Some corrections to the "article"
on
DeMuDi Linux
·
· Score: 5
The project is led by Guenter Gieger who is the developer of the Linux low latency driver for the RME Hammerfall 96xx series. Combined
with realtime kernel patches, the Hammerfall card in Linux can achieve hardware level latencies.
Not "realtime" patches to the kernel.
The patches in question are "low latency"
patches that decrease scheduler
latency. They have nothing to do with
RTLinux or "hard realtime" programming.
Guenter didn't write the "native"
digi9652 driver, Winfried Reitsch did
that. Guenter hacked Pd to use the driver.
There is also an ALSA driver for the
Hammerfall (I wrote it) that was based
loosely on Winfried's driver.
"hardware latencies" is a loose term.
lets just say that a properly written
application can use any audio interface
via ALSA to get audio i/o latencies that
are bounded only by the h/w design.
I don't think Guenter started Demudi,
but I think he may be those most active
developer involved in its evolution. Most active
by far, in fact.
If you are a developer working on or interested in
Linux audio software, I also encourage you
to check out the Linux Audio Development web site.
sounds like he not only hasn't seen 2001, he
doesn't even known his own company's acronyms. last time i looked, HAL stood for Hardware
Abstraction Layer.
this is so wrong its hard to know where to begin.
"life" is a human classification that concerns certain kinds of behaviour in the world. the
details of the classification (and they vary from era to era and person to person) explain why we consider a rock to be dead and a bacteria to be alive.
the question provoked by the "strong a-life" proponents such as grand is not over whether or not you can create a potato or a bacteria inside a computer - you obviously cannot since the definition of a potato or bacteria include certain material properties that a computer could not have. no, the question is whether or not, given a particular definition of what "life" means, its possible to organize the behaviour of a computer so that is satisfies the definition.
what generally happens when someone makes this claim in the positive (and it was being made very effectively even in 1994 and before), some objection is raised to the definition or new clauses are added to it.
the reason a tornado simulation is not a tornado is simple: the definition of a tornado involves things that simply don't happen in the simulation.
but go take a look at a college textbook's definition of "life". typically, there are about 10 characteristics used to define "life". its not that hard to create a computer system that satisfies every part of the definition, even though it shares nothing in common with what we typically consider "life".
If you tried to use code that I had written
and released under the GPL in the way you
suggest, you could plan on a (hopefully
short) legal process, initiated by me and with the intent of you being forced to release all your own software under the GPL.
I use the GPL specifically to prevent
exactly the scenario you suggest. Its true that the dynamic link element makes the GPL v2.0
a little fuzzy in this area, but you can bet that
I would purse anyone I knew to be using my GPL'ed
code in this way. My "layman's GPL" runs something like: you can use this code in your own software, but if you do, i expect you to release your
software under the GPL too. My definition of "use", which is not completely clear in the GPL, is "create a program that may at some point
cause the execution of instructions produced by
compiling my source code within the same
execution context and/or address space as the
program itself."
In the long run, who cares. These are all
lossy compression techniques, and as
bandwidth expands, downloading actual
CD quality audio (perhaps viewed as a compressed
form of 24bit/96kHz "DVD" audio) will be
feasible. Likewise for portable players
affected by storage media costs and size.
I would wager that in 10 years, nobody will
listen to MP-anything, except perhaps MP4-SA,
which is effectively a sound synthesis and
processing language rather than a compression
technique.
There's actually some evidence that they
are using GTK. Don't assume that by using GTK
you are forced into the disgustingly boring
looking GUI's that we see all the time with
Linux apps. I have several apps that use GTK
to do heavily pixmap-based GUI's, and I happen
to think they are much better for it. I don't
want my software to look like everything
else on your desktop, though I do let
you use GTK themeing to alter its appearance
without recompilation.
Nobody has noted the real virtue of a 64 bit
address space, even if the Itanium itself
only supports about 50 for VM.
With a 64 bit address space, there is no
longer any need to run applications in their
own address space. You can finally recognize
that protection is orthogonal to addressing,
and start to gain the benefits of not
having to invalidate the TLB and other parts
of the VM system when you do a context switch. That is, all processes run in the same address space, so they can share memory with no
effort whatsoever, and you use an explicit protection
mechanism to avoid memory stomping.
Opal was an experimental system that tried
to explore some of these ideas. It was a
PhD thesis at the University of Washington.
The tech report notes that with a 64MB address space, you can allocate 1MB/sec and not run
out of VM space for a period of time larger than
the estimated current life of the sun.
The real benefits of 64 bit addressing have
little to do with increasing the data width.
Avoiding a TLB flush when doing a context switch will provide one of the most dramatic speedups for multi-tasking systems that you can imagine.
