User Tag List

Results 1 to 17 of 17

Thread: Determining RAM Performance (By Vanguard)

  1. #1
    Member
    • Franklin's Gadgets
      • Motherboard:
      • Gigabyte z97P D3
      • CPU:
      • Intel Core i5 - 4590 processor
      • RAM:
      • GSkill Ripjaws-X 2x8GB 1600mhz
      • Hard Drive:
      • Toshiba 2TB 5400 rpm And Transcend 370 128GB
      • Graphics Card:
      • MSI TF V GTX 960 Gaming 2 GB
      • Display:
      • LG Flatron M227WA LCD TV Monitor
      • Sound Card:
      • Integrated
      • Speakers/HPs:
      • NIA Wireless Headset
      • Keyboard:
      • A4 tech
      • Mouse:
      • A4Tech Bloody V3M
      • Controller:
      • PS3 Wireless Controller
      • Power Supply:
      • Cooler Master v750W
      • Optical Drive:
      • Asus DVD RW 24x
      • USB Devices:
      • Transcend Jetflash v70 8 GB
      • UPS:
      • Apollo 800VA
      • Operating System:
      • Windows 8.1 Pro 64 bit (Ingeniune)
      • Comment:
      • Finally A Rig Off Of My Own Money. :)
      • ISP:
      • DOT
      • Download Speed:
      • 512 kBps with a B
      • Upload Speed:
      • 512 kBps with a B
      • Console:
      • 16
    Franklin's Avatar
    Join Date
    Jul 2011
    Location
    Universe
    Posts
    1,958

    Default Determining RAM Performance (By Vanguard)

    RAM performance solely depends on the time it takes to read (writing is equivalent for ram because it is a volatile memory, it does not store anything permanently) a byte. Thats why manufacturers are fighting for increasing the RAM frequency (mhz) or reducing the CL (cache latency). Here is the law:

    2000x (CL / Bus Speed) = time ------[It is 2000 because as i said before, RAM can perform simultaneous reading and writing, actually it is 2x1000 =2000]

    Now suppose if you have a generic 1333Mhz DDR3 ram with CL 9 ( 9-9-9-24, the most typical scenario),
    Now according to the law,

    2000 x (9 / 1333) = 13.50 ns (nano second) THE LOWER IS BETTER

    Lets have a look at a Transcend/Corsair XMS 1600Mhz DDR3 ram with CL11 (11-11-11-30)

    2000 x (11 / 1600) = 13.75 ns

    For Corsair Vengeance 1600Mhz DDR3 Premium RAM with CL9 ( it is also a 9-9-9-24) ***remember, the performance of a 9-9-9-24 module is same as a 8-10-9-24 module, ( I will discuss it later)

    2000 x (9 / 1600) =
    11.25 ns

    So far, we can see, Corsair Vengeance is better than previous two modules we have tested.

    Now lets have a look at what im using, a pair of TwinMOS DDR3 1333Mhz of CL 7 (7-7-7-20)

    2000 x ( 7 / 1333) = 10.50 ns

    KA-BOOOOM!!!!

    Remember what i said above, this will help you throughout your life when buying RAMs.


    P.S. Do not rep me up. Rep up Vanguard vaia.
    Last edited by Franklin; February 5th, 2013 at 14:57.

  2. #2
    Member
    • VANGUARD's Gadgets
      • Motherboard:
      • ASUS ROG Maximus X Hero
      • CPU:
      • Intel Core i7 8700K 5.2Ghz @Corsair H115i
      • RAM:
      • Gskill Ripjaws V 3466MHz 2x8GB
      • Hard Drive:
      • Zotac SSD 480GB Premium + Seagate Barracuda 2TB
      • Graphics Card:
      • Zotac GTX 1080Ti Amp! Extreme
      • Display:
      • ASUS VX24AH, 1440p
      • Sound Card:
      • ROG SupremeFX
      • Speakers/HPs:
      • Microlab M-910/Logitech UE 200vm
      • Keyboard:
      • Corsair STRAFE-CherryRED
      • Mouse:
      • Logitech G602 @Tt Pyrrhus
      • Controller:
      • Xbox One Wireless
      • Power Supply:
      • Antec EDGE 650w
      • Optical Drive:
      • Samsung SH-224FB
      • USB Devices:
      • Everything
      • UPS:
      • Micro UVL-1500 Long Backup
      • Operating System:
      • Windows 10 Pro x64
      • Comment:
      • The workhorse
      • ISP:
      • The Net Heads
    VANGUARD's Avatar
    Join Date
    Jun 2009
    Location
    Dhaka
    Posts
    10,096

