Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Ok, so ive been looking around, trying to learn and understand the way that ram works. Ive gotten one answer that said
"The addressing is best for 2 sticks, and when you use 4; it slows down"

Another answer said something like:
Theres bank/channel interleave that makes the memory read like one stick
Also I read something about the memory density also being a factor.

I dug further and found out that theres a higher speed limit on my board for 2 sticks vs 4, so now im trying to put an image in my head of how and why, and... pfft.

Can anyone explain, or recommend a resource that would answer these questions?

share|improve this question
up vote 2 down vote accepted

There is no 'ideal' answer since to much depends on what someone considers ideal.

Memory access speed depends on a lot of things, including:

  1. Type of Memory (Fast page RAM, SDRAM, DDR, DDR2, DDR3, ...)
  2. Speed of the RAMs IO (e.g. 667MHz for DDR3-1333)
  3. Efficiency of the RAM (DDR 1,2 and 3 are twice as efficient as SDRAM, which is why you often see RAM sold at higher IO speeds than they actually run at. E.g. DDR1-400 runs at 200MHz, DDR3-1333 at 667Mhz).
  4. Latencies on the RAM. Most often specified as DDRX-ABCD A-B-C-D.

Highest speed is a combination of these, but since you are usually stuck with a specific type of memory points 2 and 4 are the ones to check. Point 2 (memory IO speed) makes the biggest difference.

I dug further and found out that there is a higher speed limit on my board for 2 sticks vs 4.

This is because your memory controller needs to do 'more work' if it has to address more DIMMS. I realise this is an imprecise definition of work. For details check the posts on memory ranks.

Usually, it comes down to this.

  • If your memory controller has one channel and you plug in an average rank 2 DIMM then it just works.
  • If your memory controller has one channel which has to drive more than 2 ranks then it may slow down. On consumer board this is often 2 DIMMs for a total of 4 ranks.
    This can differ. E.g. on the Intel 5100 chipset it can drive up to 2 dual rank DIMMs at full speed on only has to slow down when a third DIMM is added to the memory controller.

  • If your board has two memory controllers then using 1 rank 2 DIMM on one channel is just fine.
  • If your board has two memory controllers then using 1 rank 2 DIMM on both channel is also fine. This often allows for the use of dual channel mode which is slightly faster.
  • If your board has two memory controllers then using 2 stick on both channels is usually means more work and the IO frequency is slowed down. (Same situation as 2 DIMMs per single controller, but now twice the same situation).

In the case of your AM3 setup you can gain a slight speed gain by using only two DIMMs over 4 DIMMs with an equal memory size. But more RAM (even more RAM at a slower speed) usually results in a faster experience. So go for more RAM first, and if the price difference is small then consider using the 2 DIMM solution.

share|improve this answer
Thats impressive... – TardisGuy Nov 16 '12 at 11:36

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .