"Just need to solve hills where the driver may need throttle and brake simultaneously to start moving, and it should work."
That problem has already been solved, too. Some manual transmission cars have a "hill start assist" that hold the brakes for a couple seconds to allow easy starts up hills, without rolling backwards. The brakes are released the instant any forward force is detected. It makes starting uphill virtually identical to starting on a level surface. Seems like that should be applicable to auto transmissions as well, which can also roll back a bit before the torque converter locks up.
I like the idea of having a purely mechanical way of removing power to the wheels, which can be done instantaneously in a panic; and that is why I don't trust cars with fewer than 3 pedals!
Brake fade is what happens when the brakes get overheated, they become less effective.
However, what happens when the engine is at wide open throttle is the same thing that happens when the engine is off: you lose vacuum assist. You'll have enough for maybe 2 pumps of the pedal and that's it. Once your vacuum assist is gone, you're relying 100% on the pressure of your foot on the brake via the hydraulic system to stop the car. If you've ever tried to use the brake pedal when coasting with the engine off, you know how hard that is.
So if you are ever in a "unintended acceleration" situation, push the brake down as hard as you can and do NOT let it back up. You will probably destroy your brakes in the process but that's better than the alternative.
The whole move to electronics is somewhat disconcerting. Computer software will always have bugs, and modern cars have computer software that controls the throttle, and the transmission shifter. Always make sure you know how to shift into neutral in a panic. On my car, it's easy: just push down the clutch pedal.
If you listen to people who don't do tech work talk about techies, you'll quickly realize that a lot of them do in fact put techies on roughly the same level as mechanics or bricklayers.
Except they have no problems asking tech people to do free work.
"Oh, you're a bricklayer? Hey, can you stop by sometime and replace the bricks on my front sidewalk? I'll give you a beer...."
That's not the right CPU... Apple's site states that the system uses dual 3.0GHz Clovertown processors, than means they must be a Xeon X5365, which is the first quad-core Xeon by Intel that runs at 3.0Ghz.
Funny thing is, that particular CPU doesn't ship until Q3 of this year.
So is Apple overclocking the 2.66GHz part or what? Or did Apple get some super-secret exclusive deal with Intel? Anybody have firm details on the exact chip used in the new octo-mac?
The rear-facing SRB view is more interesting in that you can see the ground pull away. And the camera stays above the water once the booster lands (they are bottom-heavy so the top part stays out of the water).
On this news, I have changed my mind... I think it's better to buy that Escalade, and reprogram the fuel mixture to run rich all the time. For the good of the planet.
From the apple store... which is very slow right now, it took me about 15 minutes to find out this info:
The new mini uses DDR2 SO-DIMM's. Must be installed in pairs, comes with 2x256 by default but is upgradeable to 2x1024. $188 to upgrade from 512MB to 2GB, which is slightly more than Newegg pricing when you consider you don't get any credit for the original 512-- but still, nowhere near as bad as the old ripoff memory pricing.
5400rpm SATA drives-- but you can upgrade to a 120GB drive for another $118 vs. the standard 80GB
64MB **Shared video memory.** Nuts. Intel GMA950 graphics chipset. This chipset is better suited for home A/V use though.
Wifi carries digital data. This means it is comprised of 1's and 0's. So instead of being gentle, flowing waves, the radio signals are sharp and pointy (the 1's more than the 0's anyway)-- so it's obvious that they would cause more cell damage.
Apple contracts out construction of their laptop, so in a way you're correct that it goes down the same assembly line. The difference is that they contract out to highly respected manufactures.
In fact, the MacBook Pro is actually produced by Asus Technologies, who have been in the x86 business for many years, and have been well known for producing high quality notebooks (albeit not very mainstream) for some time.
Oddly enough, Asus will also be producing Lamborghini branded notebooks. I don't understand the reasons for wanting an auto manufacturer emblem on a notebook (I mean, is there anything really "lambo" about it?) but it's still very cool, especially when you can get OSX running on it.
Acer has their Ferrari notebooks, and I don't know if they are any better in quality than the regular Acer notebooks, but... I think I'd rather have a Lamborghini than a Ferrari.
