Overclock.net - An Overclocking Community

Overclock.net - An Overclocking Community (https://www.overclock.net/forum/)
-   AMD - General (https://www.overclock.net/forum/13-amd-general/)
-   -   Ryzen 3000 Memory / Fabric (X370/X470/X570) (https://www.overclock.net/forum/13-amd-general/1728878-ryzen-3000-memory-fabric-x370-x470-x570.html)

elmor 07-08-2019 04:08 AM

Ryzen 3000 Memory / Fabric (X370/X470/X570)
 
5 Attachment(s)
This is a brief summary with limited testing to highlight changes to memory and fabric behavior on Zen 2 (Matisse).

As you've probably seen, Ryzen 3000 is capable of much higher memory speeds than previous generations. The main reason for this is added circuitry on-chip that allows running the memory clock (MCLK) asynchronous from the fabric clock (FCLK). While this enables higher memory frequency, it also increases access latency because the data travelling across different frequency domains has to be synchronized.

The below tests try to answer the following questions:

  • How high memory frequency can be expected with Ryzen 3000 on X370/X470/X570?
  • How high fabric clock can be expected?
  • Is there a benefit to high memory frequency and high fabric clock?
  • How big is the latency penalty when running MCLK != 2*FCLK?

An issue with these tests is that BIOS releases are at this time very unpolished. There are new AGESA versions and patches released frequently which board vendors struggle to keep up with. Especially X370 and X470 versions are problematic for memory overclocking. Both the AMD memory initialization method and vendor DRAM settings seem not to not be working or lack tuning. At best these tests will be able to hint at how things will be for end users.

CPUs

  • Ryzen 5 3600X
  • Ryzen 9 3900X

Motherboards

  • ROG Crosshair VI Hero (X370) BIOS 7106 AGESA 1.0.0.2
  • ROG Crosshair VII Hero (X470) BIOS 2406 1.0.0.2
  • ROG Crosshair VIII Formula (X570) BIOS 7702 1.0.0.3

Memory kits

  • Gskill F4-3200C14D-16GTR (2x8GB Samsung B Single-Rank)

Settings

  • CPU: Default in RamTest, 4.0G 1.25V in AIDA64 tests
  • SOC Voltage: 1.25V
  • DRAM Voltage: 1.40V (DRAM Boot Voltage = 1.40V)
  • Timings CL16: 16-16-16-16-45-1T-GD En
  • Timings CL18: 18-18-18-18-45-1T-GD En


Issues

  • C6H 7106: DRAM Voltage at the beginning of POST is always 1.200V which limits the max memory frequency. It's possible to work around by first booting with lower DRAM Frequency and higher voltage, then only increasing DRAM Frequency in steps without the board shutting down.
  • C7H 2406: DRAM Vboot is always 1.200V by default, it can be manually set instead. However the setting is lost after standby power is removed from the motherboard.
  • C6H 7106 + C7H 2406: After failing memory overclocking you get stuck at C5 POST code which never seems to recover. The only way to get back is to clear CMOS.

Results and discussion

https://www.overclock.net/forum/atta...1&d=1562580924

This shows the highest DRAM Frequency that would pass 100% in RamTest with 2x8GB Samsung B-die. 1.25V SOC and 1.40V DRAM Voltage was used in all cases. These settings are not considered 100% stable and are only an indicator of what's possible without much tuning. I plan to test and update with additional configurations but am currently lacking the time. Above 4000 MHz with 2x16GB has been demonstrated on C8F. The tests were completed at 1733 MHz FCLK, anything higher would cause stability issues. Up to 1900 MHz FCLK is possible on some chips, which will depend on the chip quality. Adjusting SOC and VDDG voltages may help to stabilize higher speeds. The C7H was getting surprisingly close to the C8F, both using "Daisy Chain" memory layout topology. The C6H with its "T" topology is doing much worse, likely because of less time spent tuning signaling parameters at this point. These numbers are expected to improve with the next couple of BIOS releases.