    Default

    Similar sort of thread posted back years ago -_-

    http://banglagamer.com/showthread.ph...ht=RAM+timings

    So i afraid this thread may get closed soon

  3. #3
    Member ian25's Avatar
    Join Date
    Oct 2012
    Location
    Dhaka
    Posts
    246

    Default

    vai valo post but amr mto novice er jonno aro kisu jana dorkar 1st of all :- 2000 ki always fixed thakbe naki variable? 2nd ram er packet e cl 9 lekha thakle ki bujbo je eta 9-9-9-24 ? or ram 9-9-9-24 na 7-7-7-20 bujhar upay ase? r 7-7-7-20 ba jegular latency kom segula usual gula theke ki besi expensive ? hole kto tuku? tnx ....question gula hoyto khub hassokor hoyeche bt ai bepare kisu jani na to tai ask krlam ...tnx

  4. #4
    Member
    • Trave160's Gadgets
      • Motherboard:
      • MSI B350M Mortar AM4
      • CPU:
      • AMD Ryzen 5 1600
      • RAM:
      • 16GB DDR4 3000Mhz
      • Hard Drive:
      • 3TB Western Digital red
      • Graphics Card:
      • ASUS Strix GTX 960 4G
      • Display:
      • Dell S2216H 22 Inch monitor
      • Sound Card:
      • None
      • Speakers/HPs:
      • Logitech Speaker
      • Keyboard:
      • TtEsports Knucker
      • Controller:
      • PS4 Dualshock 4 controller
      • Power Supply:
      • Thermaltake Smart SE 630W Modular
      • Optical Drive:
      • Samsung 24X DVD-RW Drive(Defunct)
      • USB Devices:
      • Pendrive 4GB
      • UPS:
      • Power Tree 1250VA
      • Operating System:
      • Windows 10 64bit
      • Comment:
      • Toaster rig arisen to Ultra toaster with chisel abs.
      • ISP:
      • NunuCom Ltd, Penus group
      • Download Speed:
      • Super San 200kbps
      • Upload Speed:
      • 105*2
      • Console:
      • 130
    Trave160's Avatar
    Join Date
    Aug 2009
    Location
    A Secure Area
    Posts
    7,742

    Default

    Now, now he still has new stuff to look at. Test Ripjaws and Team RAMs.

  5. #5
    Member
    • VANGUARD's Gadgets
      • Motherboard:
      • ASUS ROG Maximus X Hero
      • CPU:
      • Intel Core i7 8700K 5.2Ghz @Corsair H115i
      • RAM:
      • Gskill Ripjaws V 3466MHz 2x8GB
      • Hard Drive:
      • Zotac SSD 480GB Premium + Seagate Barracuda 2TB
      • Graphics Card:
      • Zotac GTX 1080Ti Amp! Extreme
      • Display:
      • ASUS VX24AH, 1440p
      • Sound Card:
      • ROG SupremeFX
      • Speakers/HPs:
      • Microlab M-910/Logitech UE 200vm
      • Keyboard:
      • Corsair STRAFE-CherryRED
      • Mouse:
      • Logitech G602 @Tt Pyrrhus
      • Controller:
      • Xbox One Wireless
      • Power Supply:
      • Antec EDGE 650w
      • Optical Drive:
      • Samsung SH-224FB
      • USB Devices:
      • Everything
      • UPS:
      • Micro UVL-1500 Long Backup
      • Operating System:
      • Windows 10 Pro x64
      • Comment:
      • The workhorse
      • ISP:
      • The Net Heads
    VANGUARD's Avatar
    Join Date
    Jun 2009
    Location
    Dhaka
    Posts
    10,096