I've been writing software for 14 years. I
know one person in the entire universe of
my friends, family and acquaintances who plays computer or video games.
Jon, what on earth do you mean "no longer a
subculture" ?
Could you offer some examples of these cooperatives that are outperforming other corporations? I look at the Fortune 500 and feel pretty confident none of them are cooperative work environments.
Here's two:
Puget Consumers Coop: highest retail sales per square foot in the Seattle area
REI: largest member owned coop; dominates sales of outdoor recreation gear in every market where it has a store
Both of these are member owned, not employee owned, however.
Re:The best point of the article...
on
Hacking The City
·
· Score: 1
except that according to the article, he's not.
he is quoted as saying "its not as if i've wanted to run a club all my life".
After a year as employee #2 at a phenomenally successful internet
retailer, I didn't end up with the uber-cash that jwz and others have
made from the.com bubble, but i did make enough to not have to work
anymore.
Instead of trying out some other field, I decided to keep hacking,
only this time its my own software: free pro-audio software for Linux.
This is a market dominated by a handful of companies whose software
runs on unstable and inadequate operating systems. Its the best
programming I've done, and takes place with no marketing, no
bullshit. If I want to do it over, I do it over. Every day I can wake
up inspired by the (vague) prospect of wiping out an entire market
segment of the proprietary software world.
Sadly, it seems that most rich programmers who carry on programming
seem to want to do so in an environment where they stand a chance of
making even more money. I wish more of us would take the cash and use
it as an opportunity to break the hold closed, binary-only software
has on the world. Until someone works out a viable business model for
actually selling open source software, that means working without
pay. Then, when we've done that, we can take a look at world peace,
poverty, hunger and injustice. There might be time later for a little
clubbing, if we're lucky.
what *are* you talking about ? how does a computer give an inner-city, impoverished, malnourished and not particularly loved child "go where they want and do as they please". there is no evidence that exposure to computers in high school has anything to do with career success (something of which i am a great example, actually). good *education* can offer children the promise of greater freedom, but it doesn't help them right now, and it doesn't have to do with computers.
Jon thinks they "all relate to technology".
Maybe for the educated, technologically literate and technologically immersed set. But I would have said that child poverty, malnutrition, the concentration of wealth and power in the hands of the few and in corporations, the emergence of two working members of a household as a financial necessity, appalling educational systems in our
inner cities (in particular), pollution of the air and the water, destruction of natural landscapes, increasing levels of domestic violence, the power and debt relationship between different countries of the world, the debate over abortion, racism: theses are the important issues, and they have as much to do with technology as any other part of life - not much. Whether you think that DeCSS and the DCMA are very significant issues or not (and I do), they pale in comparison to the problems, injustices and inequalities faced by millions of Americans and many more people around the world.
This hysterical pandering has nothing to do with the reality of children's lives, or their welfare. If either Bush, Cheney, Gore or Lieberman cared a whit about children, they would shriek instead about the
paucity of decent Internet access -- and even decent computers -- in America's public elementary and middle schools.
Anyone who cared one whit about children would
never equate anything to related to computers
with their general welfare. If you care about
kids and their schools, talk about class size,
books, inspirational and loving teachers,
safety from violence, interesting curricula,
engaging and nurturing environments, respect from adults and children; computers are just about
at the bottom of the list of what any
informed educator sees as making a difference
in children's experience of school. Many trendy
and ill-informed people see them as a remedy
for all kinds of things, but to my knowledge,
no research into the area has ever suggested that computers aid learning or the educational experience to anything like the degree that the things I mention above do.
the GPL isn't a governmant mandate, and it
isn't going to be. the closest it could
get would be a requirement that software
written for government agencies be
covered by the GPL. that seems unlikely to be, but irrelevant.
the GPL is about *me* (software developer)
telling you (user of my software) that i do not
want you using my work to release your own
proprietary software.
the BSD license is about *you* (software developer) telling me (user of your software) that you don't care what i do with it.
ok, so you don't care and i do. whats the problem, and how does Korea or Communism fit into this picture ?
As 3D sound controllers become more complex (like Aureal's controllers) the OS will need a good way to get huge amounts of data and commands to the sound cards
I use the ALSA mmap access mode to shunt 24 channels of 24bit/48kHz (4.5MB/sec) data directly into the h/w buffer of my digital audio interface without any interaction with the OS, and no library calls except memcpy and memset. No mere game is going to get close to this requirement, so you're basically missing the boat on this one. --p
Spend $40-$60 on a Trident based card, such as the Hoontech Soundwave-NX, and you'll have efficient full duplex PCM, 32 streams h/w mixing, a sensibly implemented MIDI port, and more. ALSA has a slightly better driver than the current OSS/Free driver in the kernel, but the latter is derived from the former anyway. For $60 you should be able to pick up a version with both optical and RCA S/PDIF output.
