But disk devices are connected via host adtoptar to the PCI bus of an computer - even USB controllers are attached to an PCI bus. To make it short: Really *ALL* i/o devices in these days desktop computers are connected to an PCI bus, sometimes with additional bridges, but in the end all i/o data streams go through PCI to the CPU(s).
So you can't kill PCI if you don't have an alternative - wich USB *IS NOT*.
Is this Linux-News? No, not really: It's not Linux specific at all.
This affects other projects as well---projects without the commercial support of IPO'd distributors or big companies---so IMHO it shouldn't have been posted in the Linux-section where a lot of people won't see it.
So please Sun, make me a believer and ship a 1000 CPU box to me as well. If this is not possible right now, a 100 CPU box would be fine for the start. Or 10...
> But until batteries get better and/or they stop putting moving parts (i.e., hard drives) in > laptops, there's a niche for machines like the z50.
Hard drives are not the problem these days. Old laptops with old NiCD-Batteries (~1994) had harddrives as well. Now we have modern hardrives wich need slightly less power and modern batteries wich can store much more energy (more than three times as much) but average battery times haven't improved at all (I have the feeling, it has gotten worse.)
Why? Because those "modern" CPUs you find in most laptops and the bigger displays have eaten up all advantages made in battery-developement.
Some numbers: IBM's latest and greates Travelstar 25GS (25MB): 2.5W(reading), 2.9W(writing), 2W (idle but rotating). Intel's Mobile-Pentium-III 500MHz: 8W
> you can pick up 2nd hand Indies for anywhere between $200 and $1500,- > on ebay and in 2ndhand computer stores, so most people can afford to run a desktop on something
Yes, I know. I'm looking for a used Sun Sparc or SGI (and even VAX!) for quite some time now, but it's not that easy here. (But I want one more for the hack value than for real work.)
BTW: I've used ARM2 and later StrongARM-driven Computers before I had to switch to x86 1 1/2 years ago. I still have them, but a NetBSD "make build" needs nearly one day on a 277MHz StrongARM (overclocked).
> else then x86... Oh, and it runs linux as well if you want..
As a second choice maybe.
> Anyway, RISC is not new, its in fact quite likely to be older then the average age of the posters on > slashdot, so do not call it new technology or such
I haven't said that RISC is new. I 've just called x86 old crap.
> maybe newer then plain cisc, but hybrid cpus have been around since at least end 70s and early 80s
Some dates: First commercial CPU: 1971 (Intel 4004). First commercial RISC CPU: 1987 (Acorn ARM2).
So RISC is still quite new in the microprocessor universe.
> Not that risc is not cool, it is, but it isn't the coolest thing since sliced bread or whatever,
I haven't said that. I've said "Crusoe is by far not as cool as those pure RISC CPUs".
This makes me smile :-)
on
SSH v. SRP
·
· Score: 1
If I ever find two lifes, I promise, I will give one to you. (fingers crossed)
And to babelfish something back:
On the search for the missing persons the space cruiser " the Unbesiegbare " pulls comrades by the ALL. On Regis III the astronauts meet puzzling remnants of a high-technical culture. Dead subject, according to long senseless rules become programmed, pursues and sells the intruders.
A crueso+the more of RAM it needs is bigger and sucks much more energy than a MIPS or [Strong]ARM.
And don't forget, the Crusoe is just s a dull and boring x86. It has some interesting internals, but from the outside it's not much more than a pentium compatible CPU wich needs extra RAM to run.
So from my POV Crusoe is *older* and by far not as "cool" as those pure RISC CPUs.
The x86/IA32-compatibility is the only advantage Crusoe has to MIPS or StrongARM. But we don't need - no - we don't *WANT* x86-compatibility for our PDAs. Let's get rid of this old crap: x86: DIE, DIE, DIE!
It's too bad that most of us (including me) can't afford something better than x86 for our desktops at the moment, but don't let this cancer spread to the PDAs.
(And no, that Crusoe was designed by a company where Linus works doesn't make it any better!)
And for those of you who really don't speak German
on
SSH v. SRP
·
· Score: 1
This is actually a German translation of this article made with babelfish.altavista.com.
Trolls aren't very creative these day I must say. What do we get next?
It is acutally a "real" plant, but most of its time it behaves exactly like a roboplant (does nothing and watering it too much kills it.)
