Measuring CPU speeds ( was RE: Why isn't Apple releasing patches for the Month of Apple )
13 Feb
2007
13 Feb
'07
5:52 p.m.
--- arcarlini at iee.org wrote:
Tony Duell wrote:
> However, 'What is an instruction?'. I've heard i
t
said that the 11/780
s
averaged 500,000 of its machine instructions per
second, but that that
was about as much processing as 10^6 of other
machine's instructions
The VAX-11/780 cranked out about as many Dhrystone as an
IBM System/370 158 which did (apparently) chew
through about
one million instructions per second.
**>> snip <<**
Antonio
Speaking of Dhrystones.... what is one?
Also, what does the MHz, or GHz, measure
exactly?
I understand MIPS (Million Instructions Per
Second), FLOPS (Floating-point Operations Per
Second) and G-FLOPS (Giga-FLOPS) but the
aforementioned two are a mystery to me.
Regards,
Andrew D. Burton
aliensrcooluk at yahoo.co.uk
13 Feb
13 Feb
6:19 p.m.
New subject: Measuring CPU speeds ( was RE: Why isn't Apple releasing patches for the Month of Apple )
At 04:52 PM 2/13/2007 -0600, you wrote:
--- arcarlini at iee.org wrote:
When one specifies or discusses a cpu and the associated clock speed in MHz
or GHz, I would assume one is referring to the cpu clock, something which
can be measured by a frequency counter on a specified pin, and probably set
in in the BIOS, or on older machines, by dip switches, or on still older
machines, set by the manufacturer.
I think.
[Philosophy] Maybe we can eventually make language a complete
impediment to understanding. (sig of Matthew Austern matt at ph
ysics.berkeley.edu)
--... ...-- -.. . -. ----. --.- --.- -...
tpeters at nospam.mixcom.com (remove "nospam") N9QQB (amateur radio)
"HEY YOU" (loud shouting) WEB ADDRESS http//www.mixweb.com/tpeters
43? 7' 17.2" N by 88? 6' 28.9" W, Elevation 815', Grid Square
EN53wc
WAN/LAN/Telcom Analyst, Tech Writer, MCP, CCNA, Registered Linux User 385531
Tony Duell wrote:
> However, 'What is an instruction?'. I've heard i
t
said that the 11/780
s
averaged 500,000 of its machine instructions per
second, but that that
was about as much processing as 10^6 of other
machine's instructions
The VAX-11/780 cranked out about as many Dhrystone as an
IBM System/370 158 which did (apparently) chew
through about
one million instructions per second.
**>> snip <<**
Speaking of Dhrystones.... what is one?
Also, what does the MHz, or GHz, measure
exactly? 
6:49 p.m.
New subject: Measuring CPU speeds ( was RE: Why isn't Apple releasing patches for the Month of Apple )
Tom Peters wrote:
But as said, it's utterly meaningless. The only thing that matters is whether
the CPU is fast enough for the required application(s)...
Speaking of
Dhrystones.... what is one?
Also, what does the MHz, or GHz, measure
exactly?
When one specifies or discusses a cpu and the associated clock speed in
MHz or GHz, I would assume one is referring to the cpu clock, something
which can be measured by a frequency counter on a specified pin, and
probably set in in the BIOS, or on older machines, by dip switches, or
on still older machines, set by the manufacturer. 
11:27 p.m.
New subject: Measuring CPU speeds ( was RE: Why isn't Apple releasing patches for the Month of Apple )
aliensrcooluk at yahoo.co.uk wrote:
Speaking of Dhrystones.... what is one?
There is no "one". Like most benchmarks, Dhrystone is best used as a
means of measuring performance differences in like environments (ie. "I
increased the clock speed and now I get XX more dhrystones" or "I
optimized the compiler and got a few more dhrystones", etc.). It is not
a platform-agnostic absolute measure.
Also, what does the MHz, or GHz, measure
exactly?
The number of machine cycles per second.
I understand MIPS (Million Instructions Per
Second), FLOPS (Floating-point Operations Per
Second) and G-FLOPS (Giga-FLOPS) but the
aforementioned two are a mystery to me.
Well, I don't like MIPS that much for the same reasons outlined in my
first paragraph. For example, most 6502 instructions run in fewer
clocks than like instructions in 8088, but 8088 has 16-bit operations
for "free"... etc.
