For those interested, UT2004 is using an open-source codec for voice communication between players. It's also probably the first game to support wideband (16 kHz) communication.
Thanks for pointing that out. I'm tired of people implying that the speed of the signal is the same as the speed of the electrons (average drift speed) in the wire.
Then you have to rewrite the whole translation system. I'd guess it's a *huge* job. At least if you want the resulting CPU to run faster than an x86 (that has been optimized by the Transmeta engineers).
What model does he have? I own a Dell Latitude D600 (Pentium-M 1.6 GHz), and I've been a bit disappointed. I can't get more than 3 hours, even with the CPU running at 600 MHz, the display at low power and the disk spinning down when unused.
Though heap overflows are harder to exploit, I don't think heap memory allocation is really random. Because of virtual memory (each process has its own address space), the same binary will tend to always allocate things at the same address.
Actually, there's a step before "calling the lawyer" or doing it yourself. You can just ask the clause to be removed or made less restrictive. That way, their lawyer can choose the wording, while you still get what you want. Of course, that only applies when the modifications are minor. In my case, I had to sign an NDA for my master project. I only asked that a clause saying "anything I develop" be changed to "anything I develop that's related to my master". They got the change it themselves and it worked.
The only thing I'd like to add also is to *NEVER* trust an employer (or anyone else) that says "sure it's in the contract, but we never really apply it".
Who says I own an iPod (I don't). BTW, the "mysterious" reason you can now change the battery someone (sorry don't have the link) made them enough bad publicity. But the thing is that cunsumer electronics is no longer designed to last more than 3-5 years. That applies to iPod, but other handheld (and non-handheld) devices. They are also not designed to be repaired, but replaced by a new one.
You're assuming an iPod lasts 15 years. That's not the case. Until recently, if your battery died, you had to buy a new iPod. Consider that the rest isn't much more reliable and you'll see how they can keep selling iPods even when everybody has one.
Note sure I like the idea, but I think it would really be more secure in one case: worms. That's because they can't really sniff the knocking, unlike a real intruder.
What is so wrong about paying for a resource you are using?
Note that you're already paying for the service by paying your ISP for connection + bandwidth. What's wrong is that it's a completely artificial cost, that will likely cause more problems that it solves (what about mailing lists, spam virus using your machine,...).
Would the advent of the 64-bit Athlon mean that a PowerPC chip could finally be emulated?
Completely unrelated. You could probably write a PowerPC emulator for a 386 if you wanted to. Extra registers may (or may not) make the emulator a bit faster, but that's all. If there's no PPC emulator yet, it's probably because nobody cared to write one.
Well google isn't revealing any non-public information. All that info was already available and google just helped find it. If something is private, it shouldn't be on the web in the first place.
...except the BBC is complicit in the death of Dr. Kelly and the 'sexing up' of the Iraq dossier...
Actually, the BBC was citing Dr. Kelly to criticize the 'sexing up' done by the british government. Then the government revealed the name of Dr. Kelly as the source, leading to pressures on him. So I don't think the BBC is really complicit in his death and it's definitely haven't 'sexed up' the Iraq dossier.
Of course, I still find what they're doing with google questionable at best.
Actually, instead of buying a Centrino laptop, I just bought a Pentium-M laptop, i.e. the CPU is fine, all you have to do is get a non-Intel card with it.
Actually, current DSPs aren't *that* fast. With x86 CPUs that have a *theoretical* performance in the order of 10 gflops, the DSPs have lost ground. Not only that, but they're much more complicated to program. Believe me, I'm doing all kinds of audio processing and if you give me a CPU that's 10x faster, I'll make use of it in a minute. There's still so many things you can't do right now with audio (even more true with video) because it would be too slow.
I read somewhere that on the P4, when an instruction is already in the L1 cache, the pipeline gets shortened. That's because the L1 instruction cache stores pre-decoded instructions (micro-ops). This means that when the instruction is reached again, the decoding (and branch prediction?) steps are already done, shortening the pipeline. When the instruction is not in cache, there's already a big hit anyway. With that in mind, we'll need to see whether the extra pipeline stages in Prescott will still be there when the instruction is in the L1.
There is currently something like that. When the processor has to guess the branch with no prior data, it'll assume backward branches (loops) are taken and forward branches aren't. With this in mind, I think you can control the prediction behaviour by inverting the "then" and the "else".
Slashdot Mirror
Microsoft is rumored to prepare a lawsuit regarding misappropriation of Win2k source code in Linux kernel version 2.0.
SCO UNIX(R) has a Committed, Well-Defined Roadmap
Translation: We hit the wall within a year.
AFAIK, UT2004 still uses Vorbis for music. Speex is only for low bit-rate (4.8 kbps narrowband, 9.6 kbps wideband) voice communication.