Whatch this place for the announcement of the incredible cycas cam! Watch a real cycas revoluta *not* growing, it's amazing! (When ever new leaves come out of it, we promise to exchange it with a "sleeping" exemplar.)
The FreeBSD web server is "www.freebsd.org". It runs apache, but it could even run thttpd, ncsa, w3c-httpd - it doesn't matter at all. "www.freebsd.org" is still the FreeBSD web server.
FUD, it's no problem to dualboot *BSD and Linux, there are tons of How-Tos.
Since we are talking of FreeBSD:
Go to www.freebd.org, goto "FAQ" and geep going on to "Installation", the question "Can I have more than one operation system on my PC?" and finaly to "the multi-OS page".
I'm no FreeBSD user, I don't know the FreeBSD web server very well, but I found that document in less than 2 minutes.
> OpenBSD and NetBSD are almost exactly the same OS, but configured differently.
No, they're not. They're still similar in many respects but not "almost exactly the same" at all. E.g. OpenBSD and NetBSD use a completely different VM.
> BSD/dev is quite different to Linux. BSD doesn't have/proc.
Wrong. It's just not mounted or enabled as default in NetBSD and OpenBSD for security reasons.
> The biggest cause of differences is differing versions of the standard C libraries. All GNU/Linuxes, > and GNU/HURD, use glibc. BSD has its own libc. As SunOS, HP-UX... have.
> This means that a fair few standard functions work quite differently. In particular, GNU > extensions, like the %a flag for scanf, will be missing from BSD.
Wich means that GNU does the same as Microsoft when they add their own "features" to an official standard. The BSDs' scanf conforms to ANSI X3.159-1989 aka "ANSI C" so it's not their fault if some Linux software won't compile or run because of GNU weakening standards.
> As for their Aerospace expertise, ever heard of Ariane? It can launch heavier payloads to higher > orbits than the shuttle. That's pretty important for geosyncronous orbits/communications > satellites. The only thing the US has in that department are old Saturn rockets from the 70's.
I'm a big fan of the Ariane, but all of this is wrong.
1. The shuttle can launch much more than the Ariane V into LEO: 24.4t (204km orbit) vs 18t (185km)
For higher orbits (i.e. geosynchronous) satellites needs much bigger kick-engines when launched with the shuttle, since Ariane already inserts its payload into eliptic GTOs, the shuttle can't do that. This makes ~5.9t (into GTO) for the shuttle vs 6.8t for Ariane V.
2. The American Titan 4B can launch more than the Ariane V: LEO 21.64t (185) GTO (8620).
3. The US has no old Saturn rockets that could be launched anymore and can't build new ones:-).
The big pluses of Ariane that makes Arianespace the leading launcher are: 1. Price 2. Reliability of service (e.g. no hurricanes in Kourou wich could delay launch (and probably cost the contraction $$s)) 3. Precision of orbit injection (wich saves satellite fuel and enhances satellite lifetime) 4. "contract to orbit" time But not power.
Have a look at http://www.friends-partners.org/~mwade/spaceflt. htm
It's a great page with detailled information about all known launch systems in the world.
BTW: France as nation (the French government etc) is not involved *AT*ALL* in this lawsuit. So people calling France stupid and hypocritical because of this must be *really* stupid since they can't even read and understand a short fragment of text.
Slashdot Mirror
But disk devices are connected via host adtoptar to the PCI bus of an computer - even USB controllers are attached to an PCI bus. To make it short: Really *ALL* i/o devices in these days desktop computers are connected to an PCI bus, sometimes with additional bridges, but in the end all i/o data streams go through PCI to the CPU(s).
So you can't kill PCI if you don't have an alternative - wich USB *IS NOT*.
> The entire idea behind USB is to kill off PCI. :-)
Really: I doubt that
Connect SCSI/IDE host adaptors or 100TX NICs with USB to your motherboard?
No, thank you.
Is this Linux-News?
No, not really: It's not Linux specific at all.
This affects other projects as well---projects without the commercial support of IPO'd distributors or big companies---so IMHO it shouldn't have been posted in the Linux-section where a lot of people won't see it.
Yes, I have doubts too!
So please Sun, make me a believer and ship a 1000 CPU box to me as well. If this is not possible right now, a 100 CPU box would be fine for the start. Or 10...
