New Ram kit performing worse than old Ram on same frequency/timings

TheBlock

Reputable
Jan 4, 2016
23
0
4,520
Hello,

I just purchased 2x4GB of Corsair Vengeance Low Profile Ram (1600MHz, CL9) (Amazon Link) to replace/extend my old 2x2GB Elixir Nanya Ram (1333MHz CL9) (Nanya Technology M2Y2G64CB8HC5N-CG)

Before I replaced the Elixir Sticks, I ran MaxxMEM2 as a benchmark to see how large the improvements in speed are going to be (I upgraded mostly for the capacity, but I thought I would gain some (benchmark-)performance too). I also overclocked the old Elixir Ram to the standard specs of the Corsair Ram, to see if there would be any difference in speed/how much faster the corsair would be. I was also surprised by how far I was able to overclock this Elixir set, as I was able to get to 1600MHz, 8-8-8-24-1T, on these 1333MHz CL9 listed sticks. (I stopped overclocking here, so maybe it would have gone even further)

After I replaced the old sticks with the Corsair Memory (onto same slots on Mobo), I ran MemTest86 for 3 passes (no errors) and then I performed the same benchmark on this new set, first with the stock listed frequency and latency, then underclocked to the stock frequency and latency of the old Elixir Ram, and then all others that I have tested, and it turned out that the old Elixir Nanya Ram performed better than the new Corsair Ram in all tests!


Why is this? Is it possible that the larger capacity of the sticks will decrease the performance?



MaxxMem2 performance of both sticks on the same frequency and timings


(I will soon try to extend my Memory, by installing both the 2x4GB Corsair and 2x2GB Elixir onto my motherboard. Im not sure if it will work because they are so different, but if it won't its not a problem at all, because my housemate has a computer with similar memory, so he can have them.)


CPU: AMD Phenom II x4 965
CPU Cooler: ARCTIC Freezer Xtreme Rev. 2
Mainboard: GA-880GA-UD3H
 
Solution
Please don't mix and match two different sets of ram. In essence it will show up as 12GB of available ram though both kits have varying internal timings and frequencies. The reason you're seeing a difference in performance with similar frequencies is because one kit was made to run at CL9 while the other kit ran at CL8 and at 1600MHz the lower numbers mean tighter timings. To add food for thought, if you considered ram being manufactured off an assembly line, any changes in temperature can cause minor changes in its internal timings and thus its overall performance. The Vengeance series of ram's are considered to be mass produced with laxed QC measures put in place so although they may be faster then most out there they aren't...

Lutfij

Titan
Moderator
Please don't mix and match two different sets of ram. In essence it will show up as 12GB of available ram though both kits have varying internal timings and frequencies. The reason you're seeing a difference in performance with similar frequencies is because one kit was made to run at CL9 while the other kit ran at CL8 and at 1600MHz the lower numbers mean tighter timings. To add food for thought, if you considered ram being manufactured off an assembly line, any changes in temperature can cause minor changes in its internal timings and thus its overall performance. The Vengeance series of ram's are considered to be mass produced with laxed QC measures put in place so although they may be faster then most out there they aren't binned with a high standard. The Dominator Platinum series is where timings are more tighter with respect to their frequency's and due to their stricter binning measures are priced higher. Might I ask at what voltages you ran both kits on your separate tests? Have you made sure your BIOS is up to date prior to populating your slots with the new kit?

The other side of the story can be luck of the draw as most have owned and even overclocked on a similar set of Elixir rams and noticed nothing but heat being performed compared to newer kits in the market.
 
Solution