Sorry, but although I am sympathetic to your points about an overly pro-corporate environment, I think your perspective on this tax is skewed.
I telecommute from suburban Philadelphia to... well, thats a good question. My employer has an office in NYC that is home to about 6 people, but it also has an office in LA and Nashville, as well as London, Paris, Berlin and Tokyo. Even if I used my company's computing equipment (I don't), it would be hard to pin point exactly what is special about the NYC office other than its my nominal "designated office".
My home neighbourhood is the one that faces expenses from my presence here - theoretically I will drive to and from home more often given than I work at home, I will use more local services because I am at home etc.
But wait. If I commuted to work, I'd be driving those same roads every day whereas by telecommuting I drive them hardly ever. Moreoever, if I returned to my life as a stay-at-home parent, I'd be using local services even more (parks, libraries, stores and more), and paying nothing "extra" to do so. So who is winning and losing here? NYC incurs costs that are asymptotically zero from my designated office location being there. My home neighbourhood incurs costs that are either identical with or lower than the ones that would be involved in me not working at home. I pay state taxes anyway, and am glad to pay for the provision of state and local government services. I just can't see how my telecommuting has anything to do with my specific case. I live and work in Pennsylvania for a company with a world wide presence and an office in NYC. I pay taxes to Pennsylvania (gladly!) and that helps pay for the services I use, and would continue to use even if I were unemployed and paying no taxes.
What's the claim that NYC should collect taxes from me?
given the investment that anyone makes in a computer system designed for gaming, how it is a "most exciting year" to be faced with the possibility of yet another set of continuing reasons to spend more money on yet more gear? wouldn't a really exciting year in gaming have nothing to do with new hardware and everything to do with cool, inspired and inspiring new games?
hmm. the lot is filled with trees, yet you expect to get enough solar exposure to generate a substantive fraction of your consumption? sounds like the lot will be less filled with trees by the time you are done.
instead of contributing to the steady spread of the eastern megalopolis, why not pick up the split level heap of crapboard, demolish it and build something in a place where people have already trashed the landscape? you'll have neighbours, amenities, and probably lots and lots of solar exposure available without cutting down any more trees.
just to be clear, what you describe was once my dream too.
Who really wants to be able to listen to the same tired filler that happens to be broadcast over the radio spectrum on an IP-connected device? Maybe a few people, but not many.
What is much more interesting are internet-only stations like the ones run by Soma FM, which provide fantastic music, no ads, no DJ, and since you're already online, you can instantly look up the musicians via google.
This type of stream totally changed my world in the last 2 years; I used to be a 40 year old guy whose passion for music had died because I just wasn't hearing anything new; now I'm a 42 year old guy who is buying more CD's than ever and who comes across a dozen or so new pieces a week that I love. Throw in my ability to listen to those few radio-spectrum stations that still play
music that is innovative and energetic (sometimes), and its a pigs heaven!
I can't mention my favorite internet-only station because they only support 16 listeners:)
Crichton can be a really nicely-worded misleader when he wants to be. Yes, there are crazed idiots within the Christian faith who believe that the End Days can be accelerated by promoting destabilization in the middle east. And there are idiots in the environmental movement who believe that the problem is doom, the destruction and the rest of the crap that he cites.
Meanwhile, there are Christians working hard to improve the lives of various people on a day to day basis, and there are environmentalists who understand that the issue at hand is not the survival of life on this planet or even humanity, but the question of what kind of life will be possible in world of reduced ecological complexity and altered chemistry/climate.
Ignoring the sane core of a movement is a common tactic, just like ignoring the message by critiquing the messenger. For those whose environmental beliefs approach fundamentalist proportions, maybe Crichton has a point. But for the rest of us, he's talking about a different set of people as a way of avoiding what we are trying to say.
Sure, we are at a common impasse. I claim that black people suffer from racial discrimination, and that compensatory steps need to be taken.
Just out of curiosity, are there any other forms of compensatory steps you can imagine that could be taken rather than affirmative action? I am actually in favor of AA, but I am also wary of a compensatory step that is fundamentally rooted in the same race-based identification (albeit as a final step post-qualification analysis) that got us into the situation to begin with.
i beg to differ. when i quit working to raise my daughter, there was absolutely no expectation that amazon would ever pay off. we relied and planned to rely on her mother's income as a science researcher. it was really very suprising when amazon did as well as it did. i left amazon when there were only 8-10 employees, and the company was still trying to figure out how to raise enough cash to keep operating for another year.
the original poster appeared to me to be claiming that people taking time off to raise children (possibly the most fundamental task most of us will ever do) automatically terminates their utility as software developers etc. the original poster is wrong. my post wasn't trying to say any more than that.
I'm a guy. I'm a software engineer. I left the business for a lot more than 1-2 years so that I could raise my daughter. I'm back in the business, and I have people falling over themselves to hire me. Google call? check. Microsoft? check. various startups in the valley, even though i live in philadelphia? check.
the difference? once my daughter was in preschool, i got pretty serious about writing pro-audio software for linux, and giving it away. result? a couple of paid trips each year to europe to talk about my work, the respect of geek hackers everywhere and hopefully soon some real cash from various companies who want to use my work.
What does my case prove? Nothing, other than that having children has no particular meaning for a person's ability to continue functioning as a productive, skilled software engineer. It might be involved in that process for some people, and some of them might be male, or female (and in this age, perhaps even both at the same time).
And yeah, ok, so I made a million or two from being the #2 hire at amazon.com, but that has little do to with the story:)
I am suprised that the iAudio X5 was not mentioned: its a superb device, marred by only 2 possible flaws that I know of:
silly "color sound" logo on front screen
audio jack is side-mounted, though it does make it more secure than a top-mounted one.
Maybe the fact that you cannot buy it in retail stores was a problem for the reviewers. Even so, video support, Ogg, USB host, full USB mass storage implementation, long battery life... its hard not to gush.
My point is that if you're the type to be out buying prosumer M-Audio cards, you're not also buying Apogee converters. If you're buying Apogee converters, your budget is clearly much higher and you will have already purchased way higher-end audio equipment.
If you are using a computer for digital audio, what else are you going to plug an apogee into? the RME is the standard for bringing 24 channels of ADAT into a computer, regardless of where the ADAT bits come from. there is no "high end" alternative to do this: RME is the high end for this task. Ditto for MADI.
Pro Tools uses a 48-bit internal mixer, optimized to preserve certain high- and low-end frequencies, and this will contribute to sonic clarity. I also have been noticing reviews raving about Sonar's new 64-bit engine which does the same. If you have a lot of tracks and buses, you will hear a difference.
someone posted a link in this discussion to a great paper by lavry, the guy who designs high end audio converters.
although his argument was about sample rate, not mixer design, i am fairly certain that he would call digi, cakewalk and you out on this issue. if you internal mixes never exceed 0dB, the format and structure of the mixer makes no difference. if they go above 0dB, the loss of resolution when using a 32 bit float does *not* have frequency specific effects. claims to the contrary are, AFAIK, utter marketing BS, just like the claims that lavry demolishes about 192kHz SR.
>One of our beta testers regularly records 32 tracks live on a small laptop, and runs sessions with 80 tracks.
With full virtual instruments and effects?
No, live instruments (large drum rigs + multiperson bands + room mics). Real music:)
On a laptop, I can completely understand the role of the CPU-based host sequencer. But in a studio...would you record a full-piece orchestra entirely with your CPU?
Absolutely! It requires hardly any CPU power at all! Recording live acoustic instruments is totally disk i/o bound.
Its recording all those virtual instruments that costs so much in DSP terms.
I read your link, and your argument basically amounts to "CPUs have gotten faster." Yes, they have, but they still can't match a hardware-based solution in a great number of cases, especially as plug-ins go 64-bit and eventually
I can run plugins on my system (not the laptop) that are 64 bit too, already. But its beside the point: look at the cost of a dual opteron system with fairly fast processors. its floating point performance is enough to allow it to keep up with all but the most current DSP farms. those DSP farms are moving forward much more slowly than general purpose CPUs.
128-bit for unparalleled sonic clarity
This is just ludicrous. It suggests you are a user and consumer of plugins, not a DSP programmer.
There's a lot of bitterness from a lot of amateur bedroom recorders because Pro Tools is the big standard, and I don't get why they don't like it.
all i can you is why i don't like it: i asked digi for source code so i could make it better. they laughed. so i decide i should write my own.
there is a lot of misinformation and misunderstanding going around about audio quality and DAWs.
there is a little evidence that some DAWs are not "bits in == bits out", but most are, at least if you are running with 0dB gain and no FX or panning. Ardour has gone to great pains to guarantee that under those conditions, the bits it will send to your interfaces outputs are *IDENTICAL* to the ones you put in. therefore, in this somewhat simplistic circumstance, ardour has no effect on audio quality whatsoever. I suspect the same is true of most other DAWs although there are some that apparently still fail a polarity inversion ("phase") test.
now, once you start mixing, the story is a little different because we are combining signals. but again, if you are working with relatively quiet signals that do not sum to more than 0dB, Ardour (and most other DAWs) offer bit-for-bit fidelity: there is absolutely no loss of any information at all in the operations performed.
if your summed signals exceed 0dB, things get more interesting, and it is in this area that some DAWs will produce better (or worse) results than some others. those using double precision floats, or 56 bit fixed point mixing code will do better under these conditions than those using single precision floats or small precision fixed point.
whether you can actually hear the difference is another question, however. what you can definitely hear are the results of things that people often do not talk about: A/D converter quality and sample clock jitter. the better the converters and the smaller the jitter, the better the sound. this is completely independent of the DAW, and if you use an all digital audio interface, its independent of the audio interface too (unless said interface is utterly braindead in its design, and there have been one or two of them, no doubt).
basically, this talk of "sound quality" and DAWs needs to be replaced by references to converters and sample clocks. after that there is room for a small discussion about mixing stages and any always-on filters that some DAWs are rumored to put on their master outs. but not a lot....