Digidesign's H/W isn't really all that special. There are lots of A/D, D/A boxes out there that can do at least as good a job as the 8810 and its cousins. The DSP Farm is just a bunch of stock DSP chips - the Creamware Pulsar is vastly more interesting for DSP work. No, the golden jewel in Digidesign/Avid's toolchest is the ProTools SOFTWARE. Its slick, but more importantly, it reflects man years of accumulated studio experience. Very little else comes close (though the Ensoniq Paris stuff, and Sonic Foundry's ACID, even if its way more limited, come close).
Slashdot Mirror
If you are a developer working on or interested in Linux audio software, I also encourage you to check out the Linux Audio Development web site.
--pSounds like a good time to link to my diatribe about Iomega, the Jaz drive (which is also subject to the CoD) and Linux.
sounds like he not only hasn't seen 2001, he doesn't even known his own company's acronyms. last time i looked, HAL stood for Hardware Abstraction Layer.
this is so wrong its hard to know where to begin. "life" is a human classification that concerns certain kinds of behaviour in the world. the details of the classification (and they vary from era to era and person to person) explain why we consider a rock to be dead and a bacteria to be alive. the question provoked by the "strong a-life" proponents such as grand is not over whether or not you can create a potato or a bacteria inside a computer - you obviously cannot since the definition of a potato or bacteria include certain material properties that a computer could not have. no, the question is whether or not, given a particular definition of what "life" means, its possible to organize the behaviour of a computer so that is satisfies the definition. what generally happens when someone makes this claim in the positive (and it was being made very effectively even in 1994 and before), some objection is raised to the definition or new clauses are added to it. the reason a tornado simulation is not a tornado is simple: the definition of a tornado involves things that simply don't happen in the simulation. but go take a look at a college textbook's definition of "life". typically, there are about 10 characteristics used to define "life". its not that hard to create a computer system that satisfies every part of the definition, even though it shares nothing in common with what we typically consider "life".
... gain root access ...
/proc/kcore > ~/my-copy-of-memory
cat
If you tried to use code that I had written and released under the GPL in the way you suggest, you could plan on a (hopefully short) legal process, initiated by me and with the intent of you being forced to release all your own software under the GPL. I use the GPL specifically to prevent exactly the scenario you suggest. Its true that the dynamic link element makes the GPL v2.0 a little fuzzy in this area, but you can bet that I would purse anyone I knew to be using my GPL'ed code in this way. My "layman's GPL" runs something like: you can use this code in your own software, but if you do, i expect you to release your software under the GPL too. My definition of "use", which is not completely clear in the GPL, is "create a program that may at some point cause the execution of instructions produced by compiling my source code within the same execution context and/or address space as the program itself."
In the long run, who cares. These are all lossy compression techniques, and as bandwidth expands, downloading actual CD quality audio (perhaps viewed as a compressed form of 24bit/96kHz "DVD" audio) will be feasible. Likewise for portable players affected by storage media costs and size. I would wager that in 10 years, nobody will listen to MP-anything, except perhaps MP4-SA, which is effectively a sound synthesis and processing language rather than a compression technique.
There's actually some evidence that they are using GTK. Don't assume that by using GTK you are forced into the disgustingly boring looking GUI's that we see all the time with Linux apps. I have several apps that use GTK to do heavily pixmap-based GUI's, and I happen to think they are much better for it. I don't want my software to look like everything else on your desktop, though I do let you use GTK themeing to alter its appearance without recompilation.
Nobody has noted the real virtue of a 64 bit address space, even if the Itanium itself only supports about 50 for VM. With a 64 bit address space, there is no longer any need to run applications in their own address space. You can finally recognize that protection is orthogonal to addressing, and start to gain the benefits of not having to invalidate the TLB and other parts of the VM system when you do a context switch. That is, all processes run in the same address space, so they can share memory with no effort whatsoever, and you use an explicit protection mechanism to avoid memory stomping. Opal was an experimental system that tried to explore some of these ideas. It was a PhD thesis at the University of Washington. The tech report notes that with a 64MB address space, you can allocate 1MB/sec and not run out of VM space for a period of time larger than the estimated current life of the sun. The real benefits of 64 bit addressing have little to do with increasing the data width. Avoiding a TLB flush when doing a context switch will provide one of the most dramatic speedups for multi-tasking systems that you can imagine.
I've been writing software for 14 years. I know one person in the entire universe of my friends, family and acquaintances who plays computer or video games. Jon, what on earth do you mean "no longer a subculture" ?
