• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Dram Calculator not showing results for the rated speed.

steveol

New Member
Joined
Jan 13, 2021
Messages
5 (0.00/day)
Hi,

The ram is TEAMGROUP T-Force Dark Za (Alpha) 32 GB Kit (2x16GB) DDR4 Dram 3600MHz (PC4-28800) CL18 for AMD Ryzen - TDZAD432G3600HC18JDC01
CPU is 4650G
motherboard is gigabyte itx 520i ac with the latest bios and drivers.
I entered the information required for my micron B-die ram listed as a 3600 cl 18 ram, chose 570 type motherboard since I don't see the option for a 520 motherboard, hopefully correctly.
At 3600, I don't have numbers showing. If I dial down to 3466 speed then the lower timings I want will show. Does this mean that I won't be able to run this kit at 3600 cl 16 being a micron chip?
I have seen good reviews of this ram however they are with the hynix die, with lower timings achieved at the rated speed of the ram.
Thanks
 

Attachments

  • at 3466.png
    at 3466.png
    69.2 KB · Views: 224
  • calc not showing.png
    calc not showing.png
    453.1 KB · Views: 280
  • hardware info.png
    hardware info.png
    229.2 KB · Views: 239
Joined
Sep 3, 2019
Messages
2,965 (1.78/day)
Location
Thessaloniki, Greece
System Name PC on since Aug 2019, 1st CPU R5 3600 + ASUS ROG RX580 8GB >> MSI Gaming X RX5700XT (Jan 2020)
Processor Ryzen 9 5900X (July 2022), 150W PPT limit, 79C temp limit, CO -9~14
Motherboard Gigabyte X570 Aorus Pro (Rev1.0), BIOS F37h, AGESA V2 1.2.0.B
Cooling Arctic Liquid Freezer II 420mm Rev7 with off center mount for Ryzen, TIM: Kryonaut
Memory 2x16GB G.Skill Trident Z Neo GTZN (July 2022) 3600MHz 1.42V CL16-16-16-16-32-48 1T, tRFC:288, B-die
Video Card(s) Sapphire Nitro+ RX 7900XTX (Dec 2023) 314~465W (366W current) PowerLimit, 1060mV, Adrenalin v24.2.1
Storage Samsung NVMe: 980Pro 1TB(OS 2022), 970Pro 512GB(2019) / SATA-III: 850Pro 1TB(2015) 860Evo 1TB(2020)
Display(s) Dell Alienware AW3423DW 34" QD-OLED curved (1800R), 3440x1440 144Hz (max 175Hz) HDR1000
Case None... naked on desk
Audio Device(s) Astro A50 headset
Power Supply Corsair HX750i, 80+ Platinum, 93% (250~700W), modular, single/dual rail (switch)
Mouse Logitech MX Master (Gen1)
Keyboard Logitech G15 (Gen2) w/ LCDSirReal applet
Software Windows 11 Home 64bit (v23H2, OSB 22631.3155)
From ThaiphoonBurner after you click "Read" and select 1 stick, then click "Report" next to it. Scroll all the way down to the end and click "Show delays in nanoseconds".
Click "File" on top bar and "Export to..." >> "Complete HTML Report". Save the file and then import it in DRAMcalculator with its "Import XMP" button.

This will enable more profiles on the calculator but its not sure they will work, since you have a low budget board/chipset that is most likely not tested.
Although 4000series APUs have great memory abilities for UCLK/FCLK.
 

steveol

New Member
Joined
Jan 13, 2021
Messages
5 (0.00/day)

Thanks Zach_01, I appreciate your response. I could import the html file and I got readings this time different from the initial ones missing the necessary input.

However I still have readings only at 3466 speed and below, I tried those numbers after setting the memory multiplier to 36 but it did not boot, which makes me think either
I need the correct numbers that correspond to the 3600 speed or as you said the chipset is not good for overclocking. I believed the same myself before giving this
an attempt after watching a video from buildzoid rambling about the 520 motherboards and he considered this board as a good memory overclocker, which made
me consider this attempt.
At least I tried.
Thanks again.
 