>The same PT HD setup that crashed for Maria Carey before she sang in the superbowl, so they had to transfer the stuff onto a RADAR system (with their own proprietary audio interfaces that sound better than almost anything) ?
That's not what happened at all. I'm surprised this got modded up. But I guess it makes sense...there's this trendy thing on the web by amateurs to hate Pro Tools for some reason, belovedly clutching their Cubase and Cakewalk installations.
Since you appear to know, would you like to tell us what happened? I heard this story from the RADAR operator who was there....
What do Apogee converters have to do with the prosumer cards that were listed?
You plug them into those cards. Digital data moves between them. The sound is phenomenal (mostly because
95% of audio quality issues arise from the sample clock and related issues, and apogee have probably the
best clock in the business.
Oh...okay, I'll believe what "most reviewers" say.:) Let me know when you name them.
i never saw a single review of HD that was really glowing about the sound quality unless it was clearly just pulling from the PR. people like it, but nobody in Mix, EQ, TapeOP or SoundOnSound thought it was that compelling, at either 96 or 192 kHz, especially when compared to other systems at the same SR.
> Yeah, probably the same PT HD setup that you paid US$10-20,000 for, to get some overpriced DSP power that a dual opteron can walk over in its sleep?
Haha. Try recording 80 simultaneous live tracks as someone else posted about. Your dual opteron will never "walk over in its sleep" hardware-based DSP. Or do you play your 3D games entirely on CPU? No, you use a dedicated 3D card.
One of our beta testers regularly records 32 tracks live on a small laptop, and runs sessions with 80 tracks. People have used Ardour to record 100 tracks simultaneously onto a RAID5. Simultaneous track count for recording is disk-io limited, not DSP related. For playback, it obviously depends on the FX level, but see below for a link to my take on this.
Pro Tools doesn't even have a "freeze track" feature. It doesn't need one, like the other DAWs do. DSP is processed off the CPU so you can keep working without having to stop what you're doing and keep your computer from coughing blood when you're pushing Ivory, Rebirth, BFD, Ozone, etc.
I love how anyone who points out that cheesy little prosumer products don't compare with the high-end stuff are suddenly "junkies" or "shills," which tells me you don't know how to argue in a debate. Ended with the classic "Do some research." Why don't you offer me some research? You're the one claiming I'm wrong.
If all those cards have really exceeded and matched today's top studios, nobody would be using Pro Tools as the industry standard. You just can't beat Pro Tools, and it's a standard for a reason...get over it.
I never called you a junkie or a shill, and I actually regret the tone this has taken on. But seriously, PT h/w is nothing particularly special, and everyone I've spoken too who knows anything about their technology agrees. In fact I find it interesting that I've never met anyone who actually likes PT at all, even though I've met many people who use it. PT's h/w is acceptable, but supports the profit margins digidesign needs, not what smaller studios and other organizations should be paying. Their s/w's audio capabilities have always been excellent, the MIDI is so-so and getting better, but there is very little in PT that isn't done better by someone else (problem is, its always different other systems). Studios that I know who care about quality sound use apogee converters and skip the PT h/w for that functionality entirely. Studios who care about modularity, flexibility and lack of vendor lock in certainly don't go the PT route, they use Nuendo, Sonar or others that can be used with various h/w. I've not heard any of them complaining that their stuff is worse quality than PT, in fact, I've heard the opposite.
The same PT HD setup that crashed for Maria Carey before she sang in the superbowl, so they had to transfer the stuff onto a RADAR system (with their own proprietary audio interfaces that sound better than almost anything) ?
Or the same PT HD setup that can't touch apogee converters with a 10 foot pole? Or the same PT HD setup that most reviewers don't think is actually that much better than a mid-level A/D-D/A setup?
Oh, and is this same PT HD that is marketed to waste 2 times the disk space without a single verifiable double blind test showing 192kHz SR's to be detectably different from 96kHz?
Yeah, probably the same PT HD setup that you paid US$10-20,000 for, to get some overpriced DSP power that a dual opteron can walk over in its sleep?
That must be the one. Now I know why it costs so much.
The "prosumer" cards (coupled with appropriate A/D-D/A converters, of course) that you dismiss with a wave match or exceed the quality and specifications in use in any top end studio worldwide as of 5 years ago; they match what almost all but the most capital-rich studios have today. Stop being such a junkie for Digi's marketing BS, and do some research.