Our coding standard is a little like this: Write clean code that can be easily understood by reading it. That is, good variable and function names, try not to make any absurdly complicated statements, and have your comments explain the logic of the operations. As for style, try to stick with the style that the original author started with. And finally, all people who use Hungarian notation are locked in the basement and given menial tasks until they repent their sinful ways
Well that's a good start, with good intentions, but you need to have a standard definition of what constitutes good function names and good variable names. If you have 5 different programmers on a project, you'll have 5 different opinions on what good names are.
Make sure your coding standards are DOCUMENTED.
If it's a java project, the best source would be Sun's java coding standards. A very useful tool for this is Checkstyle. You can decide which rules to enforce (some of the ones enabled by default are more annoying than anything) but if you take the time to get your code to where Checkstyle likes it, you'd be amazed how easy it is for humans to read.
As for my department, we use CVS for version control. Every time code is checked into CVS, it is formatted by Jalopy. So, it'll look nice and neat the next time it's checked out. Also, we have a script that does nightly builds, and then emails the result to everybody on the team. So if you checked in something that breaks the build, everybody knows about it the following morning.:-)
We have a regular release schedule. All work is done on the main CVS branch, but when it's time for a code freeze, the new version is branched off and tagged in CVS. During QA testing, bugs are fixed in the branch and the mainline. New features are only added to the mainline.
When we are ready to deploy, we tag the release in CVS. The deployment script checks the tag out of CVS, builds it, and packages everything up into the relevant.ear files which Operations can then take.
This is all a very strict process, but things rarely fall through the cracks this way. If you don't have any processes in place now, it's best to implement them a step at a time. Get everybody used to working with CVS or some other version control, get them used to the notion of tagging and branching, and make sure there's actually a document detailing whatever processes you have.
And lastly, have code reviews every week or two. Review a different person's code each time and make sure everybody on the team is allowed to have input. If you're not at the coding stage yet, have design reviews.
Yeah, I'd agree mostly with what you say. The Brass samples are good except for the trumpet sounds. I like the trombones and tuba.
The brass is good when you need a mellow sound. I particularly like the french horn. GPO just isn't good at aggressive brass though. GPO Advanced should fix that since it's integrating ProjectSAM brass samples.
However, the system requirements for GPO are insane. I can't get more than 5 or 6 samples playing on my 2.5 Gigahertz Athlon. And that is when it's being slaved from my Mac.
Athlon64? I have a 3000+ (1.8GHz) and it's capable of playing well more than 5 or 6 samples at once. Typical I have 3 or 4 instances of GPO loaded, with 4 or 5 instruments on each, and no problems. I tend to run out of memory when loading samples before running out of CPU. My scores generally have 6-8 GPO instances total (in addition to 1 or 2 stormdrum instances or whatever other plugins I'm using), but I freeze the tracks I'm not working on to free up CPU and memory. I need to upgrade from 1GB to 2GB... the disk streaming extension eats more CPU cycles. Make sure you are using your FX modules on a bus rather than individual tracks, since running multiple FX instances will kill your CPU. If you're using Garritan Ambience, you can set the quality to low to save some CPU cycles until you're ready for the final mixdown.
I've honestly never heard of someone making that mistake in my life. Are you using some sort of speech to text programme? If not, I think you have an as yet undiscovered form of dyslexia--you could get a disease named after you, CausticPuppy!
Eye hope your write, halving a disease named after me wood bee swell.
Even for $200, Garritan Personal Orchestra has some of the lowest quality I've ever heard in this kind of a commercial sample library. Honestly, how does that company get this much positive accolade? Sure, some of their solo instruments (violins, piano, harps) sound pretty good, but many of the other instruments do not (the brasses are atrocious!!!).
The reason they get so much positive accolade is because the strings ARE very good. Most of the woodwinds are very good as well, as is the percussion, except for a couple minor glitches. It's also very playable, you can get much more expression in realtime due to GPO's innovative (non-standard) use of MIDI controllers. There's nothing that comes close for $200. I agree that the brass lacks punch, that's my biggest complaint, and the symphonic grand marimba has some hiss in the samples.
