Most of your open source software that you'll find relating to this issue will be based on adaptive filtering, which when considering time delays over 1 second, this becomes infeasible, even at a 8000 Hz sampling (I imagine we have a few channels to process on each end). As a result, you're going to have to either buy some more hardware or write it yourself. Or you could find some geek willing to write a small package that does the following:
Generate an impulse test signal on one end, and using the mic/speaker set up at all the other points of the VTC, measure the return. This will give you a bunch of ranges in time delays that you're interested in. You might want to do it for each other destination seperately. Then, to be more accurate, send out some sort of PSM sequence (this is only slightly more complicated than calling random() ). Correlate this signal against itself for a while (to account for any variations) to get a precise range of delays.
You might be suprised to find the delays are extremely constant, if not dependant on the destination. A simple delay, invert, attenuate and mix will suffice at this point. Otherwise, you need to use an adaptive algorithm, but since you know the delay (roughly), you can just buffer up a bunch of data but only process a small window of it.
That's a brilliant idea and I'm mad I never thought of it. I had a different idea (but related) involving viewing tar files as mounted filesystems. But this is just a cool idea for that sort of thing.
If you're using a computer to post the message... then you're seeing particles tunnel. Many junctions that form the gates in the chips in your computer work by tunneling electrons.
Maybe not. Win3.11 never evolved into XP. NT 3.1 was born from ex-VMS engineers, which begat 3.5, which begat 4.0, which begat 2000 (5.0), which begat XP (5.1). At 5.0, some of the Win9x features were rolled in for better gaming and hardware support.
I have a dual PII Xeon w/ GeForce3 Ti200 OC and I have no problems whatsoever in Linux and Win2k. I'm using the latest reference drivers as of about 3 months ago. What's up with your system?
While SMP isn't useful for Q3A, Tribes2, etc. it is nice to play MP3s, etc. while playing without a performance hit.
Correct. However, it is a good [back]ronym, and even it's misinterpreted form is useful./usr/bin are binaries for users;/usr/lib are the corresponding libraries, and so on. I think/home is a better prefix for home directories in my opinion. It is/usr/local that definitely needs to be dropped, and I'm glad that I see it used less and less [I mean/usr/local/bin? Isn't that supposed to be/bin if it's local?].
Some enterprising system administrators have gone that extra step and liberated/usr to a network share, not including myself. The idea is that/{bin,etc,lib} are supposed to be enough of a Unix system to get/usr off the network anyway.
The problem with the better physics/better AI problem is not that we don't have enough power. The problem is programmers don't have the time to implement all those things they could implement with all that power. Just having the extra horsepower doesn't make those things happen by themselves. All they know what to do is push all the buttons on the 3d graphic controls in the newest SDK to the max (Where else can we use a cubic environment map...). Compare a game that pays great attention to detail with one that is thrown together quickly.
Sort of. Except that since the Pentium Pro core completely rewrites and rearranges each instruction into microops, you never get to write code to exploit the three pipes. The original pentium directly translated each instruction for each execution unit, hence you could order each instruction for parallel executions. With the Pentium Pro, you can't really guarantee anything since the instruction rewriter is the one doing all the ordering for parallelism. You can actually make it's job harder by trying to help it out. P5 optimized code tends to stall out the P6 series.
It (HT Xeon) does use two seperate TLBs. Not only that, each core has it's own register set including segments and TSS, LDT, the whole shebang. In fact, an issue mentioned earlier in the discussion and the article was that the two threads would sometimes conflict in usage of the cache (wanting dispersed regions of memory), thus increasing the cache miss rate than if they were two seperate cores. Furthermore, stock versions of Windows were used. They are not aware of the SMT nature of the processor, they just see two CPUs, and act accordingly. How could they know not to use threads that use different memory regions when it's never previously been an issue?
By that point, the moon's orbit would have decayed so either a) we'd have rings b) we'd be squished by the satellite. While we'd have the same revolution period as the moon's, we phyiscally wouldn't be in lock with anything.
The model is still the same, this is true. A lot of shell scripts and code will still work on a new Linux system. If you must ask why, because it was pretty good to begin with. Can anyone think of a good reason to get rid of the BSD sockets API?
On the other hand everything under the hood has been changed and is decidedly not backwards compatible. The jump from Linux 1.0 to 2.0, the jump from libc5 to GLIBC 2.x, these are all breaking points. All new driver architectures, APIs, even executable formats. I don't think Linux is being held back. Having old API compatibility is not a bad thing, but even then, they often must say: This is it, this old stuff is deprecated, we're not supporting it anymore. For example, the RAID system is neither forwards nor backwards compatible in any way between 2.2 and 2.4. Stuff like that.
Plus, there's a difference between designing an OS to have legacy support and actually emulating the legacy system. Forcing someone to recompile or edit the aging code slightly can give someone freedom to implement the compatibility layer any way they see fit, instead of having to keep it "in place" and organize new features "around it".