Attachments

  • new calculations.png
    new calculations.png
    73.1 KB · Views: 116
Joined
Sep 3, 2019
Messages
2,965 (1.78/day)
Location
Thessaloniki, Greece
System Name PC on since Aug 2019, 1st CPU R5 3600 + ASUS ROG RX580 8GB >> MSI Gaming X RX5700XT (Jan 2020)
Processor Ryzen 9 5900X (July 2022), 150W PPT limit, 79C temp limit, CO -9~14
Motherboard Gigabyte X570 Aorus Pro (Rev1.0), BIOS F37h, AGESA V2 1.2.0.B
Cooling Arctic Liquid Freezer II 420mm Rev7 with off center mount for Ryzen, TIM: Kryonaut
Memory 2x16GB G.Skill Trident Z Neo GTZN (July 2022) 3600MHz 1.42V CL16-16-16-16-32-48 1T, tRFC:288, B-die
Video Card(s) Sapphire Nitro+ RX 7900XTX (Dec 2023) 314~465W (366W current) PowerLimit, 1060mV, Adrenalin v24.2.1
Storage Samsung NVMe: 980Pro 1TB(OS 2022), 970Pro 512GB(2019) / SATA-III: 850Pro 1TB(2015) 860Evo 1TB(2020)
Display(s) Dell Alienware AW3423DW 34" QD-OLED curved (1800R), 3440x1440 144Hz (max 175Hz) HDR1000
Case None... naked on desk
Audio Device(s) Astro A50 headset
Power Supply Corsair HX750i, 80+ Platinum, 93% (250~700W), modular, single/dual rail (switch)
Mouse Logitech MX Master (Gen1)
Keyboard Logitech G15 (Gen2) w/ LCDSirReal applet
Software Windows 11 Home 64bit (v23H2, OSB 22631.3155)
Did you follow all suggested settings for timings and others?
What about DRAM voltage, GearDownMode, ProcODT?
 

steveol

New Member
Joined
Jan 13, 2021
Messages
5 (0.00/day)
I went back at it this weekend, glad to say I could achieve the results I was aiming for. I was stuck at the way I was going through the process, from what I saw on youtube, that was to input all the values from the calculator and up the voltage as needed. The better way that worked for me was to go one by one and adjust as needed, for example the TRCDRD wouldn't reach 20 as suggested even at 1.45v, at 21 it was happy with 1.42v. It took a while but worth it. I could do a cinebench r15 of 1525 from the bios' optimized default result of 1482. I had to up the voltage to 1.43 because I had random system error popping up. since then it is running smoothly.
Thanks again Zach_01
 

Attachments

  • ZenTimings_Screenshot1.png
    ZenTimings_Screenshot1.png
    31.8 KB · Views: 113
  • cinebench best score.png
    cinebench best score.png
    33.2 KB · Views: 112
Last edited:
Joined
Sep 3, 2019
Messages
2,965 (1.78/day)
Location
Thessaloniki, Greece
System Name PC on since Aug 2019, 1st CPU R5 3600 + ASUS ROG RX580 8GB >> MSI Gaming X RX5700XT (Jan 2020)
Processor Ryzen 9 5900X (July 2022), 150W PPT limit, 79C temp limit, CO -9~14
Motherboard Gigabyte X570 Aorus Pro (Rev1.0), BIOS F37h, AGESA V2 1.2.0.B
Cooling Arctic Liquid Freezer II 420mm Rev7 with off center mount for Ryzen, TIM: Kryonaut
Memory 2x16GB G.Skill Trident Z Neo GTZN (July 2022) 3600MHz 1.42V CL16-16-16-16-32-48 1T, tRFC:288, B-die
Video Card(s) Sapphire Nitro+ RX 7900XTX (Dec 2023) 314~465W (366W current) PowerLimit, 1060mV, Adrenalin v24.2.1
Storage Samsung NVMe: 980Pro 1TB(OS 2022), 970Pro 512GB(2019) / SATA-III: 850Pro 1TB(2015) 860Evo 1TB(2020)
Display(s) Dell Alienware AW3423DW 34" QD-OLED curved (1800R), 3440x1440 144Hz (max 175Hz) HDR1000
Case None... naked on desk
Audio Device(s) Astro A50 headset
Power Supply Corsair HX750i, 80+ Platinum, 93% (250~700W), modular, single/dual rail (switch)
Mouse Logitech MX Master (Gen1)
Keyboard Logitech G15 (Gen2) w/ LCDSirReal applet
Software Windows 11 Home 64bit (v23H2, OSB 22631.3155)
Keep in mind that Cinebench R15/20 scores dont benefit much from DRAM speed and performance.
You need memory benchmarks like AIDA64 has, or some gaming benchmarks to observe the before and after scores.
 

