If your computer runs out of memory, take a snapshot of its state and continue on a computer with more memory. Keep this up on newer generations of computers -- pass it off to your ancestors -- continue until the sun turns into a red giant if need be. Hopefully we will have developed quantum computer which can hold more memory than there are particles in the universe. Keep it going. Will it ever halt?
If you can determine if the following program will ever halt then you will have solved a problem that no mathematician has solved!
main() {
bigint n, a, b, i;
for (n = 1; ; n++) {
a = n*n;
b = (n+1)*(n+1);
for (i = a+1; i< b && !prime(i); i++)
;
if (i == b) break;/* no prime found, halt program */
} }
Anyone wish to explain why the energy in a giant tidal wave does not attenuate as it travels across the ocean? It just seems that the wave would greatly diminish in size as it propagates.
"...[if] you have a risk of dying in the next year of well under one in 1,000, which means that if you stayed that way forever you would have a 50/50 chance of living to over 1,000. "
Bad math.
0.999^1000 =.3677
Really you would only have slightly better than a third of a chance. I can't trust a scientist who can't do math.
why can't they bust these perps?
on
Gone Phishing?
·
· Score: 0
It seems like it should be easy to follow the trail to catch these guys. What's the typical way a perp sets up a fake site? I assume they hijack a web site, but I have seen some where they have a TLD.org url. Can't they find those who registered the domain name? Why is it so impossible to catch these people?
This is a really good point. The problem of proving you actually used a particular algorithm in your closed source binary would be difficult (probably theoretically impossible, i.e., halting problem kind of thing). So the answer is to be a LONER and SECRETIVE!
isn't it just plain obvious that there is an infinite number of primes Why ? Because there is an infinite number of possible numbers. Sure, primes get farther distanced from each other as the number increases in magnitude, but there's always one around the corner.
Not really that obvious. For example, even though there are infinite number of positive integers, no 3 of them satisfy x^n + y^n = z^n when n > 2.
man page.
e.g., man glBegin.
Actually, hypertexting through this the docs found
here
is even better.
gl pipeline not for raytracing
on
The State of OpenGL
·
· Score: 2, Informative
Most frames in Pixar movies are rendered using some form of ray-tracing. While it is possible to use vertex and fragment shaders in uncoventional ways to do ray tracing, this is *not* what the OpenGL pipeline is designed for. Great for games, but ray-tracing will still be done using render farms (and not in real time).
Yeah, the description is ambiguous. If code is embedded in a true MP3 file, how does that code get executed? If its just an application with an MP3 extension, I don't really call that a very novel trojan horse. Can anyone flush out more details?
There are plenty of "Billie Bob's" in the country where I run who feel the need to own highly aggressive dogs. The best way to deal with many of them is to get on one knee with palms up -- let them be the alpha dog -- I don't care. Of course, if this doesn't work you may have only seconds to protect you jugular -- I haven't had this problem yet.
You can do this using flat address and paging. Mark all stack pages as non-executable -- there is already support for this on the protected mode Intel chips. Segmentation not needed.
The bug involves an inconsistency with the recorded and actual buffer size. The buffer is allocated off the heap, but an exploitable overflow really only works for buffers allocated off of the stack (where you can overwrite a subroutine's return address and redirect program flow). I guess if there are subroutine addresses in struct's dynamically allocated off of the heap, then you could redirect program flow, but I am suspicious of this.
I had a load of research money for purchasing an SGI a couple of years ago. SGI made it very difficult to buy from them. Generally they will put you in contact with a reseller, and then you get quotes from them. This seemed to take awhile and was awkward. I just decided not to buy from them after a while. I am pretty happy with OpenGL performance on my Mac, and the price/performance of an SGI is really not that good. We'll see what SGI has at SIGGRAPH this year. I can't believe they haven't tanked yet.
Back in 1990 I worked for a small company that built graphics boards and my first task was to debug the "polygon fill" routine in their firmware. It turns out they use their own "home brewed" algorithm that was slow, memory hungry, and didn't handle degenerate cases correctly. If anyone in the company would have taken the time to pull any one of the graphics textbooks off their shelf (e.g., Foley, van Dam) they would find a much better solution.
I ended up rewriting the module myself using the classic solution -- it was faster, used little memory, and handled degenerate cases reasonably.
It was my experience that everything was a badly reinvented wheel when I worked there.
Slashdot Mirror
I imagine any CS dept (and maybe other technical departments) will support Linux.
Outside of that, its probably potluck between Windows and OS X.
If your computer runs out of memory, take a snapshot
of its state and continue on a computer with more memory.
Keep this up on newer generations of computers -- pass
it off to your ancestors -- continue until the sun turns
into a red giant if need be. Hopefully we will have developed
quantum computer which can hold more memory than there
are particles in the universe. Keep it going. Will it ever halt?
