OT Wrong advertised specifications

Page 2 of 6  


Guess again. AGP is what is gone. PCIe is the new standard. AGP died years ago. PICe is factors faster than AGP.

Nope. The bandwidth is identical.

Nope. It is not. Shortness means nothing, speed means everything.

You are partially right here. What you describe is potential. What out get out of the box is not the same.
Paul
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

You think the length of the signal path has nothing to do with speed?
--
Regards,
Doug Miller (alphageek at milmac dot com)
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Doug Miller wrote:

Waves hand!
I do! I do! Pick me!
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

You are aware, aren't you, that the speed of signal propagation is finite?
--
Regards,
Doug Miller (alphageek at milmac dot com)
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

First, you have something called a clock in the computer. All computers have a clock, they cannot run without one. Second, the signals can only be passed during a clock cycle. The speed of light is far faster than any clock we can employ therefore we are not dealing with theoretical limits we are dealing with practical limits i.e. the duration of each clock cycle. So in the case of a 2 inch wire trace, it would not matter if the trace were 1 inch because you can't get the data into the CPU any faster than it already is.
Paul
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

You think so, do you?
1GHz clock rate = 1 nanosecond cycle length. How far do you suppose light moves in a nanosecond?

I won't argue that the difference between one inch and two doesn't matter at all -- YET -- but I'll leave it as an exercise for you to compute the approximate clock speed at which the difference between two inches and three *does*, and then invite you to explore the availability of existing processors in that range.
--
Regards,
Doug Miller (alphageek at milmac dot com)
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

speed?
finite?
be
So
1
already
at
three
processors
Doug, you lost the argument. You claimed that the shorter bus length made for a faster data transfer. If we were talking photon switches (a theoretical possibility) then you'd be right. Someday, someday - you will be right. For today, you are wrong. The bottleneck in any computer is the CPUs ability to stay cool while you ramp up the clock speed. Silicon melts into a puddle of molten glass at the temperature generated by just the speeds we are talking about today. Try running your computer without a heat sink and cooling fan and you'll see what I mean.
We are nowhere near, not even close, to being able to run CPUs so fast they can run at the speed of light *per* channel. Think of 186,000 mps raised to the 32nd power then raise it by factors of 5286. That's faster than a horny Republican denying he's gay on national TV.
Paul
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

[Lack of response noted]

[Lack of substantive response noted]

No, I didn't. I disagreed -- and still do -- with your claim that "shortness means nothing".

Again:
How far do you suppose light moves in a nanosecond?
At what clock speed, approximately, does the difference between a two-inch and three-inch signal path make a difference?
What is the clock speed of the fastest processor on the market today?
--
Regards,
Doug Miller (alphageek at milmac dot com)
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

11.8 inches I happen to have one of Grace Hoppers nanoseconds. It is a length of wire 11.8 inches long. I got it from her when I attended a speech she gave at the DODARPA office I worked at in 1985.

What you should not is that you do not understand what I am saying because you do not know what you are talking about. Did I mention I studied computer science in college? We learned all kinds of stuff.

You lost the argument. Your claim is patently incorrect. It is wrong. It sufferes from a dearth of correctnes. It is truth challenged. It is factually insufficient. It's BS. You made a statement that was just plain wrong.

11.8 inches

186,000 *2^(-32) That should get close enough.

Which manufacturer? AMD and Intel are not the only manufacturers, you know?

Doug, I'm done with your game. You made a claim that was wrong. Get over it.
Paul
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Of course, this was done for illustration purposes. That length is correct for an electromagnetic signal travelling in a vacuum (or air, which is nearly identical) only. In real computers, signals are carried in microstrip transmission lines (signal trace on one face of a PC board with an internal ground plane, or ground on the other face), or twisted-pair transmission lines, or even coaxial cable transmission lines. In those, the velocity is somewhat lower, and a nanosecond takes you a shorter distance - about 60-80% of the distance in a vacuum.
    Dave
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Oh, the old "argument from authority" fallacy. Too bad that formal logic wasn't part of *your* computer science curriculum; it was in *mine*.

So say you. You've provided nothing to back that up, though.

Lack of accurate response noted.

Lack of response noted.
Give it up, Paul. You've lost the argument.
--
Regards,
Doug Miller (alphageek at milmac dot com)
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
wrote:

any
speech
and
because
made
It
plain
will
raised
two-inch
Yeah, ok whatever. Paul
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

[Continued lack of response noted -- without surprise, this time]
--
Regards,
Doug Miller (alphageek at milmac dot com)
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@milmac.com says...

On a motherboard, about eight inches (a foot in air). Not that that fact has anything to do with the maximum clock frequency achievable.

The difference does matter. The length of a PCI clock on a plug-in card, for instance, has to be within .1" (2.5" +/- .1", IIRC). The *difference* in trace lengths is more important than length.
--
Keith

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

You might want to rethink that notion...
--
Regards,
Doug Miller (alphageek at milmac dot com)
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
snipped-for-privacy@milmac.com says...

It doesn't. Latency <> bandwidth.
--
Keith

Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Doug Miller wrote:

Yes. And has nothing to do with the signal path.
Whether it's a 1/4 mile drag strip or the Indy 500, the top speed of my '57 VW Bug is the same. The length of the track or, as you put it, the signal path, has nothing to do with the speed.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Length of path x speed = time required to traverse the path. Since the propagation speed of electrical signals has an inherent physical upper limit, the length of the signal path places an upper limit on the speed of any device that is depending on those signals.
--
Regards,
Doug Miller (alphageek at milmac dot com)
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Doug Miller wrote:

Right. But speed is independent of the route, the length, or number of beers per mile.
The length of the signal path has no bearing on the speed of the signal. The length affects the time it takes for the signal to transverse the path, but "speed" is independent of both time and distance. V = dx/dt as dt -> 0
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Of course not -- but it *does* have bearing on the throughput ["speed"] of the device that's *processing* that signal.
--
Regards,
Doug Miller (alphageek at milmac dot com)
  Click to see the full signature.
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Related Threads

HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.