I won't go into too much detail but...
on
NY Times on Anime
·
· Score: 1
for the most part it was intended to be a psychological treatise from the beginning. The hook (and the animation budget) was the awesome mecha/end-of-the-world storyline. I've heard accusations that this was director Hideaki Anno's way of putting his own emotions and feelings into a show. Also, I heard that the end was not the intended one... it was shorter, and in retaliation it was given a shot of LSD by the production team due to political/financial factors. It gave Anno a chance to rant back at Gainax [replace "Shinji" with "Hideaki" and watch those episodes again, it starts to make sense! (-: ]
I for one really enjoyed it... wonderful characters, animation, action, drama... it was somewhat mysterious at times, frustrating at others, but in retrospect it makes sense. You like it more two weeks _after_ seeing the whole thing.
Check out some of the lists that people post here for stuff to see... cover art and blurbs on the back don't really give you an idea of what you're in for.
If you can get your hands on "Vandread" or "Martian Successor Nadesico", you'll probably appreciate them. (Mecca in space, nice animation, and funny as hell)
...considering that the Pentium processor is covered (and later iterations of the Pentium line used almost the exact same core).
No. The transition from P5 to P6 (Pentium Pro) is the introduction of an entirely new core.
The P5 core had two instruction pipes (thank god), the U and V pipes. I remember he discussed how you can optimize assembly for this in one of the chapters. Basically, you would schedule instructions that didn't step on each other's feet one after the other. However, the P6 core (Pentium Pro-Pentium III) is quite different. It has a RISC core and a variable (chip dependant) number of ALUs and FPUs. It is nowhere nearly as predictable as the P5, thus optimizing it is a tricky, black art. None of the old optimization tricks work, instead you basically follow guidelines for emitting code that won't break the longish pipeline.
Slashdot Mirror
Most of your open source software that you'll find relating to this issue will be based on adaptive filtering, which when considering time delays over 1 second, this becomes infeasible, even at a 8000 Hz sampling (I imagine we have a few channels to process on each end). As a result, you're going to have to either buy some more hardware or write it yourself. Or you could find some geek willing to write a small package that does the following:
Generate an impulse test signal on one end, and using the mic/speaker set up at all the other points of the VTC, measure the return. This will give you a bunch of ranges in time delays that you're interested in. You might want to do it for each other destination seperately. Then, to be more accurate, send out some sort of PSM sequence (this is only slightly more complicated than calling random() ). Correlate this signal against itself for a while (to account for any variations) to get a precise range of delays.
You might be suprised to find the delays are extremely constant, if not dependant on the destination. A simple delay, invert, attenuate and mix will suffice at this point. Otherwise, you need to use an adaptive algorithm, but since you know the delay (roughly), you can just buffer up a bunch of data but only process a small window of it.
"Two of the ten directors will be ... and Mahiro Maeda (Blue Submarine No. Six). "
Ahhhh.... god damn that was a cool (if not short) series. I can't wait.
It's easier to reform methane than water (less energy in required to get H2 out). Note that CH4 burns when put a match near it but H20 does not.
You're not into THAT kind of Ecchi are you?
:P
That's a brilliant idea and I'm mad I never thought of it. I had a different idea (but related) involving viewing tar files as mounted filesystems. But this is just a cool idea for that sort of thing.
:-D
Gotta remember that one.
LOL
Best strategy ever.
I have to find that car!
(I'm in Reston...)
If you're using a computer to post the message... then you're seeing particles tunnel. Many junctions that form the gates in the chips in your computer work by tunneling electrons.
:-D
Proof enough for me. Also, in my experience, the predictions of Lloyd Schumner Sr. Retired Machinist and A.A.P.B.-Certified Astrologer have been as accurate for me as any horoscope in the paper.
Maybe not. Win3.11 never evolved into XP. NT 3.1 was born from ex-VMS engineers, which begat 3.5, which begat 4.0, which begat 2000 (5.0), which begat XP (5.1). At 5.0, some of the Win9x features were rolled in for better gaming and hardware support.
OTH, NT 3.x to NT 5.0 was a GIANT leap.
:-D
I have a dual PII Xeon w/ GeForce3 Ti200 OC and I have no problems whatsoever in Linux and Win2k. I'm using the latest reference drivers as of about 3 months ago. What's up with your system?
While SMP isn't useful for Q3A, Tribes2, etc. it is nice to play MP3s, etc. while playing without a performance hit.
username: onetimeslashdot
password: onetimeslashdot
For every machine I use, no fuss, no muss. It's like a reflex now.
Correct. However, it is a good [back]ronym, and even it's misinterpreted form is useful. /usr/bin are binaries for users; /usr/lib are the corresponding libraries, and so on. I think /home is a better prefix for home directories in my opinion. It is /usr/local that definitely needs to be dropped, and I'm glad that I see it used less and less [I mean /usr/local/bin? Isn't that supposed to be /bin if it's local?].
/usr to a network share, not including myself. The idea is that /{bin,etc,lib} are supposed to be enough of a Unix system to get /usr off the network anyway.
Some enterprising system administrators have gone that extra step and liberated
Unix System Resources
Open Source is a strategy, not a solution.
I think that brings the point home for this and a number of other discussions above.