Well, the article itself touches on a few of my reasons. Ardour, specifially, is very "Linuxy" in its interface layout and design, reminding me in many ways of the old Dos version of 3D Studio. It definitely looks like a programmer-designed UI, it's very stark and bare-bones, and things are never quite where you expect them to be. It's clearly a Cubase/Logic inspired design and layout, but without the years of fine-tuning those have had to get to their current states. I prefer Ableton's more unorthodox approach anyway, but that's just me:)
Ardour's UI is based almost entirely on ProTools, which most casual users of audio s/w have never used, and many have never even seen. The people who use ardour professionally (and there are a few!) comment that its UI is the most efficient they have used, including ProTools, which most people say is the most efficient in the proprietary world because of its extensive use of keyboard shortcuts. Ardour's development and design has been geared toward learning as much as possible from the years of fine tuning done with other DAWs, although we have been a little hampered by some issues with our GUI toolkit (GTK+ v1). We are currently about 60% done porting ardour to GTK2, and plan to be quite focused on usability issues after that (among many other things).
Re: h/w DSP support: first, DAC's don't have anything to do with this, and even when they are internal to the audio interface, they use no CPU cycles - they are always h/w! But more generally, see: my position on this issue.
partial linux support, if you read the fine print. as in "a few easy classes work on linux, but none of the hard ones. i hope someone will find the time to implement them for linux".
its interesting that this was said about 1 or 2 "industry standard" video editing suites when apple released final cut (pro). final cut pro is now probably the most widely used video editing suite, even including all the big video studios. it has simply evolved to the point where it pushed the existing "industry standards" out of the way.
i doubt that ardour can do this (and i wrote ardour so i know what i am talking about), but we'll give it our best shot, ok?
Unfortunately you don't really know what you're talking about. Or maybe fortunately.
RME Hammerfall and HDSP series (26 channels), M-Audio Delta 1010 (10/12 channels), AudioScience (8 channels) and at least 4 others fully and well supported on Linux are at least equal to the quality of ProTools HD. In fact are generally up with the best you can buy (for all digital interfaces, quality is most defined by your A/D + D/A converters, which have nothing to do with what you install in the computer. They cost significantly less than PT HD hardware. I leave it up to you to figure out why that is.
Linux does have a gaping hole right now with Fireware-based external audio interfaces, which is soon to be filled in by the FreeBob project. Linux also cannot support h/w from several manufacturers who refuse to provide information required for drivers (MOTU is a particularly blatant example). Note that you cannot use your PT h/w with non-PT software, at least until very recently and even then only on OS X with particular caveats. Wanna take another guess at why it costs so much?
Disclaimer: author of Ardour, the RME Hammerfall & HSP drivers, and an RME reseller
So if a bunch of Nigerians, Chinese, Pakistanis and Mayans marched into your own particular culture, wiped out most of the existing cultural and historical power structures, ripped natural resources from your land for a few hundred years, routinely sold your society's members into slavery overseas, took control of extraction resources like oil (largely through offering bribes and control to politicians), built and then abandoned a civil society again and again and so on and so forth, that we should blame you and your neighbours for failing to throw off the past? I am not a big fan of historical determinism (the idea that the past determines the future), but fer chrissakes, what has been done to Africa over the last several centuries goes beyond "well why can't they just get *their* act together?"
I don't like to be rude, but you are one amazingly arrogant and potentially ignorant dimwit.
The african continent has been raped of resources, of stable social structures, of people, of its own history, for many centuries. "Civilizations" from Europe, the middle and far East and more recently from the Americas have all played a role in this.
If you want to get a handle on how this particular imbalance of power might have happened, read "Guns, Germs and Steel".
Blaming Africans for the state of their continent after what has been done to that continent for so long is not just ignorant, its callous and disgusting.
This happened to me recently as well. I am not as well known as ESR in general linux circles, but those of you who mess around with audio software on Linux probably know me as the author of both JACK and Ardour. MS called me 10 days ago about a job, and emailed me again yesterday. The caller indicated that he knew all about my work on linux audio, and my feelings about MS, but assured me that "MS was changing". I was sent a URL for an PR/newswire "article" suggesting that MS was moving "toward open source".
Like ESR, I indicated to them that Microsoft was a company that I could never consider working for, under any circumstances whatsoever.
unfortunately, this is a somewhat misplaced enthusiam. multitrack digital audio recording will indeed burn large quantities of disk, but not at the rate cited above (even adjusting for the possibility that "track" means "stereo"). 18 tracks at 2 bytes per sample at 44100 samples per second is about 1MB/sec, not 100MB/sec.
more importantly, as others have pointed out, you can massively increase disk data rates by using RAID, which provides much better bytes/sec/<monetary-unit> and is likely to for some time.
anyway, my point was really that current leading edge disk rates already provide more than enough bandwidth for all but the largest scale multitrack projects (e.g. movie soundtrack work with 60+ tracks). Track counts of 30-50 tracks are easily achievable with today's technology.
