* Aaron Trumm <aaron(a)nquit.com> [Oct 06 03 21:21]: Of course, think about what is happening: The incoming data is being buffered by the sound card then delivered to ardour. Ardour then typically re routes this back via jack to soundcard. You have three options: 1) lower the latency settings for jack (-p -n) 2) Use a card that supports some form of HW monitoring 3) Don't monitor using ardour. Note that having a low latency kernel doesn't decrease latency, it just makes things work more reliably. HTH. --ant

On Tue, 2003-10-07 at 08:54, Aaron Trumm wrote: Aaron, That's quite large. I run at -p 128 most of the time, and sometimes at 64. Mark

On Tue, 2003-10-07 at 09:27, Aaron Trumm wrote: Yes, give it a try. If your machine is set up well and the card has a good IRQ (9, 10 or 11 with nothing generating lots of interrupts n front of it) the my experience with both of my RME cards under Linux is taht they work quite well. :-( I'm having trouble with one Windows ME installation right now that has my Hammerfall Lite in it. I have a triple boot setup with three separate installations on it: 1) Planet CCRMA Linux - no xruns, works great 2) Win ME running GigaStudio - works great. no clicks and pops 3) Win ME running Reaktor Session - clicks and pops all over the place Reaktor takes a lot of CPU power, and this is my slowest box (P3 500) so I'm thinking that it's just underpowered, but it goes to show that lots of things effect how well a box works... :-( Cheers, Mark

On Wed, 2003-10-08 at 00:25, Rick Taylor wrote: Yes, I've used it, but it's not even in the same game, much less the same league. I think the much more similar might be something more like AMS, which I admit I haven't used. It looks quite good though. It does raise an interesting point for me though. The reason I went with Reaktor Session and not Reaktor is that I do NOT want to program sounds. I just wanted to get a bunch of preprogrammed synths and find some new sounds that I could use, which I was able to do in about 10 minutes of installation. The neat thing (for me anyway) about Reaktor is the on-line user contributed library of about 1500 prebuilt synths that come with, on average, maybe 10 preprogrammed patches. That's a lot of stuff to get on day one. Why isn't something like that happening in the Linux soft synth area? Instead of having so many soft synths, which seems more a focus on writing software, why not more sounds for some existing soft synth, which seems more a focus on making music? I think we Linux soft synth users should try to make that happen. - Mark

On Wed, 2003-10-08 at 03:51, Patrick Shirkey wrote: I'd forgotten about that page. Thanks Patrick. I think I even found one of my patches for SSM there. ;-)

Dave Phillips wrote: ^^^^^^^^^^^^^^^^^^^ Actually I think Mark was the original poster...

I think the strength of Reaktor is the resources in that central program. Right now in Linux we have Pd which is probably the closet to it and a bunch of synths that are seperate and different from each other. It seems we need a similar program in terms of functionality. Ams is probably close, but its not multi-timbral, multi-channel, or is it. Like it or not, Reaktor works but I wouldn't go back into that big money arena again if you gave me the money! Ken Mark Knecht wrote:

ahhh ha yes true true but that implies that it's a given you can LAND those jobs, which makes me giggle :)

Hallo, Robert Jonsson hat gesagt: // Robert Jonsson wrote: There now also is Pd patch and track sharing on http://pure-data.org/community/ which I thereby announce here... ciao -- Frank Barknecht _ ______footils.org__

Dave Griffiths wrote: I know it's anathema to some folks here, but I'll suggest it anyway: patch randomizers. If PD or SSM or AMS or ZASF each had patch randomization controls I'd find it easier to create more patches. Constrained patch randomization is an especially nice touch. (My apologies to the authors of those synths if such controls already exist.) I created hundreds of useful sounds for my TX802 and TX81Z by using such controls in editors like Voyetra's Sideman series. Programming synth patches from scratch implies a lot of knowledge regarding any particular synthesis method. It takes time, and I consider it a craft, definitely worth pursuing *if you have the time to do so*. But using a patch randomizer to generate, say, a dozen variants on a basic piano sound, that doesn't require much beyond an ear capable of nice judgement and some knowledge about general synthesis basics (amplitude envelope functions, velocity response, etc.) The same applies to effects patches, and IIRC there's now a randomization control available for LADSPA. Maybe Steve H. can elaborate on its utility ? Most of the members of this list are aware of this page: http://linux-sound.org/music.html I'll be adding a couple new items to it next release, and I welcome any new links or updates. Some of the titles there are pretty cool, definitely show-off stuff for people who want to hear music made with Linux audio tools. There's never enough... ;) == dp

