Wrong! Yes, the CPU will sit in an idle loop and will execute instructions. However with a multiple issue processor with a big instruction window and lots of functional units, a proper idle loop shouldn't fill the window and utilize all the functional units. Laptop processors already automatically lower the clock rate...all processors will do the same in the near future
If you measured the power consumed during idle and compare it to something like SETI@home, I guarantee you will see a non-trivial difference with any modern processor (P4, Athlon, etc.).
You list of Solaris "flaws" makes me laugh. Nobody cares if Solaris doesn't include gawk! If you need gawk, then install it! Some problem eh?
Solaris is far more scalable than Linux, more reliable (yes, I did say this...I can point you to research papers), has has more enterprise features (does Linux have Intimate Shared Memory?), and as you say it, "the list goes on".
1) Get rid of 20-stage pipeline, it's too long for anything serious.
No its not. It enables a high clock rate and with good branch prediction and selective replays, it is just fine.
2) As a follow up to 1, try to actually get some work done in a clock cycle.
Read some studies about the available ILP (instruction-level parallelism) in common applications. There really isn't much of it unless instruction windows are made huge which isn't feasible. This is why simulaneous-multithreading (hyperthreading) made it into an actual chip because it takes advantage of low ILP.
3) Throw out the x86 ISA.
Pentium4 is the leader of SpecINT. Not AMD, not Sun (RISC), not IBM (RISC), not MIPS (RISC). Some players, such as MIPS, don't have the resources to compete however IBM does.
And look whats happening to Itanium? Disaster even with its oh so elegant ISA.
4) Look at the MIPS ISA.
What about it? Yes, very clean and orthogonal. Intel and AMD have proven that an ISA is irrelevant to achieving high performance.
5) Realize that it's actually possible to understand the MIPS arch, and that it still works great for multimedia, math, and general use.
Realize that undergraduate computer architecture is simply an introduction.
6) Buy the rights to the MIPS ISA, make small improvements (get rid of branch delay slot, load delay slot), speed it up, and design new Intel processors from the improved ISA.
Unnecessary. Besides, I like software compatibility and binary translation just doesn't work at this level.
7) Release versions of processors with 4MB Cache (2MB each I$, D$) for consumers, and 24MB Cache (8MB I$, 16MB D$) for servers/clustering/etc.
2MB each for I$ and D$? Then you must be referring to L1 caches which are the only ones typically separated into data and instruction. And how many cycles would it take to access those caches?
8) Release Motherboards for 1, 2, and 4 CPU configurations. 9)... 10) Profit!
Read Intel's annual report. They are quite profitable already and don't need the advice from someone with a B.S. in EE or CS and marvels at how great the MIPS ISA is.
i'm wondering if there's really that much demand out there to play moh:aa now, since that game was released over a year ago, and i personally haven't heard of any friends playing it lately. insert your diatribe below...
Yeah, and nobody plays Counterstrike anymore either.
In multiplayer popularity according to Gamespy, MOHAA is the second most popular behind Counterstrike (although a distant second).
Personally, I've purchased numerous other first-person shooters since buying MOHAA over a year ago. I've tried, but I keep going back to MOHAA for my daily fix of multiplayer action.
I hate using framemaker. Poorly-ported Mac software is always the worst. Or maybe I'm thinking of pagemaker. Who cares, it sucks.
Framemaker was originally developed on Unix.
OS/2 Laptop is provided with new mainframe
on
Bill Gates On Linux
·
· Score: 1
Ever new IBM Mainframe sold, even ones with the newly developed 390/G5 CMOS processor, comes with a laptop running OS/2 to provide system management functions.
The OP says that the the floppy disk driver NEEDS free access to every users process memory space.
This is correct. QNX runs in a flat memory space I believe.
And yes, Intel has far too many levels of run-time execution modes. A supervisor and user mode is all that is necessary.
And he is also right by saying that microkernels do too much work for simple things. This is why they are slow. However QNX gets away with it because they have no memory protection.
And no, the NT kernel is not a true "microkernel". It has kernel modules just like Linux, however device drivers run priveleged.
It's nice to see AMD, IBM, and Apple working together to defeat Intel.
Yes, lets all pray that Intel is defeated so that we have a different company that has a monopoly on mainstream microprocessors. Therefore, the existing competition that has driven down the cost of microprocessors, will disappear.
