating (as far as we are concerned) a dead end technology. That is, once you got the basic Har-Bal plugin there is very little more you can do effectively in a plugin straight jacket whereas the standalone product will grow into something far more powerful because it can.
We would have hoped that VST was essentially one standard and one standard only though early results confirmed this not to be the case. What we are getting at here is that the definition of the VST standard was loose enough to enable the various VST host applications to develop there own unique interpretations of what a host should do. Hence, in our development experience we found a great variety in the messaging sent to the plugin from the host which, from our perspective, is problematic because it opens up the potential for a product support nightmare. With such a loose definition of the standard we would essential have to test and design any plugin to work on all hosts since they all behave differently. This is a software developerís nightmare because you have to do release testing on all hosts rather than the idea of testing on the standard and knowing it will run on all hosts. Past experiences with ActiveX hosting doesn't inspire confidence in DirectX either.
Our experiences with VST over the last few months have clearly shown us that to create a VST plugin of comparable quality to Har-Bal standalone is a difficult ask, not the least because of the variation in host behavior. We found that most of our development time, instead of being spent on Har-Bal development, was concentrated on working around idiosyncrasies between the various hosts, some of which we didn't find a solution to. Contrast this to the development of Har-Bal standalone where the majority of our time is spent on new feature coding rather than obscure bug fixing and it became patently clear to us that continuing with the VST development was not in our best interests.
In summary, we are suspending the development of a VST version of Har-Bal so that we can, firstly complete a port of Har-Bal standalone to a common Mac / PC code base and then develop it into the product we want it to be. Believe us when we say that what Har-Bal is now is only a shadow of what it will turn out to be functionally, and the more time we waste on a dead end VST port the longer that wait will be. Our sincerest apologies for those who had there hearts set on a VST version of Har-Bal but we feel that we cannot do justice to harmonic balancing with a plugin and don't wish to release a half baked problematic child to help destroy our hard won reputation.
Earle & Paavo."
:: sigh ::
[ Back to Top ]
Mar 20, 2005 03:38 pm Yeah.. for the best of it, let it be....
Since: May 15, 2004
The current format has helped shortcutting EQ adjustment time (god knows how long that process took me previously) so it's not so troubling importing & expoting files in & out of Har-Bal.
At least I think so.
Mar 20, 2005 04:07 pm I dunno, I am personally seriously disappointed by the decision, but still, like you say, it saves me major time in mastering and is still a great tool. I was, however, looking very forward to the plugin for more than just mastering, I was looking forward to using it in Sonar as well. It seriously bums me out...Still the #1 tool for what it does though...
vdalehubbardLost for words with all to say.Contributor
Mar 20, 2005 04:08 pm My heart dropped some when I saw the headline, but after reading it I totally understand.
Since: Sep 12, 2003
Mar 20, 2005 05:13 pm I agree with there decision. I was never that hot on the VST development anyway. From the sounds of it it wasn't going to work in real time so it would have been of limited use anyway. There are plenty of EQ alternatives already for real time use while mixing. I think putting the effort into HB being best stand alone program it can be is time much better spent.
Since: Jul 02, 2003
Noize2uCzar of MidiAdministrator
Mar 20, 2005 05:20 pm After working on the beta dB and I are currently involved with. I guess I now have a better grasp of just how much time it does consume to get it right. dB can attest to haveing some major issues which some may stem from one problem being fixed, and another pops up because of the fix.
Since: Apr 04, 2002
In fact just last night I reported 2 bugs which I suspected were being casued by changes made to fix the way a curser reacts to left and right mouse clicks in the application. Low and behold, indeed the developers found it at the same time I did.
Though indeed it does sadden me, as I really looked forward to not having to export a single track one at a time if they needed fixing and then import them back in. I can understand now why with only Paavo and Earle coding that app they decided against it.
Here is to a bigger and better HarBal in the future.
Mar 21, 2005 12:49 pm what about offering, some kinda "har-bal-like" plug in?....basically a trimmed down verson of the standalone...
Mar 22, 2005 04:54 am Yes I know dB, you were pretty much excited about the plugin thing in previous forum messages.
Since: May 15, 2004
It is quite easy actually if to use it real time in Wavelab so compression, EQ, stereo expansion, phasing and maximing will just be done once in a file.
We dont have any other choices right now, but who knows in the future?
OD:"There are plenty of EQ alternatives already for real time use while mixing. "
Yep OD, too many of that, and they are all look the same to me; Har-Bal is way too different. So we're gonna treat the app differently.
Good effort Har-Bal team!
If you would like to participate in the forum discussions, feel free to register for your free membership.