[Ardour-Users] a few thoughts
ampetrosillo at gmail.com
Tue Dec 4 09:35:09 PST 2012
2012/12/4 Paul Davis <paul at linuxaudiosystems.com>
>>> Mixbus is based on Ardour 2, so no MIDI, but anyway, I find Mixbus
>>>> limited and limiting, I don't know about regular channel-strips, but there
>>>> are only 4 group buses,
>>> mixbus2 has 8 mix busses.
>> It is still potentially "limiting", it's more of a theorical aspect
>> rather than a practical one: you still have to conform to Harrison's idea
>> of a console.
> its not harrison's idea of a console. its a workflow that has worked for
> hundreds or thousands of highly skilled audio engineers for years, and you
> display an incredible level of arrogance in casually dismissing it. in
> fact, it is the same level of arrogance that i did when i started working
> on ardour and thought "oh, all that silly mixer design ... its all just h/w
> limitations and we should just ignore it". It took me a decade to fully
> understand WHY mixers have the design that they do, how this design is NOT
> just a result of "hardware limitations" and how the best consoles actually
> represent what people *want to accomplish* rather than being some limiting
> factor in their work.
I'm not dismissing it, I'm only saying that having a limited number of
busses, and an interface which is closely tied to Harrison's DSP is not
exactly a prime example of versatility... which I understand isn't an aim
for the Mixbus project, rather it's exactly the opposite, since Mixbus
wants to provide a complete solution for mixing, and I can see how it can
be a very good idea. Maybe not the best for all kinds of projects, but
certainly useful. But you have to AGREE with Harrison's idea of an EQ, of a
What I'm advocating for Ardour is a Mixbus-like solution, simply more free
and versatile: it could be useful for the DAW to have an (optional!) set of
bundled plugins (say, a set of *recommended* plugins by the Ardour team -
say, ONE good track compressor, with sidechain options, ONE good parametric
EQ with "analog" feel, ONE good digital linear phase EQ, ONE good
multi-mode filter, etc. - with a consistent GUI across the package in order
to make it seem cohesive...), and it could be useful for users, since most
always use the same tools anyway (apart from few people, who may use a
different compressor for each track to impart them their unique
"flavour"... I don't really believe in it, but still), so for example, I
could easily make a few track templates, presetting each to use a specific
set of plugins, and visually exposing only the most relevant parameters for
each plugin, and tweaking parameters using Ardour's mixer (with "real"
knobs, as in a real mixer) instead of having to access every plugin's
interface just to boost lows or cut highs, which I often find distracting,
you may prefer different plugins, with different layouts, and thus make
your own track templates to use, as if they were modules, in your own
projects, and so on. Then, having bundled plugins and preset track
templates can be a good start for absolute newcomers to Ardour, or to
mixing in general, knowing that they can run Ardour and, with a few clicks,
make their own console with an interface you can easily use.
> now of course, not everybody agrees that the harrison approach with mixbus
> is right or that their particular compressor is the best sounding one, etc.
> this is one of the main reasons why ardour does NOT attempt close
> integration with "specific plugins" because opinions vary as to what to
> use, and as soon as a different specific plugin is used, the level of
> integration becomes hard to maintain (because the plugin may have different
> controls or the same controls but with different sematnics.
Bundled plugins would only serve the purpose to have something to start
with (and maybe use, why not? That's why I said "high-quality" in the first
place, then, if a user prefers third-party plugins, he is free to use
them... just like it happens on ANY other DAW, really). And since many
plugins have weird conflicts with certain DAWs as opposed to others, or
don't work as well as they should, having a tried and tested set of
fully-functioning plugins is a nice way to ensure you can always make a mix
on Ardour (plus, if you stick to bundled plugins - or recommended optional
plugins, as you wish - it is MUCH easier to move projects around).
> and I don't really get the reason to "port" the limitations of hardware
>>>> mixers in the digital domain, as there is no real "need" and it's most
>>>> probably a deliberate desire to limit the application's functionality (for
>>>> marketing purposes, maybe?).
>>> one person's limitation is another person's rapid workflow, based on 30
>>> years of actual experience of how to mix many different styles of music.
>> Tough, it still equates to forcing "another person's rapid workflow" on
>> the user,
> its not forcing anything. you don't have to buy mixbus. but everything
> about mixbus is centered on integration and workflow. if you don't like
> what they've done, don't use it and figure out a way to get someone else
> (or yourself) to invest the amount of hours that they (and I) put in to
> getting it to where it is today, but with your design decisions. the source
> is all available.
That's why many people won't buy Mixbus. Fine, you don't have to make
everyone happy. But if you can make MORE people happy without necessarily
having to compromise on your original idea, I can't really see why you
shouldn't do it. There are design choices, which can only be implemented in
an exclusive way because they clash with other possible approaches, and
there are design choices which can be bent enough without departing from it.
