Closed Solved

RAM Latency vs. Clock speed: which and why?

My ram is starting to go bad and i need a replacement, the only problem is which do i choose?

The two i'm considering are

RAM A: http://www.newegg.com/Product/Product.aspx?Item=N82E16820231689
and
RAM B: http://www.newegg.com/Product/Product.aspx?Item=N82E16820231428

So ram A is 2133mhz, but has CAS latency timings of 9-11-11-31 and a 1.6-1.65v.
Whereas ram B is 1600mhz, but has CAS latency timings of 9-9-9-24 and 1.5v

I know those are the deciding factors in choosing ram, and am under the impression that lower latency timings can outweigh the actual speed of the ram. Can anyone clarify?

This is primarily for a gaming PC, but how would the use change the outcome?

Thanks!
28 answers Last reply Best Answer
More about ram latency clock speed
  1. Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.
  2. Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?
  3. thefoxer said:

    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    well, that's will be long explanation..:D
    you can read here for complete tutorial of memory timing => http://www.hardwaresecrets.com/article/Understanding-RAM-Timings/26
  4. There are also faster 1600Mhz kits than just 9-9-9-24.

    You can buy them in Cas7 even buy 1866Mhz kits in CAS8. 2133Mhz seems to be sweet spot for integrated GPUs at least for Haswell and Richland.
  5. Quaddro said:
    thefoxer said:

    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    well, that's will be long explanation..:D
    you can read here for complete tutorial of memory timing => http://www.hardwaresecrets.com/article/Understanding-RAM-Timings/26


    Thanks for the info, you've been incredibly insightful!
  6. Immaculate said:
    There are also faster 1600Mhz kits than just 9-9-9-24.

    You can buy them in Cas7 even buy 1866Mhz kits in CAS8. 2133Mhz seems to be sweet spot for integrated GPUs at least for Haswell and Richland.


    I understand this, and what i'm unsure of is when that performance won't make a difference, in my case, for gaming.
  7. Best answer
    thefoxer said:
    Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    It may seem like it makes sense but most of what he wrote is incorrect.

    DDR3-2133 is a memory module rated for a transfer rate of 2133 megabits per second on each IO pin. Data on the IO pin is exchanged on both the rising edge and the falling edge of the DRAM IO bus reference clock. Thus, a module rated for DDR3-2133 has an IO bus frequency of 1066Mhz, not 2133Mhz. The measure of IO pin transfer rate is in transfers per second, not cycles per second, so a DDR transfer rate of 2133MT/s has a bus frequency of 1066Mhz. The same holds true for DDR, DDR2, and DDR3.

    The actual DRAM modules themselves operate at an even slower rate. What do DDR-400, DDR2-800, and DDR3-1600 all have in common? The core memory clock is 200Mhz.

    DDR-400 has an IO bus frequency of 200Mhz, and a 1:1 ratio between the IO bus reference clock and the core memory clock.

    DDR2-800 has an IO bus frequency of 400Mhz, and a 2:1 ratio between the IO bus reference clock and the core memory clock.

    DDR3-1600 has an IO bus frequency of 800Mhz, and a 4:1 ratio between the IO bus reference clock and the core memory clock.

    This pattern is an example of one of the most misunderstood aspects of modern DRAM. It has gotten wider and deeper -- which allows for greater capacity and larger prefetch buffers -- not faster which allows for lower latency. Latency measured in real time (nanoseconds) hasn't changed significantly over the past decade.

    The most commonly cited DRAM performance metric aside from the IO bus transfer rate is the CAS latency. The CAS latency is the duration (in clock cycles) that it takes for the DRAM module performing a read operation to register a column address and produce a stable output on the IO pins. This delay is measured in IO bus clock cycles, which as noted above scale between generations. This is why DDR2-800C (CL4) has the same first word delay of 10ns as DDR3-1600G (CL8). Where DDR3 really improves over its predecessors is in fourth and eighth word delays, this is accomplished through a deeper prefetch buffer (outside of the scope of this explanation).

    So it should be no surprise then that DDR3-2133 with CAS 9 is faster than DDR3-1600 with CAS9. This is nice, but somewhat immaterial. Why is that?

    x86 memory channels data paths are 64 bits wide, and the DRAM modules are arranged to fit this. If you look at a DRAM module without the heatsink attached (eg, this one) you'll see 8 integrated circuits on one side of the PCB. Each of these ICs is an 8 bit DRAM module and they are all tied together with common command and address lines, but with data lines formed into a 64 bit bus. Arranged together, they form what is called a rank. DIMMs that have 8 modules on each side arranged into two ranks are called dual-rank. Servers often use compact modules which have 4 or even 8 ranks crammed onto a single PCB for extreme amounts of memory.

    The memory controller can only access one rank at a time, even if two or more ranks are installed in the channel on one or more DIMMs. However, the individual DRAM chips themselves are broken down into individual banks, each of which can be active and working independently at the same time. DDR3 modules have 8 individual banks on each IC. So, even with high DRAM latencies, the DRAM controller can keep the entire channel busy nearly 100% of the time simply by switching to another bank after a command has been issued, and switching back when the output is ready. Thus, high speed memory almost always outperforms low latency memory; this is especially true when high speed memory is accompanied by low latency (often at the cost of higher supply voltage).

    This does not necessarily mean that DDR3-2133 is necessary or even a good idea. I run 32GiB of it in my PC but it requires some extra tweaking to obtain stability.