> But until batteries get better and/or they stop putting moving parts (i.e., hard drives) in
> laptops, there's a niche for machines like the z50.
Hard drives are not the problem these days.
Old laptops with old NiCD-Batteries (~1994) had harddrives as well. Now we have modern hardrives wich need slightly less power and modern batteries wich can store much more energy (more than three times as much) but average battery times haven't improved at all (I have the feeling, it has gotten worse.)
Why? Because those "modern" CPUs you find in most laptops and the bigger displays have eaten up all advantages made in battery-developement.
Some numbers:
IBM's latest and greates Travelstar 25GS (25MB): 2.5W(reading), 2.9W(writing), 2W (idle but rotating).
Intel's Mobile-Pentium-III 500MHz: 8W
MVS = "Multiple Virtual Storage", an OS for the IBM 390. It was later renamed to "OS/390".
VMS = "Virtuel Memory System", mad by DEC for their VAXens and later Alphas (OpenVMS).
> you can pick up 2nd hand Indies for anywhere between $200 and $1500,-
> on ebay and in 2ndhand computer stores, so most people can afford to run a desktop on something
Yes, I know. I'm looking for a used Sun Sparc or SGI (and even VAX!) for quite some time now, but it's not that easy here. (But I want one more for the hack value than for real work.)
BTW: I've used ARM2 and later StrongARM-driven Computers before I had to switch to x86 1 1/2 years ago. I still have them, but a NetBSD "make build" needs nearly one day on a 277MHz StrongARM (overclocked).
> else then x86... Oh, and it runs linux as well if you want..
As a second choice maybe.
> Anyway, RISC is not new, its in fact quite likely to be older then the average age of the posters on
> slashdot, so do not call it new technology or such
I haven't said that RISC is new.
I 've just called x86 old crap.
> maybe newer then plain cisc, but hybrid cpus have been around since at least end 70s and early 80s
Some dates:
First commercial CPU: 1971 (Intel 4004).
First commercial RISC CPU: 1987 (Acorn ARM2).
So RISC is still quite new in the microprocessor universe.
> Not that risc is not cool, it is, but it isn't the coolest thing since sliced bread or whatever,
I haven't said that. I've said "Crusoe is by far not as cool as those pure RISC CPUs".
If I ever find two lifes, I promise, I will give one to you. (fingers crossed)
And to babelfish something back:
And for our French friends:A crueso+the more of RAM it needs is bigger and sucks much more energy than a MIPS or [Strong]ARM.
And don't forget, the Crusoe is just s a dull and boring x86. It has some interesting internals, but from the outside it's not much more than a pentium compatible CPU wich needs extra RAM to run.
So from my POV Crusoe is *older* and by far not as "cool" as those pure RISC CPUs.
The x86/IA32-compatibility is the only advantage Crusoe has to MIPS or StrongARM. But we don't need - no - we don't *WANT* x86-compatibility for our PDAs. Let's get rid of this old crap: x86: DIE, DIE, DIE!
It's too bad that most of us (including me) can't afford something better than x86 for our desktops at the moment, but don't let this cancer spread to the PDAs.
(And no, that Crusoe was designed by a company where Linus works doesn't make it any better!)
This is actually a German translation of this article made with babelfish.altavista.com.
.`?
Trolls aren't very creative these day I must say.
What do we get next?
`strings netscape.core`?
`fortune -a -m
Cycas revoluta, plant of the geeks.
It is acutally a "real" plant, but most of its
time it behaves exactly like a roboplant (does nothing and watering it too much kills it.)
Whatch this place for the announcement of the incredible cycas cam! Watch a real cycas revoluta *not* growing, it's amazing! (When ever new leaves come out of it, we promise to exchange it with a "sleeping" exemplar.)
> These numbers total a mere 8688. What accounts for the missing 1312 stories? Duplicates?
No, the other 1312 stories were submitted by the authors who didn't made at into the "Top 5".
(OctobrX, CowboyNeal, emmet, jamie, JonKatz etc.).
Excuse me for not reading carefully enough.
Sorry, sorry, sorry and thousend more sorries.
(Damn, what idiot I am!)
Sorry, but putting images for this purpose in a database is plain stupid IMHO.
Store them as normal files, accessible throuh
"normal" static and *cacheable* URLs and put the filenames into the database.
I loved them! Too bad that they nearly completely disappeared from the German market a few years ago. :-(
(I bet that the swiss food-mafia (Nestle') is responsible for this!)
The FreeBSD web server is "www.freebsd.org".
It runs apache, but it could even run thttpd, ncsa, w3c-httpd - it doesn't matter at all.
"www.freebsd.org" is still the FreeBSD web server.
You are stupid, aren't you?
> Nowhere is stated it's a LINUX only ANTI-REDMOND site. Or better: PRO-Linux MEANS ANTI-MS.
> Or am I wrong?
As BSD user I've the strong feeling that "pro Linux" means "anti everything else"...
- offtopic
- trollwork to distract people from trying FreeBSD
- FUD, it's no problem to dualboot *BSD and Linux, there are tons of How-Tos.
Since we are talking of FreeBSD:Go to www.freebd.org, goto "FAQ" and geep going on to "Installation", the question "Can I have more than one operation system on my PC?" and finaly to "the multi-OS page".
I'm no FreeBSD user, I don't know the FreeBSD web server very well, but I found that document in less than 2 minutes.
> by this argument, we should all be using sysV or whatever.
"whatever" - maybe
"sysV" - no, BSD (first distribution 1978/79) is older than System V (1983), wich was strongly influenced by BSD BTW.
Wich Devil?
> OpenBSD and NetBSD are almost exactly the same OS, but configured differently.
/dev is quite different to Linux. BSD doesn't have /proc.
... have.
No, they're not. They're still similar in many respects but not "almost exactly the same" at all. E.g. OpenBSD and NetBSD use a completely different VM.
> BSD
Wrong. It's just not mounted or enabled as default in NetBSD and OpenBSD for security reasons.
> The biggest cause of differences is differing versions of the standard C libraries. All GNU/Linuxes,
> and GNU/HURD, use glibc. BSD has its own libc.
As SunOS, HP-UX
> This means that a fair few standard functions work quite differently. In particular, GNU
> extensions, like the %a flag for scanf, will be missing from BSD.
Wich means that GNU does the same as Microsoft when they add their own "features" to an official standard. The BSDs' scanf conforms to ANSI X3.159-1989 aka "ANSI C" so it's not their fault if some Linux software won't compile or run because of GNU weakening standards.
The numbers are from a Microsoft memo, so ask them to name some of those bugs.
> As for their Aerospace expertise, ever heard of
:-).
. htm
Ariane? It can launch heavier payloads to higher
> orbits than the shuttle. That's pretty important for geosyncronous orbits/communications
> satellites. The only thing the US has in that department are old Saturn rockets from the 70's.
I'm a big fan of the Ariane, but all of this is wrong.
1. The shuttle can launch much more than the Ariane V into LEO: 24.4t (204km orbit) vs 18t (185km)
For higher orbits (i.e. geosynchronous) satellites needs much bigger kick-engines when launched with the shuttle, since Ariane already inserts its payload into eliptic GTOs, the shuttle can't do that. This makes ~5.9t (into GTO) for the shuttle vs 6.8t for Ariane V.
2. The American Titan 4B can launch more than the Ariane V: LEO 21.64t (185) GTO (8620).
3. The US has no old Saturn rockets that could be launched anymore and can't build new ones
The big pluses of Ariane that makes Arianespace the leading launcher are:
1. Price
2. Reliability of service (e.g. no hurricanes in Kourou wich could delay launch (and probably cost the contraction $$s))
3. Precision of orbit injection (wich saves satellite fuel and enhances satellite lifetime)
4. "contract to orbit" time
But not power.
Have a look at
http://www.friends-partners.org/~mwade/spaceflt
It's a great page with detailled information about all known launch systems in the world.
Thank you for making "France sues" out of Times' "French to sue" headline, Hemos.
n -earth-and-can-and-will-do-everything-th ey-want
This is good journalism!
That's about the same as writing (say) "USA has killed peoples in Rome" instead of "Americans have shot two men in Rome".
And now we have the usuall France-is-stupid-and-USA-is-the-greatest-nation-o
type of discussion.
BTW: France as nation (the French government etc) is not involved *AT*ALL* in this lawsuit. So people calling France stupid and hypocritical because of this must be *really* stupid since they can't even read and understand a short fragment of text.