    Default

    Quote Originally Posted by ian25 View Post
    vai valo post but amr mto novice er jonno aro kisu jana dorkar 1st of all :- 2000 ki always fixed thakbe naki variable? 2nd ram er packet e cl 9 lekha thakle ki bujbo je eta 9-9-9-24 ? or ram 9-9-9-24 na 7-7-7-20 bujhar upay ase? r 7-7-7-20 ba jegular latency kom segula usual gula theke ki besi expensive ? hole kto tuku? tnx ....question gula hoyto khub hassokor hoyeche bt ai bepare kisu jani na to tai ask krlam ...tnx
    1. 2000 is always fixed.

    2. usually it is stated as just CL9, renowned brands often state the entire timings as 9-9-9-24-1T. But here is a thing to be understood, some company often state 8-10-9-24 or something peculiar like that and mention it as a CL8 ram, which is theoretically correct but not practically. If RAS to CAS precharge delay is 10, no matter how much you reduce CAS, performance wont increase a single bit.

    3. 7-7-7-20 ram gula onek expensive but what i do is to buy a TwinMOS CL9 1333 RAM, then reduce the timing manually to 7-7-7-20-1T or even 7-7-7-18-1T without touching volt or anything. I have been able to do this by myself for TwinMOS but failed for Transcend, ADATA, Team, Patriot and even Corsair vengeance. Those ram wont even boot if you try to reduce the latency to 7-7-7-20-1T.

    I hope this answer your questions

    - - - Updated - - -

    Quote Originally Posted by Trave160 View Post
    Now, now he still has new stuff to look at. Test Ripjaws and Team RAMs.
    i didnt test RJs but tried TEAM and the result was pathetic

  6. #6
    Member
    • fahimspider's Gadgets
      • Motherboard:
      • MSI P67A GD65(B3)
      • CPU:
      • I5 2400 @ 3.4 GHZ Cooled by CM Hyper 412 Slim
      • RAM:
      • TWINMOS 8 GB 1600 BUS
      • Hard Drive:
      • TRANSCEND 128 GB SSD, 2 TB WD SATA III BLACK, 1 TB HITACHI
      • Graphics Card:
      • Saphire ATI HD 7950 Dual-X
      • Display:
      • HP 18.5 INCH LCD
      • Sound Card:
      • Built In
      • Speakers/HPs:
      • Microlab 2.1
      • Keyboard:
      • Tt E-sports Knucker
      • Mouse:
      • Tt Theron mouse
      • Controller:
      • Razer Nostromo
      • Power Supply:
      • THERMALTAKE SMART MODULAR 650 WATT
      • Optical Drive:
      • Asus DVD RW. HP DVD ROM
      • UPS:
      • Apollo 1200 VA
      • Operating System:
      • WINDOWS 7 64 BIT
      • Benchmark Scores:
      • 7.5 WEI
      • Comment:
      • ROCKING
      • ISP:
      • SmileBD
      • Download Speed:
      • 300-600kbps
      • Upload Speed:
      • 250-450kbps
    fahimspider's Avatar
    Join Date
    Aug 2012
    Location
    Dhaka
    Posts
    768

    Default

    I would like to try this with twinmos RAM. But before that i would like to know if some thing goes wrong like, boot failure/ boot loop - then what to do?

  7. #7
    Member
    • VANGUARD's Gadgets
      • Motherboard:
      • ASUS ROG Maximus X Hero
      • CPU:
      • Intel Core i7 8700K 5.2Ghz @Corsair H115i
      • RAM:
      • Gskill Ripjaws V 3466MHz 2x8GB
      • Hard Drive:
      • Zotac SSD 480GB Premium + Seagate Barracuda 2TB
      • Graphics Card:
      • Zotac GTX 1080Ti Amp! Extreme
      • Display:
      • ASUS VX24AH, 1440p
      • Sound Card:
      • ROG SupremeFX
      • Speakers/HPs:
      • Microlab M-910/Logitech UE 200vm
      • Keyboard:
      • Corsair STRAFE-CherryRED
      • Mouse:
      • Logitech G602 @Tt Pyrrhus
      • Controller:
      • Xbox One Wireless
      • Power Supply:
      • Antec EDGE 650w
      • Optical Drive:
      • Samsung SH-224FB
      • USB Devices:
      • Everything
      • UPS:
      • Micro UVL-1500 Long Backup
      • Operating System:
      • Windows 10 Pro x64
      • Comment:
      • The workhorse
      • ISP:
      • The Net Heads
    VANGUARD's Avatar
    Join Date
    Jun 2009
    Location
    Dhaka
    Posts
    10,096