    I hope that this answered your question
  8. thefoxer said:
    Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    Kinda late but FYI:
    "Memory timings or RAM timings measure the performance of DRAM memory using four parameters: CL, tRCD, tRP, and tRAS in units of clock cycles; they are commonly written as four numbers separated with dashes, e.g. 7-8-8-24. " src: http://en.wikipedia.org/wiki/Memory_timings
  9. Pinhedd said:
    thefoxer said:
    Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    It may seem like it makes sense but most of what he wrote is incorrect.

    DDR3-2133 is a memory module rated for a transfer rate of 2133 megabits per second on each IO pin. Data on the IO pin is exchanged on both the rising edge and the falling edge of the DRAM IO bus reference clock. Thus, a module rated for DDR3-2133 has an IO bus frequency of 1066Mhz, not 2133Mhz. The measure of IO pin transfer rate is in transfers per second, not cycles per second, so a DDR transfer rate of 2133MT/s has a bus frequency of 1066Mhz. The same holds true for DDR, DDR2, and DDR3.

    The actual DRAM modules themselves operate at an even slower rate. What do DDR-400, DDR2-800, and DDR3-1600 all have in common? The core memory clock is 200Mhz.

    DDR-400 has an IO bus frequency of 200Mhz, and a 1:1 ratio between the IO bus reference clock and the core memory clock.

    DDR2-800 has an IO bus frequency of 400Mhz, and a 2:1 ratio between the IO bus reference clock and the core memory clock.

    DDR3-1600 has an IO bus frequency of 800Mhz, and a 4:1 ratio between the IO bus reference clock and the core memory clock.

    This pattern is an example of one of the most misunderstood aspects of modern DRAM. It has gotten wider and deeper -- which allows for greater capacity and larger prefetch buffers -- not faster which allows for lower latency. Latency measured in real time (nanoseconds) hasn't changed significantly over the past decade.

    The most commonly cited DRAM performance metric aside from the IO bus transfer rate is the CAS latency. The CAS latency is the duration (in clock cycles) that it takes for the DRAM module performing a read operation to register a column address and produce a stable output on the IO pins. This delay is measured in IO bus clock cycles, which as noted above scale between generations. This is why DDR2-800C (CL4) has the same first word delay of 10ns as DDR3-1600G (CL8). Where DDR3 really improves over its predecessors is in fourth and eighth word delays, this is accomplished through a deeper prefetch buffer (outside of the scope of this explanation).

    So it should be no surprise then that DDR3-2133 with CAS 9 is faster than DDR3-1600 with CAS9. This is nice, but somewhat immaterial. Why is that?

    x86 memory channels data paths are 64 bits wide, and the DRAM modules are arranged to fit this. If you look at a DRAM module without the heatsink attached (eg, this one) you'll see 8 integrated circuits on one side of the PCB. Each of these ICs is an 8 bit DRAM module and they are all tied together with common command and address lines, but with data lines formed into a 64 bit bus. Arranged together, they form what is called a rank. DIMMs that have 8 modules on each side arranged into two ranks are called dual-rank. Servers often use compact modules which have 4 or even 8 ranks crammed onto a single PCB for extreme amounts of memory.

    The memory controller can only access one rank at a time, even if two or more ranks are installed in the channel on one or more DIMMs. However, the individual DRAM chips themselves are broken down into individual banks, each of which can be active and working independently at the same time. DDR3 modules have 8 individual banks on each IC. So, even with high DRAM latencies, the DRAM controller can keep the entire channel busy nearly 100% of the time simply by switching to another bank after a command has been issued, and switching back when the output is ready. Thus, high speed memory almost always outperforms low latency memory; this is especially true when high speed memory is accompanied by low latency (often at the cost of higher supply voltage).

    This does not necessarily mean that DDR3-2133 is necessary or even a good idea. I run 32GiB of it in my PC but it requires some extra tweaking to obtain stability.

    I hope that this answered your question


    Excuse me sir, would be so kind to inform me what you would recommend in regards to 32g kit or dual 16g kits with 2133hmz and/or 2400hmz memory for an Asus Rampage IV Extreme mobo? I'm also running a i7 3930k OC'd to 4.8ghz stable.
  10. WhiteFigs said:
    Pinhedd said:
    thefoxer said:
    Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    It may seem like it makes sense but most of what he wrote is incorrect.

    DDR3-2133 is a memory module rated for a transfer rate of 2133 megabits per second on each IO pin. Data on the IO pin is exchanged on both the rising edge and the falling edge of the DRAM IO bus reference clock. Thus, a module rated for DDR3-2133 has an IO bus frequency of 1066Mhz, not 2133Mhz. The measure of IO pin transfer rate is in transfers per second, not cycles per second, so a DDR transfer rate of 2133MT/s has a bus frequency of 1066Mhz. The same holds true for DDR, DDR2, and DDR3.

    The actual DRAM modules themselves operate at an even slower rate. What do DDR-400, DDR2-800, and DDR3-1600 all have in common? The core memory clock is 200Mhz.

    DDR-400 has an IO bus frequency of 200Mhz, and a 1:1 ratio between the IO bus reference clock and the core memory clock.

    DDR2-800 has an IO bus frequency of 400Mhz, and a 2:1 ratio between the IO bus reference clock and the core memory clock.

    DDR3-1600 has an IO bus frequency of 800Mhz, and a 4:1 ratio between the IO bus reference clock and the core memory clock.

    This pattern is an example of one of the most misunderstood aspects of modern DRAM. It has gotten wider and deeper -- which allows for greater capacity and larger prefetch buffers -- not faster which allows for lower latency. Latency measured in real time (nanoseconds) hasn't changed significantly over the past decade.