Hallo, Michal Seta hat gesagt: // Michal Seta wrote: It's Sourceforge. This problem should be resolved soon, as SF is installing and enabling new servers, but until then you don't have much choice but to retry it several times. Like: $ while ! cvs update; do sleep 5 ; done or so. ciao -- Frank Barknecht _ ______footils.org__

Hi, Hmmm, it seems Noziefarm only accepts patches that are in a text-only format. Probably technical reasons... This does however mean that ZynAddSubFX patches can't be uploaded... Any possibility of fixing this, anyone? /Robert

Hi Mark: I think you raise some good points. They've been raised before, maybe many times over the last few years, but they do indicate how far we still have to go before the average *musician* will forsake his Win/Mac tools for the Linux kit. As another "odd man out", I really don't worry much about it, probably because most of my music-making is still done as a live performer with tools like a guitar, a PA, and my voice, along with other live players (the dead ones never make the gigs ;). I don't use the computer on my jobs, nor do I plan to. However, obviously a lot of people would like to make Linux their music-making platform of choice, but our current lack of GUI sex appeal and the general attitude of "Linux is still a bit too difficult for me" work against a swelling of the ranks in Linux audio *users*. It's probably safe to say that at this time most of the active users _are_ the developers. I like the visionary aspect of it all, but I know that the majority of computer-based music-makers are utterly non-interested in becoming developers (as you say, you'd rather develop music than software). Linux audio developers might best be considered as the architects of a foundation for a wonderful future, which is hardly consolation for the average musical Joe who wants to turn on, tune in, and rock *now*. All that being said, I am very impressed with the whole JACK development strategy, the evolution of Ardour, Csound going LGPL, Kjetil's vstserver, Steve Harris's LADSPA plugins, Miler's awesome PD, PlanetCCRMA/AGNULA, and even some of the newer Linux softsynths. IMO we're already far beyond the "toy" stage and we're moving forward at a respectable rate. Much of the available Linux audio software is usable right now, though there is indeed still a long path ahead before it attains the transparency that Win/Mac users are accustomed to. Anyway, I'm rambling. Sinus infection, wrecked laptop, hard-disk failure, bad tooth, WINE headaches... maybe I'll just switch back to Win98 and end it all... ;) [OT] Btw, has anyone else noted that if Arnie becomes governor of CA, that'll make two members of the original cast of "Predator" who have become state governors ? Who knows, maybe Carl Weathers is next... :) Best, == dp Mark Knecht wrote:

On Wed, Oct 08, 2003 at 10:02:46AM -0700, Mark Knecht wrote: True, but not a complete description. As a user, you have direct contact via email and mailing lists with all of these developers. And while they are perfectly free to ignore you, there is a cultural bias in the free software world in favor of being responsive to one's user communty :-) Remember that no one is going to *make* Steinberg, Digidesign, etc etc do anything either. They have a vested interest in giving you enough of what you want that you keep coming back for more; but if you need something that isn't a popular feature you're out of luck and it is impossible to do anything about it. At least with free software there is a chance that you can find somebody to help you get what you need. I was once on a very active user group for a Yamaha sampler that I owned. We collected a detailed list of features and improvements we'd like to see in a revised version of the ROM for the sampler. Some of these were extravagant, but some were quite simple useability enhancements that should not have been technically difficult at all. We managed to get in touch with some people at Yamaha. They basically told us we were out of luck, but if we'd like about a quarter of the fixes on our list, we could always fork over twice as much money for the replacement device they were coming out with, which would probably never do everything we wanted either, and would probably have a similar lifecycle: get some of what you want... expensive upgrade which *removes* some of your favorite features while adding others... repeat infinitely. Thank god free software doesn't work that way. Sure. The major limiting factor is developer time. -- Paul Winkler http://www.slinkp.com Look! Up in the sky! It's MICRO-MEGATON BEER MONKEY! (random hero from isometric.spaceninja.com)

