You gain many of these advantages with Ruby+FastCGI, btw. It's (Ruby) server based, so you've got persistant objects, threading, etc to play with, along with Ruby's kickass object model, exceptions, easy-as-pie C/++ extensions, and really quite decent performance.
The same can surely be said for Perl/Python+FastCGI.. I suspect PHP lacks the flexibility to make good use of it, though.
This can actually be a serious issue if you sleep in the same room as a computer. I've moved my bed so I'm no longer in direct line of sight with my cable modem and server LED's... and still the ethernet activity lights lighting up the wall behind them annoys me.
Forget Turbo switches; we need cases with buttons to turn off all the damn lights:)
... pirates get to wait an extra couple of hours for a crack, and paying users get yet another annoying "protection" to get in their way. Yet again, those who aren't paying are getting a *better* product (no faffing about keeping the CD in the drive, no mandatory registration) than those who are. That's great! No, really...
Are they at least going to offset the annoyance factor by forgoing the CD checks? I have 300GB of disk space - I don't want to have to screw about hunting for a disk I can't copy properly which should be in storage somewhere safe.
Screw Steam too; post it to Usenet and P2P, and give me a way of buying a cheap license which factors in the fact that I didn't cost anything to distribute the game to. It's going to appear there anyway; it might as well do so legitimately.
These two people will also need to be using the same drives, and a good quality CD and/or secure mode ripper (EAC), and drive offset correction or the same drive. Pre-scratched ("Copy Protected") CD's are even more hairy, because they break the error correction and detection codes, making it difficult enough to get two identical rips using identical setups; never mind two entirely seperate systems.
Two different people using different drives with different offsets are really relatively unlikely to come out with identical files, especially ripping in burst mode (like most do).
Of course, those who use offset correction and secure mode ripping (i.e. clueful users) are *much* more likely to produce identical rips. But then these people are also more likely to tag files uniquely (%DISC% vs%DISK%, %SINGLETRACK%, %ALBUM ARTIST% vs %VARIOUS%, etc), apply ReplayGain/WavGain/mp3Gain, and so on, which counteracts this somewhat.
But... none of this really helps anyone. If you come out with identical MD5's, all it shows is two users have the same rip. It doesn't mean one got it off another; maybe they just have the same drives, same setup, or just got lucky. Maybe they even used a ripper which compares checksums with other rips.
Identical MD5's do not mean they're both from one sharing user; differing MD5's do not mean they aren't. Now, someone please go tell the lawyers.
It means any port which uses it needs to be updated too (unless you use portupgrade and have it preserve old libraries), and that's quite a lot. Not sure what it's doing here though;)
The Seagate 7200.7's are almost as quiet as the 'cuda IV, but come in 160G versions with 8MB cache -- seek time's still sub-par, but max transfer rate is lovely (>50MB/s sustained read).
(BTW, another factor in the 'cuda IV's favour is it's been around for a while, and is known to be a very reliable drive; hopefully the 7200.7's will behave similarly).
(Yes, I know the answer is that nobody actually needs these new CPUs, but you know Microsoft and Intel won't stand for that...)
I run a 1.4GHz Thunderbird; yup, hottest Athlon until the 3000XP. I'm increasingly finding myself tapping my feet waiting for it to do things I'm doing rather often.
ReplayGaining my music, for a start; I want to listen to a new album, but find it's not ReplayGained -- I have to wait a few minutes while foobar2k churns away. If I've downloaded it, maybe I'll be waiting even longer while I transcode from APE to FLAC or so.
Years ago, I used to listen to MP3's on an Amiga, about 50 times slower than my current system. ReplayGain was but a glint in somebody's eye (it would have taken hours), lossless was inconceivable (1 album using 1/3rd of my drive space? No thanks), and who would have thought I'd be running a 200+ line playlist formatting string now? Hell, even volume control now is a DSP done at 64bit precision -- ok, so I don't need a 3.5GHz CPU for that, but there *are* tasks I do a lot which would benefit from being several times faster.
And of course, if we had that sort of power, maybe the awfully CPU-heavy Lossless Audio codec would be a sensible format to use -- maybe my playlist formatting string will evolve into a stateful 1000+ line script! Hell, we might even be able to start seriously looking at the full MPEG-4 spec...
The point is, even things that are typically concidered as being CPU friendly end up benefiting greatly from more power in ways that nobody concidered. You get similar things from increases in storage space, in memory, in fillrates and resolution. There are plenty of things we'd love to do now, but simply don't have the power or storage to do it, and even more things we haven't even thought about because the power isn't there to back them.
Until we literally have more space than we know what to do with; until we can perform complex transformations and searches on that space (without pansy-ass indexes;) in fractions of a second; until we can render photorealistic 3D images of anything we can think of and do it individually targeted at each eye of every observer (potentially thousands); until we can simulate reality faster and bigger than reality itself WE DO NOT HAVE ENOUGH POWER.
Yes -- I indeed will not be happy until our computers (or whatever higher-level machine we come up with to replace them) have to be stored in purpose built universes connected to this one via wormholes. And even then...
What constitutes a "Critical Update"? I'm running the very latest WHQL nVidia drivers (4523); yet Windows Update has "NVIDIA display software update released on May 02 2003." in the critical updates section.
All these resources.. they can't even get this right. *shudder*.
Latest nVidia drivers are a shade under 8MB. It's only the international version which approaches 30MB, and that serves those people right for not using the One True Language;)
(Yeah, 8MB is still pretty huge for a driver and control panel stuff, but.. meh, it supports about 10 generations of products, what do you expect?:)
GPL is Free Software(R) -- Open source is a conciderably wider target than that. A quick look over the site doesn't show anything GPL-specific, just Linux-specific.
BSDers need love too!
With Vorbis being tuned to higher bitrates, and with it scaling up bitrate for complex music, I'd be testing up to the 0.5Mbps level; that's still easily half the size of FLAC, and will ensure it's not going to fall over on particularly complex tracks.
Now, we just need MusePack, ReplayGain, APEv2 tag support on MP3's, foobar2000-alike customisable playlist displays, and we'll be getting somewhere...
DivX/XviD require a tiny, easy to install, open source decoder like ffdshow in Win; similar systems exist for other OS's.
QuickTime requires a huge bloated badly written hacked up custom player which likes to try to take over half your machine and bugs you to pay for it every time you run it.
Now, why on earth would you choose the latter over the former?
Obviously, having enough memory is preferable to *having* to swap (I have half a gig in this machine, it has 160k of swap in use; a similar 2G system has 116k in use), but that doesn't make swapping a bad idea, and it doesn't reduce performance in the way you suggest. Preemptively swapping doesn't reduce performance, and being able to reuse clean pages straight away because of that is *massively* preferable to the system simply running out of memory and having to nuke tasks at random to satisfy requests.
Uhm, paging actually improves memory efficiency when implemented properly (see: FreeBSD) by keeping rarely used pages out of limited physical memory (or more accurately by preemtively writing out rarely hit dirty pages to swap and leaving the pages in memory until something else wants more memory, at which point the already swapped pages can be dropped immediately).
Even Windows understands this to some extent; by aggressively swapping unused things out it can ensure physical memory is available for running applications and caching. It's implementation and tuning may leave something to be desired, of course:)
Slashdot Mirror
I thought it was 97 watts each.. maybe it was total; anyone have a reference?
You gain many of these advantages with Ruby+FastCGI, btw. It's (Ruby) server based, so you've got persistant objects, threading, etc to play with, along with Ruby's kickass object model, exceptions, easy-as-pie C/++ extensions, and really quite decent performance.
The same can surely be said for Perl/Python+FastCGI.. I suspect PHP lacks the flexibility to make good use of it, though.
This can actually be a serious issue if you sleep in the same room as a computer. I've moved my bed so I'm no longer in direct line of sight with my cable modem and server LED's... and still the ethernet activity lights lighting up the wall behind them annoys me.
:)
Forget Turbo switches; we need cases with buttons to turn off all the damn lights
... pirates get to wait an extra couple of hours for a crack, and paying users get yet another annoying "protection" to get in their way. Yet again, those who aren't paying are getting a *better* product (no faffing about keeping the CD in the drive, no mandatory registration) than those who are. That's great! No, really...
Are they at least going to offset the annoyance factor by forgoing the CD checks? I have 300GB of disk space - I don't want to have to screw about hunting for a disk I can't copy properly which should be in storage somewhere safe.
Screw Steam too; post it to Usenet and P2P, and give me a way of buying a cheap license which factors in the fact that I didn't cost anything to distribute the game to. It's going to appear there anyway; it might as well do so legitimately.
These two people will also need to be using the same drives, and a good quality CD and/or secure mode ripper (EAC), and drive offset correction or the same drive. Pre-scratched ("Copy Protected") CD's are even more hairy, because they break the error correction and detection codes, making it difficult enough to get two identical rips using identical setups; never mind two entirely seperate systems.
Two different people using different drives with different offsets are really relatively unlikely to come out with identical files, especially ripping in burst mode (like most do).
Of course, those who use offset correction and secure mode ripping (i.e. clueful users) are *much* more likely to produce identical rips. But then these people are also more likely to tag files uniquely (%DISC% vs%DISK%, %SINGLETRACK%, %ALBUM ARTIST% vs %VARIOUS%, etc), apply ReplayGain/WavGain/mp3Gain, and so on, which counteracts this somewhat.
But... none of this really helps anyone. If you come out with identical MD5's, all it shows is two users have the same rip. It doesn't mean one got it off another; maybe they just have the same drives, same setup, or just got lucky. Maybe they even used a ripper which compares checksums with other rips.
Identical MD5's do not mean they're both from one sharing user; differing MD5's do not mean they aren't. Now, someone please go tell the lawyers.
I find it has a very bitter aftertaste. Earl Gray kicks it's ass
It means any port which uses it needs to be updated too (unless you use portupgrade and have it preserve old libraries), and that's quite a lot. Not sure what it's doing here though ;)
FreeBSD was using 0.11.5 before this change, btw.