--
Jim Leonard (trixter at oldskool.org) http://www.oldskool.org/
Help our electronic games project: http://www.mobygames.com/
Or check out some trippy MindCandy at http://www.mindcandydvd.com/
A child borne of the home computer wars: http://trixter.wordpress.com/
14 Feb
14 Feb
2:08 a.m.
New subject: Measuring CPU speeds ( was RE: Why isn't Apple releasing patches for the Month of Apple )
Jim Leonard wrote:
aliensrcooluk at yahoo.co.uk wrote:
It's the fastest clock you can claim to have anywhere
in your machine while keeping a straight face :-)
Acually, to be fair, manufacturers weren't quite that
bad (usually) but identifying the fastest clock is
actually quite hard, even if you are trying to be honest!
As CPU cycle times dropped, it was quite often the case
that the external clock might be running at say 20MHz
so that the rest of the slow external world could keep up
but that was doubled inside the chip to perhaps 40MHz.
In other cases the external clock might be quite high
but would be split internally to provide the phases
required to clock the various CPU stages.
Sometimes the manufacturer would publish a cycle time,
but as pipelined architectures came to the fore those
cycle times became less useful, so then they started
to publish instructions per tick or clocks per instruction.
Antonio
Also, what does the MHz, or GHz, measure
exactly?
The number of machine cycles per second. 
8:59 a.m.
New subject: Measuring CPU speeds ( was RE: Why isn't Apple releasing patches for the Month of Apple )
On 14 Feb 2007 at 7:08, arcarlini at iee.org wrote:
It's the fastest clock you can claim to have
anywhere
in your machine while keeping a straight face :-)
...And sometimes, (particularly in the Pentium I look-alike game) it
can mean a number that bears no relationship to any particular
physical quantity.
Consider, for example the Cyrix/IBM Pentium "clones". Rated in "PR"
numbers which was a synthetic quantity that was supposed to resemble
the clock speed that an Intel Pentium *would* have had to operate at
to deliver equivalent performance. Mostly, the PR numbers were sheer
twaddle aimed at a na?ve public.
But then, most benchmarks are nearly useless once vendors get a
chance to figure them out.
In the 80's, a synthetic benchmark for PCs was the Landmark Speed
test number, which attempted to assign a value based on how much
faster one's particular machine was than a stock 5150 PC. You could
go to a show and the silly test would be running on just about every
vendor of Far East clone PCs. I think the Landmark test was mostly
notable for measuring public gullibility.
As an aside, I wonder how many Landmark PC diagnostic kits were
actually sold (I have a couple, complete with alignment discs and
port loopback connectors).
Even well-intentioned attempts at measuring computer performance
across all platforms can run aground.
Sometime in the mid 70's to early 80's, a suite of tests called
Linpack (written in FORTRAN) to measure supercomputer performance.
Once it became known that customers were using the Linpack numbers to
rate processors according to perceived speed, vendors would start
tweaking compilers and then later hardware to deliver higher numbers.
In particular, I remember a startup firm named Saxpy whose machine
architecture was designed to run the Linpack test of the same name
(essentially a sum of two vectors times a scalar constant and
probably less than 50 lines of code). I don't know what became of
the firm, but one of their boxes would certainly be a collector's
item today.
This, of course was nothing new. In the days of big iron, there were
groups within many vendor marketing organizations dedicated to
running potential customer-supplied benchmarks in order to qualify in
the bidding process, usually by any means possible.
Cheers,
Chuck
1:51 a.m.
New subject: Measuring CPU speeds ( was RE: Why isn't Apple releasing patches for the Month of Apple )
On 2/13/07, aliensrcooluk at yahoo.co.uk <aliensrcooluk at yahoo.co.uk> wrote:
--- arcarlini at iee.org wrote:
Dhrystone From Wikipedia, the free encyclopedia Jump to:
navigation<http://en.wikipedia.org/wiki/Dhrystone#column-one>ne>,
search <http://en.wikipedia.org/wiki/Dhrystone#searchInput>
*Dhrystone* is a synthetic
benchmark<http://en.wikipedia.org/wiki/Benchmark_%28computing%29>prog…
developed in
1984 <http://en.wikipedia.org/wiki/1984> by Reinhold P. Weicker intended to
be representative of system (integer) programming. The Dhrystone grew to
become representative of general processor (CPU) performance until it was
superseded by the CPU89 benchmark suite from the Standard Performance
Evaluation
Corporation<http://en.wikipedia.org/wiki/Standard_Performance_Evaluation…on>,
today known as the "SPECint <http://en.wikipedia.org/wiki/SPECint>"
suite.