Mark Knecht wrote: I like hearing that. :) Sadly (?), all attempts at commercial Linux audio software have failed, with the sole exception of 4Front (and I believe they're still here because they wisely diversified to provide audio drivers for a variety of *NIX platforms). IMO it's a classic double bind: few users mean little commercial potential, while lack of commercial apps (especially the existing powerhouses) stifles any potential for an existing user community to move to the new platform. The bells have tolled for the Ceres soundfile editor (*not* the Ceres of NoTAM fame), the Jazz++ MIDI sequencer (now GPL'd but perhaps too late for a strong development interest), and the Ultramaster RS101. The TAON soundfile editor was a promising piece back in 1999, but the authors' insistence on closed sources inhibited community involvement, and when they graduated from school the project went belly-up, with no possible life-support from its users. And God knows how long I cajoled Boris Nagels about releasing the sources for his Multitrack, to no avail. Sometimes I worry about things like Stefan Nitschke's very cool RTSynth (free, but closed-source) or the Bros. Hayward's STEEM project (though they plan on going GPL sometime in the future). Those projects are wonderful, but if their maintainers disappear or simply lose interest, we lose some very nice software. I know, we don't really lose it completely, but it dies the death as far as development is concerned. Given the dependency hell that is Linux, you never know when the next kernel release will break binary compatibility with such apps, leaving the user with the uncomfortable choice of remaining behind the OS curve or abandoning the tools. Lose/lose. But not to worry much. We have ecasound, we have Ardour, and these are fantastic tools, 'way beyond what I had any right to expect back in 1999. And things are getting better here all the time. Recently I sync'd SoundTracker, ecasound, and Hydrogen via the JACK transport. Yeh, it's got some problems, but fundamentally the scheme works, and the particular developers are all dedicated to making it *really* work. I'm getting happier... One of my favorite cautionary tales re: closed sources regards the beautiful version of M produced by Voyetra in the late 80s. Apparently the original coder simply walked off with the sources, so even Voyetra no longer has access to the code. And has anyone else ever wondered what happened to the code base for Opcode's Studio Vision ? Btw, I've been testing some Windows apps for WINE compatibility. I have Buzz working now, and the WINE AppDB lists at least one user who claims to have run Reaktor under some version of WINE. Just FYI... I'll have to check for a Reaktor demo to test... Maybe we need to lobby the folks at Crossover or WineX to consider versions of their products optimized for running Windows audio/MIDI apps ? Obviously I'd prefer users to simply switch to the Linux equivalents, but at least you wouldn't have to write off your considerable investment in Windows software. Best, == dp

Opening Pandora's box again... Wednesday 08 October 2003 22.01 skrev Dave Phillips: An example that is of particular interest to us free software people is that of Buzz. A remarkable piece of free audio software for Windows, though closed source. Anyway, the story has it that the author, for some reason, lost the sources. Thus effectively cutting the long term viability of the branch that was Buzz... Buzz still lives on, for now... but imagine if it had been open... /Robert

"Mark Knecht" <mknecht(a)controlnet.com> wrote: Same here {all counts} I don't think the development model works that way though. If there were just a way of funneling the cash from server sales to linux audio...

On Thu, 2003-10-09 at 06:31, Daniel James wrote: :-((( I burned a copy and tried booting my office PC with this. This machine has 3 1394 controllers in it. The boot process hung loading the ohci driver. I'll check it out on another box or two, probably over the weekend. Cheers, Mark

Hallo, Mark Knecht hat gesagt: // Mark Knecht wrote: [...] I'm working on making exactly this happen with Pd. At least it is one of my current goals. Technically Pd allows for a very interesting thing with its graph-on-parent patches. Basically this is a Pd patch that carries its own GUI. You can see some very early screenshots of these in my angriff drum machine for Pd here: http://footils.org/cgi-bin/cms/?show_article=13 What GOP allows is, that more experienced users could build real easy-to-use instruments for the casual, musician type users. I already coined a marketing term for this: (drum roll...) RRAD-ical Pd where RRAD stands for "Reusable and Rapid Audio Development". This is only half joked. Even Max/MSP, which is far more complex than Reaktor, has much more users than any Linux tool, and the main reason, the secret is, that the many users of Max do build and share (or sell) many useful instruments. The same applies to Csound. I want to help build the same for open source software. Pd is a natural choice for me, because I like and use it a lot, but it is also one of the more widely used "synth" applications available for Linux, maybe because it also is crossplatform and available on Win and OS-X, too. Pd might not be the prettiest software or have the cleanest app design, but it is there, it works and - most importantly - it has a community around it. As none of the open source audio tools really is backed by a company which could force its employees to build patches, write docs or which could sponsor musicians to use its products, the community is the main and only 'selling' point. ciao -- Frank Barknecht _ ______footils.org__

??? One's done with a mouse and the other is done with 88 keys? ;-)