microprocessors
-
he, what makes one mp more intelligent than others.the instruction set or the electronics part because in the alien ad of intel they say it is the smartest? technical- does non intel mps too have the data segment,code segment etc?is the address resolution as complicated? bye.
-
he, what makes one mp more intelligent than others.the instruction set or the electronics part because in the alien ad of intel they say it is the smartest? technical- does non intel mps too have the data segment,code segment etc?is the address resolution as complicated? bye.
The ad is 100% pure marketing. It has absolutely nothing to do with the actual capabilities of the CPU. The segmented addressing model of IA32 processors (Intel, AMD, VIA, Transmeta or any other) is an useless heritage from the age of 16-bit CPUs. Non-IA32 processors (even those from Intel) don't use segmenting.
-
The ad is 100% pure marketing. It has absolutely nothing to do with the actual capabilities of the CPU. The segmented addressing model of IA32 processors (Intel, AMD, VIA, Transmeta or any other) is an useless heritage from the age of 16-bit CPUs. Non-IA32 processors (even those from Intel) don't use segmenting.
I disagree totally. In my opinion, segmented addressing is one of the absolutely best aspects of the x86 architecture. That is, now that segments can be up to 4GB. I think this because segmentation allows the processor to implement address protection in the hardware which I consider to be a very good thing. The software alternatives are considerably slower. However, I also think that segmentation was something of a pain when the segment size was limited to 16 bits (64K). The reason I think protected address spaces are important is that I have seen far too much software running on older *nix systems crash for no apparent reason and it is very difficult to track down the cause of the crash when one process with a stray pointer can cause complete havoc.
-
I disagree totally. In my opinion, segmented addressing is one of the absolutely best aspects of the x86 architecture. That is, now that segments can be up to 4GB. I think this because segmentation allows the processor to implement address protection in the hardware which I consider to be a very good thing. The software alternatives are considerably slower. However, I also think that segmentation was something of a pain when the segment size was limited to 16 bits (64K). The reason I think protected address spaces are important is that I have seen far too much software running on older *nix systems crash for no apparent reason and it is very difficult to track down the cause of the crash when one process with a stray pointer can cause complete havoc.
Rick York wrote: I disagree totally. In my opinion, segmented addressing is one of the absolutely best aspects of the x86 architecture. You haven't been doing much general programming on x86 the last 10 years, have you? I think this because segmentation allows the processor to implement address protection in the hardware which I consider to be a very good thing. Ehhh, like the MMU has been doing since the days of the 80386 you mean? The reason I think protected address spaces are important is... Here you are switching to a completely different matter. Of course any general purpose OS is to have protected address spaces, but that in no way implies they should have the horrible segmented addressing!
-
Rick York wrote: I disagree totally. In my opinion, segmented addressing is one of the absolutely best aspects of the x86 architecture. You haven't been doing much general programming on x86 the last 10 years, have you? I think this because segmentation allows the processor to implement address protection in the hardware which I consider to be a very good thing. Ehhh, like the MMU has been doing since the days of the 80386 you mean? The reason I think protected address spaces are important is... Here you are switching to a completely different matter. Of course any general purpose OS is to have protected address spaces, but that in no way implies they should have the horrible segmented addressing!
Mike Nordell wrote: like the MMU has been doing since the days of the 80386 you mean Those nice old TLB* tables?....(shudder) :eek: I had to write a boot section for one of these in assembler once. Not nice. Especially as it was all 1st generation proprietry silicon, that was yet to be fully debugged. You could never tell whether the problem was your code or the hardware! * Thats a Translate Lookaside Buffer for those who don't know the anacronym. Roger Allen Sonork 100.10016 If I had a quote, it would be a very good one.