https://www.overclock.net/forum/atta...1&d=1562750953

3600X / 1xCCD

Setting MCLK = 2*FCLK (1:1) yields by far the highest read bandwidth at FCLK / MCLK = 1800 / 3600. Increasing DRAM Frequency by one step to 3666 MT/s while keeping the fabric at 1800 MHz reveals the penalty when using asynchronous clock domains. The read bandwidth immediately drops 1.7 GB/s. At higher memory speeds and CAS Latency, it can be seen how fabric and memory clocks scale. Increasing DRAM Frequency from 4266 MT/s to 4333 MT/s at the same 1733 MHz fabric clock doesn't increase read bandwidth. Instead with DRAM Frequency fixed at 4333 MT/s, increasing FCLK from 1733 MHZ to 1800 MHz immediately shows a gain of 1.6 GB/s.

3900X / 2xCCD

With 2xCCD read bandwidth is no longer penalized in asynchronous mode but keeps scaling with both fabric and memory clock.



https://www.overclock.net/forum/atta...1&d=1562750953

3600X / 1xCCD

Due to Infinity Fabric redesign between CCD and IOD the memory write bandwidth is half on CPUs with 1xCCD (<=8C) compared to previous generations (see https://hexus.net/tech/reviews/cpu/1...ryzen-7-3700x/). This is clearly reflected in these results, where the bandwidth only seem to depend on the fabric speed.

3900X / 2xCCD

As expected with 2xCCD write bandwidth is approximately doubled. Similarly to the read bandwidth scenario, asynchronous mode does not longer seem to reduce the write bandwidth.



https://www.overclock.net/forum/atta...1&d=1562750953

3600X / 1xCCD

Moving away from synchronous memory/fabric clocks doesn't hurt copy bandwidth by much, and it seems to scale nicely both with increased memory and fabric speeds.

3900X / 2xCCD

Copy bandwidth gets a nice boost from the increased write bandwidth with 2xCCD.



https://www.overclock.net/forum/atta...1&d=1562750953

As mentioned in AMD's own presentation, the main issue with asynchronous clock domains is the latency penalty. Changing settings from 1800 / 3600 synchronous to 1800 / 3666 asynchronous increases latency by over 10 ns. The difference is slowly reduced as memory and fabric clocks are increased, but never fully recovers.


Summary

  • X370/X470 BIOS is still immature when using Ryzen 3000 (depending on the board)
  • Highest achieved memory frequency was 4333 MT/s on X570, 4200 MT/s on X470 and 3666 MT/s on X370. The results will most likely get closer with BIOS updates.
  • Fabric clock target 1700-1900 MHz
  • Memory latency penalty ~10ns when moving from synchronous to asynchronous memory/fabric clock, which can be reduced by further increasing memory frequency
  • Highest possible fabric clock with memory synchronized (MCLK = 2*FCLK) yields the highest read bandwidth and lowest latency with 1xCCD
  • Write bandwidth only scales with fabric clock with 1xCCD
  • Copy bandwidth benefits from both memory and fabric clock, and is not hurt much by increased latency
  • 2xCCD read, write and copy bandwidth keeps scaling with both memory and fabric clock without being hurt by asynchronous mode. In this case, it comes down to choosing between lower latency or higher bandwidth.

Bonus: PCI-E Gen4 test on X470

With the current available C7H BIOS 2406, PCI-E Gen 4 will work with any PCI-E Gen 4 device connected to the CPU PCI-E lanes without any changes to BIOS settings. Just plug the device and install the driver. Similarly a PCI-E Gen 4 NVMe drive plugged in the CPU M.2-slot should work at Gen 4. The conclusion is that at least on AGESA 1.0.0.2, PCI-E Gen 4 is not actively blocked on non-X570 motherboards. If it will work on your board or in specific slots will be down to the board layout. The closer the slot is to the CPU (shorter trace length) the higher the chance of success. Increased PCB layer count and lack of PCI-E switches will also increase the chance of success.

Proof of concept with Radeon 5700XT:

https://www.overclock.net/forum/atta...1&d=1562696771

edit: Added detailed settings
edit2: Added note about 1/2 CCD configurations and write bandwidth
edit3: Added 3900X results and PCI-E Gen 4 test on X470

chakku 07-08-2019 04:21 AM

Quote:

Above 4000 MHz with 2x16GB has been demonstrated on C8F.
By demonstrated do you mean observed by those who were testing it or has it been published online somewhere? Would personally love to see more on this as I have my finger crossed on getting my 2x16GB B-Die kit to 3600 on the new processors with the C7H.

