I was under the impression that real time raytracing was still far from practical. I know that Pixar's Renderman uses a much more efficient algorithm, and it still takes on the order of hours to render a full-format movie frame on beefy hardware (I know that's large, but raytracing should scale fine, and we aren't talking about millions of times more pixels). IIRC they use a raytracer only when they need reflections, and those frames are extremely expensive in terms of CPU time.
Why do they use the 74HCU04 instead of the 74HCU04? I'm using a 74HC04 to drive S/PDIF output on a USB-audio device I built because I couldn't find a 74HCU04 at Frys and it seems to work fine. But would I get less jitter with a 74HCU04?
What's a dead virus?
on
HIV Vaccine
·
· Score: 3, Interesting
I always hear about vaccines involving "dead" virus material. But I thought viruses weren't alive in the first place; that they were essentially protien envelopes containing viral DNA or RNA. Can anyone explain?
Among other things, Glucas is writen in C and Prime95 is mostly x86 assembly that's heavily optimized for SSE2 and the P4.
Not to mention that you can't expect the threading to scale perfectly. I'm surprised that there are any gains at all because the LL algorithm is so sequential. I remember hearing that Glucas could have done it in half the time on that machine if it had been optimized for NUMA, though.
There are so many things wrong with this that it's difficult to start. The first is Jeff Waugh's proposal that Epiphany replace Firefox as a the de-facto Linux browser. Jeff Waugh is GNOME's most extreme evangelist and from my communications with him I get the impression that he would like nothing more than for GNOME to become integral to every Linux system. In reality, all that users would gain from Epiphany is that things might be a little more consistent *if* they happened to use GNOME. If not, too bad:
# apt-get install epiphany-browser
Reading Package Lists... Done
Building Dependency Tree... Done
The following extra packages will be installed:
docbook-xml docbook-xsl gconf2 gnome-doc-tools gnome-icon-theme
gnome-mime-data libbonobo2-0 libbonobo2-common libbonoboui2-0
libbonoboui2-common libeel2-2 libeel2-data libfam0c102 libgconf2-4
libgnome-desktop-2 libgnome2-0 libgnome2-common libgnomeui-0
libgnomeui-common libgnomevfs2-0 libgnomevfs2-common libnautilus2-2
liborbit2 libscrollkeeper0 libxslt1.1 scrollkeeper yelp
Suggested packages:
gnome-vfs-extras2
The following NEW packages will be installed:
docbook-xml docbook-xsl epiphany-browser gconf2 gnome-doc-tools
gnome-icon-theme gnome-mime-data libbonobo2-0 libbonobo2-common
libbonoboui2-0 libbonoboui2-common libeel2-2 libeel2-data libfam0c102
libgconf2-4 libgnome-desktop-2 libgnome2-0 libgnome2-common libgnomeui-0
libgnomeui-common libgnomevfs2-0 libgnomevfs2-common libnautilus2-2
liborbit2 libscrollkeeper0 libxslt1.1 scrollkeeper yelp
0 upgraded, 28 newly installed, 0 to remove and 83 not upgraded.
...And I already have GTK2 and Gecko installed. You get to keep most of the new stuff resident in memory for no gain.
By the way, even if Epiphany does not become the standard browser, I don't like the direction things are headed in with Firefox becoming the default. The designers seem to have outsourced their UI design to Redmond. Here are a few examples of the things I don't like about Firefox that I feel came directly from IE and really don't jive with my habits that come from using NSCP products for 10 years:
When you try to add a bookmark, it opens a dialog. Netscape never did this, not even in version 1.0. I much prefer the old way.
By default, Firefox zooms images that you view directly to fit the screen. When I occasionally used Windows lab machines a few years ago there was nothing I hated more than loading a screenshot and seeing it scaled down. At least you can turn this off in the preferences.
Both Mozilla and Firefox have adopted favicon.ico from IE. IMHO this is a horrible idea and never had any point. In Mozilla you can turn it off in the preferences dialog. In Firefox, you can partially turn it off using the cryptic about:config.
"Options" is under "Tools". This is unintuitive. I understand that Edit isn't the best choice either, but it's been that way for a decade.
When you try to type in a URL, a dropdown of completions appear. Most people like this. I don't. Mozilla lets you turn it off. Firefox doesn't.
And there are many more things I dislike about Firefox, but these are just the ones I feel they took directly from IE. I understand that they want it to be easy for IE users to migrate, but this inflexible browser really doesn't meet the expectations of power users like myself and many of my friends. I've tried to reason with some of the Firefox developers about some of these issues, and they firmly believe that Firefox should cater to the needs of IE converts. They don't want to add preference items for the admittedly minor things I care about because it would confuse people and possibly make Firefox's preferences UI as slow as Mozilla's (XUL doesn't seem very scalable). This is proof that only having one browser is not sufficient, especially if it's a least common denominator one. I haven't used Epiphany, but I can't see how it would be better to remove the choice between Firefox and Epipha
Because it would be better to better to implement that in the protocol. Knocking on ports is a silly way to do a secure authentication handshake. It has more overhead than sending data over a normal connection - you'd have to open dozens of connections just to transfer a few bytes. Firewalls would totally screw it up. And ultimately, it's no more secure than, say, SSH, SSH tunneling, SSL, or IPSec. Sure, some implementations of these may have undiscovered security holes, but whatever handles the port knocking has an equal chance of them.
