The Development Curve?

[Dave]

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

Does anyone know of a site that is tracking the historical pc
development curves? I mean in terms of performance, or performance per
dollar, or something like that? I'm wondering what rule-of-thumb is
most commonly used when deciding on upgrading? For example do you buy a
new pc or motherboard when the new models are clocking at 1.5x or 2x or
3x faster than what you own?

 

[Guest]

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

sounds like you are trying to do some school project or assignment




Cowly the Game player

Please click on my links
http://spacefed.com
http://gc.gamestotal.org
http://uc.gamestotal.org
http://aw.gamestotal.org
http://www.gamestotal.org
http://3700ad.gamestotal.com
http://www.spacefederation.net
http://www.gamestotal.org/news/
http://ballmonster.gamestotal.com
http://www.spacefederation.net/manual/
http://gc.gamestotal.org/i.cfm?p=aboutgc
http://uc.gamestotal.org/i.cfm?p=aboutgc
http://www.gamestotal.org/corp/
http://www.gamestotal.org/strategygames/

 

[Dave]

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

ste...@yahoo.com wrote:
>
> sounds like you are trying to do some school project or assignment

No, I'm just curious about the rate of progress. For example has the
rate of doubling pc clock speeds been close to a constant or has it
slowed down?

 

[Guest]

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

hi dave, one mans criteria (mine !) do I need it ?. or would I just like it.
If it does what you want, leave it. If it doesn`t upgrade it. But don`t
worry about trying to impress your pals with what you`ve got, that`s a
stupid race to nowhere. stay 6 months behind the latest technology, and let
others find the bugs and flaws. (and i`m only 73). best wishes..J

"Dave" <galt_57@hotmail.com> wrote in message
news:1114625897.805022.200730@g14g2000cwa.googlegroups.com...
> Does anyone know of a site that is tracking the historical pc
> development curves? I mean in terms of performance, or performance per
> dollar, or something like that? I'm wondering what rule-of-thumb is
> most commonly used when deciding on upgrading? For example do you buy a
> new pc or motherboard when the new models are clocking at 1.5x or 2x or
> 3x faster than what you own?
>

 

[JohnS]

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

> rate of doubling pc clock speeds been close to a constant or has it
> slowed down?

I look at rates over about a 3 year period .... several
rates. And since these machines I build tend to stay in
production, I can benchmark them directly. I estimate
that the overall speed increase is more like a factor
of 10. Along with that has come products that use to
be something that sort of worked, but were generally
useless, to products that really work, and are a blast
to own. It all depends on your interests, and whether
or not new products are user friendly. It seems to be
getting there. I use my computer as a Programmable
Video Recorder ( PVR ) that will record cable TV
in very high resolution mode ... much nicer than my
living room TV. It plays dvds better .. esp the sound.
I have a super sound system on this box that cost me
about $150. It will beat the tar out of a $1000 Marantz
stereo. If I were willing to pay for it, I could have
satelight lan that is super fast. Same outfit gives me
mpg from my camera ... and creates dvds from TV
shows I've recorded. I can keep them for the future.
At the same time, I'm a gamer, and this system runs
video games at the highest settings. If you have a box
more than 2 years old, it will not do that.

johns

 

[Guest]

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

maybe in the past but manufactures are now starting to focus on efficiencies
rather than simply clock speeds
"Dave" <galt_57@hotmail.com> wrote in message
news:1114638609.799797.152650@o13g2000cwo.googlegroups.com...
> ste...@yahoo.com wrote:
>>
>> sounds like you are trying to do some school project or assignment
>
> No, I'm just curious about the rate of progress. For example has the
> rate of doubling pc clock speeds been close to a constant or has it
> slowed down?
>

 

[peter]

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

In article <1114625897.805022.200730@g14g2000cwa.googlegroups.com>,
galt_57@hotmail.com says...
> Does anyone know of a site that is tracking the historical pc
> development curves? I mean in terms of performance, or performance per
> dollar, or something like that? I'm wondering what rule-of-thumb is
> most commonly used when deciding on upgrading? For example do you buy a
> new pc or motherboard when the new models are clocking at 1.5x or 2x or
> 3x faster than what you own?
>
>

Don't know if it would come up with anything relevant, but doing a
search on "moore's law" and 'computers' comes up with over 600,000 hits
in Google.

--
Pete Ives
Remove All_stRESS before sending me an email

 

[Guest]

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

On 27 Apr 2005 11:18:17 -0700, "Dave" <galt_57@hotmail.com> wrote:

>Does anyone know of a site that is tracking the historical pc
>development curves? I mean in terms of performance, or performance per
>dollar, or something like that? I'm wondering what rule-of-thumb is
>most commonly used when deciding on upgrading? For example do you buy a
>new pc or motherboard when the new models are clocking at 1.5x or 2x or
>3x faster than what you own?

I don't know of such a site. The closest thing to it may be the Case Labs
page here:

http://www.caselab.okstate.edu/research/benchmark.html

It gives a list of AMD & Intel CPUs by type and speed, run against the
same program: the STARS CFD solver with an identical mesh configuration.

This list may also be useful:

http://field.hypermart.net/CPU/cpu.htm

The bottom line is there is lots of benchmark data out there. The
problems are a) it may be hard to find data for the specific systems you're
interested in; b) the tests are usually not well controlled for independent
variables. It's hard, therefore, to pin down exactly what factors determine
the speed changes. Was it processor type? Number of processors? Amount of
main memory? Amount of cache memory? Chipset? Software version? Operating
system version? Something else?

You get the picture. A good overview of the situation is here:

http://www.realworldtech.com/page.cfm?ArticleID=RWT082901025838

HTH,

Chris