Netac NV7000 2 TB NVMe SSD Review
ASUS GeForce RTX 4080 Noctua OC Edition review
MSI Clutch GM51 Wireless mouse review
ASUS ROG STRIX B760-F Gaming WIFI review
Asus ROG Harpe Ace Aim Lab Edition mouse review
SteelSeries Arctis Nova Pro Headset review
Ryzen 7800X3D preview - 7950X3D One CCD Disabled
MSI VIGOR GK71 SONIC Blue keyboard review
AMD Ryzen 9 7950X3D processor review
FSP Hydro G Pro 1000W (ATX 3.0, 1000W PSU) review
AMD Ryzen 7 Memory And Tweaking Analysis review





In this article we'll take the new G.Skill Flare X 3200 MHz memory kit and have a closer look at the effect of memory frequencies on applications and games. Next to that we'll also address tweaking a bit so you can peek what effect faster memory and a CPU tweak has as an overall impact on Ryzen 7 performance.
Read article
Advertisement
« ASUS ROG Strix GeForce GTX 1080 Ti Review · AMD Ryzen 7 Memory And Tweaking Analysis review
· AMD Ryzen 5 1500X and 1600X review »
pages « < 7 8 9 10
Fox2232
Senior Member
Posts: 11808
Senior Member
Posts: 11808
Posted on: 06/22/2017 05:39 PM
Very few synthetic tests can truly show benefit of higher bandwidth memory access.
And Bandwidth jump from 3200MHz to 3600MHz is crazy. So, those few real world applications like databases will benefit plenty.
Not even AMD claims this update has an impact on performance. The only thing it is meant to do is increase memory compatibility, so you're not expected to notice anything. However, there was a benchmark done where they saw up to a 1% improvement in some tests. But, depending when their old results were done, that could've been due to Windows updates.
Very few synthetic tests can truly show benefit of higher bandwidth memory access.
And Bandwidth jump from 3200MHz to 3600MHz is crazy. So, those few real world applications like databases will benefit plenty.
schmidtbag
Senior Member
Posts: 7236
Senior Member
Posts: 7236
Posted on: 06/22/2017 05:52 PM
Don't you mean the other way around? Many synthetic benchmarks can show the theoretical potential of memory bandwidth whereas real-world applications often (not always) don't. Very few (again, not all) real-world applications show any benefit when adding more memory channels (look up PC World's article on dual vs quad channel). Up until Ryzen was released, RAM frequency didn't have that big of an impact either.
IGPs, however, do a very good job at taking advantage of total memory bandwidth. Its interesting to me to consider how much potential APUs have lost due to starving for memory bandwidth.
Yes, jumping from 3200 to 3600 makes a difference - I'm not denying that, and I understand this update allows for that. There has already been proof of that prior to the AGESA 1006 update. But my point was going from AGESA update itself has no impact on performance assuming the RAM speed isn't changed. In other words, if your RAM was already at 3600, going to AGESA 1006 won't really improve anything further.
Very few synthetic tests can truly show benefit of higher bandwidth memory access.
Don't you mean the other way around? Many synthetic benchmarks can show the theoretical potential of memory bandwidth whereas real-world applications often (not always) don't. Very few (again, not all) real-world applications show any benefit when adding more memory channels (look up PC World's article on dual vs quad channel). Up until Ryzen was released, RAM frequency didn't have that big of an impact either.
IGPs, however, do a very good job at taking advantage of total memory bandwidth. Its interesting to me to consider how much potential APUs have lost due to starving for memory bandwidth.
And Bandwidth jump from 3200MHz to 3600MHz is crazy. So, those few real world applications like databases will benefit plenty.
Yes, jumping from 3200 to 3600 makes a difference - I'm not denying that, and I understand this update allows for that. There has already been proof of that prior to the AGESA 1006 update. But my point was going from AGESA update itself has no impact on performance assuming the RAM speed isn't changed. In other words, if your RAM was already at 3600, going to AGESA 1006 won't really improve anything further.
Exascale
Senior Member
Posts: 390
Senior Member
Posts: 390
Posted on: 06/23/2017 04:43 AM
So the TLDR is that Infinity Fabric runs at 1/2 memory controller speed, therefore the faster your memory, the faster your chip?
So the TLDR is that Infinity Fabric runs at 1/2 memory controller speed, therefore the faster your memory, the faster your chip?
schmidtbag
Senior Member
Posts: 7236
Senior Member
Posts: 7236
Posted on: 06/23/2017 05:07 AM
Depending how you look at it, yes. But in actuality, Infity Fabric runs at the full RAM speed, not half of it. Keep in mind that memory is DDR, meaning you take the actual clock rate, multiply it by 2, and you get the effective speed (which is usually what the RAM is advertised to be at). So when your RAM is, for example, 2666MHz, that's the effective speed. The actual speed is 1333MHz, and that's the speed IF is based on.
http://www.hardwaresecrets.com/everything-you-need-to-know-about-ddr-ddr2-and-ddr3-memories/
So the TLDR is that Infinity Fabric runs at 1/2 memory controller speed, therefore the faster your memory, the faster your chip?
Depending how you look at it, yes. But in actuality, Infity Fabric runs at the full RAM speed, not half of it. Keep in mind that memory is DDR, meaning you take the actual clock rate, multiply it by 2, and you get the effective speed (which is usually what the RAM is advertised to be at). So when your RAM is, for example, 2666MHz, that's the effective speed. The actual speed is 1333MHz, and that's the speed IF is based on.
http://www.hardwaresecrets.com/everything-you-need-to-know-about-ddr-ddr2-and-ddr3-memories/
pages « < 7 8 9 10
Click here to post a comment for this article on the message forum.
Senior Member
Posts: 7236
Not even AMD claims this update has an impact on performance. The only thing it is meant to do is increase memory compatibility, so you're not expected to notice anything. However, there was a benchmark done where they saw up to a 1% improvement in some tests. But, depending when their old results were done, that could've been due to Windows updates.