    Default

    Quote Originally Posted by fahimspider View Post
    I would like to try this with twinmos RAM. But before that i would like to know if some thing goes wrong like, boot failure/ boot loop - then what to do?
    mention me your default timing and entire model name first as not ALL the TwinMOS can reach CL7. Anything too experimental may harm your modules.


    *** Usually the system restarts and things go default. But again, do with caution.

  8. #8
    Member
    • 's Gadgets
      • Motherboard:
      • GIGABYTE B75M-D3H rev1.1
      • CPU:
      • Intel Core i5 3470 Ivy bridge//Core i5-4200U
      • RAM:
      • Silicon Power 4GB DDR3 @1600 mhz(8-8-8-24-1T) // TwinMOS 4GB DDR3
      • Hard Drive:
      • WD green 500gb// Hitachi 500GB
      • Graphics Card:
      • Asus GTX 960 strix// GT 750M
      • Display:
      • AOC e2043Fsk 20" LED(1600 X 900) DCR 50,000,000:1
      • Keyboard:
      • A4Tech KRS-86
      • Mouse:
      • Thermaltake azureus
      • Power Supply:
      • Antec VP450P
      • Optical Drive:
      • Samsung DVD RW+Asus CD RW
      • Operating System:
      • Win 8.1 pro x64
    @nonymous™'s Avatar
    Join Date
    Dec 2010
    Location
    Unknown
    Posts
    1,490

    Default

    Quote Originally Posted by AAA Chosen One View Post
    RAM performance solely depends on the time it takes to read (writing is equivalent for ram because it is a volatile memory, it does not store anything permanently) a byte. Thats why manufacturers are fighting for increasing the RAM frequency (mhz) or reducing the CL (cache latency). Here is the law:

    2000x (CL / Bus Speed) = time ------[It is 2000 because as i said before, RAM can perform simultaneous reading and writing, actually it is 2x1000 =2000]

    Now suppose if you have a generic 1333Mhz DDR3 ram with CL 9 ( 9-9-9-24, the most typical scenario),
    Now according to the law,

    2000 x (9 / 1333) = 13.50 ns (nano second) THE LOWER IS BETTER

    Lets have a look at a Transcend/Corsair XMS 1600Mhz DDR3 ram with CL11 (11-11-11-30)

    2000 x (11 / 1600) = 13.75 ns

    For Corsair Vengeance 1600Mhz DDR3 Premium RAM with CL9 ( it is also a 9-9-9-24) ***remember, the performance of a 9-9-9-24 module is same as a 8-10-9-24 module, ( I will discuss it later)

    2000 x (9 / 1600) =
    11.25 ns
    the formula is entirely wrong...i.e 2000x(9/1333mhz) = 13.5 μs/micro-second, not 13.5 nano-second, 2000 x (11 / 1600) = 13.75 μs, 2000 x (9 / 1600) = = 11.25 μs....which are impossible values for CAS latency

    1333/1600 mhz are not the actual frequency, they are 666.6/800 mhz respectively. and there is nothing like a fixed multiplier '2000' lol....

    the actual formula for the cas delay of a CL9 1333mhz ddr1/2/3 memory= 1/(666.6 x exp6) x 9 = 13.6 ns. you must consider the actual frequency when calculating delay.
    Last edited by @nonymous™; February 6th, 2013 at 00:26.

