1. You are viewing our forum as a guest. For full access please Register. WindowsBBS.com is completely free, paid for by advertisers and donations.

Single sided versus double sided RAM modules

Discussion in 'PC Hardware' started by Christer, 2005/12/22.

  1. 2005/12/22
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    Hello all!

    My apologies for posting a technical question but I have bent my brain and to no avail searched the web for information. The issue concerns memory modules and "Refresh Rate ". I believed that "Refresh Rate" is the inverse of frequency (which in both cases below should give 7.5 ns) but it is obviously not that simple.

    I currently have two identical modules Samsung M366S3253ETS-C7A, single sided, according to Everest:

    Module Size: 256 MB (1 rows, 4 banks)
    Memory Speed: PC133 (133 MHz)
    Module width: 64 bit
    Refresh Rate: Reduced (7.8 µs), Self-Refresh

    Memory timings -
    - @133 MHz: 3.0-3-3-6 (CL-RCD-RP-RAS)
    - @100 MHz: 2.0-2-2-5 (CL-RCD-RP-RAS)

    According to Samsung specifications: "Auto & self refresh capability (8192 Cycles/64ms)" which gives 7.8125 ns. (Everest says µs but that is obviously false ...... :confused: ...... right?)

    I have obtained (from a scrapped computer) a module Samsung M366S3323DTS-C7A, double sided, according to Everest:

    Module Size: 256 MB (2 rows, 4 banks)
    Memory Speed: PC133 (133 MHz)
    Module width: 64 bit
    Refresh Rate: Normal (15.625 µs), Self-Refresh

    Memory timings -
    - @133 MHz: 3.0-3-3-6 (CL-RCD-RP-RAS)
    - @100 MHz: 2.0-2-2-5 (CL-RCD-RP-RAS)

    According to Samsung specifications: "Auto & self refresh capability (4096 Cycles/64ms)" which gives 15.625 ns. (Everest says µs but that is obviously false ...... :confused: ...... right?)

    What I've been trying to figure out is -

    - what is the significance of the different Refresh Rates? (All other specifications indicate that the different modules have the same performance but the respective Refresh Rate indicates that the single sided module is approximately twice as fast.)

    - what would the impact be if I added the "second hand" module to the currently installed modules (mixing)?

    - why is the Refresh Rate "Reduced" for the M366S3253ETS-C7A? (7.8 ns translates into a lower frequency than 7.8125 ns.)

    Thanks for your time,
    Christer
     
  2. 2005/12/22
    sparrow

    sparrow Inactive

    Joined:
    2004/03/21
    Messages:
    2,282
    Likes Received:
    0
    Although you posed a technical question, this is a not so technical answer. Searched a lot today and finally found something pertinent using dogpile.
    According to this page the reason may be that each rank, side, row is addressed by a single Chip-Select pulse, so two rows need 2 pulses (twice the time). Check the bottom of the page under "Definition of Memory Module Terms ".
    Suspect no impact, assuming it's still good, but try it and see (at your own risk, of course).:D Have never shunned mixing brands without a single problem.
    But probably not a significant difference.
     

  3. to hide this advert.

  4. 2005/12/22
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    Hi sparrow,
    I knew that you would stick your neck into this trap ...... ;) ...... !

    I too have searched and found a few webpages with "fragments" of information. It seems like you are on the right track. This is what I have found:

    Refresh Rate :

    Dynamic random access memory :

    The mud is a bit clearer ...... :confused: ...... or ...... :eek: ...... ?

    I now think that the Self Refresh Rate has nothing to do with the operating frequency. The SRR is there to recharge the DRAMs in a constant read-rewrite process and nothing will be changed, just written back. If so, the writing of new data is a separate, "parallel" process (running at 133 MHz regardless of refresh rate).

    If my reasoning "holds water ", then mixing the modules should have no impact on performance. (They are both Samsung PC133 SDRAM but the single sided is a later iteration with higher density DRAMs so, no mixing of brands.) I have tested the module using Windows Memory Diagnostic and the result was OK.

    About the "reduced rate ", again if I'm right, there is no impact on performance.