    The most commonly cited DRAM performance metric aside from the IO bus transfer rate is the CAS latency. The CAS latency is the duration (in clock cycles) that it takes for the DRAM module performing a read operation to register a column address and produce a stable output on the IO pins. This delay is measured in IO bus clock cycles, which as noted above scale between generations. This is why DDR2-800C (CL4) has the same first word delay of 10ns as DDR3-1600G (CL8). Where DDR3 really improves over its predecessors is in fourth and eighth word delays, this is accomplished through a deeper prefetch buffer (outside of the scope of this explanation).

    So it should be no surprise then that DDR3-2133 with CAS 9 is faster than DDR3-1600 with CAS9. This is nice, but somewhat immaterial. Why is that?

    x86 memory channels data paths are 64 bits wide, and the DRAM modules are arranged to fit this. If you look at a DRAM module without the heatsink attached (eg, this one) you'll see 8 integrated circuits on one side of the PCB. Each of these ICs is an 8 bit DRAM module and they are all tied together with common command and address lines, but with data lines formed into a 64 bit bus. Arranged together, they form what is called a rank. DIMMs that have 8 modules on each side arranged into two ranks are called dual-rank. Servers often use compact modules which have 4 or even 8 ranks crammed onto a single PCB for extreme amounts of memory.

    The memory controller can only access one rank at a time, even if two or more ranks are installed in the channel on one or more DIMMs. However, the individual DRAM chips themselves are broken down into individual banks, each of which can be active and working independently at the same time. DDR3 modules have 8 individual banks on each IC. So, even with high DRAM latencies, the DRAM controller can keep the entire channel busy nearly 100% of the time simply by switching to another bank after a command has been issued, and switching back when the output is ready. Thus, high speed memory almost always outperforms low latency memory; this is especially true when high speed memory is accompanied by low latency (often at the cost of higher supply voltage).

    This does not necessarily mean that DDR3-2133 is necessary or even a good idea. I run 32GiB of it in my PC but it requires some extra tweaking to obtain stability.

    I hope that this answered your question


    Excuse me sir, would be so kind to inform me what you would recommend in regards to 32g kit or dual 16g kits with 2133hmz and/or 2400hmz memory for an Asus Rampage IV Extreme mobo? I'm also running a i7 3930k OC'd to 4.8ghz stable.


    Hi,

    I'm running two sets of these

    http://www.newegg.com/Product/Product.aspx?Item=N82E16820226274

    32GiB stable using just the XMP profile for the memory. I have to boost the IMC and SA voltage slightly though.
  11. Pinhedd said:
    WhiteFigs said:
    Pinhedd said:
    thefoxer said:
    Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    It may seem like it makes sense but most of what he wrote is incorrect.

    DDR3-2133 is a memory module rated for a transfer rate of 2133 megabits per second on each IO pin. Data on the IO pin is exchanged on both the rising edge and the falling edge of the DRAM IO bus reference clock. Thus, a module rated for DDR3-2133 has an IO bus frequency of 1066Mhz, not 2133Mhz. The measure of IO pin transfer rate is in transfers per second, not cycles per second, so a DDR transfer rate of 2133MT/s has a bus frequency of 1066Mhz. The same holds true for DDR, DDR2, and DDR3.

    The actual DRAM modules themselves operate at an even slower rate. What do DDR-400, DDR2-800, and DDR3-1600 all have in common? The core memory clock is 200Mhz.

    DDR-400 has an IO bus frequency of 200Mhz, and a 1:1 ratio between the IO bus reference clock and the core memory clock.

    DDR2-800 has an IO bus frequency of 400Mhz, and a 2:1 ratio between the IO bus reference clock and the core memory clock.

    DDR3-1600 has an IO bus frequency of 800Mhz, and a 4:1 ratio between the IO bus reference clock and the core memory clock.

    This pattern is an example of one of the most misunderstood aspects of modern DRAM. It has gotten wider and deeper -- which allows for greater capacity and larger prefetch buffers -- not faster which allows for lower latency. Latency measured in real time (nanoseconds) hasn't changed significantly over the past decade.

    The most commonly cited DRAM performance metric aside from the IO bus transfer rate is the CAS latency. The CAS latency is the duration (in clock cycles) that it takes for the DRAM module performing a read operation to register a column address and produce a stable output on the IO pins. This delay is measured in IO bus clock cycles, which as noted above scale between generations. This is why DDR2-800C (CL4) has the same first word delay of 10ns as DDR3-1600G (CL8). Where DDR3 really improves over its predecessors is in fourth and eighth word delays, this is accomplished through a deeper prefetch buffer (outside of the scope of this explanation).

    So it should be no surprise then that DDR3-2133 with CAS 9 is faster than DDR3-1600 with CAS9. This is nice, but somewhat immaterial. Why is that?

    x86 memory channels data paths are 64 bits wide, and the DRAM modules are arranged to fit this. If you look at a DRAM module without the heatsink attached (eg, this one) you'll see 8 integrated circuits on one side of the PCB. Each of these ICs is an 8 bit DRAM module and they are all tied together with common command and address lines, but with data lines formed into a 64 bit bus. Arranged together, they form what is called a rank. DIMMs that have 8 modules on each side arranged into two ranks are called dual-rank. Servers often use compact modules which have 4 or even 8 ranks crammed onto a single PCB for extreme amounts of memory.