  9. #9
    Member
    • fahimspider's Gadgets
      • Motherboard:
      • MSI P67A GD65(B3)
      • CPU:
      • I5 2400 @ 3.4 GHZ Cooled by CM Hyper 412 Slim
      • RAM:
      • TWINMOS 8 GB 1600 BUS
      • Hard Drive:
      • TRANSCEND 128 GB SSD, 2 TB WD SATA III BLACK, 1 TB HITACHI
      • Graphics Card:
      • Saphire ATI HD 7950 Dual-X
      • Display:
      • HP 18.5 INCH LCD
      • Sound Card:
      • Built In
      • Speakers/HPs:
      • Microlab 2.1
      • Keyboard:
      • Tt E-sports Knucker
      • Mouse:
      • Tt Theron mouse
      • Controller:
      • Razer Nostromo
      • Power Supply:
      • THERMALTAKE SMART MODULAR 650 WATT
      • Optical Drive:
      • Asus DVD RW. HP DVD ROM
      • UPS:
      • Apollo 1200 VA
      • Operating System:
      • WINDOWS 7 64 BIT
      • Benchmark Scores:
      • 7.5 WEI
      • Comment:
      • ROCKING
      • ISP:
      • SmileBD
      • Download Speed:
      • 300-600kbps
      • Upload Speed:
      • 250-450kbps
    fahimspider's Avatar
    Join Date
    Aug 2012
    Location
    Dhaka
    Posts
    768

    Default

    @Vanguard I have 2 options -
    1. Twinmos Ocean Blue 2x2 GB 1333 bus DDR3(9-9-9-24)
    2. Twinmos twister 4x2 GB 1600 bus running at 1333 (9-9-9-24)

  10. #10
    Member
    • VANGUARD's Gadgets
      • Motherboard:
      • ASUS ROG Maximus X Hero
      • CPU:
      • Intel Core i7 8700K 5.2Ghz @Corsair H115i
      • RAM:
      • Gskill Ripjaws V 3466MHz 2x8GB
      • Hard Drive:
      • Zotac SSD 480GB Premium + Seagate Barracuda 2TB
      • Graphics Card:
      • Zotac GTX 1080Ti Amp! Extreme
      • Display:
      • ASUS VX24AH, 1440p
      • Sound Card:
      • ROG SupremeFX
      • Speakers/HPs:
      • Microlab M-910/Logitech UE 200vm
      • Keyboard:
      • Corsair STRAFE-CherryRED
      • Mouse:
      • Logitech G602 @Tt Pyrrhus
      • Controller:
      • Xbox One Wireless
      • Power Supply:
      • Antec EDGE 650w
      • Optical Drive:
      • Samsung SH-224FB
      • USB Devices:
      • Everything
      • UPS:
      • Micro UVL-1500 Long Backup
      • Operating System:
      • Windows 10 Pro x64
      • Comment:
      • The workhorse
      • ISP:
      • The Net Heads
    VANGUARD's Avatar
    Join Date
    Jun 2009
    Location
    Dhaka
    Posts
    10,096

    Default

    Quote Originally Posted by fahimspider View Post
    @Vanguard I have 2 options -
    1. Twinmos Ocean Blue 2x2 GB 1333 bus DDR3(9-9-9-24)
    2. Twinmos twister 4x2 GB 1600 bus running at 1333 (9-9-9-24)
    Im afraid that none of those two supports CL7. I use basic non-heatsink DDR3 1333Mhz and i did that with it.

    - - - Updated - - -

    Quote Originally Posted by @nonymous™ View Post
    the formula is entirely wrong...i.e 2000x(9/1333mhz) = 13.5 μs
    , not 13.5 nano second
    , 2000 x (11 / 1600) = 13.75 μs, 2000 x (9 / 1600) = = 11.25 μs....which are impossible values for CAS latency


    1333/1600 mhz are not the actual frequency, they are 666.6/800 mhz respectively. and there is nothing like a fixed multiplier '2000' lol....

    the actual formula for the cas delay of a CL9 1333mhz ddr1/2/3 memory= 1/(666.6 x exp6)
    x 9 = 13.6 ns. you must consider the actual frequency when calculating delay.
    http://www.anandtech.com/show/6372/memory-performance-16gb-ddr31333-to-ddr32400-on-ivy-bridge-igp-with-gskill

    Read thoroughly, then make random comments.