Also, a port-knocking-based handshake will not automatically encrypt your session, which is essential if you care about security. Not only that, but it seems like it would be highly vulnerable to man-in-the-middle attacks.
This extension is proof that extensions can make Fire* suck less. Here are some others that I would really like to see:
Disable the @#$@#% popdown menu that appears when I start typing URLs. I know most people like completion menus; I don't. You can't even turn this off in about:config in Firefox.
Map control-Q to File:Quit. Simple, huh? I don't know why this doesn't work by defaults. Obviously the developers feel that their application is too useful to ever be quit;).
Disable all the "security" restrictions on ports. I have a RAID controller daemon running on localhost:1080. Neither Mozilla nor Firefox will let me connect to it for "security reasons". I know it's possible to open individual ports (by mucking around with javascript commands), but AFAIK there's no way to turn off this silly mechanism entirely.
BTW, the flashblock extension is awesome, even though it hasn't been working on every page for me.
This shouldn't have been modded down to troll. I was about to post the same sentiments. I'm rather disappointed that slashdot is running flash ads. They're a LOT more annoying than even animated gifs. Fortunately, the mozilla extension saves my eyes.
It's reassuring to have a hypothesis of what's wrong with you. When I make guesses based on medical information available on the internet, I'm often right. Having a hypothesis makes it easier for a doctor to diagnose the problem.
So get a 4 proc box instead with the extra $2300 and run Linux on it. I'd be surprised if it didn't have higher throughput, especially if you use 2.6 which scales quite well.
DES is vulnerable to both linear and differential cryptanalysis. I believe the differential attack is singificantly easier than brute force if you can handle the volume of information it requires.
The General Number Field Sieve has four different stages, and only the first is very parallelizable. http://www.nfsnet.org/ is a distributed computing project that distributes the first stage of GNFS. The last 3 are run in-house (at Microsoft, in fact).
...using a supernova to complete in a reasonable amount of time, that is. Power is not energy. Unfortuately I don't have the book handy so I don't know what "reasonable" meant.
Yes, it is directly proportional to the temperature and Boltzmann's constant is the constant of proportionality. Schneier did some calculations in Applied Cryptography about how much energy it would take to brute force a 256 bit symmetric key and still not violate any laws of the Universe. I believe it was on the order of needing a supernova as a power source.
Much of the spam we get comes from mailing lists. This kind of scheme would require every list admin to submit all their mailing list addresses to some stupid opt out lists. There are many examples of this not being practical, such as the Debian bug tracking system which has a different email address of each bug (and there are over 200k). FWIW, it does receive spams that clutter up bug audit trails and are extremely annoying. Being allowed to spam should not be the default.
Slashdot Mirror
What optical discs can store huge quantities of hi-def content without compression?
Seriously. Why just pop?
Pretty much, yeah.
I was under the impression that real time raytracing was still far from practical. I know that Pixar's Renderman uses a much more efficient algorithm, and it still takes on the order of hours to render a full-format movie frame on beefy hardware (I know that's large, but raytracing should scale fine, and we aren't talking about millions of times more pixels). IIRC they use a raytracer only when they need reflections, and those frames are extremely expensive in terms of CPU time.
And then P2P programs will start encrypting traffic. What's the point?
Why do they use the 74HCU04 instead of the 74HCU04? I'm using a 74HC04 to drive S/PDIF output on a USB-audio device I built because I couldn't find a 74HCU04 at Frys and it seems to work fine. But would I get less jitter with a 74HCU04?
I always hear about vaccines involving "dead" virus material. But I thought viruses weren't alive in the first place; that they were essentially protien envelopes containing viral DNA or RNA. Can anyone explain?
Maybe if you live in France.
Where do you think your electricity comes from?
Force BERKELEY to do that?! You're out of your mind.
Backslashes? DOS-style dir listings? UGH, I thought the idea was to remove obsolescence.
Also, this is even worse than ACPI from a needless complexity standpoint.
Among other things, Glucas is writen in C and Prime95 is mostly x86 assembly that's heavily optimized for SSE2 and the P4.
Not to mention that you can't expect the threading to scale perfectly. I'm surprised that there are any gains at all because the LL algorithm is so sequential. I remember hearing that Glucas could have done it in half the time on that machine if it had been optimized for NUMA, though.