However, GPO:Advanced coming out later this year will feature ProjectSAM brass samples which, to my ear, are better than EastWest's. Another reason I like GPO is that Gary Garritan is very involved and helpful in the community. For example he corrected the vibraphone pedal behavior in the latest GPO patch because I brought it up on his discussion board earlier this year.
EWQLSO is good as long as you have the platinum version, the lower priced versions don't have nearly enough articulations to be useful, and only the platinum version has all 3 mic placings. EWQLSO is good for those big, epic sounding scores, but I think GPO is better for more intimate, expressive music, particularly with the strings.
My first choice, when I was looking for symphonic libraries, was going to be GigaStudio. It includes some VSL libraries, but that package has the most insane system requirements of all of them. You pretty much have to run Gigastudio on its own dedicated box.
Samples, yes, you need huge file sizes for it to work well.
Wavetable synthesis, however, you don't.
The idea behind wavetable synthesis is to take short samples- one cycle of the waveform- for different stages of the sound. Loop and manipulate them appropriately.
The reason for this form of synthesis is to conserve memory.
These days, every sampler that I know of is capable of this form of wavetable synthesis. It's standard. There will be an attack sample, a sustain sample (which can be looped), and a release sample, for example. Some instruments work well with this, but others don't. Instruments that are capable of sustaining indefinitely work well for this, although longer loops will generally sound better than shorter loops if done correctly. In a lot of cases there are multiple attack samples to be used for various articulations.
Another trick is to only sample maybe 3 notes per octave and then speed up or slow down the sample playback for the other pitches.
A really good piano sample can't have looped samples because the sound of a single note is constantly changing in subtle ways over the course of its decay. Harmonics come in and out, resonances change within the body of the piano itself, etc. You can get by with shorter looped samples (wavetable), and have a somewhat decent sounding piano (like Roland's) which is fine for using in a mix of other instruments. But this has its limits. The [i]best[/i] sampled pianos have all 88 keys sampled individually, at multiple velocities, for the full decay of the note. This can add up to a few gigs, just for the piano alone if the samples are recorded at 24/96. The sampled piano that I use is "only" about 500 megabytes, which is small enough to store in RAM rather than stream from disk like many samplers do, and it beats any other looped wavetable piano that I've heard.
It seems like MIDI discussion come up every so often on Slashdot.
First, a couple things to get cleared up:
MIDI is just a serial protocol, nothing more. It's been around since the early 80's. The protocol defines 128 MIDI notes, on 16 channels, and 128 controllers that have values of 0 - 127. That's basically it, along with a few other things like channel change and bank change messages. MIDI itself does not define any instruments, because MIDI is used to control non-instrument devices like effects boxes too.
Now, when most normal computer people think of MIDI, what they are actually thinking of is GM or "General MIDI." GM defines a standard set of instruments, for example instrument 1 is always a piano, instrument 74 is always a flute, etc. It's up to the hardware or software to actually implement these instruments, usually done with wavetable samples. The idea is that a MIDI file played through any "GM compatible" device will sound roughly the same on any other GM device, although the quality of the samples varies widely. Roland's GS is an extension of GM.
GM used to be used for games primarily (think Doom1 and Doom2!) but has fallen by the wayside now that everybody is using full audio tracks for music.
But most of the music created for video games these stays was still created using MIDI! The file format is specific to the studio application, but MIDI is still used internally to communicate with various synthesizers and samplers including virtual synths that run on the local machine. So if you were to get the original data files, you would need to also have the sample libraries-- which are VERY high quality, and can cost several thousand dollars. And you need to be using software that works with these libraries, which rules out free/OSS software-- you're gonna NEED something like Sonar, Logic Audio, etc.
Almost all video games and most TV shows that have symphonic music are actually MIDI based, but use enormous sample libraries like EastWest symphony orchestra. In fact I believe that the Return to Castle Wolfenstein soundtrack was created mostly with that sample library.
Other examples, the "fire baby" sequence in The Incredibles is created with Voices of the Apocalypse so even realistic choirs can be created using MIDI.
You don't have to spend THAT much though-- the libraries I use the most are Storm Drum and Garritan Personal Orchestra, both of which are very affordable but good enough that they are often used in hollywood. All of these are plugins that can be used in many different software packages on both OSX and Windows, but not linux that I'm aware of.