  11. #11
    Member
    • 's Gadgets
      • Motherboard:
      • GIGABYTE B75M-D3H rev1.1
      • CPU:
      • Intel Core i5 3470 Ivy bridge//Core i5-4200U
      • RAM:
      • Silicon Power 4GB DDR3 @1600 mhz(8-8-8-24-1T) // TwinMOS 4GB DDR3
      • Hard Drive:
      • WD green 500gb// Hitachi 500GB
      • Graphics Card:
      • Asus GTX 960 strix// GT 750M
      • Display:
      • AOC e2043Fsk 20" LED(1600 X 900) DCR 50,000,000:1
      • Keyboard:
      • A4Tech KRS-86
      • Mouse:
      • Thermaltake azureus
      • Power Supply:
      • Antec VP450P
      • Optical Drive:
      • Samsung DVD RW+Asus CD RW
      • Operating System:
      • Win 8.1 pro x64
    @nonymous™'s Avatar
    Join Date
    Dec 2010
    Location
    Unknown
    Posts
    1,490

    Default

    Quote Originally Posted by Vanguard View Post
    http://www.anandtech.com/show/6372/m...gp-with-gskill

    Read thoroughly, then make random comments.
    2000 x (CL / Speed ) = Latency in nanoseconds (ns)
    time & time in nano second are not same thing.
    the op stated '2000 x (CL / Speed ) = time ' which is entirely wrong. anand basically made a simplified form to make it easier to calculate......
    Quote Originally Posted by Vanguard View Post
    If RAS to CAS precharge delay is 10, no matter how much you reduce CAS, performance wont increase a single bit.
    and where did you got that line? post that link too.....
    Last edited by @nonymous™; February 6th, 2013 at 01:29.

  12. #12
    Member
    • VANGUARD's Gadgets
      • Motherboard:
      • ASUS ROG Maximus X Hero
      • CPU:
      • Intel Core i7 8700K 5.2Ghz @Corsair H115i
      • RAM:
      • Gskill Ripjaws V 3466MHz 2x8GB
      • Hard Drive:
      • Zotac SSD 480GB Premium + Seagate Barracuda 2TB
      • Graphics Card:
      • Zotac GTX 1080Ti Amp! Extreme
      • Display:
      • ASUS VX24AH, 1440p
      • Sound Card:
      • ROG SupremeFX
      • Speakers/HPs:
      • Microlab M-910/Logitech UE 200vm
      • Keyboard:
      • Corsair STRAFE-CherryRED
      • Mouse:
      • Logitech G602 @Tt Pyrrhus
      • Controller:
      • Xbox One Wireless
      • Power Supply:
      • Antec EDGE 650w
      • Optical Drive:
      • Samsung SH-224FB
      • USB Devices:
      • Everything
      • UPS:
      • Micro UVL-1500 Long Backup
      • Operating System:
      • Windows 10 Pro x64
      • Comment:
      • The workhorse
      • ISP:
      • The Net Heads
    VANGUARD's Avatar
    Join Date
    Jun 2009
    Location
    Dhaka
    Posts
    10,096

    Default

    Quote Originally Posted by @nonymous™ View Post
    time & time in nano second are not same thing.
    the op stated '2000 x (CL / Speed ) = time ' which is entirely wrong. anand basically made a simplified form to make it easier to calculate......

    and where did you got that line? post that link too.....
    I wrote time to make it simple for him.


    And you read this too,

    http://banglagamer.com/showthread.ph...ht=RAM+timings

  13. #13
    Member
    • 's Gadgets
      • Motherboard:
      • GIGABYTE B75M-D3H rev1.1
      • CPU:
      • Intel Core i5 3470 Ivy bridge//Core i5-4200U
      • RAM:
      • Silicon Power 4GB DDR3 @1600 mhz(8-8-8-24-1T) // TwinMOS 4GB DDR3
      • Hard Drive:
      • WD green 500gb// Hitachi 500GB
      • Graphics Card:
      • Asus GTX 960 strix// GT 750M
      • Display:
      • AOC e2043Fsk 20" LED(1600 X 900) DCR 50,000,000:1
      • Keyboard:
      • A4Tech KRS-86
      • Mouse:
      • Thermaltake azureus
      • Power Supply:
      • Antec VP450P
      • Optical Drive:
      • Samsung DVD RW+Asus CD RW
      • Operating System:
      • Win 8.1 pro x64
    @nonymous™'s Avatar
    Join Date
    Dec 2010
    Location
    Unknown
    Posts
    1,490

    Default

    Quote Originally Posted by Vanguard View Post
    I wrote time to make it simple for him.