Slashdot Mirror
I telecommute from suburban Philadelphia to ... well, thats a good question. My employer has an office in NYC that is home to about 6 people, but it also has an office in LA and Nashville, as well as London, Paris, Berlin and Tokyo. Even if I used my company's computing equipment (I don't), it would be hard to pin point exactly what is special about the NYC office other than its my nominal "designated office".
My home neighbourhood is the one that faces expenses from my presence here - theoretically I will drive to and from home more often given than I work at home, I will use more local services because I am at home etc.
But wait. If I commuted to work, I'd be driving those same roads every day whereas by telecommuting I drive them hardly ever. Moreoever, if I returned to my life as a stay-at-home parent, I'd be using local services even more (parks, libraries, stores and more), and paying nothing "extra" to do so. So who is winning and losing here? NYC incurs costs that are asymptotically zero from my designated office location being there. My home neighbourhood incurs costs that are either identical with or lower than the ones that would be involved in me not working at home. I pay state taxes anyway, and am glad to pay for the provision of state and local government services. I just can't see how my telecommuting has anything to do with my specific case. I live and work in Pennsylvania for a company with a world wide presence and an office in NYC. I pay taxes to Pennsylvania (gladly!) and that helps pay for the services I use, and would continue to use even if I were unemployed and paying no taxes.
What's the claim that NYC should collect taxes from me?
given the investment that anyone makes in a computer system designed for gaming, how it is a "most exciting year" to be faced with the possibility of yet another set of continuing reasons to spend more money on yet more gear? wouldn't a really exciting year in gaming have nothing to do with new hardware and everything to do with cool, inspired and inspiring new games?
Oh, you mean like their already announced and available collaboration with Apogee to produce an ieee1394 interface to work with logic?
hmm. the lot is filled with trees, yet you expect to get enough solar exposure to generate a substantive fraction of your consumption? sounds like the lot will be less filled with trees by the time you are done.
instead of contributing to the steady spread of the eastern megalopolis, why not pick up the split level heap of crapboard, demolish it and build something in a place where people have already trashed the landscape? you'll have neighbours, amenities, and probably lots and lots of solar exposure available without cutting down any more trees.
just to be clear, what you describe was once my dream too.
Who really wants to be able to listen to the same tired filler that happens to be broadcast over the radio spectrum on an IP-connected device? Maybe a few people, but not many.
What is much more interesting are internet-only stations like the ones run by Soma FM, which provide fantastic music, no ads, no DJ, and since you're already online, you can instantly look up the musicians via google. This type of stream totally changed my world in the last 2 years; I used to be a 40 year old guy whose passion for music had died because I just wasn't hearing anything new; now I'm a 42 year old guy who is buying more CD's than ever and who comes across a dozen or so new pieces a week that I love. Throw in my ability to listen to those few radio-spectrum stations that still play music that is innovative and energetic (sometimes), and its a pigs heaven!
I can't mention my favorite internet-only station because they only support 16 listeners :)
Crichton can be a really nicely-worded misleader when he wants to be. Yes, there are crazed idiots within the Christian faith who believe that the End Days can be accelerated by promoting destabilization in the middle east. And there are idiots in the environmental movement who believe that the problem is doom, the destruction and the rest of the crap that he cites.
Meanwhile, there are Christians working hard to improve the lives of various people on a day to day basis, and there are environmentalists who understand that the issue at hand is not the survival of life on this planet or even humanity, but the question of what kind of life will be possible in world of reduced ecological complexity and altered chemistry/climate.
Ignoring the sane core of a movement is a common tactic, just like ignoring the message by critiquing the messenger. For those whose environmental beliefs approach fundamentalist proportions, maybe Crichton has a point. But for the rest of us, he's talking about a different set of people as a way of avoiding what we are trying to say.
Just out of curiosity, are there any other forms of compensatory steps you can imagine that could be taken rather than affirmative action? I am actually in favor of AA, but I am also wary of a compensatory step that is fundamentally rooted in the same race-based identification (albeit as a final step post-qualification analysis) that got us into the situation to begin with.
i beg to differ. when i quit working to raise my daughter, there was absolutely no expectation that amazon would ever pay off. we relied and planned to rely on her mother's income as a science researcher. it was really very suprising when amazon did as well as it did. i left amazon when there were only 8-10 employees, and the company was still trying to figure out how to raise enough cash to keep operating for another year.
the original poster appeared to me to be claiming that people taking time off to raise children (possibly the most fundamental task most of us will ever do) automatically terminates their utility as software developers etc. the original poster is wrong. my post wasn't trying to say any more than that.