So, nowadays MIDI is still an integral part of even the most modern studios, but General MIDI is nowhere in sight. GM still has a place in cell phone ringers.
Evolution is a phenomenon. It can be observed easily, even in something as trivial and obvious as dog breeding.
I just wanted to expand on this point before somebody jumps on and says that dog breeding is NOT evolution because it's controlled by humans....
Evolution is not only driven by natural selection. Evolution is simply the change in genome over time. Natural selection has been the primary mechanism by which evolution occurs. The rate of evolution by natural selection depends on the environment, which includes predators. However, human selection of traits, and specifically breeding for them (as in dog breeding) causes evolution to happen at a much faster rate. Domestic dogs have traits nowadays that are desireable to humans, but would not allow a dog to survive in the wild, so most dogs are completely dependent on us for survival. Natural selection is not the primary driver of dog evolution any more. If humans keep any two dog breeds separated for long enough (many generations), eventually they will become genetically incompatible as the two paths of genetic mutations diverge without any mixing of the gene pools. This is how new species are created... what was one species eventually branches into two, or more, but some of the branches sometimes die off.
On a slightly related note, evolution by natural selection for humans has just about stopped, because technology can compensate for genetic traits that are detrimental for survival. For example, asthma. It's very common in humans today because we have found effective ways to treat it. But how many wild animals have asthma? How many are even genetic carriers of the disorder?
Please demonstrate that Intelligent Design is wrong. I'm curious.
More to the point, ID is wrong because it is not a valid scientific theory. It is not based on the scientific method, and it is not falsifiable (although I have seen some interesting claims of falsifiability from ID proponents, which can be reduced to nothing more than circular logic).
A scientific theory is a logical model, an abstraction of reality, a collection of knowledge about how processes work. Intelligent Design is none of those things. It is argument that essentially boils down to "Darwinism can't be right, therefore we win by default." Intelligent design is nothing more than a flawed argument from people who refuse to accept the fact that they are genetically related to monkeys.
The most aggravating argument of all coming from ID proponents is that evolution is "Just a theory." Clearly they don't even understand what a theory is. They might as well be saying that electromagnetism is "just a theory." Otherwise, why would we have something called electromagnetic theory?
As we see more and understand more of how our world works, that means (logically) that god is less and less powerful. Right now (according to ID), god is directly responsible for "X" amount of the world around us, where "X" is everything we don't understand, or haven't observed directly. As we are constantly learning, that means that god is less and less responsible for the world around us, up until the point where we understand everything, and hence god (to quote Douglas Adams) disappears in a puff of logic.
FYI, this is a common argument from the creationists, known as the God of the Gaps.
I agree C# is better than Java but I don't like both, I know you don't need to cover your traces when dealing with memory and dynamic allocation, but you are sacrificing quite a bit of processor time and system memory in the process. Garbage collectors and every other tool meant to make programming easy costs heavily in time and hardware. Most probably you are going to say that now days computers are fast enough to pay the toll but that is true only for certain cases. Think about AI, speech recognition (even an OS) and many other processor demanding tasks, you would like to use wisely every tick of the clock... As always the method to solve the problem is in the problem itself, but I personally would like to spend my days developing something more significant than a monthly savings report for McDonalds so I'm glad to take my time, use my brain and continue developing on C++.
So, you use C++ for apps where speed and efficiency are important, while you use C# or Java for apps where maintainability and development time are important. Nobody said you have to use one tool for everything. C# and Java are particularly well suited to server applications, web services, etc. and you can program a LOT of functionality in a relatively small amount of time. We use Java for most things (primarily server-side apps) and they are quite a bit more significant than savings reports.
Slashdot Mirror
"Just need to solve hills where the driver may need throttle and brake simultaneously to start moving, and it should work."
That problem has already been solved, too. Some manual transmission cars have a "hill start assist" that hold the brakes for a couple seconds to allow easy starts up hills, without rolling backwards. The brakes are released the instant any forward force is detected. It makes starting uphill virtually identical to starting on a level surface. Seems like that should be applicable to auto transmissions as well, which can also roll back a bit before the torque converter locks up.
