Motherboard bus speed physical maximum

Click For Summary

Discussion Overview

The discussion revolves around the physical limitations of motherboard bus speeds, particularly focusing on the effects of high frequencies on signal integrity and the implications for processor performance. Participants explore concepts related to transmission line theory, signal degradation, and the use of multipliers in processors.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested

Main Points Raised

  • One participant suggests that high motherboard bus speeds may lead to signal disappearance, likening the bus wire to an antenna at high frequencies.
  • Another participant speculates that advancements in optronics may address some of these limitations.
  • A different participant doubts the notion of signals disappearing, proposing instead that they may degrade due to effects like skin effect, especially given the small size of motherboard traces.
  • One participant explains that motherboard traces behave like transmission lines, emphasizing the importance of impedance and capacitance, and notes the limitations imposed by the physical size of components.
  • Another participant provides a specific example of propagation speed on motherboards and discusses the critical dimensions for high-speed circuitry, highlighting the importance of waveform integrity over clock speed.

Areas of Agreement / Disagreement

Participants express differing views on the nature of signal integrity at high frequencies, with some agreeing on the relevance of transmission line theory while others question the initial claim about signals disappearing. The discussion remains unresolved regarding the implications of these factors on motherboard design and performance.

Contextual Notes

Participants reference various technical concepts such as skin effect, transmission line theory, and waveform degradation, but do not reach a consensus on the specifics of how these factors limit bus speeds. The discussion includes assumptions about the physical properties of motherboard traces and the implications for signal integrity.

DwithQs
Messages
11
Reaction score
0
I read somewhere (article, messageboard, i can't find it anymore) that motherboard bus speeds are limited because if the frequency gets too high then the signal will disappear. I guess I understood it to mean that the bus wire will become like and antenna if the frequency is too high.

They then went on to explain that it is the reason why processors use multipliers to run at higher speeds even though the bus isn't capable.

Is that true?
 
Engineering news on Phys.org
I don't know. Maybe that's why people are working so hard to get optronics up and running.
 
I doubt if they meant the signal will "disappear" but likely they meant it will degrade/weaken. Possibly due to "skin effect" if the motherboard "wires" are very small (which they are) and the frequency gets high enough for the skin effect to be significant.
 
The signals on the motherboard traces act as if on a transmission line, with impedance, capacitance and all that fun stuff to take into account. Since motherboards can reduced to a certain size to accept components, and copper traces can only but placed so many per inch, propagation delays result. Someone better versed in transmission line theory would be able to a better description than myself, but for starters you might want to consider this PDF.
Most likely you can make better sense of it than I can.
http://www.polarinstruments.com/support/cits/Critical_length.pdf
 
The transmission line concept is valid, but at the current speed of microprocessors the dimensions are extremely critical.

Example: On a typical motherboard the propagation speed of electrical waveforms is about 20cm/ns. If the clock frequency of your Core 5 is 2.5GHz, this corresponds to a wavelength of 8cm.This again means that the layout of your CPU should have no dimension of this order (otherwise the internal signals would be out of phase just due to differences in traveled distances. A good rule of thumb is to keep the high-speed circuitry within 10% of the wavelength - which means 8mm circuit size.

If we are talking about transmission lines, the situation is different. Here we know that the signal will take whatever time is necessary to reach the other end of the line. In this case we are concerned about waveform degradation, reflections and line termination. The critical parameter for a transmission line is not so much the clock speed of the signals as the waveform rise and fall times.
 

Similar threads

  • · Replies 14 ·
Replies
14
Views
5K
  • · Replies 9 ·
Replies
9
Views
4K
  • · Replies 7 ·
Replies
7
Views
4K
  • · Replies 5 ·
Replies
5
Views
2K
Replies
5
Views
4K
Replies
5
Views
3K
  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 13 ·
Replies
13
Views
645
  • · Replies 22 ·
Replies
22
Views
4K
  • · Replies 42 ·
2
Replies
42
Views
5K