Also glad to see a post from yourself again!

elmor 07-08-2019 04:31 AM

2 Attachment(s)
Quote:

Originally Posted by chakku (Post 28031076)
By demonstrated do you mean observed by those who were testing it or has it been published online somewhere? Would personally love to see more on this as I have my finger crossed on getting my 2x16GB B-Die kit to 3600 on the new processors with the C7H.

Also glad to see a post from yourself again!

I've seen it myself, though it was at benchmark settings with high memory voltage. See below. I would have tested it, but one of the sticks in my Dual-Rank kit didn't want to play along.

edit: Booted up the system with one stick and works with 4000CL18 1.4V. But it's not 100% stable.

edit2: MSI lists the 4000 MT/s 2x16GB kit CMT32GX4M2K4000C19 in their QVL: https://www.msi.com/Motherboard/supp...support-mem-19

edit3: Similarly Asus lists a 4000 MT/s 4x8GB kit CMK32GX4M4K4000C19 in their QVL: https://dlcdnets.asus.com/pub/ASUS/m...Memory_QVL.pdf . I believe the "Spectek" listing in the Asus QVL refers to Micron E-die.

chakku 07-08-2019 04:44 AM

Quote:

Originally Posted by elmor (Post 28031090)
I've seen it myself, though it was at benchmark settings with high memory voltage. See below. I would have tested it, but one of the sticks in my Dual-Rank kit didn't want to play along.

edit: Booted up the system with one stick and works with 4000CL18 1.4V. But it's not 100% stable.

Looks very promising! Also yeah can't imagine the voltage you'd need for 4000+ at CL12 is usable daily.

Do you still need GDM enabled for dual rank sticks or do they play better with true 1T command rates on the new IMC?

elmor 07-08-2019 04:48 AM

Quote:

Originally Posted by chakku (Post 28031104)
Looks very promising! Also yeah can't imagine the voltage you'd need for 4000+ at CL12 is usable daily.

Do you still need GDM enabled for dual rank sticks or do they play better with true 1T command rates on the new IMC?

~2.0 :) I've not tested that, but even the SR results above are using the default setting 1T Geardown=Enabled.

Streetdragon 07-08-2019 05:33 AM

so tldr:
1800/3600 and tight the timings as low as possible?

VPII 07-08-2019 07:03 AM

@elmor I noticed on some review that showed the memory performance that the 3900X had double the write speed of the 3700X. Is it possibly to do with 3900X being two chips and thus dual or two IF?

Sent from my SM-G960F using Tapatalk

elmor 07-08-2019 08:10 AM

1 Attachment(s)
Quote:

Originally Posted by Streetdragon (Post 28031146)
so tldr:
1800/3600 and tight the timings as low as possible?


Yes, unless you want higher copy bandwidth.


Quote:

Originally Posted by VPII (Post 28031222)
@elmor I noticed on some review that showed the memory performance that the 3900X had double the write speed of the 3700X. Is it possibly to do with 3900X being two chips and thus dual or two IF?

Sent from my SM-G960F using Tapatalk


I suppose theoretically, since each has 16-bit width lanes for writes (32 bits for reads) which makes 2x16=32 bits total. It will depend on the application and if it uses both dies to access memory. From what I remember AIDA64 always tests memory using core 0 which wouldn't show any improvement, but that could have changed. I re-do these tests with 3900X when I get my hands on one.

https://www.overclock.net/forum/atta...1&d=1562598373

Jackalito 07-08-2019 11:13 AM

Thank you so much for sharing your wisdom with us, elmor.
It's great seeing you so active around here again! :specool:

AlphaC 07-08-2019 11:24 AM

Thanks for confirming what AMD wrote: 3600C16 is probably the best compromise for most people


All times are GMT -7. The time now is 05:38 AM.

Powered by vBulletin® Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.

User Alert System provided by Advanced User Tagging (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.
vBulletin Security provided by vBSecurity (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.

vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2019 DragonByte Technologies Ltd.