With Dhrystone, Weicker gathered meta-data from a broad range of software -
including programs written in FORTRAN, PL/1, SAL, ALGOL 68, and Pascal. He
then characterized these programs in terms of various common constructs -
procedure calls, pointer indirections, assignments, etc. From this he wrote
the Dhrystone benchmark to correspond to a representative mix. Dhrystone was
published in Ada, with the C version for Unix developed by Rick Richardson
("version 1.1") greatly contributing to its popularity.
Contents [hide <javascript:toggleToc()>]
- 1 Dhrystone vs
Whetstone<http://en.wikipedia.org/wiki/Dhrystone#Dhrystone_vs_Whetstone&…
- 2 Issues addressed by
Dhrystone<http://en.wikipedia.org/wiki/Dhrystone#Issues_addressed_by_Dhr…
- 3 Results <http://en.wikipedia.org/wiki/Dhrystone#Results>
- 4 See also <http://en.wikipedia.org/wiki/Dhrystone#See_also>
- 5 References <http://en.wikipedia.org/wiki/Dhrystone#References>
[edit<http://en.wikipedia.org/w/index.php?title=Dhrystone&action=edit…
] Dhrystone vs Whetstone
The Dhrystone benchmark contains no floating
point<http://en.wikipedia.org/wiki/Floating_point>operationsations, thus the
name is a pun on the then-popular
Whetstone <http://en.wikipedia.org/wiki/Whetstone_%28benchmark%29> benchmark
for floating point operations. The output from the benchmark is the number
of Dhrystones per second (the number of iterations of the main code loop per
second).
Both Whetstone and Dhrystone are *synthetic* benchmarks, meaning that they
are simple programs that are carefully designed to statistically mimic some
common set of programs. Whetstone, developed in 1972, originally strove to
mimic typical Algol 60 programs based on measurements from 1970, but
eventually became most popular in its Fortran version. Whetstone thus
reflected the highly numerical orientation of computing in the 1960s.
[edit<http://en.wikipedia.org/w/index.php?title=Dhrystone&action=edit…
] Issues addressed by Dhrystone
Dhrystone's eventual importance as an indicator of general-purpose
("integer") performance of new computers made it a target for commercial
compiler writers. Various modern compiler techniques (such as dead code
elimination) make the use and design of synthetic benchmarks more difficult.
Version 2.0 of the benchmark, released by Weicker and Richardson in March of
1988, had a number of changes intended to foil a range of compiler
techniques. Yet it was carefully crafted so as not to change the underlying
benchmark. This effort to foil compilers was only partly successful.
Dhrystone 2.1, released in May of the same year, had some minor changes and
remains the current definition of Dhrystone.
Other than issues related to compiler optimization, various other issues
have been cited with the Dhrystone. Most of these were understood at the
time of its publication in 1984 - including the small code size and small
data set size. More subtle is the slight over-representation of string
operations, which is largely language related: both Ada and Pascal have
strings as first class citizens in the language, whereas C does not, so what
was simple variable assignments in reference benchmarks became buffer copy
operations in the C library.
Dhrystone remains remarkably resilient as a simple benchmark. It is easy to
use, well documented, is fully self-contained, is well understood, and can
be made to work on almost any system. In particular, it has remained in
broad use in the embedded computing world, though the recently developed *
EEMBC <http://en.wikipedia.org/wiki/EEMBC>* benchmark suite is likely to
supersede it in that role just as SPEC's CPU89 did for the general computing
and server markets. Still, 20 years of continued use is quite a testament to
Weicker's careful design and foresight.
[edit<http://en.wikipedia.org/w/index.php?title=Dhrystone&action=edit…
] Results
Dhrystone tries to represent the result more meaningfully than MIPS (million
instructions per second), because MIPS cannot be used across different
instruction sets (e.g. RISC <http://en.wikipedia.org/wiki/RISC> vs.
CISC<http://en.wikipedia.org/wiki/CISC>)
for the same computation requirement from users. Thus, the main score is
just Dhrystone loops per second. Another common representation of the
Dhrystone benchmark is the *DMIPS* - Dhrystone
MIPS<http://en.wikipedia.org/wiki/Million_instructions_per_second>-
obtained when the Dhrystone score is divided by 1,757 (the number of
Dhrystones per second obtained on the VAX
11/780<http://en.wikipedia.org/wiki/VAX>AX>,
nominally a 1 MIPS machine).
