The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    M17x R1 RAM question

    Discussion in 'Alienware 17 and M17x' started by TiMN8R, Jun 5, 2011.

  1. TiMN8R

    TiMN8R Notebook Enthusiast

    Reputations:
    0
    Messages:
    23
    Likes Received:
    0
    Trophy Points:
    5
    Hey guys!

    Edit::I think the problem is resolved, if it even was a problem. If you're curious, do read on, but I don't think anything can be done.

    I selected 4GB of 1333MHz RAM when I purchased this machine. After a recent upgrade to a quad-core (best thing I ever did, btw), I had Windows re-evaluate its WEI, and to my amazement, everything shot upwards at least .4 points, except for the RAM score, which holds the low at 5.9. Well, I don't put much trust in that dumb benchmark, but it made me wonder, so I looked at the results from HWiNFO32, and it claims that the maximum transfer rate is between 2650MB/s and 3200MB/s. (each time its bench runs, it gives a slightly different result) CPU-Z also shows this, and the same with PassMark software. They all also agree that the timing is 9-9-9-24, except for poor ol' HWiNFO32, which is confused and thinks it's 2-9-9-24. :p But in short, Alienware didn't just take the cheapest thing they could find, but instead, they gave something with decent latency numbers, especially for a laptop.

    Now, why on Earth is the data rate averaging 2.9GB/s, when it's PC3-10800 RAM (10.8GB/s theoretical capacity), as reported by CPU-Z? By the way, that statistic is stored in the SPD chip on each RAM sticks, and both modules independently report the same thing. So I know it's real. That's less than one third of the theoretical capacity for a single module. But then again, there's two modules, and it is a dual-channel architecture. So... 10.8GB/s * 2 = 21.6GB/s total theoretical capacity. Then why's it at 2.9-ish?

    Any ideas? Is anyone else having this same problem? (If you don't know, you may want to look into it. 13.4% efficiency is pathetic!!)

    Thanks!
    ~TiMN8R
     
    TiMN8R, Jun 5, 2011
    #1
  2. TiMN8R

    TiMN8R Notebook Enthusiast

    Reputations:
    0
    Messages:
    23
    Likes Received:
    0
    Trophy Points:
    5
    And now, it's looking like those benchmarks I used only ever write or read from a single module. I just ran AIDA64 (Everest's replacement), the RAM and cache bench, and it reported something much more reasonable, at 6600MB/s. That is comparable to other DDR3 speeds that I've found online, so I guess the problem is resolved because it isn't a problem. Now, why is it 60% efficient at full load? Just a question for the hardware guys, I guess.
     
    TiMN8R, Jun 5, 2011
    #2