    And you read this too,

    http://banglagamer.com/showthread.ph...ht=RAM+timings
    i actually asked for the link in which you got that 'reducing CL doesn't have any effect if CAS to RAS precharge delay is 10 for a 8-10-9-24 module!

  14. #14
    Member
    • VANGUARD's Gadgets
      • Motherboard:
      • ASUS ROG Maximus X Hero
      • CPU:
      • Intel Core i7 8700K 5.2Ghz @Corsair H115i
      • RAM:
      • Gskill Ripjaws V 3466MHz 2x8GB
      • Hard Drive:
      • Zotac SSD 480GB Premium + Seagate Barracuda 2TB
      • Graphics Card:
      • Zotac GTX 1080Ti Amp! Extreme
      • Display:
      • ASUS VX24AH, 1440p
      • Sound Card:
      • ROG SupremeFX
      • Speakers/HPs:
      • Microlab M-910/Logitech UE 200vm
      • Keyboard:
      • Corsair STRAFE-CherryRED
      • Mouse:
      • Logitech G602 @Tt Pyrrhus
      • Controller:
      • Xbox One Wireless
      • Power Supply:
      • Antec EDGE 650w
      • Optical Drive:
      • Samsung SH-224FB
      • USB Devices:
      • Everything
      • UPS:
      • Micro UVL-1500 Long Backup
      • Operating System:
      • Windows 10 Pro x64
      • Comment:
      • The workhorse
      • ISP:
      • The Net Heads
    VANGUARD's Avatar
    Join Date
    Jun 2009
    Location
    Dhaka
    Posts
    10,096

    Default

    Quote Originally Posted by @nonymous™ View Post
    i actually asked for the link in which you got that 'reducing CL doesn't have any effect if CAS to RAS precharge delay is 10 for a 8-10-9-24 module!
    "
    tRAS determines the length of time between initialization and the requirement for the memory row to recharge. Within this tRAS we need a tRCD to initialize the column of the row from which we would like to read. After the tRCD is the CL, which provides a read latency. There are also other features which allow for reads across multiple columns within the tRAS, however in order to move to the next row the tRAS needs to end and the tRP allows the next row to precharge.
    All this means that:
    If tRAS is a low number, it is quick to read from different rows. If it is a high number, reading from different columns is easier.
    If CL is a low number, reading from within a row (and the columns) is quicker.
    If tRCD is low, more CLs can be initialized inside the tRAS.
    If tRP is low, then the overall time (tRAS+tRP) to jump between row reads is quicker."

  15. #15
    Member
    • 's Gadgets
      • Motherboard:
      • GIGABYTE B75M-D3H rev1.1
      • CPU:
      • Intel Core i5 3470 Ivy bridge//Core i5-4200U
      • RAM:
      • Silicon Power 4GB DDR3 @1600 mhz(8-8-8-24-1T) // TwinMOS 4GB DDR3
      • Hard Drive:
      • WD green 500gb// Hitachi 500GB
      • Graphics Card:
      • Asus GTX 960 strix// GT 750M
      • Display:
      • AOC e2043Fsk 20" LED(1600 X 900) DCR 50,000,000:1
      • Keyboard:
      • A4Tech KRS-86
      • Mouse:
      • Thermaltake azureus
      • Power Supply:
      • Antec VP450P
      • Optical Drive:
      • Samsung DVD RW+Asus CD RW
      • Operating System:
      • Win 8.1 pro x64
    @nonymous™'s Avatar
    Join Date
    Dec 2010
    Location
    Unknown
    Posts
    1,490