Rip on them all you want, but overall, Intel has been good for the mainstream computer industry. They generally participate in standards groups and for the most part, have an open architecture. Otherwise AMD wouldn't have a compatible chip. Yes, I realize that Intel fought this and would love to have a monopoly on x86 chips. But with the way the chips have fallen (no pun intended), they don't have this monopoly.
Intel has also generally given Linux more support than AMD has. Remember the AGP cache coherence bug on the Athlon? AMD reported the problem to Microsoft nearly immediately with a fix that appeared in Win2k SP1. However they didn't even take the time to send an e-mail to the Linux kernel mailing list and the problem wasn't discovered nor fixed for another year and a half. Anyone remember weird crashes when playing TuxRacer and an AGP card with an Athlon T-bird? I do.
And lets not forget that Jerry Sanders, the CEO (or former CEO) of AMD, supported Microsoft in the anti-trust case.
Like it or not, they are both businesses. However AMD is no more "friendly" than Intel is.
I'd say it's more a function of the service (DHCP?) than anything. You can't much simpler than DHCP. I'd never expect to see an NT domain controller, file/print server, Exchange, or IIS server make it more than a couple of months without a reboot.
I'd be happy to send you a screenshot of an NT4 box showing an uptime of over 600 days. It is used as a SQL server, file/print sharing, and the primary domain server for a small business.
In my neighborhood, in a college town, DSL is faster than cable 75% of the time. The only time my friends get decent data rates is from 2-8am. From 4-6pm, they often complain that their cable runs the same speed as a 56k modem
With my 768/128 DSL connection, I always get 80kbps and my connection has never gone down in the last 18 months.
The main reason the Opteron is a good thing is because 1) it provides MORE registers, allowing the compiler to make smarter register allocations, which can provide drastic performance improvements,
Having more logical registers in the ISA will not make much of a difference. 32-bit x86 processors from Intel and AMD already provide dozens of internal physical registers that are dynamically renamed on the fly.
A game compiled for x86-64 will run significantly faster since there are more general purpose registers available to it. So even if it doesn't make use of 64-bit ops, it will still run faster.
I'm sorry, but the 32-bit x86 processors from Intel and AMD already have dozens of internal registers that are dynamically renamed. This argument is invalid. The true advantage of 64-bit is the increased size of address space.
Every time I talk about an American innovation, a European always tells me that they were actually first. Transistors, integrated circuits, televisions, you name it!
The facts are usually twisted to give one country the credit over another. In this case, The Swedish Telecom likely had a radio system patched into land-lines. Can you verify that it was truly a cellular system?
Its not broken. It does its job, and does it well. I'll say it again: "If it ain't broken, don't fix it". Something is broken if it doesn't do a job it was intended for.
I don't use the server to store confidential data on an open network.
I'm not at all claiming that NT4 is just as robust/stable as Linux. Look, my NT4 machine will likely run for another 1600 days if the hardware doesn't fail and I don't bring it down. I'm sure someone could do the same running Linux 1.x if it exists
Certain IBM mainframe machines have uptimes of over 10 years.
The system was installed in late 2000. Its running SP5 and I haven't patched it since as its behind a firewall and is in a small organization where employees are trusted for the most part.
To be honest, it does do much beyond file serving.
The dual power supplies are on separate UPS systems.
I've got a Dell server running NT4 with an uptime of over 500 days. The nice thing about such an old OS is that it doesn't get updated every 2-6 months! And because I'm behind a firewall, I don't need to worry about the recent vulnerability.
Slashdot Mirror
Wrong! Yes, the CPU will sit in an idle loop and will execute instructions. However with a multiple issue processor with a big instruction window and lots of functional units, a proper idle loop shouldn't fill the window and utilize all the functional units. Laptop processors already automatically lower the clock rate...all processors will do the same in the near future
If you measured the power consumed during idle and compare it to something like SETI@home, I guarantee you will see a non-trivial difference with any modern processor (P4, Athlon, etc.).
You list of Solaris "flaws" makes me laugh. Nobody cares if Solaris doesn't include gawk! If you need gawk, then install it! Some problem eh?