[edit<http://en.wikipedia.org/w/index.php?title=Dhrystone&action=edit…
] See also
Benchmark (computing)<http://en.wikipedia.org/wiki/Benchmark_%28computing%29>
Whetstone (benchmark)<http://en.wikipedia.org/wiki/Whetstone_%28benchmark%29>
[edit<http://en.wikipedia.org/w/index.php?title=Dhrystone&action=edit…
] References
- Newsgroup posting for calculation of
DMIPS<http://groups.google.com/groups?q=DMIPS&hl=en&group=comp.b…
- http://www.netlib.org/benchmark/dhry-c c version of Dhrystone in a
sh file.
Tony Duell wrote:
> However, 'What is an instruction?'. I've heard i
t
said that the 11/780
s
averaged 500,000 of its machine instructions per
second, but that that
was about as much processing as 10^6 of other
machine's instructions
The VAX-11/780 cranked out about as many Dhrystone as an
IBM System/370 158 which did (apparently) chew
through about
one million instructions per second.
**>> snip <<**
Antonio
Speaking of Dhrystones.... what is one?
Regards,
Andrew D. Burton
aliensrcooluk at yahoo.co.uk
15 Feb
15 Feb
2:09 a.m.
New subject: Measuring CPU speeds
On 2/14/07, Jay West <jwest at classiccmp.org> wrote:
Andrew wrote....
oh ok sorry
> Speaking of Dhrystones.... what is one?
To which Hex Star replied....
Dhrystone From Wikipedia, the free encyclopedia
Jump to:
...snip...
Just a url link would have been fine :)
Jay
10:11 a.m.
New subject: Measuring CPU speeds
Hex Star wrote:
Although personally I'd prefer a KB or so of text delivered right before my
eyeballs rather than having to wander off and fire up a web browser to look at
something, but that's just me.
Doubtless it depends on the speeds of network connection as to which is
preferred...
If someone tried to post a few MB of text [1] about something that I wasn't
even interested in anyway, *then* I'd get upset :-)
[1] presumably there's a few-KB limit on list message size anyway?
To which Hex
Star replied....
oh ok sorry Dhrystone From Wikipedia, the free encyclopedia
Jump to:
...snip...
Just a url link would have been fine :)
Jay
9:08 p.m.
New subject: Measuring CPU speeds
On 2/15/07, Jules Richardson <julesrichardsonuk at yahoo.co.uk> wrote:
Hex Star wrote:
[1] presumably there's a few-KB limit on list message size anyway?
Wanna try and find out? I got some 5mb or so files... XD
To which
Hex Star replied....
oh ok sorry Dhrystone From Wikipedia, the free encyclopedia
Jump to:
...snip...
Just a url link would have been fine :)
Jay
7:43 a.m.
On 14/02/2007, at 9:52 AM, aliensrcooluk at yahoo.co.uk wrote:
Speaking of Dhrystones.... what is one?
Also, what does the MHz, or GHz, measure
exactly?
Well obviously :-) Dhrystones are related to Whetstones. The
Whetstone was a synthetic benchmark written originally in Algol-60
and the performance of this benchmark using the Whetstone Algol-60
compiler was used to compare Algol-60 compilers (in the days when
this had some major relevance).
More information can be found in "Algol 60 Compilation and
Assessment" by B. A. Wichmann. My copy is around somewhere but not
close enough to hand to quote from directly.
Huw Davies | e-mail: Huw.Davies at kerberos.davies.net.au
Melbourne | "If soccer was meant to be played in the
Australia | air, the sky would be painted green"
8:37 a.m.
New subject: Measuring CPU speeds ( was RE: Why isn't Apple releasing patches for the Month of Apple )
Speaking of
Dhrystones.... what is one?
Also, what does the MHz, or GHz, measure
exactly?
Well obviously :-) Dhrystones are related to Whetstones.
6371
days inactive
6374
days old
12 comments
10 participants
participants (10)
-
aliensrcooluk＠yahoo.co.uk -
arcarlini＠iee.org -
cclist＠sydex.com -
hexstar＠gmail.com -
huw.davies＠kerberos.davies.net.au -
julesrichardsonuk＠yahoo.co.uk -
jwest＠classiccmp.org -
spectre＠floodgap.com -
tpeters＠mixcom.com -
trixter＠oldskool.org