Date: 10 Apr 1996 From: Jaap van Ganswijk To: Multiple recipients of list Subject: Re: speed comparison between 68HC11 and PC-XT At 06:29 PM 4/9/96 -0500, you wrote: >Greetings > >What I am after is a simple processing power/speed comparison between a 2MHz >68HC11 and an XT, without getting bogged down in the usual issues, ie, is >the benchmark program computational or does it do a lot of IO, etc. This is >for the delectation and diversion of some upper management types when I >point at a little controller and say that this contains the power of 2 XT's >(or whatever). > >Hopefully this is not comparing chalk & cheese and will not spark a controversy. I personally propose these numbers: 2 (internal) MHz 6811 is 0.5 (8 bits) Mips 2 (internal) MHz 6809 is 0.5 (8 bits) Mips 10 (internal) MHz Zilog Super8 is 1 (8 bits) Mips 4.77 MHz 8088 is 1 (8 bits) Mips (?) 16 MHz 80286 is 3 (8..16 bits) Mips (?) 40 MHz 80386 is 9 (8..32 bits) Mips (?) 33 MHz 486 is 16 (8..32 bits) Mips 166 Mhz DEC Alpha is 83 (8..64 bits) Mips Notes: Some of the early 386's were slower than the 80286 at the same frequency. Above a certain frequency cache efficiency starts playing a role. You can do a lot with a 0.5 Mips 6811 when your software is efficient and your memory need is below 64 Kbyte... Groeten, Jaap -- Chip Directory -- http://www.hitex.com/chipdir/ - USA -- http://www.xs4all.nl/~ganswijk/chipdir/ - Europe -- And other international mirror sites... New: -- http://ftp.unina.it/pub/chipdir/ - Napels Date: 10 Apr 1996 From: Tom G. Brusehaver/Consultant Euler Solutions To: Multiple recipients of list Subject: Re: speed comparison between 68HC11 and PC-XT >>Hopefully this is not comparing chalk & cheese and will not spark a >controversy. > >I personally propose these numbers: > >2 (internal) MHz 6811 is 0.5 (8 bits) Mips >2 (internal) MHz 6809 is 0.5 (8 bits) Mips >10 (internal) MHz Zilog Super8 is 1 (8 bits) Mips >4.77 MHz 8088 is 1 (8 bits) Mips (?) >16 MHz 80286 is 3 (8..16 bits) Mips (?) >40 MHz 80386 is 9 (8..32 bits) Mips (?) >33 MHz 486 is 16 (8..32 bits) Mips >166 Mhz DEC Alpha is 83 (8..64 bits) Mips > Couldn't someone rate it on mips per watt (ARM does this already). So something like this could be reasonable: (system power) 68HC11 (miniboard) 0.5 mips 5watt = 0.1 PC/XT 1.0 mips 85watt = 0.01 40MHz '386 9.? mips 250watt = 0.036 Alpha 83.? mips toaster oven I ask this because depending on your application, this may be a better way to think about cost when comparing dissimiliar systems. A PC bus (ISA/PCI/PC104) computer has to be rated as a system if compared against a microcontroller, assuming you want digital and analog IO. Of course if you need disk drives, you might have reason to compare things the other way. -- Date: 10 Apr 1996 From: Grant Beattie To: Multiple recipients of list Subject: Re: speed comparison between 68HC11 and PC-XT On Tue, 9 Apr 1996, HC11 Tom Harris wrote: > Greetings > > What I am after is a simple processing power/speed comparison between a 2MHz > 68HC11 and an XT, without getting bogged down in the usual issues, ie, is > the benchmark program computational or does it do a lot of IO, etc. This is > for the delectation and diversion of some upper management types when I > point at a little controller and say that this contains the power of 2 XT's > (or whatever). Well, the HC11 is faster at basic operation. Just do a cycle-by-cycle comparison of a few instructions: ldaa #val vs mov AH,val ldd #val vs mov AX,val ldaa addr vs mov AL,[addr] ldd addr vs mov AX,[addr] : etc The HC11 blows it away. Now, of course you can find instructions that demonstrate the opposite.... just don't mention those :) GB