    Default

    Quote Originally Posted by Vanguard View Post
    "
    tRAS determines the length of time between initialization and the requirement for the memory row to recharge. Within this tRAS we need a tRCD to initialize the column of the row from which we would like to read. After the tRCD is the CL, which provides a read latency. There are also other features which allow for reads across multiple columns within the tRAS, however in order to move to the next row the tRAS needs to end and the tRP allows the next row to precharge.
    All this means that:
    If tRAS is a low number, it is quick to read from different rows. If it is a high number, reading from different columns is easier.
    If CL is a low number, reading from within a row (and the columns) is quicker.
    If tRCD is low, more CLs can be initialized inside the tRAS.
    If tRP is low, then the overall time (tRAS+tRP) to jump between row reads is quicker."

    still my question is not answered. anyway for a single memory instruction on a single row or memory word total delay = CL+(tRCD+tRP+tRAS). 8-10-9-24-1T & 9-9-9-24-1T will yield same amount of delay but 7-10-9-24-1T must be faster.....
    Last edited by @nonymous™; February 8th, 2013 at 19:45.

  16. #16
    Member
    • VANGUARD's Gadgets
      • Motherboard:
      • ASUS ROG Maximus X Hero
      • CPU:
      • Intel Core i7 8700K 5.2Ghz @Corsair H115i
      • RAM:
      • Gskill Ripjaws V 3466MHz 2x8GB
      • Hard Drive:
      • Zotac SSD 480GB Premium + Seagate Barracuda 2TB
      • Graphics Card:
      • Zotac GTX 1080Ti Amp! Extreme
      • Display:
      • ASUS VX24AH, 1440p
      • Sound Card:
      • ROG SupremeFX
      • Speakers/HPs:
      • Microlab M-910/Logitech UE 200vm
      • Keyboard:
      • Corsair STRAFE-CherryRED
      • Mouse:
      • Logitech G602 @Tt Pyrrhus
      • Controller:
      • Xbox One Wireless
      • Power Supply:
      • Antec EDGE 650w
      • Optical Drive:
      • Samsung SH-224FB
      • USB Devices:
      • Everything
      • UPS:
      • Micro UVL-1500 Long Backup
      • Operating System:
      • Windows 10 Pro x64
      • Comment:
      • The workhorse
      • ISP:
      • The Net Heads
    VANGUARD's Avatar
    Join Date
    Jun 2009
    Location
    Dhaka
    Posts
    10,096

    Default

    Quote Originally Posted by @nonymous™ View Post
    *1 still my question is not unnswered. anyway for a single memory instruction on a single row or memory word total delay = CL+(tRCD+tRP+tRAS). 8-10-9-24-1T & 9-9-9-24-1T will yield same amount of delay but *2 7-10-9-24-1T must be faster.....
    1. Glad i could end your curiosity.

    2. Exactly my point.
    Last edited by VANGUARD; February 6th, 2013 at 14:14.

  17. #17
    Member
    • QuadStorm's Gadgets
      • Motherboard:
      • Gigabyte H81M-S2PV
      • CPU:
      • Core i5-4460
      • RAM:
      • Avexir Core Series 8GB (4*2) 1600MHz
      • Hard Drive:
      • Toshiba 1TB
      • Graphics Card:
      • Sapphire R7 265 Dual-X 2GB Gddr5
      • Display:
      • Asus VX229H
      • Keyboard:
      • A4tech
      • Mouse:
      • ProLink Evotis
      • Controller:
      • Xbox 360 Wireless Controller For Windows
      • Power Supply:
      • GameMax GM-650
      • USB Devices:
      • Banglalion Modem
      • Operating System:
      • Windows 8.1 64bit
      • ISP:
      • Banglalion
      • Download Speed:
      • 120
      • Upload Speed:
      • 8

    Join Date
    Mar 2012
    Location
    Chittagong
    Posts
    625

    Default

    Vai, amar ram lagbe...

    nicher eita nile ki CL 7 e ana jabe? @Vanguard

    http://www.ryanscomputers.com/index....em_Details/353

Similar Threads

  1. We want vanguard bro
    By Ghost Riley in forum Member Central
    Replies: 25
    Last Post: January 28th, 2013, 21:44

Tags for this Thread

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Page generated in 0.28863 seconds with 14 queries.