I'm a guy. I'm a software engineer. I left the business for a lot more than 1-2 years so that I could raise my daughter. I'm back in the business, and I have people falling over themselves to hire me. Google call? check. Microsoft? check. various startups in the valley, even though i live in philadelphia? check.
the difference? once my daughter was in preschool, i got pretty serious about writing pro-audio software for linux, and giving it away. result? a couple of paid trips each year to europe to talk about my work, the respect of geek hackers everywhere and hopefully soon some real cash from various companies who want to use my work.
What does my case prove? Nothing, other than that having children has no particular meaning for a person's ability to continue functioning as a productive, skilled software engineer. It might be involved in that process for some people, and some of them might be male, or female (and in this age, perhaps even both at the same time).
And yeah, ok, so I made a million or two from being the #2 hire at amazon.com, but that has little do to with the story :)
US$500 ? i don't think so. I paid US$289 for the one we have, at amazon.com.
I am suprised that the iAudio X5 was not mentioned: its a superb device, marred by only 2 possible flaws that I know of:
Maybe the fact that you cannot buy it in retail stores was a problem for the reviewers. Even so, video support, Ogg, USB host, full USB mass storage implementation, long battery life ... its hard not to gush.
My point is that if you're the type to be out buying prosumer M-Audio cards, you're not also buying Apogee converters. If you're buying Apogee converters, your budget is clearly much higher and you will have already purchased way higher-end audio equipment.
If you are using a computer for digital audio, what else are you going to plug an apogee into? the RME is the standard for bringing 24 channels of ADAT into a computer, regardless of where the ADAT bits come from. there is no "high end" alternative to do this: RME is the high end for this task. Ditto for MADI.
Pro Tools uses a 48-bit internal mixer, optimized to preserve certain high- and low-end frequencies, and this will contribute to sonic clarity. I also have been noticing reviews raving about Sonar's new 64-bit engine which does the same. If you have a lot of tracks and buses, you will hear a difference.
someone posted a link in this discussion to a great paper by lavry, the guy who designs high end audio converters. although his argument was about sample rate, not mixer design, i am fairly certain that he would call digi, cakewalk and you out on this issue. if you internal mixes never exceed 0dB, the format and structure of the mixer makes no difference. if they go above 0dB, the loss of resolution when using a 32 bit float does *not* have frequency specific effects. claims to the contrary are, AFAIK, utter marketing BS, just like the claims that lavry demolishes about 192kHz SR.
>One of our beta testers regularly records 32 tracks live on a small laptop, and runs sessions with 80 tracks.
With full virtual instruments and effects?
No, live instruments (large drum rigs + multiperson bands + room mics). Real music :)
On a laptop, I can completely understand the role of the CPU-based host sequencer. But in a studio...would you record a full-piece orchestra entirely with your CPU?
Absolutely! It requires hardly any CPU power at all! Recording live acoustic instruments is totally disk i/o bound. Its recording all those virtual instruments that costs so much in DSP terms.
I read your link, and your argument basically amounts to "CPUs have gotten faster." Yes, they have, but they still can't match a hardware-based solution in a great number of cases, especially as plug-ins go 64-bit and eventually
I can run plugins on my system (not the laptop) that are 64 bit too, already. But its beside the point: look at the cost of a dual opteron system with fairly fast processors. its floating point performance is enough to allow it to keep up with all but the most current DSP farms. those DSP farms are moving forward much more slowly than general purpose CPUs.
128-bit for unparalleled sonic clarity
This is just ludicrous. It suggests you are a user and consumer of plugins, not a DSP programmer.
There's a lot of bitterness from a lot of amateur bedroom recorders because Pro Tools is the big standard, and I don't get why they don't like it.
all i can you is why i don't like it: i asked digi for source code so i could make it better. they laughed. so i decide i should write my own.
there is a lot of misinformation and misunderstanding going around about audio quality and DAWs. there is a little evidence that some DAWs are not "bits in == bits out", but most are, at least if you are running with 0dB gain and no FX or panning. Ardour has gone to great pains to guarantee that under those conditions, the bits it will send to your interfaces outputs are *IDENTICAL* to the ones you put in. therefore, in this somewhat simplistic circumstance, ardour has no effect on audio quality whatsoever. I suspect the same is true of most other DAWs although there are some that apparently still fail a polarity inversion ("phase") test.
now, once you start mixing, the story is a little different because we are combining signals. but again, if you are working with relatively quiet signals that do not sum to more than 0dB, Ardour (and most other DAWs) offer bit-for-bit fidelity: there is absolutely no loss of any information at all in the operations performed.
if your summed signals exceed 0dB, things get more interesting, and it is in this area that some DAWs will produce better (or worse) results than some others. those using double precision floats, or 56 bit fixed point mixing code will do better under these conditions than those using single precision floats or small precision fixed point.
whether you can actually hear the difference is another question, however. what you can definitely hear are the results of things that people often do not talk about: A/D converter quality and sample clock jitter. the better the converters and the smaller the jitter, the better the sound. this is completely independent of the DAW, and if you use an all digital audio interface, its independent of the audio interface too (unless said interface is utterly braindead in its design, and there have been one or two of them, no doubt).
basically, this talk of "sound quality" and DAWs needs to be replaced by references to converters and sample clocks. after that there is room for a small discussion about mixing stages and any always-on filters that some DAWs are rumored to put on their master outs. but not a lot ....