The problem with the better physics/better AI problem is not that we don't have enough power. The problem is programmers don't have the time to implement all those things they could implement with all that power. Just having the extra horsepower doesn't make those things happen by themselves. All they know what to do is push all the buttons on the 3d graphic controls in the newest SDK to the max (Where else can we use a cubic environment map...). Compare a game that pays great attention to detail with one that is thrown together quickly.
Sort of. Except that since the Pentium Pro core completely rewrites and rearranges each instruction into microops, you never get to write code to exploit the three pipes. The original pentium directly translated each instruction for each execution unit, hence you could order each instruction for parallel executions. With the Pentium Pro, you can't really guarantee anything since the instruction rewriter is the one doing all the ordering for parallelism. You can actually make it's job harder by trying to help it out. P5 optimized code tends to stall out the P6 series.
Conjecture, exactly.
It (HT Xeon) does use two seperate TLBs. Not only that, each core has it's own register set including segments and TSS, LDT, the whole shebang. In fact, an issue mentioned earlier in the discussion and the article was that the two threads would sometimes conflict in usage of the cache (wanting dispersed regions of memory), thus increasing the cache miss rate than if they were two seperate cores. Furthermore, stock versions of Windows were used. They are not aware of the SMT nature of the processor, they just see two CPUs, and act accordingly. How could they know not to use threads that use different memory regions when it's never previously been an issue?
1) U-V pipe only was used in the original pentium. Pentium pro and up use an out-of-order executing RISC core.
2) That L2 cache is probably shared between the
cores, if so, it'll still suffer some.
[see my other comment]
Since it's the same server, it would be more difficult to not cross-post.
[root@loki /root]# nslookup www.slashdot.org
/root]# nslookup apple.slashdot.org
/root]# nslookup bsd.slashdot.org
-snip-
Name: slashdot.org
Address: 64.28.67.150
Aliases: www.slashdot.org
[root@loki
-snip-
Name: slashdot.org
Address: 64.28.67.150
Aliases: apple.slashdot.org
[root@loki
-snip-
Name: slashdot.org
Address: 64.28.67.150
Aliases: bsd.slashdot.org
They're just aliases, a neat way to filter out topics without doing any extra work.
By that point, the moon's orbit would have decayed so either a) we'd have rings b) we'd be squished by the satellite. While we'd have the same revolution period as the moon's, we phyiscally wouldn't be in lock with anything.
Yes and no.
The model is still the same, this is true. A lot of shell scripts and code will still work on a new Linux system. If you must ask why, because it was pretty good to begin with. Can anyone think of a good reason to get rid of the BSD sockets API?
On the other hand everything under the hood has been changed and is decidedly not backwards compatible. The jump from Linux 1.0 to 2.0, the jump from libc5 to GLIBC 2.x, these are all breaking points. All new driver architectures, APIs, even executable formats. I don't think Linux is being held back. Having old API compatibility is not a bad thing, but even then, they often must say: This is it, this old stuff is deprecated, we're not supporting it anymore. For example, the RAID system is neither forwards nor backwards compatible in any way between 2.2 and 2.4. Stuff like that.
Plus, there's a difference between designing an OS to have legacy support and actually emulating the legacy system. Forcing someone to recompile or edit the aging code slightly can give someone freedom to implement the compatibility layer any way they see fit, instead of having to keep it "in place" and organize new features "around it".
for the most part it was intended to be a psychological treatise from the beginning. The hook (and the animation budget) was the awesome mecha/end-of-the-world storyline. I've heard accusations that this was director Hideaki Anno's way of putting his own emotions and feelings into a show. Also, I heard that the end was not the intended one... it was shorter, and in retaliation it was given a shot of LSD by the production team due to political/financial factors. It gave Anno a chance to rant back at Gainax [replace "Shinji" with "Hideaki" and watch those episodes again, it starts to make sense! (-: ]
I for one really enjoyed it... wonderful characters, animation, action, drama... it was somewhat mysterious at times, frustrating at others, but in retrospect it makes sense. You like it more two weeks _after_ seeing the whole thing.
Check out some of the lists that people post here for stuff to see... cover art and blurbs on the back don't really give you an idea of what you're in for.
If you can get your hands on "Vandread" or "Martian Successor Nadesico", you'll probably appreciate them. (Mecca in space, nice animation, and funny as hell)
...considering that the Pentium processor is covered (and later iterations of the Pentium line used almost the exact same core).
No. The transition from P5 to P6 (Pentium Pro) is the introduction of an entirely new core.
The P5 core had two instruction pipes (thank god), the U and V pipes. I remember he discussed how you can optimize assembly for this in one of the chapters. Basically, you would schedule instructions that didn't step on each other's feet one after the other. However, the P6 core (Pentium Pro-Pentium III) is quite different. It has a RISC core and a variable (chip dependant) number of ALUs and FPUs. It is nowhere nearly as predictable as the P5, thus optimizing it is a tricky, black art. None of the old optimization tricks work, instead you basically follow guidelines for emitting code that won't break the longish pipeline.
I'm betting dollars-to-donuts Peter St. George is the only one who works at ZeoSync, and he wants your gullible-ass money.