View Single Post
Old 01-30-03, 07:03 PM   #8 (permalink)
merle
Aximsite Prospect
 
Join Date: Jan 2003
Posts: 17
Thanked 0 Times in 0 Posts
Quote:
Originally posted by cageyjames@Jan 30 2003, 04:14 PM
But this is a moot point if the OS doesn't support the processor isn't it?
Actually, the performance increase in the new chip should not depend on OS optimization. The old XScale chip has a 100MHz bus. Because the XScale chip runs at 300/400MHz (much faster than the bus), and program instructions and data come via the bus, the chip has to wait for instructions to arrive. Thus, the bus is a major bottleneck. The new bus of 200MHz should improve the situation greatly. No software changes are needed.

Note also that the old StrongARM chip runs at 200MHz but uses a 103MHz bus. That's a slightly faster bus than that the current XScale uses. So not surprisingly the StrongARM can outperform the XScale on certain memory-intensive applications. For example, several benchmarks have shown that the current XScale is much slower than the old StrongARM on "memory move" operations. If you know anything about programming, you know that "memory move" is a very common operation.

At least that's how I understand it from what I've read. Not that I've done any benchmarking myself.

Here's a more detailed article about the bus problem. Looks like Intel agreed.

From http://www.pocketpcthoughts.com/articles.p...ion=expand,5432

"In reality the speed factor between CPU and the memory bus is the opposite of what we just described. XScale Pocket PCs running at 400 MHz, 300 MHz and 200 MHz all use a 100 MHz bus, and the 206 MHz StrongARM use a 103 MHz bus.

I guess you just spotted the main bottleneck, and the reason why XScale running at 400 MHz, 300 MHz and 200 MHz get almost identical benchmark results for tests that involve shuffling memory around – typical applications are graphics and multimedia. (Note: some Pocket PCs incorporate graphics accelerators that might confuse this picture a little bit). And since the StrongARM use a bus that is 3 % faster than the bus used in XScale, we also find a logical explanation for why StrongARM based Pocket PC’s are sometimes slightly faster than XScale based Pocket PC’s in some tests."
merle is offline   Reply With Quote