    The memory controller can only access one rank at a time, even if two or more ranks are installed in the channel on one or more DIMMs. However, the individual DRAM chips themselves are broken down into individual banks, each of which can be active and working independently at the same time. DDR3 modules have 8 individual banks on each IC. So, even with high DRAM latencies, the DRAM controller can keep the entire channel busy nearly 100% of the time simply by switching to another bank after a command has been issued, and switching back when the output is ready. Thus, high speed memory almost always outperforms low latency memory; this is especially true when high speed memory is accompanied by low latency (often at the cost of higher supply voltage).

    This does not necessarily mean that DDR3-2133 is necessary or even a good idea. I run 32GiB of it in my PC but it requires some extra tweaking to obtain stability.

    I hope that this answered your question


    Excuse me sir, would be so kind to inform me what you would recommend in regards to 32g kit or dual 16g kits with 2133hmz and/or 2400hmz memory for an Asus Rampage IV Extreme mobo? I'm also running a i7 3930k OC'd to 4.8ghz stable.


    Hi,

    I'm running two sets of these

    http://www.newegg.com/Product/Product.aspx?Item=N82E16820226274

    32GiB stable using just the XMP profile for the memory. I have to boost the IMC and SA voltage slightly though.


    Alright though would that be the best possible option for my mobo and cpu? I assumed there can be issues with cpu compatibility in regards to 2400 freq memory.
  12. WhiteFigs said:
    Pinhedd said:
    WhiteFigs said:
    Pinhedd said:
    thefoxer said:
    Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    It may seem like it makes sense but most of what he wrote is incorrect.

    DDR3-2133 is a memory module rated for a transfer rate of 2133 megabits per second on each IO pin. Data on the IO pin is exchanged on both the rising edge and the falling edge of the DRAM IO bus reference clock. Thus, a module rated for DDR3-2133 has an IO bus frequency of 1066Mhz, not 2133Mhz. The measure of IO pin transfer rate is in transfers per second, not cycles per second, so a DDR transfer rate of 2133MT/s has a bus frequency of 1066Mhz. The same holds true for DDR, DDR2, and DDR3.

    The actual DRAM modules themselves operate at an even slower rate. What do DDR-400, DDR2-800, and DDR3-1600 all have in common? The core memory clock is 200Mhz.

    DDR-400 has an IO bus frequency of 200Mhz, and a 1:1 ratio between the IO bus reference clock and the core memory clock.

    DDR2-800 has an IO bus frequency of 400Mhz, and a 2:1 ratio between the IO bus reference clock and the core memory clock.

    DDR3-1600 has an IO bus frequency of 800Mhz, and a 4:1 ratio between the IO bus reference clock and the core memory clock.

    This pattern is an example of one of the most misunderstood aspects of modern DRAM. It has gotten wider and deeper -- which allows for greater capacity and larger prefetch buffers -- not faster which allows for lower latency. Latency measured in real time (nanoseconds) hasn't changed significantly over the past decade.

    The most commonly cited DRAM performance metric aside from the IO bus transfer rate is the CAS latency. The CAS latency is the duration (in clock cycles) that it takes for the DRAM module performing a read operation to register a column address and produce a stable output on the IO pins. This delay is measured in IO bus clock cycles, which as noted above scale between generations. This is why DDR2-800C (CL4) has the same first word delay of 10ns as DDR3-1600G (CL8). Where DDR3 really improves over its predecessors is in fourth and eighth word delays, this is accomplished through a deeper prefetch buffer (outside of the scope of this explanation).

    So it should be no surprise then that DDR3-2133 with CAS 9 is faster than DDR3-1600 with CAS9. This is nice, but somewhat immaterial. Why is that?

    x86 memory channels data paths are 64 bits wide, and the DRAM modules are arranged to fit this. If you look at a DRAM module without the heatsink attached (eg, this one) you'll see 8 integrated circuits on one side of the PCB. Each of these ICs is an 8 bit DRAM module and they are all tied together with common command and address lines, but with data lines formed into a 64 bit bus. Arranged together, they form what is called a rank. DIMMs that have 8 modules on each side arranged into two ranks are called dual-rank. Servers often use compact modules which have 4 or even 8 ranks crammed onto a single PCB for extreme amounts of memory.

    The memory controller can only access one rank at a time, even if two or more ranks are installed in the channel on one or more DIMMs. However, the individual DRAM chips themselves are broken down into individual banks, each of which can be active and working independently at the same time. DDR3 modules have 8 individual banks on each IC. So, even with high DRAM latencies, the DRAM controller can keep the entire channel busy nearly 100% of the time simply by switching to another bank after a command has been issued, and switching back when the output is ready. Thus, high speed memory almost always outperforms low latency memory; this is especially true when high speed memory is accompanied by low latency (often at the cost of higher supply voltage).

    This does not necessarily mean that DDR3-2133 is necessary or even a good idea. I run 32GiB of it in my PC but it requires some extra tweaking to obtain stability.

    I hope that this answered your question


    Excuse me sir, would be so kind to inform me what you would recommend in regards to 32g kit or dual 16g kits with 2133hmz and/or 2400hmz memory for an Asus Rampage IV Extreme mobo? I'm also running a i7 3930k OC'd to 4.8ghz stable.


    Hi,

    I'm running two sets of these

    http://www.newegg.com/Product/Product.aspx?Item=N82E16820226274

    32GiB stable using just the XMP profile for the memory. I have to boost the IMC and SA voltage slightly though.


    Alright though would that be the best possible option for my mobo and cpu? I assumed there can be issues with cpu compatibility in regards to 2400 freq memory.


