h a l f b a k e r yRenovating the wheel
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
|
I know I wouldn't want to drive a Model T at 180mph. |
|
|
Back to the idea for just a second... an 8086 wouldn't run at 3Ghz because of the way it's manufactured. |
|
|
In a way, I have a bit of sympathy for the author here. I'll give [pieterboots] a little credit for at least thinking about it, but unfortunately, he/she seems to lack some rather basic and fundamental understanding of how microprocessors actually work. |
|
|
To answer the question presented in the idea. In a word... no. |
|
|
If I understand the idea, you're calling for a return to simpler processors, but RISC processors exist. |
|
|
Modern x86s have much better architectures in addition to faster clock speeds. For example, quicker arithmetic units, much better pipelining and multiple instruction decoders. |
|
|
Also, it would be difficult to perform a fair comparison as modern cpus are designed to run in protected mode, which the 8086 didn't support. |
|
|
Modern multiple-cpu systems (or multiple cpu cpu's, a la hyperthreading) rely on the software to be running in multiple threads, which in turn (to be fast and reliable) really requires some hardware assistance (protected mode again). |
|
|
I wonder if we'll ever get access to the underlying risc hardware of modern x86s (ie, at the 'microcode' level)? |
|
|
I thought that once you got much higher then the 25 processor mark, you have to start assigning specific processors specific tasks otherwise the communication between the processors gets so great it becomes unworthwhile. |
|
|
My client has 30 processors and 6 of them we have to explicitly say 'Don't do any of the normal jobs, do special jobs instead'. Benchmarking showed 30 processors as being slower then 25!!! |
|
| |