I like the idea of having a purely mechanical way of removing power to the wheels, which can be done instantaneously in a panic; and that is why I don't trust cars with fewer than 3 pedals!
Brake fade is what happens when the brakes get overheated, they become less effective.
However, what happens when the engine is at wide open throttle is the same thing that happens when the engine is off: you lose vacuum assist. You'll have enough for maybe 2 pumps of the pedal and that's it. Once your vacuum assist is gone, you're relying 100% on the pressure of your foot on the brake via the hydraulic system to stop the car. If you've ever tried to use the brake pedal when coasting with the engine off, you know how hard that is.
So if you are ever in a "unintended acceleration" situation, push the brake down as hard as you can and do NOT let it back up. You will probably destroy your brakes in the process but that's better than the alternative.
The whole move to electronics is somewhat disconcerting. Computer software will always have bugs, and modern cars have computer software that controls the throttle, and the transmission shifter. Always make sure you know how to shift into neutral in a panic. On my car, it's easy: just push down the clutch pedal.
If you listen to people who don't do tech work talk about techies, you'll quickly realize that a lot of them do in fact put techies on roughly the same level as mechanics or bricklayers.
Except they have no problems asking tech people to do free work.
"Oh, you're a bricklayer? Hey, can you stop by sometime and replace the bricks on my front sidewalk? I'll give you a beer...."
Affordable health care is not insane.
It's just that insane health care "reform" is... insane.
Apple still has a "spare" Steve available... though he's currently only a part-time employee and probably has no interest in becoming CEO.
But maybe he (Woz) can deliver keynotes? The fanbase will love him.
That's not the right CPU... Apple's site states that the system uses dual 3.0GHz Clovertown processors, than means they must be a Xeon X5365, which is the first quad-core Xeon by Intel that runs at 3.0Ghz.
Funny thing is, that particular CPU doesn't ship until Q3 of this year.
So is Apple overclocking the 2.66GHz part or what? Or did Apple get some super-secret exclusive deal with Intel?
Anybody have firm details on the exact chip used in the new octo-mac?
the link is on this page...
d ia/index.html
http://www.nasa.gov/mission_pages/shuttle/multime
The rear-facing SRB view is more interesting in that you can see the ground pull away. And the camera stays above the water once the booster lands (they are bottom-heavy so the top part stays out of the water).
On this news, I have changed my mind... I think it's better to buy that Escalade, and reprogram the fuel mixture to run rich all the time. For the good of the planet.
From the apple store... which is very slow right now, it took me about 15 minutes to find out this info:
The new mini uses DDR2 SO-DIMM's. Must be installed in pairs, comes with 2x256 by default but is upgradeable to 2x1024. $188 to upgrade from 512MB to 2GB, which is slightly more than Newegg pricing when you consider you don't get any credit for the original 512-- but still, nowhere near as bad as the old ripoff memory pricing.
5400rpm SATA drives-- but you can upgrade to a 120GB drive for another $118 vs. the standard 80GB
64MB **Shared video memory.** Nuts. Intel GMA950 graphics chipset. This chipset is better suited for home A/V use though.
How's wifi different than any other radio signal?
Wifi carries digital data. This means it is comprised of 1's and 0's. So instead of being gentle, flowing waves, the radio signals are sharp and pointy (the 1's more than the 0's anyway)-- so it's obvious that they would cause more cell damage.
Apple contracts out construction of their laptop, so in a way you're correct that it goes down the same assembly line. The difference is that they contract out to highly respected manufactures.
In fact, the MacBook Pro is actually produced by Asus Technologies, who have been in the x86 business for many years, and have been well known for producing high quality notebooks (albeit not very mainstream) for some time.
Oddly enough, Asus will also be producing Lamborghini branded notebooks.
I don't understand the reasons for wanting an auto manufacturer emblem on a notebook (I mean, is there anything really "lambo" about it?) but it's still very cool, especially when you can get OSX running on it.
Acer has their Ferrari notebooks, and I don't know if they are any better in quality than the regular Acer notebooks, but... I think I'd rather have a Lamborghini than a Ferrari.
Did I just go off-topic? Damn.
The skill to write well is a thousand times more valuable today than most people give it credit for.
Did you just end a sentence with a preposition?