    I also have to admit that Everest is right, it is 7.8 µs, not 7.8 ns (0.064 s / 8192 = 0.0000078125 s = 7.8125 µs) which means that the "refresh frequency" is approximately 1/1000 of the operating frequency.

    I think that I understand now ...... :p ...... if I'm right, that is. Nothing beats a discussion to clear the thoughts!

    Christer
     
    Last edited: 2005/12/22
  5. 2005/12/23
    mattman

    mattman Inactive Alumni

    Joined:
    2002/06/10
    Messages:
    8,198
    Likes Received:
    63
    Hello Christer,

    I think you have probably answered your own question.

    When speed is mentioned it generally refers to frequency. If you overclock the RAM it will go faster (also, create heat and risk failure). The CAS Latency is next most important (the figures 3.0-3-3-6 after the frequency). The CAS Latency is an internal speed like the refresh rate as you have found, but more important to how well the RAM works. High quality RAM will have a lower CAS Latency than low quality RAM.

    AMD and Intel now have different ways of utilising RAM...DDR or DDR2. You will not be able to compare those because they use different systems.

    I have not researched much more.

    I don't research RAM sticks any more, there are too many variables like the double-sided "difference ". I just put it in and see if it works. The only hazard I know of is to put/force a RAM stick of the wrong voltage into a slot. Early 168 pin SDRAM had lugs/grooves that were only slightly different to later sticks of a different voltage. It was/is possible to force the stick into the wrong slot with enough force. Check if the lugs/grooves match by putting the two sticks edge to edge with one reversed to the other.

    Apart from that...the motherboard may "like" it or it doesn't.

    Matt
    ps ARRRGGH you got me into the discussion!
    :D

    pps EDO or non-EDO is important, the motherboard specs will tell you which type is compatible, won't hurt to try.

    ppps(?) The "comparitors" at the RAM manufacturer's website will have a recommendation for which best/acceptable/works type of RAM is best (works?) with a particular motherboard.

    pppps ARRRRGGH
    :D
     
    Last edited: 2005/12/23
  6. 2005/12/23
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    Hi Matt,
    I was expecting your company too ...... :p ...... !

    Usually, I don't dig too deep into the mud but since Everest reported the Refresh Rate, I thought that it had an impact on performance but I found out that it has to do with conforming to design standards.

    I have learnt about Memory Timings too but ...... :rolleyes: ...... I'll stop digging now.

    Christer
     
  7. 2005/12/23
    Rockster2U

    Rockster2U Geek Member

    Joined:
    2002/04/01
    Messages:
    3,181
    Likes Received:
    9
    Had to Jump In to wish you guys a Merry Christmas. You picked a setting that I know little about but in this case, lower isn't neccessarily better.

    This from the Master (and that ain't me).

    Refresh Rates - Common Options: 7.8 µsec, 15.6 µsec, 31.2 µsec, 64 µsec, 128 µsec, Auto

    This BIOS feature allows you to set the refresh interval of the memory chips. There are three different settings as well as an Auto option. If the Auto option is selected, the BIOS will query the memory modules' SPD chips and use the lowest setting found for maximum compatibility.
    For better performance, you should consider increasing the Refresh Interval from the default values (15.6 µsec for 128Mbit or smaller memory chips and 7.8 µsec for 256Mbit or larger memory chips) up to 128 µsec. Please note that if you increase the Refresh Interval too much, the memory cells may lose their contents.

    Therefore, you should start with small increases in the Refresh Interval and test your system after each hike before increasing it further. If you face stability problems upon increasing the refresh interval, reduce the refresh interval step by step until the system is stable.


    ;)
     
  8. 2005/12/23
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    Hi Rockster2U,
    a Merry Christmas 2U too ...... :) ...... !

    I have looked into a few BIOSes but have not studied the options for memory settings. Will do later.

    Christer
     
  9. 2005/12/23
    mattman

    mattman Inactive Alumni

    Joined:
    2002/06/10
    Messages:
    8,198
    Likes Received:
    63
    I thought this topic sounded like Rockster2U bait :D

    Merry Xmas to you as well, hope things are going well.

