Is there any difference in speed 8Gb having Bus speed 1066, 1333, 1600?

[Procurepod]

Can anyone tell me the difference between the 1066, 1333, 1600 bus speed Ram of 8GB, i have attached a 2 different RAM's for having different bus speed. does it effects the the performance of RAM?

[kalniel]

Those are memory speeds, not bus speeds really. The bus speed will be determined by the memory controller/motherboard.

The difference is exactly what it says on the tin. 1066 works at the equivalent rate of 1066mhz, 1333 at 1333mhz and so on.

If you have two different RAM speeds then you usually have to choose a common speed to address to run them both at, which since you are limited by the hardware, will be the slowest set.

As for affecting performance, it depends what the task is and what the rest of the hardware is, but given you never answered that in your last thread about memory I doubt we'll be any more the wiser.

[EndlessWaves]

There is a fairly substantial difference in speed but very few applications are limited by memory speed so it doesn't tend to be seen in most common desktop uses. Gaming on integrated graphics is probably about the most common scenario where memory bandwidth makes a noticeable difference.

[HalloweenJack]

does depend on the cpu ofc! AMD cpu`s better utilise faster speeds (especially the APU`s)

[Giraffe]

The timings seem to make a lot of difference. I had some trouble with DDR3L RAM (turned out to be the MoBo, not the RAM, but going to a listed product seemed a good move)
Crucial's site advised that the DDR3L would be OK (looking later, Gigabyte's list didn't include it - oops!) so I got 1600 at, IIRC, 11-11-11-30 and the Windows Experience Index gave, again IIRC, about 4.8 with the RAM as the lowest.
With Crucial Ballistix 1600 at 9-9-9-24, the WEI has gone to 5.9; RAM is 7.7 (out of 7.9) with the SATA3 2.5" HGST 500GB HDD being the limiting factor.

To me, this shows that there's v. little gain by buying super-duper wham-bam-goodbye-mam RAM and, , a good excuse to get an SSD!

[kalniel]

If you've gone from a WEI RAM score of 4.8 to 7.7 for the same amount of RAM with just a tweak of timings then something was wrong with your previous set!

[satrow]

If you've gone from a WEI RAM score of 4.8 to 7.7 for the same amount of RAM with just a tweak of timings then something was wrong with your previous set!

If the 'board wasn't setting the XMP/Dual channel, etc. correctly for the DDR3L, that kind of difference is to be expected.

[Giraffe]

Probably - perhaps the board was marginal with the L type. It did seem a big difference for a small change. "Ballistix" or plain vanilla, the same frequency and not particularly aggressive timings surely would be marginal.

[Lee H]

It depends on the architecture, i.e what platform your using.

With some DDR3 based systems it did make a little improvement, but for the newer skylake platforms from intel, there are various benchmarks all over the place showing 1 fps difference between 2133 MHz and 3000 Mhz with better returns by overclocking your graphics card by a little.

[Bonebreaker777]

When I was playing with timings and checking any measurable differences, I was using AIDA64 3x repeats for each change in settings (including secondary timings). Nailed down possible primary timings pretty soon. 1T-8-8-8-20 @ 1600Mhz using 1.35V.

Never really thought of Windows scoring as a performance metre - maybe that's why its been abandoned in later versions.