Yes, that is the case. Is there really a need to punish the other countries for it? If they an opinion, then the other countries should respect it and get on with it. Face it, the chances of us all agreeing on something are basically zero.
This has also happened to the US. When WW2 started in the 1939 - America did not join the war because public opinion was against it. Sure, you helped Europe with supplies but you didn't get directly involved.
Now France doesn't get involved due to public opinion. That is their opinion, and they are entitled to it!
You only joined WW2 because Japan attacked you, and German declared war on you. I don't "hold it against" the US, as I know I'd really not like to get involved myself in such a situation.
The key point I was trying to make was to explain that the Opteron should have a performance boost in 64-bit mode, the G5 is the same so isn't as likely. Therefore if the Opteron is better at 32-bit, then it is likely to be even better at 64-bit.
I understand the RISC/CISC thing. The Itanium 2 has 128 registers I think! The thing I've found with RISC code is that because of the nature of it if is longer, which puts more strain on the cache. At work I deal with SPARC & x86 binaries, and the x86 binaries are/always/ a lot smaller. I'm not sure about the G5 though.
Also due to the AMD & Intel understanding x86 very well despite all of it's flaws and qwirks, they are able to get processors running it very quickly. The G5 is a cleaner architecture, but it isn't as well exposed as x86 so between AMD and x86 compilers they can do a very good job of making a bad ISA run well.
I think the decision to go with 16 registers is that it was a lot more hacky to go with more than 2x. AMD decided 16 was enough for most things, and went with that.
As for Intel, they do have an x86-64 license so they could use it if they want. Microsoft have said they will only support IA64 and x86-64, so I think that'll stop Intel from making x86-64-IA!
x86 mode has to remain compatabile with every other x86 processor since the 386. As such it has 8 int and 8 floating point registers. When AMD's chips are in 64-bit mode, you get 16 int and 16 fp registers.
The G5 doesn't get any 64-bit benefit, the Athlon64 does. Also remember, 64-bit more is normally harder on caches as the bit-length of everything (like pointers) is doubled also.
Of course, in my experience, that's often way too much to ask of IT.
If you are asking IT then you are probably not in the IT dept. If you are in IT, then perhaps you should be asking why you can't do it yourself?
If you're too busy to support standards, and/or cannot distinguish such things from platforms, then it's obvious you're not qualified to be a sys-admin.
I'm sorry, this is just crap. I do support standards, the corporate desktop standard. I work for a company with 20000 thousand PCs, do you presume we just allow them to run anything they want, and let them install it only for us to try and fix their problems? If we did, we'd spend all of our time fixing them!
We do run 3 OSes, and I probably could handle any other Unix-esq. But by your definition, if someone brings in a BeOS/Amiga/whatever and I can't bring it onto the network then I'm crap.
Ultimately it is pointless, as I wouldn't allow them to do it because:
1. I work in R&D, we don't allow people to bring stuff onto our network at their will. Our IP is too important. They could easily take a copy of it on their machine and walk out.
2. They could have viruses on their machines.
3. They could be running packet sniffer, or launching $LATEST_OS_VULN against my machines.
4. If it was MacOSX, I don't support it.
If one of our corporate desktops was not picking up a DHCP lease then I would fix it. But as long as the production machines are working, I'm not going to care whatever problems are occuring with people's various stuff.
Well I presume that the desktop PCs are getting IP addresses just fine, otherwise nothing would work.
It is obvious that you are not a sys-admin from your above comment also.
As a sys-admin, if someone bought a Mac in they'd be on their own. I support 3 OSes as it is, and I'd rather my time was taken up trying to learn 3 BSDs, BeOS, Mac OSX, etc when we I've got a damned lot of work to do as well.
Yes, please let me drop everything to support $DIFFERENT_OS as I've not got enough to be going on with anyway.
Lord Of The Rings gets released everywhere on the same day doesn't it?
Whatever the differences with the books are, the first two have still been pretty damned good! By youe definition, anything like that would be instantly marked as rubbish whether it is or not!
Use the Red Hat Network Update Agent sometime if you want to see an updating process that's REALLY a pain in the ass.
Yeah, I mean, cos typing "up2date -u -f --nox" to update all of my machines packages in just one reboot is really difficult to do.
Unlike multiple-reboot Windows update. I hope you don't try to update SP1, IE6 or WMP at the same time. 'cos it doesn't allow that!
Actually Sun do do that! We've got some Sun Ultra60s that we are giving away to a University. They are going to have be wiped and we'll let the Uni sort out the OS.
Of course, they can just go and d/l Solaris anyway so it isn't a big issue as long as that remains. But like a lot of things, it appears that Solaris is not transferable!
In the high end markets, RISC CPUs from HP, SGI, IBM and Sun still dominate. x86 has never been able to reach these performance levels even though they are sometimes a process generation or two ahead. RISC vendors will always be able to make a faster, smaller CPUs. Intel however can make many more CPUs for less.
Lets see. HP and SGI have sold themselves to the Itanium 2 which is ok but is EPIC, not RISC. IBM have the Power4+ and Power5 on the way which are pretty damned good. Sun have the US3 and US3i.
Sun certainly don't have a performance lead. I've benchmarked an Opteron against the US3 and US3i and it isn't pretty for Sun. The Opteron is actually MORE efficient clock for clock in 95% of tests I ran. And yes, all the tests in question were real world programs running real world data.
So I disagree. Just because it is RISC, doesn't mean it is faster. Chips like the Alpha prove it is possible, but with the rate of developlment of x86 and it's compilers it is becoming more difficult for them to keep up.