    Christer, don't stop "digging" on my account. RAM research drove me crazy when I started, mainly due to the fact that there were so many different types and there was little or no information about individual models. The industry is standardizing a lot of things now so it will be easier to at least find out some information.

    I cannot add much more, but I am interested to read your findings.

    Matt
     
  10. 2005/12/24
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    Matt,
    one of my thoughts was that Refreshing interferes with Writing and Rockster2U gave food for that thought.

    However, imrovement from tweaking memory settings would be minimal (if noticable) and the reseach priority is low.

    Christer
     
  11. 2005/12/24
    Rockster2U

    Rockster2U Geek Member

    Joined:
    2002/04/01
    Messages:
    3,181
    Likes Received:
    9
    Perhaps this explains it a little better and satisfies that inquiring (gotta love it) mind. Again, from the Master's own words:

    Memory cells normally need to be refreshed every 64 msec. However, simultaneously refreshing all the rows in a typical memory chip will cause a big surge in power requirements. In addition, a simultaneous refresh causes all data requests to stall, which greatly impacts performance.

    To avoid both problems, refreshes are normally staggered according to the number of rows. Since a typical memory chip contains 4096 rows, the memory controller usually refreshes a different row every 15.6 µsec (64,000 µsec / 4096 rows = 15.6 µsec). This reduces the amount of current used during each refresh and it allows data to be accessed from rows that are not being refreshed.

    Usually, memory modules that use 128Mbit or smaller memory chips have 4096 rows while memory chips with higher capacity (256Mbit and above) will have 8192 rows. For memory chips that come with 8192 rows, the refresh interval will need to be halved to 7.8 µsec because there are now twice as many rows to serviced within the stipulated 64 msec for the entire chip.

    Therefore, the typical refresh interval for 128Mbit (not MB!) or smaller memory chips would be 15.6 µsec while those for 256Mbit or larger memory chips would be 7.8 µsec. Please note that if you are using a mix of 128Mbit and 256Mbit memory modules, the fail-safe Refresh Interval would be 7.8 µsec, not 15.6 µsec.

    Although JEDEC standards call for a 64 msec refresh cycle, memory chips these days can actually hold data for longer than that. So, using a longer refresh cycle is quite possible. With a longer refresh cycle, the memory chips are refreshed less often, reducing both the amount of bandwidth wasted on refreshes and the amount of power consumed (which is great for laptops and other portable devices).


    ;)
     
  12. 2005/12/24
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    Thanks, Rockster2U!

    Who is the "Master "? Any site reference?

    Christer
     
  13. 2005/12/24
    charlesvar

    charlesvar Inactive Alumni

    Joined:
    2002/02/18
    Messages:
    7,024
    Likes Received:
    0
    This has been a fascinating discussion, nothing to add technically, just wanted to say Merry Christmass Guys!

    And welcome back Rock :) Its been a long time.

    Regards - Charles
     
  14. 2005/12/24
    Rockster2U

    Rockster2U Geek Member

    Joined:
    2002/04/01
    Messages:
    3,181
    Likes Received:
    9
    Adrian Wong - third year medical student from Malaysia. Has authored the best BIOS Guide available anywhere at any cost in my opinion. His site is Adrian's Rojak Pot - named after a mixture of local Malaysian foods. There is a free version and as of about 3 years ago, he started a paid subscription version which covers subjects unavailable anywhere else on the Internet. (At least I've never found them anywhere else) He gets my contribution every year.

    Charles - leave it to Chister to come up with a topic like this, huh? It grabbed my interest too and I have to admit, I didn't know much about refresh intervals either.

    ;)
     
  15. 2005/12/24
    Christer

    Christer Geek Member Staff Thread Starter

    Joined:
    2002/12/17
    Messages:
    6,585
    Likes Received:
    74
    Charles,
    a Merry Christmas to you too!

    There is a saying: "Curiosity killed the cat" ...... :eek: ...... and I've only got one life ...... :D ...... !

    Rockster2U,
    I'll have a look at his site later. Now, it's time for the first Christmas Beer!

    Christer
     

Share This Page

  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.