- Puget Consumers Coop: highest retail sales per square foot in the Seattle area
- REI: largest member owned coop; dominates sales of outdoor recreation gear in every market where it has a store
Both of these are member owned, not employee owned, however.except that according to the article, he's not. he is quoted as saying "its not as if i've wanted to run a club all my life".
After a year as employee #2 at a phenomenally successful internet retailer, I didn't end up with the uber-cash that jwz and others have made from the .com bubble, but i did make enough to not have to work
anymore.
Instead of trying out some other field, I decided to keep hacking,
only this time its my own software: free pro-audio software for Linux.
This is a market dominated by a handful of companies whose software
runs on unstable and inadequate operating systems. Its the best
programming I've done, and takes place with no marketing, no
bullshit. If I want to do it over, I do it over. Every day I can wake
up inspired by the (vague) prospect of wiping out an entire market
segment of the proprietary software world.
Sadly, it seems that most rich programmers who carry on programming
seem to want to do so in an environment where they stand a chance of
making even more money. I wish more of us would take the cash and use
it as an opportunity to break the hold closed, binary-only software
has on the world. Until someone works out a viable business model for
actually selling open source software, that means working without
pay. Then, when we've done that, we can take a look at world peace,
poverty, hunger and injustice. There might be time later for a little
clubbing, if we're lucky.
This seems like the idea oppurtunity to publicize a rant of mine. See Why I don't like Iomega JAZ drives.
what *are* you talking about ? how does a computer give an inner-city, impoverished, malnourished and not particularly loved child "go where they want and do as they please". there is no evidence that exposure to computers in high school has anything to do with career success (something of which i am a great example, actually). good *education* can offer children the promise of greater freedom, but it doesn't help them right now, and it doesn't have to do with computers.
Jon thinks they "all relate to technology". Maybe for the educated, technologically literate and technologically immersed set. But I would have said that child poverty, malnutrition, the concentration of wealth and power in the hands of the few and in corporations, the emergence of two working members of a household as a financial necessity, appalling educational systems in our inner cities (in particular), pollution of the air and the water, destruction of natural landscapes, increasing levels of domestic violence, the power and debt relationship between different countries of the world, the debate over abortion, racism: theses are the important issues, and they have as much to do with technology as any other part of life - not much. Whether you think that DeCSS and the DCMA are very significant issues or not (and I do), they pale in comparison to the problems, injustices and inequalities faced by millions of Americans and many more people around the world.
Anyone who cared one whit about children would never equate anything to related to computers with their general welfare. If you care about kids and their schools, talk about class size, books, inspirational and loving teachers, safety from violence, interesting curricula, engaging and nurturing environments, respect from adults and children; computers are just about at the bottom of the list of what any informed educator sees as making a difference in children's experience of school. Many trendy and ill-informed people see them as a remedy for all kinds of things, but to my knowledge, no research into the area has ever suggested that computers aid learning or the educational experience to anything like the degree that the things I mention above do.
yes, i did read the wrong article. thats because of the page layout. you might want to consider redoing this.
they didn't review performance under Linux in any way shape or form. they say this plan to do this later. the review is entirely Windows-centric.
the GPL isn't a governmant mandate, and it isn't going to be. the closest it could get would be a requirement that software written for government agencies be covered by the GPL. that seems unlikely to be, but irrelevant. the GPL is about *me* (software developer) telling you (user of my software) that i do not want you using my work to release your own proprietary software. the BSD license is about *you* (software developer) telling me (user of your software) that you don't care what i do with it. ok, so you don't care and i do. whats the problem, and how does Korea or Communism fit into this picture ?
A forgotten computer scientist, DEC PA Research, circa 1993.
Spend $40-$60 on a Trident based card, such as the Hoontech Soundwave-NX, and you'll have efficient full duplex PCM, 32 streams h/w mixing, a sensibly implemented MIDI port, and more. ALSA has a slightly better driver than the current OSS/Free driver in the kernel, but the latter is derived from the former anyway. For $60 you should be able to pick up a version with both optical and RCA S/PDIF output.
Digidesign's H/W isn't really all that special. There are lots of A/D, D/A boxes out there that can do at least as good a job as the 8810 and its cousins. The DSP Farm is just a bunch of stock DSP chips - the Creamware Pulsar is vastly more interesting for DSP work. No, the golden jewel in Digidesign/Avid's toolchest is the ProTools SOFTWARE. Its slick, but more importantly, it reflects man years of accumulated studio experience. Very little else comes close (though the Ensoniq Paris stuff, and Sonic Foundry's ACID, even if its way more limited, come close).