0
\$\begingroup\$

The aim is to increase data transfer rate.

I understand each may have some drawbacks and advantages. Frankly I'm not knowledgeable enough to know of many so I'd like some input from others. I know that doubling bus width requires making sure memory can handle receiving double the bits at a time. I also know that running things at higher frequency than they're designed for could be damaging. If you can't tell I'm not that aware of the technical details.

As for advantages I understand that a wider bus width may be more advantageous in streaming as you have a greater bandwidth stream and capability to load more bits at a time. I also understand that increasing clock frequency may increase the speed of read operations from memory, so you can load an operand more quickly in that case. I'd like some more tangible input from more knowledgeable people. For example, how do they compare in:

  1. practicality/convenience
  2. cost
  3. Advantages in different uses. and so on

Thanks.

\$\endgroup\$

closed as too broad by Andy aka, Bimpelrekkie, Elliot Alderson, RoyC, awjlogan Oct 12 '18 at 12:32

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • \$\begingroup\$ This resembles the CISC vs RISC debate. \$\endgroup\$ – Harry Svensson Oct 10 '18 at 18:24
  • \$\begingroup\$ You should make a list of the (I hope obvious) advantages and disadvantages of both. For example: a wider bus needs needs a doubled amount of electronics. A faster bus needs faster electronics. Those are exactly the reasons why one or the other is chosen because both strategies are still in use. What is chosen depends on the advantages and disadvantages in that list. \$\endgroup\$ – Bimpelrekkie Oct 10 '18 at 18:38
  • \$\begingroup\$ It also resembles the difference between FPGA (relatively low speed & highly parallel) and CPU (like your typical i7). Or GPU (wider bus, slower clock) vs CPU (smaller bus, higher clock) - I'm not the one downvoting this question, but as you can see... there's no right answer. - If you got a specific question then you'd be met much more positively. \$\endgroup\$ – Harry Svensson Oct 10 '18 at 18:38
  • \$\begingroup\$ I do realize there are some advantages to both, but I am aware of only a few, and was hoping some more knowldgeable people can give more insight into unique advantages. I understand the question is to an extent faulty, but this is actually a textbook question as I'm a Computer engineering student. I'll try to add some of the details I know to the question description. \$\endgroup\$ – M Zein Oct 10 '18 at 18:59
1
\$\begingroup\$

The only advantage to a wider, slower bus is slightly simpler interface electronics. When receiving data, for instance, a parallel bus requires n gates or flip-flops to accept the data. A serial bus handling the same data will require (typically) an n-bit shift register and then another n flip-flops as holding registers. Plus, the serial bus will require a synchronization process to inform the receiving unit when it has accepted n bits. Much the same considerations apply when transmitting data.

On the other hand, the parallel bus will require n bus drivers, which typically take more silicon and power than internal logic, so a serial bus will probably take less power. The parallel bus will, of course, require a physically bigger cable (more conductors) and a bigger connector footprint, which may be an issue on a crowded board. With only a single signal path being used for the serial bus, differential signalling becomes quite reasonable, which both reduces EMI and increases speed.

\$\endgroup\$

Not the answer you're looking for? Browse other questions tagged or ask your own question.