>The same PT HD setup that crashed for Maria Carey before she sang in the superbowl, so they had to transfer the stuff onto a RADAR system (with their own proprietary audio interfaces that sound better than almost anything) ?
That's not what happened at all. I'm surprised this got modded up. But I guess it makes sense...there's this trendy thing on the web by amateurs to hate Pro Tools for some reason, belovedly clutching their Cubase and Cakewalk installations.
Since you appear to know, would you like to tell us what happened? I heard this story from the RADAR operator who was there ....
What do Apogee converters have to do with the prosumer cards that were listed?
You plug them into those cards. Digital data moves between them. The sound is phenomenal (mostly because 95% of audio quality issues arise from the sample clock and related issues, and apogee have probably the best clock in the business.
Oh...okay, I'll believe what "most reviewers" say. :) Let me know when you name them.
i never saw a single review of HD that was really glowing about the sound quality unless it was clearly just pulling from the PR. people like it, but nobody in Mix, EQ, TapeOP or SoundOnSound thought it was that compelling, at either 96 or 192 kHz, especially when compared to other systems at the same SR.
> Yeah, probably the same PT HD setup that you paid US$10-20,000 for, to get some overpriced DSP power that a dual opteron can walk over in its sleep?
Haha. Try recording 80 simultaneous live tracks as someone else posted about. Your dual opteron will never "walk over in its sleep" hardware-based DSP. Or do you play your 3D games entirely on CPU? No, you use a dedicated 3D card.
One of our beta testers regularly records 32 tracks live on a small laptop, and runs sessions with 80 tracks. People have used Ardour to record 100 tracks simultaneously onto a RAID5. Simultaneous track count for recording is disk-io limited, not DSP related. For playback, it obviously depends on the FX level, but see below for a link to my take on this.
Pro Tools doesn't even have a "freeze track" feature. It doesn't need one, like the other DAWs do. DSP is processed off the CPU so you can keep working without having to stop what you're doing and keep your computer from coughing blood when you're pushing Ivory, Rebirth, BFD, Ozone, etc.
My take on DSP vs. native.
I love how anyone who points out that cheesy little prosumer products don't compare with the high-end stuff are suddenly "junkies" or "shills," which tells me you don't know how to argue in a debate. Ended with the classic "Do some research." Why don't you offer me some research? You're the one claiming I'm wrong.
If all those cards have really exceeded and matched today's top studios, nobody would be using Pro Tools as the industry standard. You just can't beat Pro Tools, and it's a standard for a reason...get over it.
I never called you a junkie or a shill, and I actually regret the tone this has taken on. But seriously, PT h/w is nothing particularly special, and everyone I've spoken too who knows anything about their technology agrees. In fact I find it interesting that I've never met anyone who actually likes PT at all, even though I've met many people who use it. PT's h/w is acceptable, but supports the profit margins digidesign needs, not what smaller studios and other organizations should be paying. Their s/w's audio capabilities have always been excellent, the MIDI is so-so and getting better, but there is very little in PT that isn't done better by someone else (problem is, its always different other systems). Studios that I know who care about quality sound use apogee converters and skip the PT h/w for that functionality entirely. Studios who care about modularity, flexibility and lack of vendor lock in certainly don't go the PT route, they use Nuendo, Sonar or others that can be used with various h/w. I've not heard any of them complaining that their stuff is worse quality than PT, in fact, I've heard the opposite.
The same PT HD setup that crashed for Maria Carey before she sang in the superbowl, so they had to transfer the stuff onto a RADAR system (with their own proprietary audio interfaces that sound better than almost anything) ?
Or the same PT HD setup that can't touch apogee converters with a 10 foot pole? Or the same PT HD setup that most reviewers don't think is actually that much better than a mid-level A/D-D/A setup?
Oh, and is this same PT HD that is marketed to waste 2 times the disk space without a single verifiable double blind test showing 192kHz SR's to be detectably different from 96kHz?
Yeah, probably the same PT HD setup that you paid US$10-20,000 for, to get some overpriced DSP power that a dual opteron can walk over in its sleep?
That must be the one. Now I know why it costs so much.
The "prosumer" cards (coupled with appropriate A/D-D/A converters, of course) that you dismiss with a wave match or exceed the quality and specifications in use in any top end studio worldwide as of 5 years ago; they match what almost all but the most capital-rich studios have today. Stop being such a junkie for Digi's marketing BS, and do some research.