Solaris is far more scalable than Linux, more reliable (yes, I did say this...I can point you to research papers), has has more enterprise features (does Linux have Intimate Shared Memory?), and as you say it, "the list goes on".
1) Get rid of 20-stage pipeline, it's too long for anything serious.
No its not. It enables a high clock rate and with good branch prediction and selective replays, it is just fine.
2) As a follow up to 1, try to actually get some work done in a clock cycle.
Read some studies about the available ILP (instruction-level parallelism) in common applications. There really isn't much of it unless instruction windows are made huge which isn't feasible. This is why simulaneous-multithreading (hyperthreading) made it into an actual chip because it takes advantage of low ILP.
3) Throw out the x86 ISA.
Pentium4 is the leader of SpecINT. Not AMD, not Sun (RISC), not IBM (RISC), not MIPS (RISC). Some players, such as MIPS, don't have the resources to compete however IBM does.
And look whats happening to Itanium? Disaster even with its oh so elegant ISA.
4) Look at the MIPS ISA.
What about it? Yes, very clean and orthogonal. Intel and AMD have proven that an ISA is irrelevant to achieving high performance.
5) Realize that it's actually possible to understand the MIPS arch, and that it still works great for multimedia, math, and general use.
Realize that undergraduate computer architecture is simply an introduction.
6) Buy the rights to the MIPS ISA, make small improvements (get rid of branch delay slot, load delay slot), speed it up, and design new Intel processors from the improved ISA.
Unnecessary. Besides, I like software compatibility and binary translation just doesn't work at this level.
7) Release versions of processors with 4MB Cache (2MB each I$, D$) for consumers, and 24MB Cache (8MB I$, 16MB D$) for servers/clustering/etc.
2MB each for I$ and D$? Then you must be referring to L1 caches which are the only ones typically separated into data and instruction. And how many cycles would it take to access those caches?
8) Release Motherboards for 1, 2, and 4 CPU configurations.
9)
10) Profit!
Read Intel's annual report. They are quite profitable already and don't need the advice from someone with a B.S. in EE or CS and marvels at how great the MIPS ISA is.
UWisc was just named the #2 party school in the nation. I wouldn't feel so sorry for them!
i'm wondering if there's really that much demand out there to play moh:aa now, since that game was released over a year ago, and i personally haven't heard of any friends playing it lately. insert your diatribe below...
Yeah, and nobody plays Counterstrike anymore either.
In multiplayer popularity according to Gamespy, MOHAA is the second most popular behind Counterstrike (although a distant second).
Personally, I've purchased numerous other first-person shooters since buying MOHAA over a year ago. I've tried, but I keep going back to MOHAA for my daily fix of multiplayer action.
I hate using framemaker. Poorly-ported Mac software is always the worst. Or maybe I'm thinking of pagemaker. Who cares, it sucks.
Framemaker was originally developed on Unix.
Ever new IBM Mainframe sold, even ones with the newly developed 390/G5 CMOS processor, comes with a laptop running OS/2 to provide system management functions.
Wow. Why this is moderated "Troll" blows me away.
The OP says that the the floppy disk driver NEEDS free access to every users process memory space.
This is correct. QNX runs in a flat memory space I believe.
And yes, Intel has far too many levels of run-time execution modes. A supervisor and user mode is all that is necessary.
And he is also right by saying that microkernels do too much work for simple things. This is why they are slow. However QNX gets away with it because they have no memory protection.
And no, the NT kernel is not a true "microkernel". It has kernel modules just like Linux, however device drivers run priveleged.
It's nice to see AMD, IBM, and Apple working together to defeat Intel.
Yes, lets all pray that Intel is defeated so that we have a different company that has a monopoly on mainstream microprocessors. Therefore, the existing competition that has driven down the cost of microprocessors, will disappear.
Rip on them all you want, but overall, Intel has been good for the mainstream computer industry. They generally participate in standards groups and for the most part, have an open architecture. Otherwise AMD wouldn't have a compatible chip. Yes, I realize that Intel fought this and would love to have a monopoly on x86 chips. But with the way the chips have fallen (no pun intended), they don't have this monopoly.