If you can determine if the following program will
ever halt then you will have solved a problem that
no mathematician has solved!
main() {
bigint n, a, b, i;
for (n = 1; ; n++) {
a = n*n;
b = (n+1)*(n+1);
for (i = a+1; i< b && !prime(i); i++)
;
if (i == b) break;
}
}
How about a tool that will tell me if my program will
eventually halt or not for a given input? I'd pay big money for that!
After all, he did invent
the "AlGoreIthm."
sputter, choke, cough
Anyone wish to explain why the energy in a giant
tidal wave does not attenuate as it travels across
the ocean? It just seems that the wave would greatly
diminish in size as it propagates.
"...[if] you have a risk of dying in the next year of well under one in 1,000, which means that if you stayed that way forever you would have a 50/50 chance of living to over 1,000. "
.3677
Bad math.
0.999^1000 =
Really you would only have slightly better than a third of a chance. I can't trust a scientist who can't do math.
It seems like it should be easy to follow
the trail to catch these guys. What's the
typical way a perp sets up a fake site?
I assume they hijack a web site, but I have
seen some where they have a TLD.org url.
Can't they find those who registered the
domain name? Why is it so impossible to
catch these people?
This is a really good point. The problem of proving
you actually used a particular algorithm in your closed
source binary would be difficult (probably theoretically
impossible, i.e., halting problem kind of thing).
So the answer is to be a LONER and SECRETIVE!
Summary.
FreeBSD, in its stable version, lacks the advance features
of Linux. The advanced versions are unstable. Use Linux instead.
NetBSD is not secure and thus shouldn't be considered for
serious computing environments.
OpenBSD is tightly controlled by a madman, thus should
be avoided.
Darwin -- only available for Mac hardware or specific Intel architectures. (aside: Buy a mac, has the best OS out there by a *long* shot).
Did anyone else notive the Intel advert.
that appeared when you held your mouse
over the "compilling results" chart?
It actually covered the results!
I am not Todd Walters, but does anyone know
****HOW**** code embedded in the image
gets executed?
No one is giving any technical details.
Toooo much ****NOISE****, not enough ****INFO****.
Actually, the flaming in usenet and elsewhere demonstrates
how badly people behave if they think they are anonymous.
..175-pound Samoan attorneys.
what was that all about anyway?
--w
Not really that obvious. For example, even though there
are infinite number of positive integers, no 3 of them satisfy
x^n + y^n = z^n when n > 2.
man page.
e.g., man glBegin.
Actually, hypertexting through this the docs found here is even better.
Most frames in Pixar movies are rendered using some form of ray-tracing. While it is possible to use vertex and fragment shaders in uncoventional ways to do ray tracing, this is *not* what the OpenGL pipeline is designed for. Great for games, but ray-tracing will still be done using render farms (and not in real time).
Yeah, the description is ambiguous. If code is embedded in a true MP3 file, how does that code get executed? If its just an application with an MP3 extension, I don't really call that a very novel trojan horse. Can anyone flush out more details?
There are plenty of "Billie Bob's" in the country where I run who feel the need to own highly aggressive dogs. The best way to deal with many of them is to get on one knee with palms up -- let them be the alpha dog -- I don't care. Of course, if this doesn't work you may have only seconds to protect you jugular -- I haven't had this problem yet.
140 miles per week!? That 20 miles per day on average!?!
No kidding? I was impressed with my lowly 30 miles
per week.
I agree -- No mp3 for me. Just shoes, socks, shorts,
and a watch. Sometimes I feel like a need pepper spray
for dogs.
--w
You can do this using flat address and paging.
Mark all stack pages as non-executable -- there
is already support for this on the protected mode
Intel chips. Segmentation not needed.
The bug involves an inconsistency with
the recorded and actual buffer size.
The buffer is allocated off the heap,
but an exploitable overflow really only
works for buffers allocated off of
the stack (where you can overwrite
a subroutine's return address and redirect
program flow). I guess if there are subroutine
addresses in struct's dynamically allocated
off of the heap, then you could redirect
program flow, but I am suspicious of this.
I had a load of research money for purchasing
an SGI a couple of years ago. SGI made it very
difficult to buy from them. Generally they will
put you in contact with a reseller, and then you
get quotes from them. This seemed to take awhile
and was awkward. I just decided not to buy from
them after a while. I am pretty happy with OpenGL
performance on my Mac, and the price/performance
of an SGI is really not that good. We'll
see what SGI has at SIGGRAPH this year. I can't believe
they haven't tanked yet.
Back in 1990 I worked for a small company that built
graphics boards and my first task was to debug
the "polygon fill" routine in their firmware.
It turns out they use their own "home brewed"
algorithm that was slow, memory hungry, and didn't
handle degenerate cases correctly. If anyone in
the company would have taken the time to pull
any one of the graphics textbooks off their shelf
(e.g., Foley, van Dam) they would find a much better
solution.
I ended up rewriting the module myself using
the classic solution -- it was faster, used little memory,
and handled degenerate cases reasonably.
It was my experience that everything was a badly
reinvented wheel when I worked there.
It's certainly not a text file.
Is the PDF format proprietary?
--w