1. Your brain uncritically accepts the first information it gets in any new subject area as correct, whether it is or not.
So how do I know whether this statement is itself correct??
The southern icecap on Mars is melting because it is spring there
FYI, a quick google search returns the following results:
Mars global warming
The warming trend appears to be something longer than seasonal.
Our coding standard is a little like this:
:-)
.ear files which Operations can then take.
Write clean code that can be easily understood by reading it. That is, good variable and function names, try not to make any absurdly complicated statements, and have your comments explain the logic of the operations. As for style, try to stick with the style that the original author started with. And finally, all people who use Hungarian notation are locked in the basement and given menial tasks until they repent their sinful ways
Well that's a good start, with good intentions, but you need to have a standard definition of what constitutes good function names and good variable names. If you have 5 different programmers on a project, you'll have 5 different opinions on what good names are.
Make sure your coding standards are DOCUMENTED.
If it's a java project, the best source would be Sun's java coding standards. A very useful tool for this is Checkstyle. You can decide which rules to enforce (some of the ones enabled by default are more annoying than anything) but if you take the time to get your code to where Checkstyle likes it, you'd be amazed how easy it is for humans to read.
As for my department, we use CVS for version control.
Every time code is checked into CVS, it is formatted by Jalopy. So, it'll look nice and neat the next time it's checked out.
Also, we have a script that does nightly builds, and then emails the result to everybody on the team. So if you checked in something that breaks the build, everybody knows about it the following morning.
We have a regular release schedule. All work is done on the main CVS branch, but when it's time for a code freeze, the new version is branched off and tagged in CVS. During QA testing, bugs are fixed in the branch and the mainline. New features are only added to the mainline.
When we are ready to deploy, we tag the release in CVS. The deployment script checks the tag out of CVS, builds it, and packages everything up into the relevant
This is all a very strict process, but things rarely fall through the cracks this way. If you don't have any processes in place now, it's best to implement them a step at a time. Get everybody used to working with CVS or some other version control, get them used to the notion of tagging and branching, and make sure there's actually a document detailing whatever processes you have.
And lastly, have code reviews every week or two. Review a different person's code each time and make sure everybody on the team is allowed to have input. If you're not at the coding stage yet, have design reviews.
Yeah, I'd agree mostly with what you say. The Brass samples are good except for the trumpet sounds. I like the trombones and tuba.
The brass is good when you need a mellow sound. I particularly like the french horn. GPO just isn't good at aggressive brass though. GPO Advanced should fix that since it's integrating ProjectSAM brass samples.
However, the system requirements for GPO are insane. I can't get more than 5 or 6 samples playing on my 2.5 Gigahertz Athlon. And that is when it's being slaved from my Mac.
Athlon64? I have a 3000+ (1.8GHz) and it's capable of playing well more than 5 or 6 samples at once. Typical I have 3 or 4 instances of GPO loaded, with 4 or 5 instruments on each, and no problems. I tend to run out of memory when loading samples before running out of CPU.
My scores generally have 6-8 GPO instances total (in addition to 1 or 2 stormdrum instances or whatever other plugins I'm using), but I freeze the tracks I'm not working on to free up CPU and memory.
I need to upgrade from 1GB to 2GB... the disk streaming extension eats more CPU cycles. Make sure you are using your FX modules on a bus rather than individual tracks, since running multiple FX instances will kill your CPU. If you're using Garritan Ambience, you can set the quality to low to save some CPU cycles until you're ready for the final mixdown.
So thanks for clearing all that up but asfofar I don't think it was necessary.
It's primarily for the benefit of others who are reading the thread.
I've honestly never heard of someone making that mistake in my life. Are you using some sort of speech to text programme? If not, I think you have an as yet undiscovered form of dyslexia--you could get a disease named after you, CausticPuppy!
Eye hope your write, halving a disease named after me wood bee swell.
Even for $200, Garritan Personal Orchestra has some of the lowest quality I've ever heard in this kind of a commercial sample library. Honestly, how does that company get this much positive accolade? Sure, some of their solo instruments (violins, piano, harps) sound pretty good, but many of the other instruments do not (the brasses are atrocious!!!).