Slashdot Mirror
Yes, that is the case. Is there really a need to punish the other countries for it? If they an opinion, then the other countries should respect it and get on with it. Face it, the chances of us all agreeing on something are basically zero.
This has also happened to the US. When WW2 started in the 1939 - America did not join the war because public opinion was against it. Sure, you helped Europe with supplies but you didn't get directly involved.
Now France doesn't get involved due to public opinion. That is their opinion, and they are entitled to it!
You only joined WW2 because Japan attacked you, and German declared war on you. I don't "hold it against" the US, as I know I'd really not like to get involved myself in such a situation.
Only on slashdot would people mod-up a post with "MAJOR SPOILER" in it!!!
Got to say, great post. Very well thought out.
/always/ a lot smaller. I'm not sure about the G5 though.
The key point I was trying to make was to explain that the Opteron should have a performance boost in 64-bit mode, the G5 is the same so isn't as likely. Therefore if the Opteron is better at 32-bit, then it is likely to be even better at 64-bit.
I understand the RISC/CISC thing. The Itanium 2 has 128 registers I think! The thing I've found with RISC code is that because of the nature of it if is longer, which puts more strain on the cache. At work I deal with SPARC & x86 binaries, and the x86 binaries are
Also due to the AMD & Intel understanding x86 very well despite all of it's flaws and qwirks, they are able to get processors running it very quickly. The G5 is a cleaner architecture, but it isn't as well exposed as x86 so between AMD and x86 compilers they can do a very good job of making a bad ISA run well.
I think the decision to go with 16 registers is that it was a lot more hacky to go with more than 2x. AMD decided 16 was enough for most things, and went with that.
As for Intel, they do have an x86-64 license so they could use it if they want. Microsoft have said they will only support IA64 and x86-64, so I think that'll stop Intel from making x86-64-IA!
Yes, because x86 and x86-64 ARE NOT the same.
x86 mode has to remain compatabile with every other x86 processor since the 386. As such it has 8 int and 8 floating point registers. When AMD's chips are in 64-bit mode, you get 16 int and 16 fp registers.
The G5 doesn't get any 64-bit benefit, the Athlon64 does. Also remember, 64-bit more is normally harder on caches as the bit-length of everything (like pointers) is doubled also.
We do run 3 OSes, and I probably could handle any other Unix-esq. But by your definition, if someone brings in a BeOS/Amiga/whatever and I can't bring it onto the network then I'm crap.
Ultimately it is pointless, as I wouldn't allow them to do it because:
1. I work in R&D, we don't allow people to bring stuff onto our network at their will. Our IP is too important. They could easily take a copy of it on their machine and walk out.
2. They could have viruses on their machines.
3. They could be running packet sniffer, or launching $LATEST_OS_VULN against my machines.
4. If it was MacOSX, I don't support it.
If one of our corporate desktops was not picking up a DHCP lease then I would fix it. But as long as the production machines are working, I'm not going to care whatever problems are occuring with people's various stuff.
Well I presume that the desktop PCs are getting IP addresses just fine, otherwise nothing would work.
It is obvious that you are not a sys-admin from your above comment also.
As a sys-admin, if someone bought a Mac in they'd be on their own. I support 3 OSes as it is, and I'd rather my time was taken up trying to learn 3 BSDs, BeOS, Mac OSX, etc when we I've got a damned lot of work to do as well.
Yes, please let me drop everything to support $DIFFERENT_OS as I've not got enough to be going on with anyway.
I accept your point, but....
Lord Of The Rings gets released everywhere on the same day doesn't it?
Whatever the differences with the books are, the first two have still been pretty damned good! By youe definition, anything like that would be instantly marked as rubbish whether it is or not!
Well, decade = 10 years. Therefore decades has to be at least 20 years ago.
Sun were formed in Feburary 82, so by September 83 they must have had a 64-bit OS. Pretty good going!
</ULTRA_PENDANTIC>
This is a totally pointless post... but I just love the parent post! Too cool.
On the top of our alien overlords....
I, for one, welcome the stay of our brain eating alien overlords!
In CAMBRIDGE ENGLAND, Tesco experiment on YOU!
I, for one, welcome our new emerald overlords!
Wow! BA use Athlons?!
I'm only flying with them in the future now!
(even though I don't like flying, I'll overcome that if they use Athlons!)
Actually Sun do do that! We've got some Sun Ultra60s that we are giving away to a University. They are going to have be wiped and we'll let the Uni sort out the OS.
Of course, they can just go and d/l Solaris anyway so it isn't a big issue as long as that remains. But like a lot of things, it appears that Solaris is not transferable!
More stable, yes, as long as the hardware is ok.
As for faster? No, I don't believe it. What is it actually faster at?
I've done a lot of tests with various x86 and SPARC CPUs on lots of various EDA tools run real-world EDA data and the x86 wins easily currently.
SPARC/Solaris may have many things going for it, but out-right speed isn't one of them.
I can't resist!
.....
5.
6. Profit!!! (hopefully at M$'s expense of course)
From xiph.org....
The best reference I've found is Dave Jones' website..... Linux 2.5 core updates.
Mr T's van is a helluva fast foo!
Not only that, the DVD was out about a month ago as well.
Time the folks in the US to look-up on-line stores that ship region 2 stuff to the US!
All AIs sound the same to me.
If you've not played Deus Ex, this will not make sense. Go and play Deus Ex, it is damn fine game.
Well done that AC! Nothing beats the weekly Strong Bad email.
Especially when it is.... Trogdor The Burninator!
You can play the game to... Trogdor The Game!
What about the AMD Athlon64/Opteron? Soon Windows will have a 64-bit consumer version, and Linux already does have.
:-)
And you get x86 compatability too. I'll leave it upto the reader if that is a good/bad thing!