    There's always the risk of incompatibility or having to tweak settings when configuring memory outside of the official parameters specified by Intel or AMD. Mushkin offers excellent bang for your buck.
  13. Pinhedd said:
    thefoxer said:
    Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    It may seem like it makes sense but most of what he wrote is incorrect.

    DDR3-2133 is a memory module rated for a transfer rate of 2133 megabits per second on each IO pin. Data on the IO pin is exchanged on both the rising edge and the falling edge of the DRAM IO bus reference clock. Thus, a module rated for DDR3-2133 has an IO bus frequency of 1066Mhz, not 2133Mhz. The measure of IO pin transfer rate is in transfers per second, not cycles per second, so a DDR transfer rate of 2133MT/s has a bus frequency of 1066Mhz. The same holds true for DDR, DDR2, and DDR3.

    The actual DRAM modules themselves operate at an even slower rate. What do DDR-400, DDR2-800, and DDR3-1600 all have in common? The core memory clock is 200Mhz.

    DDR-400 has an IO bus frequency of 200Mhz, and a 1:1 ratio between the IO bus reference clock and the core memory clock.

    DDR2-800 has an IO bus frequency of 400Mhz, and a 2:1 ratio between the IO bus reference clock and the core memory clock.

    DDR3-1600 has an IO bus frequency of 800Mhz, and a 4:1 ratio between the IO bus reference clock and the core memory clock.

    This pattern is an example of one of the most misunderstood aspects of modern DRAM. It has gotten wider and deeper -- which allows for greater capacity and larger prefetch buffers -- not faster which allows for lower latency. Latency measured in real time (nanoseconds) hasn't changed significantly over the past decade.

    The most commonly cited DRAM performance metric aside from the IO bus transfer rate is the CAS latency. The CAS latency is the duration (in clock cycles) that it takes for the DRAM module performing a read operation to register a column address and produce a stable output on the IO pins. This delay is measured in IO bus clock cycles, which as noted above scale between generations. This is why DDR2-800C (CL4) has the same first word delay of 10ns as DDR3-1600G (CL8). Where DDR3 really improves over its predecessors is in fourth and eighth word delays, this is accomplished through a deeper prefetch buffer (outside of the scope of this explanation).

    So it should be no surprise then that DDR3-2133 with CAS 9 is faster than DDR3-1600 with CAS9. This is nice, but somewhat immaterial. Why is that?

    x86 memory channels data paths are 64 bits wide, and the DRAM modules are arranged to fit this. If you look at a DRAM module without the heatsink attached (eg, this one) you'll see 8 integrated circuits on one side of the PCB. Each of these ICs is an 8 bit DRAM module and they are all tied together with common command and address lines, but with data lines formed into a 64 bit bus. Arranged together, they form what is called a rank. DIMMs that have 8 modules on each side arranged into two ranks are called dual-rank. Servers often use compact modules which have 4 or even 8 ranks crammed onto a single PCB for extreme amounts of memory.

    The memory controller can only access one rank at a time, even if two or more ranks are installed in the channel on one or more DIMMs. However, the individual DRAM chips themselves are broken down into individual banks, each of which can be active and working independently at the same time. DDR3 modules have 8 individual banks on each IC. So, even with high DRAM latencies, the DRAM controller can keep the entire channel busy nearly 100% of the time simply by switching to another bank after a command has been issued, and switching back when the output is ready. Thus, high speed memory almost always outperforms low latency memory; this is especially true when high speed memory is accompanied by low latency (often at the cost of higher supply voltage).

    This does not necessarily mean that DDR3-2133 is necessary or even a good idea. I run 32GiB of it in my PC but it requires some extra tweaking to obtain stability.

    I hope that this answered your question


    so would you say Corsair Vengeance Pro Series 8GB (2 x 4GB) DDR3 DRAM 2400MHz C11 Memory Kit is good and will work well with a 4690k i5?
  14. aschoneveld said:
    Pinhedd said:
    thefoxer said:
    Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    It may seem like it makes sense but most of what he wrote is incorrect.

    DDR3-2133 is a memory module rated for a transfer rate of 2133 megabits per second on each IO pin. Data on the IO pin is exchanged on both the rising edge and the falling edge of the DRAM IO bus reference clock. Thus, a module rated for DDR3-2133 has an IO bus frequency of 1066Mhz, not 2133Mhz. The measure of IO pin transfer rate is in transfers per second, not cycles per second, so a DDR transfer rate of 2133MT/s has a bus frequency of 1066Mhz. The same holds true for DDR, DDR2, and DDR3.

    The actual DRAM modules themselves operate at an even slower rate. What do DDR-400, DDR2-800, and DDR3-1600 all have in common? The core memory clock is 200Mhz.

    DDR-400 has an IO bus frequency of 200Mhz, and a 1:1 ratio between the IO bus reference clock and the core memory clock.

    DDR2-800 has an IO bus frequency of 400Mhz, and a 2:1 ratio between the IO bus reference clock and the core memory clock.

    DDR3-1600 has an IO bus frequency of 800Mhz, and a 4:1 ratio between the IO bus reference clock and the core memory clock.

    This pattern is an example of one of the most misunderstood aspects of modern DRAM. It has gotten wider and deeper -- which allows for greater capacity and larger prefetch buffers -- not faster which allows for lower latency. Latency measured in real time (nanoseconds) hasn't changed significantly over the past decade.