The reason they get so much positive accolade is because the strings ARE very good. Most of the woodwinds are very good as well, as is the percussion, except for a couple minor glitches.
It's also very playable, you can get much more expression in realtime due to GPO's innovative (non-standard) use of MIDI controllers. There's nothing that comes close for $200.
I agree that the brass lacks punch, that's my biggest complaint, and the symphonic grand marimba has some hiss in the samples.
However, GPO:Advanced coming out later this year will feature ProjectSAM brass samples which, to my ear, are better than EastWest's. Another reason I like GPO is that Gary Garritan is very involved and helpful in the community. For example he corrected the vibraphone pedal behavior in the latest GPO patch because I brought it up on his discussion board earlier this year.
EWQLSO is good as long as you have the platinum version, the lower priced versions don't have nearly enough articulations to be useful, and only the platinum version has all 3 mic placings. EWQLSO is good for those big, epic sounding scores, but I think GPO is better for more intimate, expressive music, particularly with the strings.
My first choice, when I was looking for symphonic libraries, was going to be GigaStudio. It includes some VSL libraries, but that package has the most insane system requirements of all of them. You pretty much have to run Gigastudio on its own dedicated box.
Samples, yes, you need huge file sizes for it to work well.
Wavetable synthesis, however, you don't.
The idea behind wavetable synthesis is to take short samples- one cycle of the waveform- for different stages of the sound. Loop and manipulate them appropriately.
The reason for this form of synthesis is to conserve memory.
These days, every sampler that I know of is capable of this form of wavetable synthesis. It's standard. There will be an attack sample, a sustain sample (which can be looped), and a release sample, for example. Some instruments work well with this, but others don't. Instruments that are capable of sustaining indefinitely work well for this, although longer loops will generally sound better than shorter loops if done correctly.
In a lot of cases there are multiple attack samples to be used for various articulations.
Another trick is to only sample maybe 3 notes per octave and then speed up or slow down the sample playback for the other pitches.
A really good piano sample can't have looped samples because the sound of a single note is constantly changing in subtle ways over the course of its decay. Harmonics come in and out, resonances change within the body of the piano itself, etc. You can get by with shorter looped samples (wavetable), and have a somewhat decent sounding piano (like Roland's) which is fine for using in a mix of other instruments.
But this has its limits. The [i]best[/i] sampled pianos have all 88 keys sampled individually, at multiple velocities, for the full decay of the note. This can add up to a few gigs, just for the piano alone if the samples are recorded at 24/96. The sampled piano that I use is "only" about 500 megabytes, which is small enough to store in RAM rather than stream from disk like many samplers do, and it beats any other looped wavetable piano that I've heard.
It seems like MIDI discussion come up every so often on Slashdot.
First, a couple things to get cleared up:
MIDI is just a serial protocol, nothing more. It's been around since the early 80's. The protocol defines 128 MIDI notes, on 16 channels, and 128 controllers that have values of 0 - 127. That's basically it, along with a few other things like channel change and bank change messages. MIDI itself does not define any instruments, because MIDI is used to control non-instrument devices like effects boxes too.
Now, when most normal computer people think of MIDI, what they are actually thinking of is GM or "General MIDI." GM defines a standard set of instruments, for example instrument 1 is always a piano, instrument 74 is always a flute, etc.
It's up to the hardware or software to actually implement these instruments, usually done with wavetable samples. The idea is that a MIDI file played through any "GM compatible" device will sound roughly the same on any other GM device, although the quality of the samples varies widely. Roland's GS is an extension of GM.
GM used to be used for games primarily (think Doom1 and Doom2!) but has fallen by the wayside now that everybody is using full audio tracks for music.
But most of the music created for video games these stays was still created using MIDI! The file format is specific to the studio application, but MIDI is still used internally to communicate with various synthesizers and samplers including virtual synths that run on the local machine.
So if you were to get the original data files, you would need to also have the sample libraries-- which are VERY high quality, and can cost several thousand dollars. And you need to be using software that works with these libraries, which rules out free/OSS software-- you're gonna NEED something like Sonar, Logic Audio, etc.