Handy that the BBC has more bandwidth than geeks have braincells then :)
The Seagate 7200.7's are almost as quiet as the 'cuda IV, but come in 160G versions with 8MB cache -- seek time's still sub-par, but max transfer rate is lovely (>50MB/s sustained read).
(BTW, another factor in the 'cuda IV's favour is it's been around for a while, and is known to be a very reliable drive; hopefully the 7200.7's will behave similarly).
I run a 1.4GHz Thunderbird; yup, hottest Athlon until the 3000XP. I'm increasingly finding myself tapping my feet waiting for it to do things I'm doing rather often.
ReplayGaining my music, for a start; I want to listen to a new album, but find it's not ReplayGained -- I have to wait a few minutes while foobar2k churns away. If I've downloaded it, maybe I'll be waiting even longer while I transcode from APE to FLAC or so.
Years ago, I used to listen to MP3's on an Amiga, about 50 times slower than my current system. ReplayGain was but a glint in somebody's eye (it would have taken hours), lossless was inconceivable (1 album using 1/3rd of my drive space? No thanks), and who would have thought I'd be running a 200+ line playlist formatting string now? Hell, even volume control now is a DSP done at 64bit precision -- ok, so I don't need a 3.5GHz CPU for that, but there *are* tasks I do a lot which would benefit from being several times faster.
And of course, if we had that sort of power, maybe the awfully CPU-heavy Lossless Audio codec would be a sensible format to use -- maybe my playlist formatting string will evolve into a stateful 1000+ line script! Hell, we might even be able to start seriously looking at the full MPEG-4 spec...
The point is, even things that are typically concidered as being CPU friendly end up benefiting greatly from more power in ways that nobody concidered. You get similar things from increases in storage space, in memory, in fillrates and resolution. There are plenty of things we'd love to do now, but simply don't have the power or storage to do it, and even more things we haven't even thought about because the power isn't there to back them.
Until we literally have more space than we know what to do with; until we can perform complex transformations and searches on that space (without pansy-ass indexes
Yes -- I indeed will not be happy until our computers (or whatever higher-level machine we come up with to replace them) have to be stored in purpose built universes connected to this one via wormholes. And even then...
What constitutes a "Critical Update"? I'm running the very latest WHQL nVidia drivers (4523); yet Windows Update has "NVIDIA display software update released on May 02 2003." in the critical updates section.
All these resources.. they can't even get this right. *shudder*.
Latest nVidia drivers are a shade under 8MB. It's only the international version which approaches 30MB, and that serves those people right for not using the One True Language ;)
:)
(Yeah, 8MB is still pretty huge for a driver and control panel stuff, but.. meh, it supports about 10 generations of products, what do you expect?
Er, why? /bin, /sbin and /lib should all be on the same partition, and if you're really screwed, that's what /rescue is for.
SMTP supports TLS, but of course, nobody bothers setting it up but me...
GPL is Free Software(R) -- Open source is a conciderably wider target than that. A quick look over the site doesn't show anything GPL-specific, just Linux-specific. BSDers need love too!
Will someone please tell me wtf hot grits are? Every time I read it I think of someone heating gravel over a fire...
With Vorbis being tuned to higher bitrates, and with it scaling up bitrate for complex music, I'd be testing up to the 0.5Mbps level; that's still easily half the size of FLAC, and will ensure it's not going to fall over on particularly complex tracks.
Now, we just need MusePack, ReplayGain, APEv2 tag support on MP3's, foobar2000-alike customisable playlist displays, and we'll be getting somewhere...
Great; just don't go mistaking depression for introversion... that's probably a lot more damaging than the other way around.
DivX/XviD require a tiny, easy to install, open source decoder like ffdshow in Win; similar systems exist for other OS's.
QuickTime requires a huge bloated badly written hacked up custom player which likes to try to take over half your machine and bugs you to pay for it every time you run it.
Now, why on earth would you choose the latter over the former?
If this were true, Xing might actually be usable.
Uh... right, so how are you supposed to know that?
The {
Obviously, having enough memory is preferable to *having* to swap (I have half a gig in this machine, it has 160k of swap in use; a similar 2G system has 116k in use), but that doesn't make swapping a bad idea, and it doesn't reduce performance in the way you suggest. Preemptively swapping doesn't reduce performance, and being able to reuse clean pages straight away because of that is *massively* preferable to the system simply running out of memory and having to nuke tasks at random to satisfy requests.
Uhm, paging actually improves memory efficiency when implemented properly (see: FreeBSD) by keeping rarely used pages out of limited physical memory (or more accurately by preemtively writing out rarely hit dirty pages to swap and leaving the pages in memory until something else wants more memory, at which point the already swapped pages can be dropped immediately).
:)
Even Windows understands this to some extent; by aggressively swapping unused things out it can ensure physical memory is available for running applications and caching. It's implementation and tuning may leave something to be desired, of course