Well, the article itself touches on a few of my reasons. Ardour, specifially, is very "Linuxy" in its interface layout and design, reminding me in many ways of the old Dos version of 3D Studio. It definitely looks like a programmer-designed UI, it's very stark and bare-bones, and things are never quite where you expect them to be. It's clearly a Cubase/Logic inspired design and layout, but without the years of fine-tuning those have had to get to their current states. I prefer Ableton's more unorthodox approach anyway, but that's just me :)
Ardour's UI is based almost entirely on ProTools, which most casual users of audio s/w have never used, and many have never even seen. The people who use ardour professionally (and there are a few!) comment that its UI is the most efficient they have used, including ProTools, which most people say is the most efficient in the proprietary world because of its extensive use of keyboard shortcuts. Ardour's development and design has been geared toward learning as much as possible from the years of fine tuning done with other DAWs, although we have been a little hampered by some issues with our GUI toolkit (GTK+ v1). We are currently about 60% done porting ardour to GTK2, and plan to be quite focused on usability issues after that (among many other things).
Re: h/w DSP support: first, DAC's don't have anything to do with this, and even when they are internal to the audio interface, they use no CPU cycles - they are always h/w! But more generally, see: my position on this issue.
partial linux support, if you read the fine print. as in "a few easy classes work on linux, but none of the hard ones. i hope someone will find the time to implement them for linux".
its interesting that this was said about 1 or 2 "industry standard" video editing suites when apple released final cut (pro). final cut pro is now probably the most widely used video editing suite, even including all the big video studios. it has simply evolved to the point where it pushed the existing "industry standards" out of the way.
i doubt that ardour can do this (and i wrote ardour so i know what i am talking about), but we'll give it our best shot, ok?
Unfortunately you don't really know what you're talking about. Or maybe fortunately.
RME Hammerfall and HDSP series (26 channels), M-Audio Delta 1010 (10/12 channels), AudioScience (8 channels) and at least 4 others fully and well supported on Linux are at least equal to the quality of ProTools HD. In fact are generally up with the best you can buy (for all digital interfaces, quality is most defined by your A/D + D/A converters, which have nothing to do with what you install in the computer. They cost significantly less than PT HD hardware. I leave it up to you to figure out why that is.
Linux does have a gaping hole right now with Fireware-based external audio interfaces, which is soon to be filled in by the FreeBob project. Linux also cannot support h/w from several manufacturers who refuse to provide information required for drivers (MOTU is a particularly blatant example). Note that you cannot use your PT h/w with non-PT software, at least until very recently and even then only on OS X with particular caveats. Wanna take another guess at why it costs so much?
Disclaimer: author of Ardour, the RME Hammerfall & HSP drivers, and an RME reseller
So if a bunch of Nigerians, Chinese, Pakistanis and Mayans marched into your own particular culture, wiped out most of the existing cultural and historical power structures, ripped natural resources from your land for a few hundred years, routinely sold your society's members into slavery overseas, took control of extraction resources like oil (largely through offering bribes and control to politicians), built and then abandoned a civil society again and again and so on and so forth, that we should blame you and your neighbours for failing to throw off the past? I am not a big fan of historical determinism (the idea that the past determines the future), but fer chrissakes, what has been done to Africa over the last several centuries goes beyond "well why can't they just get *their* act together?"
I don't like to be rude, but you are one amazingly arrogant and potentially ignorant dimwit.
The african continent has been raped of resources, of stable social structures, of people, of its own history, for many centuries. "Civilizations" from Europe, the middle and far East and more recently from the Americas have all played a role in this.
If you want to get a handle on how this particular imbalance of power might have happened, read "Guns, Germs and Steel".
Blaming Africans for the state of their continent after what has been done to that continent for so long is not just ignorant, its callous and disgusting.
This happened to me recently as well. I am not as well known as ESR in general linux circles, but those of you who mess around with audio software on Linux probably know me as the author of both JACK and Ardour. MS called me 10 days ago about a job, and emailed me again yesterday. The caller indicated that he knew all about my work on linux audio, and my feelings about MS, but assured me that "MS was changing". I was sent a URL for an PR/newswire "article" suggesting that MS was moving "toward open source".
Like ESR, I indicated to them that Microsoft was a company that I could never consider working for, under any circumstances whatsoever.
unfortunately, this is a somewhat misplaced enthusiam. multitrack digital audio recording will indeed burn large quantities of disk, but not at the rate cited above (even adjusting for the possibility that "track" means "stereo"). 18 tracks at 2 bytes per sample at 44100 samples per second is about 1MB/sec, not 100MB/sec.
more importantly, as others have pointed out, you can massively increase disk data rates by using RAID, which provides much better bytes/sec/<monetary-unit> and is likely to for some time.
anyway, my point was really that current leading edge disk rates already provide more than enough bandwidth for all but the largest scale multitrack projects (e.g. movie soundtrack work with 60+ tracks). Track counts of 30-50 tracks are easily achievable with today's technology.