steveol

New Member
Joined
Jan 13, 2021
Messages
5 (0.00/day)
Keep in mind that Cinebench R15/20 scores dont benefit much from DRAM speed and performance.
You need memory benchmarks like AIDA64 has, or some gaming benchmarks to observe the before and after scores.
I just did the Aida64 tests, and looking at the results between the xmp alone and the xmp/ memory times tightening+overvoltage, I am wondering if there is an impactful gain there?
 

Attachments

  • aida bench1.png
    aida bench1.png
    206.1 KB · Views: 79
  • image_2021-01-18_080529.png
    image_2021-01-18_080529.png
    1.5 MB · Views: 91
Joined
Sep 3, 2019
Messages
2,965 (1.78/day)
Location
Thessaloniki, Greece
System Name PC on since Aug 2019, 1st CPU R5 3600 + ASUS ROG RX580 8GB >> MSI Gaming X RX5700XT (Jan 2020)
Processor Ryzen 9 5900X (July 2022), 150W PPT limit, 79C temp limit, CO -9~14
Motherboard Gigabyte X570 Aorus Pro (Rev1.0), BIOS F37h, AGESA V2 1.2.0.B
Cooling Arctic Liquid Freezer II 420mm Rev7 with off center mount for Ryzen, TIM: Kryonaut
Memory 2x16GB G.Skill Trident Z Neo GTZN (July 2022) 3600MHz 1.42V CL16-16-16-16-32-48 1T, tRFC:288, B-die
Video Card(s) Sapphire Nitro+ RX 7900XTX (Dec 2023) 314~465W (366W current) PowerLimit, 1060mV, Adrenalin v24.2.1
Storage Samsung NVMe: 980Pro 1TB(OS 2022), 970Pro 512GB(2019) / SATA-III: 850Pro 1TB(2015) 860Evo 1TB(2020)
Display(s) Dell Alienware AW3423DW 34" QD-OLED curved (1800R), 3440x1440 144Hz (max 175Hz) HDR1000
Case None... naked on desk
Audio Device(s) Astro A50 headset
Power Supply Corsair HX750i, 80+ Platinum, 93% (250~700W), modular, single/dual rail (switch)
Mouse Logitech MX Master (Gen1)
Keyboard Logitech G15 (Gen2) w/ LCDSirReal applet
Software Windows 11 Home 64bit (v23H2, OSB 22631.3155)
From cache & memory benchmark between XMP and OC you only gained some on "Copy" bandwidth (+1.5GB/s). But latency gone up from 73 to 76ns.
Its actually worst over-all, with such an increase in latency.

When you did the manual timings did you also let XMP profile Enabled? Also it would help to compare DRAM settings with ZenTimings software between XMP and manual.
Get latest:

Dont bother with 2400Mhz speed. Just compare XMP vs manual(tighter) timings.
You can try manual settings with XMP disabled, but then you also have to manually set DRAM speed (3600MHz or x36) and DRAM voltage to 1.35V or more if its not stable.
 

steveol

New Member
Joined
Jan 13, 2021
Messages
5 (0.00/day)
From cache & memory benchmark between XMP and OC you only gained some on "Copy" bandwidth (+1.5GB/s). But latency gone up from 73 to 76ns.
Its actually worst over-all, with such an increase in latency.

When you did the manual timings did you also let XMP profile Enabled? Also it would help to compare DRAM settings with ZenTimings software between XMP and manual.
Get latest:

Dont bother with 2400Mhz speed. Just compare XMP vs manual(tighter) timings.
You can try manual settings with XMP disabled, but then you also have to manually set DRAM speed (3600MHz or x36) and DRAM voltage to 1.35V or more if its not stable.
Yes, with xmp as a base. I redid the OC but from the default bios settings, I have comparisons between xmp and default and OC among other results in here:
No matter what I did, xmp has the best latency and with the best tightened timings I could achieve it is not spectacular vis-a-vis xmp. and that at 1.44v
May be it is better just to revert to xmp at 1.35 and call it a day I suppose.
 
Top