After all, in the beginning, Ardour wasn't even meant to feature MIDI, and
now it does, because it's a feature which doesn't affect the normal audio
functionality of Ardour and can actually improve it.
> THIS kind of limitations. One thing is "standard number of mix busses at
>> boot-up", another is "up to 8 mix busses", it smells of "Lite Edition"
> this is an unfounded suspicion. stop being so cynical.
Please stop being defensive, I'm not saying it IS a Lite Edition (maybe you
have been accused about it, in the past, I really don't think it is), I'm
just saying that track number limitations are the kind of thing which is
very common in the various "LE" editions of commercial software
>> I'm not talking about track templates which adapt to ANY plugin of a
>> certain type, I'm talking about track templates which use SPECIFIC plugins
>> (which then goes well together with having a range of bundled plugins in
>> the DAW).
> and if someone disagrees with your (or my) choices about which specific
> plugins? just like your example of the vintage warmer above. and yes, track
> templates currently already define specific plugins if the user made that
>> You said that all I want is just a bit more "chrome"... well, even if I
>> DID want more chrome, what's really wrong with it? And I personally don't
>> even WANT chrome, but many people I told about Ardour thought that "it
>> looked cheap" or "unrefined" or whatever compared to the other brand-name
>> DAWs they used.
> then they don't need to use it. this isn't a beauty competition. if there
> is one thing i've learned over the years of working on this program it is
> that there is NO appearance, or workflow, or basic program conception that
> will make all users happy (or even excited or even just interested).
> pursuing this as though there is one single design that will somehow
> convince everyone to use a given DAW is absurd. its why there is space in
> the marketplace for ProTools and Abelton Live and Garageband and Pyramix
> and Sonar and Reaper etc. etc.. well, for some definition of "space"
Well, actually, in theory, there IS a superior design which results in
being the best compromise of all, it simply is very hard to see it, but OK,
in practice, different horses for different courses. That's exactly why I'm
saying that, with the few added features I proposed, you could have more
than one horse in the stable.
> >Why keep them out, when you do "survive" on donations alone? In the long
> run, caving in on "chrome" may allow you to earn enough money from
> subscriptions to hire a full-time >developer or two to help you with the
> things that matter the most to you and actually benefiting the project.
> because i don't see any evidence of this claim at all. the idea that there
> is some vast pool of revenue waiting to flow into the project "if only <X>
> happened" is an illusion i don't pay attention to anymore. there are things
> that will hopefully produce incremental increases in revenue, but this is
> an insanely competitive niche that has seen many attempts to break into it
> either fail or barely hang on.
> >Really, I'd be delighted to contribute in first person to Ardour, it's
> not something I can already do (I'm studying electronics engineering - at
> 24, I took the plunge and went back to >studying, hoping I did the right
> choice - and I'll study quite a big deal of programming too, but I'm still
> far from being even vaguely literate about it), I might in the future, but
> seeing >that development for Ardour 3 has been a bit slow (in your own
> words!), maybe you should start considering investing in making the project
> more popular with the majority of users >(without alienating the old ones).
> oh really? its so unpopular that its probably the most financially
> successful GPL niche software so far? the development has been slow because
> there are very few developers with the ability and/or desire (and it takes
> both) to contribute to the project. money doesn't fix that, certainly not
> by itself.
This is equivalent to saying "Poland is the most prosperous country in
Eastern Europe", maybe it is, and it's probably a nice place to live, too,
but you shouldn't compare Ardour to other GPL projects, you should compare
Ardour to the other DAWs in the field. Anyway, you were defensive again,
here, and again, I understand you put a lot of effort in developing Ardour,
but I'm neither saying Ardour is "crap", nor I'm saying it isn't
successful, I'm saying that giving it a bit more visibility may attract
users, developers (who may well decide to contribute, or to base their own
commercial projects on Ardour, giving it even more visibility and,
maybe...), money (which doesn't fix lack of contribution by itself, but it
certainly helps, once you get $10000 worth of donations each month, you
certainly have many more opportunities to invest).
Just to be clear: I'm not saying you HAVE to do what I'm suggesting, my
post in the list was just an "experiment", to see if other people may be
interested in my proposals. So far, I haven't seen either "yes, that's a
good idea", nor "I don't think it's a good idea", but only half-answers
which don't really tackle the issue, and my idea of bundling plugins hasn't
even been really discussed.
> meanwhile take a look at http://ardour.org/development/post3.0 ... that
> is the list of what is slated for work post-3.0. if you really think that
> more chrome matters more than any of those items, then you should figure
> out how to get developer resources applied to the "more chrome" project,
> because i don't (which is not to say that when i find ways to make ardour
> "prettier" along the way i won't work on them).
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Ardour-Users