    The most commonly cited DRAM performance metric aside from the IO bus transfer rate is the CAS latency. The CAS latency is the duration (in clock cycles) that it takes for the DRAM module performing a read operation to register a column address and produce a stable output on the IO pins. This delay is measured in IO bus clock cycles, which as noted above scale between generations. This is why DDR2-800C (CL4) has the same first word delay of 10ns as DDR3-1600G (CL8). Where DDR3 really improves over its predecessors is in fourth and eighth word delays, this is accomplished through a deeper prefetch buffer (outside of the scope of this explanation).

    So it should be no surprise then that DDR3-2133 with CAS 9 is faster than DDR3-1600 with CAS9. This is nice, but somewhat immaterial. Why is that?

    x86 memory channels data paths are 64 bits wide, and the DRAM modules are arranged to fit this. If you look at a DRAM module without the heatsink attached (eg, this one) you'll see 8 integrated circuits on one side of the PCB. Each of these ICs is an 8 bit DRAM module and they are all tied together with common command and address lines, but with data lines formed into a 64 bit bus. Arranged together, they form what is called a rank. DIMMs that have 8 modules on each side arranged into two ranks are called dual-rank. Servers often use compact modules which have 4 or even 8 ranks crammed onto a single PCB for extreme amounts of memory.

    The memory controller can only access one rank at a time, even if two or more ranks are installed in the channel on one or more DIMMs. However, the individual DRAM chips themselves are broken down into individual banks, each of which can be active and working independently at the same time. DDR3 modules have 8 individual banks on each IC. So, even with high DRAM latencies, the DRAM controller can keep the entire channel busy nearly 100% of the time simply by switching to another bank after a command has been issued, and switching back when the output is ready. Thus, high speed memory almost always outperforms low latency memory; this is especially true when high speed memory is accompanied by low latency (often at the cost of higher supply voltage).

    This does not necessarily mean that DDR3-2133 is necessary or even a good idea. I run 32GiB of it in my PC but it requires some extra tweaking to obtain stability.

    I hope that this answered your question


    so would you say Corsair Vengeance Pro Series 8GB (2 x 4GB) DDR3 DRAM 2400MHz C11 Memory Kit is good and will work well with a 4690k i5?


    Pretty much any brand name DDR3 will be fine.
  15. Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    wow that makes sense

    so...
    1333MHz CL6 is better than 1600MHz CL9

    Because... 6/1333= 0,0045 and 9/1600=0,0056
    right?
  16. freerq said:
    Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    wow that makes sense

    so...
    1333MHz CL6 is better than 1600MHz CL9

    Because... 6/1333= 0,0045 and 9/1600=0,0056
    right?


    Not necessarily. Take a look at the memory tutorial at the top of the forum
  17. Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    The catch here of course is the fact that 1600 is NOT cycles per second bu MILLIONS of cycles per second. 1600 is 1600 MHz not 1600Hz as he proposes here. Only off by a factor of 10^6...
  18. Pinhedd said:
    thefoxer said:
    Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.


    Ok that makes a lot of sense, thanks! Now why are there 4 numbers for the CAS latency though? What do the gaps between the 2nd and 3rd number mean compared to the lower first number?


    It may seem like it makes sense but most of what he wrote is incorrect.

    DDR3-2133 is a memory module rated for a transfer rate of 2133 megabits per second on each IO pin. Data on the IO pin is exchanged on both the rising edge and the falling edge of the DRAM IO bus reference clock. Thus, a module rated for DDR3-2133 has an IO bus frequency of 1066Mhz, not 2133Mhz. The measure of IO pin transfer rate is in transfers per second, not cycles per second, so a DDR transfer rate of 2133MT/s has a bus frequency of 1066Mhz. The same holds true for DDR, DDR2, and DDR3.

    The actual DRAM modules themselves operate at an even slower rate. What do DDR-400, DDR2-800, and DDR3-1600 all have in common? The core memory clock is 200Mhz.

    DDR-400 has an IO bus frequency of 200Mhz, and a 1:1 ratio between the IO bus reference clock and the core memory clock.

    DDR2-800 has an IO bus frequency of 400Mhz, and a 2:1 ratio between the IO bus reference clock and the core memory clock.

    DDR3-1600 has an IO bus frequency of 800Mhz, and a 4:1 ratio between the IO bus reference clock and the core memory clock.

    This pattern is an example of one of the most misunderstood aspects of modern DRAM. It has gotten wider and deeper -- which allows for greater capacity and larger prefetch buffers -- not faster which allows for lower latency. Latency measured in real time (nanoseconds) hasn't changed significantly over the past decade.

    The most commonly cited DRAM performance metric aside from the IO bus transfer rate is the CAS latency. The CAS latency is the duration (in clock cycles) that it takes for the DRAM module performing a read operation to register a column address and produce a stable output on the IO pins. This delay is measured in IO bus clock cycles, which as noted above scale between generations. This is why DDR2-800C (CL4) has the same first word delay of 10ns as DDR3-1600G (CL8). Where DDR3 really improves over its predecessors is in fourth and eighth word delays, this is accomplished through a deeper prefetch buffer (outside of the scope of this explanation).

    So it should be no surprise then that DDR3-2133 with CAS 9 is faster than DDR3-1600 with CAS9. This is nice, but somewhat immaterial. Why is that?

    x86 memory channels data paths are 64 bits wide, and the DRAM modules are arranged to fit this. If you look at a DRAM module without the heatsink attached (eg, this one) you'll see 8 integrated circuits on one side of the PCB. Each of these ICs is an 8 bit DRAM module and they are all tied together with common command and address lines, but with data lines formed into a 64 bit bus. Arranged together, they form what is called a rank. DIMMs that have 8 modules on each side arranged into two ranks are called dual-rank. Servers often use compact modules which have 4 or even 8 ranks crammed onto a single PCB for extreme amounts of memory.