Almost all video games and most TV shows that have symphonic music are actually MIDI based, but use enormous sample libraries like EastWest symphony orchestra. In fact I believe that the Return to Castle Wolfenstein soundtrack was created mostly with that sample library.
Other examples, the "fire baby" sequence in The Incredibles is created with Voices of the Apocalypse so even realistic choirs can be created using MIDI.
You don't have to spend THAT much though-- the libraries I use the most are Storm Drum and Garritan Personal Orchestra, both of which are very affordable but good enough that they are often used in hollywood. All of these are plugins that can be used in many different software packages on both OSX and Windows, but not linux that I'm aware of.
So, nowadays MIDI is still an integral part of even the most modern studios, but General MIDI is nowhere in sight. GM still has a place in cell phone ringers.
Evolution is a phenomenon. It can be observed easily, even in something as trivial and obvious as dog breeding.
I just wanted to expand on this point before somebody jumps on and says that dog breeding is NOT evolution because it's controlled by humans....
Evolution is not only driven by natural selection. Evolution is simply the change in genome over time.
Natural selection has been the primary mechanism by which evolution occurs. The rate of evolution by natural selection depends on the environment, which includes predators.
However, human selection of traits, and specifically breeding for them (as in dog breeding) causes evolution to happen at a much faster rate. Domestic dogs have traits nowadays that are desireable to humans, but would not allow a dog to survive in the wild, so most dogs are completely dependent on us for survival. Natural selection is not the primary driver of dog evolution any more.
If humans keep any two dog breeds separated for long enough (many generations), eventually they will become genetically incompatible as the two paths of genetic mutations diverge without any mixing of the gene pools. This is how new species are created... what was one species eventually branches into two, or more, but some of the branches sometimes die off.
On a slightly related note, evolution by natural selection for humans has just about stopped, because technology can compensate for genetic traits that are detrimental for survival. For example, asthma. It's very common in humans today because we have found effective ways to treat it. But how many wild animals have asthma? How many are even genetic carriers of the disorder?
Please demonstrate that Intelligent Design is wrong. I'm curious.
More to the point, ID is wrong because it is not a valid scientific theory. It is not based on the scientific method, and it is not falsifiable (although I have seen some interesting claims of falsifiability from ID proponents, which can be reduced to nothing more than circular logic).
A scientific theory is a logical model, an abstraction of reality, a collection of knowledge about how processes work.
Intelligent Design is none of those things. It is argument that essentially boils down to "Darwinism can't be right, therefore we win by default." Intelligent design is nothing more than a flawed argument from people who refuse to accept the fact that they are genetically related to monkeys.
The most aggravating argument of all coming from ID proponents is that evolution is "Just a theory." Clearly they don't even understand what a theory is. They might as well be saying that electromagnetism is "just a theory." Otherwise, why would we have something called electromagnetic theory?
As we see more and understand more of how our world works, that means (logically) that god is less and less powerful. Right now (according to ID), god is directly responsible for "X" amount of the world around us, where "X" is everything we don't understand, or haven't observed directly. As we are constantly learning, that means that god is less and less responsible for the world around us, up until the point where we understand everything, and hence god (to quote Douglas Adams) disappears in a puff of logic.
FYI, this is a common argument from the creationists, known as the God of the Gaps.
I agree C# is better than Java but I don't like both, I know you don't need to cover your traces when dealing with memory and dynamic allocation, but you are sacrificing quite a bit of processor time and system memory in the process. Garbage collectors and every other tool meant to make programming easy costs heavily in time and hardware. Most probably you are going to say that now days computers are fast enough to pay the toll but that is true only for certain cases. Think about AI, speech recognition (even an OS) and many other processor demanding tasks, you would like to use wisely every tick of the clock... As always the method to solve the problem is in the problem itself, but I personally would like to spend my days developing something more significant than a monthly savings report for McDonalds so I'm glad to take my time, use my brain and continue developing on C++.
So, you use C++ for apps where speed and efficiency are important, while you use C# or Java for apps where maintainability and development time are important. Nobody said you have to use one tool for everything. C# and Java are particularly well suited to server applications, web services, etc. and you can program a LOT of functionality in a relatively small amount of time.
We use Java for most things (primarily server-side apps) and they are quite a bit more significant than savings reports.