# apt-get install epiphany-browser
Reading Package Lists... Done
Building Dependency Tree... Done
The following extra packages will be installed:
docbook-xml docbook-xsl gconf2 gnome-doc-tools gnome-icon-theme gnome-mime-data libbonobo2-0 libbonobo2-common libbonoboui2-0 libbonoboui2-common libeel2-2 libeel2-data libfam0c102 libgconf2-4 libgnome-desktop-2 libgnome2-0 libgnome2-common libgnomeui-0 libgnomeui-common libgnomevfs2-0 libgnomevfs2-common libnautilus2-2 liborbit2 libscrollkeeper0 libxslt1.1 scrollkeeper yelp
Suggested packages:
gnome-vfs-extras2
The following NEW packages will be installed:
docbook-xml docbook-xsl epiphany-browser gconf2 gnome-doc-tools gnome-icon-theme gnome-mime-data libbonobo2-0 libbonobo2-common libbonoboui2-0 libbonoboui2-common libeel2-2 libeel2-data libfam0c102 libgconf2-4 libgnome-desktop-2 libgnome2-0 libgnome2-common libgnomeui-0 libgnomeui-common libgnomevfs2-0 libgnomevfs2-common libnautilus2-2 liborbit2 libscrollkeeper0 libxslt1.1 scrollkeeper yelp
0 upgraded, 28 newly installed, 0 to remove and 83 not upgraded.
By the way, even if Epiphany does not become the standard browser, I don't like the direction things are headed in with Firefox becoming the default. The designers seem to have outsourced their UI design to Redmond. Here are a few examples of the things I don't like about Firefox that I feel came directly from IE and really don't jive with my habits that come from using NSCP products for 10 years:
And there are many more things I dislike about Firefox, but these are just the ones I feel they took directly from IE. I understand that they want it to be easy for IE users to migrate, but this inflexible browser really doesn't meet the expectations of power users like myself and many of my friends. I've tried to reason with some of the Firefox developers about some of these issues, and they firmly believe that Firefox should cater to the needs of IE converts. They don't want to add preference items for the admittedly minor things I care about because it would confuse people and possibly make Firefox's preferences UI as slow as Mozilla's (XUL doesn't seem very scalable). This is proof that only having one browser is not sufficient, especially if it's a least common denominator one. I haven't used Epiphany, but I can't see how it would be better to remove the choice between Firefox and Epipha
Because it would be better to better to implement that in the protocol. Knocking on ports is a silly way to do a secure authentication handshake. It has more overhead than sending data over a normal connection - you'd have to open dozens of connections just to transfer a few bytes. Firewalls would totally screw it up. And ultimately, it's no more secure than, say, SSH, SSH tunneling, SSL, or IPSec. Sure, some implementations of these may have undiscovered security holes, but whatever handles the port knocking has an equal chance of them.
Also, a port-knocking-based handshake will not automatically encrypt your session, which is essential if you care about security. Not only that, but it seems like it would be highly vulnerable to man-in-the-middle attacks.
- Disable the @#$@#% popdown menu that appears when I start typing URLs. I know most people like completion menus; I don't. You can't even turn this off in about:config in Firefox.
- Map control-Q to File:Quit. Simple, huh? I don't know why this doesn't work by defaults. Obviously the developers feel that their application is too useful to ever be quit
;).
- Disable all the "security" restrictions on ports. I have a RAID controller daemon running on localhost:1080. Neither Mozilla nor Firefox will let me connect to it for "security reasons". I know it's possible to open individual ports (by mucking around with javascript commands), but AFAIK there's no way to turn off this silly mechanism entirely.
BTW, the flashblock extension is awesome, even though it hasn't been working on every page for me.This shouldn't have been modded down to troll. I was about to post the same sentiments. I'm rather disappointed that slashdot is running flash ads. They're a LOT more annoying than even animated gifs. Fortunately, the mozilla extension saves my eyes.
It's reassuring to have a hypothesis of what's wrong with you. When I make guesses based on medical information available on the internet, I'm often right. Having a hypothesis makes it easier for a doctor to diagnose the problem.
It's funny you should say that because NSA's security system is part of the official kernel. It's optional, but it's there just the same.
So get a 4 proc box instead with the extra $2300 and run Linux on it. I'd be surprised if it didn't have higher throughput, especially if you use 2.6 which scales quite well.
DES is vulnerable to both linear and differential cryptanalysis. I believe the differential attack is singificantly easier than brute force if you can handle the volume of information it requires.
The General Number Field Sieve has four different stages, and only the first is very parallelizable. http://www.nfsnet.org/ is a distributed computing project that distributes the first stage of GNFS. The last 3 are run in-house (at Microsoft, in fact).
...using a supernova to complete in a reasonable amount of time, that is. Power is not energy. Unfortuately I don't have the book handy so I don't know what "reasonable" meant.
Yes, it is directly proportional to the temperature and Boltzmann's constant is the constant of proportionality. Schneier did some calculations in Applied Cryptography about how much energy it would take to brute force a 256 bit symmetric key and still not violate any laws of the Universe. I believe it was on the order of needing a supernova as a power source.
This probably doesn't apply to quantum computing.
Much of the spam we get comes from mailing lists. This kind of scheme would require every list admin to submit all their mailing list addresses to some stupid opt out lists. There are many examples of this not being practical, such as the Debian bug tracking system which has a different email address of each bug (and there are over 200k). FWIW, it does receive spams that clutter up bug audit trails and are extremely annoying. Being allowed to spam should not be the default.
you mean this?