    The memory controller can only access one rank at a time, even if two or more ranks are installed in the channel on one or more DIMMs. However, the individual DRAM chips themselves are broken down into individual banks, each of which can be active and working independently at the same time. DDR3 modules have 8 individual banks on each IC. So, even with high DRAM latencies, the DRAM controller can keep the entire channel busy nearly 100% of the time simply by switching to another bank after a command has been issued, and switching back when the output is ready. Thus, high speed memory almost always outperforms low latency memory; this is especially true when high speed memory is accompanied by low latency (often at the cost of higher supply voltage).

    This does not necessarily mean that DDR3-2133 is necessary or even a good idea. I run 32GiB of it in my PC but it requires some extra tweaking to obtain stability.

    I hope that this answered your question


    using this thread as a link for ram related questions from now on XD however on GiB that is needed on a gaming PC we can agree 32 GiB is way too much unless your running a ram disk. I personally have yet to see a game to max out even 8 GiB of ram although I have 16 GiB in my computer @1866. side note I had to clock it up to 1.65 volts to get stable even though the xmp profile was rated for 1866 @1.5 volts. I am running amd board however.
  19. Thank you Pinhedd for that fantastic information! So if you were comparing DDR3 2133Mhz to DDR4 2133Mhz, but the DDR4 has a higher CAS latency, would the DDR4 be better or worse? Wouldn't it essentially have half the IO bus frequency than the DDR3?
  20. turkey3_scratch said:
    Thank you Pinhedd for that fantastic information! So if you were comparing DDR3 2133Mhz to DDR4 2133Mhz, but the DDR4 has a higher CAS latency, would the DDR4 be better or worse? Wouldn't it essentially have half the IO bus frequency than the DDR3?


    I would also like to know this pinhead.

    I think ddr4 supports quad pumping while ddr3 only had dual pumping so it gets signals 4 times per clock cycle? I could be way off on this. anandtech did a review on this and saw ddr4 helped in graphics type applications and a few cpu ones while ddr3 was faster for cpu related stuff (skylake review)

    EDIT: all tests done the DDR4 was I think .2% faster than DDR3 and only gives like 2.7% ipc improvement over broadwell and 5.3% over haswell.
  21. DDR4 does not result in higher IPC, the processor architecture does.

    Both DDR2-1600 and DDR3-1600, for instance in my mind, would have an IO bus frequency of 800Mhz. That is a 4:1 ratio to the actual memory clock speed. Does this mean Pinhedd that every cycle of the memory 4 bits can be transferred essentially?

    And if this is the case, what makes DDR2 different from DDR3 if the IO bus frequency and memory clock speed are the same?
  22. turkey3_scratch said:
    DDR4 does not result in higher IPC, the processor architecture does.

    Both DDR2-1600 and DDR3-1600, for instance in my mind, would have an IO bus frequency of 800Mhz. That is a 4:1 ratio to the actual memory clock speed. Does this mean Pinhedd that every cycle of the memory 4 bits can be transferred essentially?

    And if this is the case, what makes DDR2 different from DDR3 if the IO bus frequency and memory clock speed are the same?


    I knew IPC was skylake improvements hence why I mentioned it just pointing out that ddr4 gave ipc a .2% boost over ddr3.

    ddr2 runs at 2.5V and ddr3 runs at 1.5V right? that's a difference.
  23. Vogner16 said:
    turkey3_scratch said:
    DDR4 does not result in higher IPC, the processor architecture does.

    Both DDR2-1600 and DDR3-1600, for instance in my mind, would have an IO bus frequency of 800Mhz. That is a 4:1 ratio to the actual memory clock speed. Does this mean Pinhedd that every cycle of the memory 4 bits can be transferred essentially?

    And if this is the case, what makes DDR2 different from DDR3 if the IO bus frequency and memory clock speed are the same?


    I knew IPC was skylake improvements hence why I mentioned it just pointing out that ddr4 gave ipc a .2% boost over ddr3.

    ddr2 runs at 2.5V and ddr3 runs at 1.5V right? that's a difference.


    True, I'm just wondering if the difference is strictly voltage.
  24. turkey3_scratch said:
    Thank you Pinhedd for that fantastic information! So if you were comparing DDR3 2133Mhz to DDR4 2133Mhz, but the DDR4 has a higher CAS latency, would the DDR4 be better or worse? Wouldn't it essentially have half the IO bus frequency than the DDR3?


    DDR4 has many advantages over DDR3, even at the expense of looser latencies. Some of these are detailed in my tutorial which can be found at the top of the forums.

    The short and sweet answer is that the real-time constraints of DDR3 puts a ceiling on both its IO capabilities and memoryt density. I don't have any hard data handy but anecdotal evidence suggests that even the best memory controllers have a hard time keeping the DDR3 IO bus busy when data rates exceed ~DDR3-2400. Data rates above this tend to be wasted because the memory controller cannot issue any meaningful commands due to timing and refresh constraints.

    Many of the design changes introduced in DDR4 make it much easier to keep the bus busy at higher data rates without negatively affecting the power consumption of the memory. This does come at a slight real-time latency cost (which will drop over time just as it did with DDR3) but this latency cost is negligible compared to the benefits of higher density and higher data rates.