For those interested, UT2004 is using an open-source codec for voice communication between players. It's also probably the first game to support wideband (16 kHz) communication.
Thanks for pointing that out. I'm tired of people implying that the speed of the signal is the same as the speed of the electrons (average drift speed) in the wire.
Then you have to rewrite the whole translation system. I'd guess it's a *huge* job. At least if you want the resulting CPU to run faster than an x86 (that has been optimized by the Transmeta engineers).
What model does he have? I own a Dell Latitude D600 (Pentium-M 1.6 GHz), and I've been a bit disappointed. I can't get more than 3 hours, even with the CPU running at 600 MHz, the display at low power and the disk spinning down when unused.
Though heap overflows are harder to exploit, I don't think heap memory allocation is really random. Because of virtual memory (each process has its own address space), the same binary will tend to always allocate things at the same address.
Actually, there's a step before "calling the lawyer" or doing it yourself. You can just ask the clause to be removed or made less restrictive. That way, their lawyer can choose the wording, while you still get what you want. Of course, that only applies when the modifications are minor. In my case, I had to sign an NDA for my master project. I only asked that a clause saying "anything I develop" be changed to "anything I develop that's related to my master". They got the change it themselves and it worked.
The only thing I'd like to add also is to *NEVER* trust an employer (or anyone else) that says "sure it's in the contract, but we never really apply it".
Who says I own an iPod (I don't). BTW, the "mysterious" reason you can now change the battery someone (sorry don't have the link) made them enough bad publicity. But the thing is that cunsumer electronics is no longer designed to last more than 3-5 years. That applies to iPod, but other handheld (and non-handheld) devices. They are also not designed to be repaired, but replaced by a new one.
You're assuming an iPod lasts 15 years. That's not the case. Until recently, if your battery died, you had to buy a new iPod. Consider that the rest isn't much more reliable and you'll see how they can keep selling iPods even when everybody has one.
Note sure I like the idea, but I think it would really be more secure in one case: worms. That's because they can't really sniff the knocking, unlike a real intruder.
What is so wrong about paying for a resource you are using?
...).
Note that you're already paying for the service by paying your ISP for connection + bandwidth. What's wrong is that it's a completely artificial cost, that will likely cause more problems that it solves (what about mailing lists, spam virus using your machine,
Well, if it continues to work and spammers eventually go out of business, then it also solves their problem.
Well, that's about the only difference. Anyway, what I mostly meant is that Fedora stability is that of a release, not a beta/community release.
Fedora is in fact the same as the Mandrake Cooker
No. Fedora is like Mandrake releases, though I think it also has a cooker-style release.
Would the advent of the 64-bit Athlon mean that a PowerPC chip could finally be emulated?
Completely unrelated. You could probably write a PowerPC emulator for a 386 if you wanted to. Extra registers may (or may not) make the emulator a bit faster, but that's all. If there's no PPC emulator yet, it's probably because nobody cared to write one.
Voters should at least be able to what got printed. Otherwise a paper receipt is useless, since the voter says X and the machine prints Y.
Well google isn't revealing any non-public information. All that info was already available and google just helped find it. If something is private, it shouldn't be on the web in the first place.
...except the BBC is complicit in the death of Dr. Kelly and the 'sexing up' of the Iraq dossier...
Actually, the BBC was citing Dr. Kelly to criticize the 'sexing up' done by the british government. Then the government revealed the name of Dr. Kelly as the source, leading to pressures on him. So I don't think the BBC is really complicit in his death and it's definitely haven't 'sexed up' the Iraq dossier.
Of course, I still find what they're doing with google questionable at best.
Actually, instead of buying a Centrino laptop, I just bought a Pentium-M laptop, i.e. the CPU is fine, all you have to do is get a non-Intel card with it.
Actually, current DSPs aren't *that* fast. With x86 CPUs that have a *theoretical* performance in the order of 10 gflops, the DSPs have lost ground. Not only that, but they're much more complicated to program. Believe me, I'm doing all kinds of audio processing and if you give me a CPU that's 10x faster, I'll make use of it in a minute. There's still so many things you can't do right now with audio (even more true with video) because it would be too slow.
I read somewhere that on the P4, when an instruction is already in the L1 cache, the pipeline gets shortened. That's because the L1 instruction cache stores pre-decoded instructions (micro-ops). This means that when the instruction is reached again, the decoding (and branch prediction?) steps are already done, shortening the pipeline. When the instruction is not in cache, there's already a big hit anyway. With that in mind, we'll need to see whether the extra pipeline stages in Prescott will still be there when the instruction is in the L1.
There is currently something like that. When the processor has to guess the branch with no prior data, it'll assume backward branches (loops) are taken and forward branches aren't. With this in mind, I think you can control the prediction behaviour by inverting the "then" and the "else".
Not that sure. Depends on what they bought exactly, who is the real UNIX copyright owner and things like that.