Intel has also generally given Linux more support than AMD has. Remember the AGP cache coherence bug on the Athlon? AMD reported the problem to Microsoft nearly immediately with a fix that appeared in Win2k SP1. However they didn't even take the time to send an e-mail to the Linux kernel mailing list and the problem wasn't discovered nor fixed for another year and a half. Anyone remember weird crashes when playing TuxRacer and an AGP card with an Athlon T-bird? I do.
And lets not forget that Jerry Sanders, the CEO (or former CEO) of AMD, supported Microsoft in the anti-trust case.
Like it or not, they are both businesses. However AMD is no more "friendly" than Intel is.
Don't forget that GPS is a CDMA-like system. Effectively blocking a large spectrum is not easy nor cheap.
I'd say it's more a function of the service (DHCP?) than anything. You can't much simpler than DHCP. I'd never expect to see an NT domain controller, file/print server, Exchange, or IIS server make it more than a couple of months without a reboot.
I'd be happy to send you a screenshot of an NT4 box showing an uptime of over 600 days. It is used as a SQL server, file/print sharing, and the primary domain server for a small business.
I've found the OEM fans included with AMD's retail chips are lousy. They spin at 5000RPM making them noisy and tend to get noiser over time and fail.
I much prefer a larger 80mm fan spinning at 2500rpm to cool my processor.
In my neighborhood, in a college town, DSL is faster than cable 75% of the time. The only time my friends get decent data rates is from 2-8am. From 4-6pm, they often complain that their cable runs the same speed as a 56k modem
With my 768/128 DSL connection, I always get 80kbps and my connection has never gone down in the last 18 months.
The main reason the Opteron is a good thing is because 1) it provides MORE registers, allowing the compiler to make smarter register allocations, which can provide drastic performance improvements,
Having more logical registers in the ISA will not make much of a difference. 32-bit x86 processors from Intel and AMD already provide dozens of internal physical registers that are dynamically renamed on the fly.
A game compiled for x86-64 will run significantly
faster since there are more general purpose
registers available to it. So even if it doesn't
make use of 64-bit ops, it will still run faster.
I'm sorry, but the 32-bit x86 processors from Intel and AMD already have dozens of internal registers that are dynamically renamed. This argument is invalid. The true advantage of 64-bit is the increased size of address space.
I have a 12-year old HP Series II laser printer that is still running. The page counter shows over a million pages printed.
"They don't make 'em like they used to"
Every time I talk about an American innovation, a European always tells me that they were actually first. Transistors, integrated circuits, televisions, you name it!
The facts are usually twisted to give one country the credit over another. In this case, The Swedish Telecom likely had a radio system patched into land-lines. Can you verify that it was truly a cellular system?
United States surrenders to Iraq!
At my previous job at a Fortune-100 company, the proprietary bug database we used kept track of the Bug ID with an unsigned 16-bit number.
Yep, to the surprise and dismay of many, we overflowed at bug# 65537
Its not broken. It does its job, and does it well. I'll say it again: "If it ain't broken, don't fix it". Something is broken if it doesn't do a job it was intended for.
I don't use the server to store confidential data on an open network.
I'm not at all claiming that NT4 is just as robust/stable as Linux. Look, my NT4 machine will likely run for another 1600 days if the hardware doesn't fail and I don't bring it down. I'm sure someone could do the same running Linux 1.x if it exists
Certain IBM mainframe machines have uptimes of over 10 years.
The system was installed in late 2000. Its running SP5 and I haven't patched it since as its behind a firewall and is in a small organization where employees are trusted for the most part.
To be honest, it does do much beyond file serving.
The dual power supplies are on separate UPS systems.
I've got a Dell server running NT4 with an uptime of over 500 days. The nice thing about such an old OS is that it doesn't get updated every 2-6 months! And because I'm behind a firewall, I don't need to worry about the recent vulnerability.
If it ain't broke, don't fix it!
Its not a war. The last declaration of war by the United States occurred December 8th, 1941.
It is a conflict!! The media is fixated on using the word "war".
Ok, I'll load my 30-year old Canon with some Kodak Technical Pan film. Lets make 16x20" enlargements and see how we compare, huh?
Or, lets take wide-angle pictures. With the cropping factor on your Nikon D1X, how can you be any wider than say 32mm (35mm equivalent).
Digital is great, but in some cases, 35mm cameras are still superior. Especially low-light and wide-angle photography.