    Furthermore, native 8-gigabit dies are realizable using the DDR4 specification whereas they are not with the DDR3 specification (they are specified, but have never been manufactured; all 8-gigabit DDR3 chips use stacked dies).
  25. turkey3_scratch said:
    DDR4 does not result in higher IPC, the processor architecture does.

    Both DDR2-1600 and DDR3-1600, for instance in my mind, would have an IO bus frequency of 800Mhz. That is a 4:1 ratio to the actual memory clock speed. Does this mean Pinhedd that every cycle of the memory 4 bits can be transferred essentially?

    And if this is the case, what makes DDR2 different from DDR3 if the IO bus frequency and memory clock speed are the same?


    Great questions. I've answered these in my memory tutorial at the top of the forum, but I'll summarize the answer here just for good measure.

    DDR3 has an 8 word prefetch. Every time a read operation is executed, the command is sent to the appropriate selected bank. The bank reads 8 words from the open row (also called the open page) in parallel. The size of the word is equal to the number of IO pins on the DRAM chip. Most personal computers use 8-bit DRAM chips, but 16-bit DRAM chips can be found in compact devices and 4-bit DRAM chips are found in servers with truckloads of memory.

    The 8 words that are selected are always right beside each other in the column's address space. The lower 3 bits of the column address that accompanies the read command determines the order that they are serialized in. If the lower 3 bits are all 0 (by far the easiest), the order is simply 0,1,2,3,4,5,6,7.

    So, a read command on an 8-bit DDR3 chip selects 8 8-bit words that are spatially adjacent to eachother for a total of 64-bits that need to be transmitted back to the memory controller that issued the read request. Once the data is selected, it is fed into the shared IO logic which is connected to the IO bus. The actual transmission process takes 4 cycles; the timing parameter for this is Tccd, or Column-Command-Delay. This is the minimum number of cycles between consecutive read commands to the open row. Issuing a read command to a bank before Tccd cycles has elapsed would result in data that is in queue for transmission getting clobbered.

    Tcas is the number of cycles between the latching of the read command by the DRAM chip and the DRAM chip presenting the first word on the IO bus. Since Tccd, which represents the shared IO time, is fixed at 4 cycles for DDR3, Tcas is lower bounded by Tccd. The lowest programmable Tcas for DDR3 is 5 cycles which leaves only one cycle for the entire decoding, fetching, and serialization process. This is only practical at very low data rates; in practice, Tcas between 9 and 11 cycles is common which leaves between 5 and 7 cycles for the backend logic to complete. I believe that JEDEC requires compliant DDR3 to support programmable Tcas between 7 and 14 cycles, with the rest being optional.

    The neat point here is that memory operations can overlap.

    So, each DDR3 DRAM chip transfers a single word each half-cycle for a total of 4 cycles as part of a single read operation. x86 DIMMs place a number of DRAM chips in parallel to form a 64-bit functional unit called a rank. On PCs this is typically (but not always) 8x8-bit chips located on one side of a PCB. Other rank forms are 4x16-bit chips found in compact devices, and 16x4-bit chips found in servers.

    64 bits per rank * 8 transfers per command = 64 bytes transferred per command

    One of the chief differences between DDR2 and DDR3 is that DDR2 has a 4 word prefetch compared to DDR3's 8 word prefetch. Similarly, Tccd for DDR2 is only 2 cycles rather than 4.

    So you are correct that DDR2-1600 (I'm not sure that any modules of this data rate were ever available, the highest that I own is DDR2-1200) would have the same command clock frequency as DDR3-1600 but the DDR2-1600 core would be operating twice as fast as the DDR3-1600 core.

    DDR4 retains the 8 word prefetch from DDR3.
  26. DDR3 2133MHz Corsair Vengence/Dominator have the best latency to bandwidth ratio on the market right now of any ram available though so really no reason to use anything above DDR3 2400MHz quite frankly at the end of the day for memory ns speed wins. Right now DDR4 has it's perks in the form of reduced power and greater capacity, but raw performance isn't it's strong suit yet at least not until they tighten up the latency to bandwidth ratio on them which eventually will happen as it matures.
  27. Lots of misinformation and some truths. Here's an easy way to understand it: http://www.crucial.com/usa/en/memory-performance-speed-latency
    Happens to be the correct understanding and right from the manufacture.
  28. Quaddro said:
    Speed and latency is the main factor to determine how fast is your memory.

    For example:
    DDR3 1600 CL 6 vs DDR3 2000 CL 9.

    To put it on simple maths:

    The frequency is expressed in Hertz, which means "cycles per second". So, the DDR3 2000 will perform 2000 cycles a second while the DDR3 1600 will do, well, 1600.

    The CAS latency is given in cycles. So, a CAS9 RAM will take 9 cycles to respond and the CAS6, 6 cycles.

    Now putting it together: the DDR3 2000 CAS9 will take 9/2000 seconds, which is equal to 0,0045 seconds, to respond while the DDR3 1600 CAS 6 will take 6/1600, which is equal to 0,0038 seconds, to respond. Thus, the 1600 one is faster.



    Wrong math man !!

    Mhz = 10^-6
    Ghz = 10^-9

    1600 Mhz CL6
    1600 Mhz = 1.6 Ghz = 1/1.6 = 0.625 ns [nano second response time]
    0.625 x 6 = 3.75 ns

    2000 Mhz CL9
    2000 Mhz = 2 Ghz = 1/2 = 0.5 ns
    0.5 x 9 = 4.5 ns

    So of course 1600 Mhz CL6 is faster than 2000 Mhz CL9, but you were off by order of millionth of a second :|
Ask a new question

Read More

Latency RAM Speed Memory