Why Does AMD always lag behind Intel in terms of memory clocking / memory performance?

storm-chaser

storm-chaser

storm-chaser

Associate
Joined
15 Jun 2012
Posts
185
I've always wondered why this issue continues to plague AMD... ever since Ryzen's market release AMD has struggled in keeping up with Intel on the memory performance front. I can excuse a couple generations of Ryzen as working out the bugs and developing the platform, but this has now been going on for years and years, with no end in sight, because AMDs latest releases also suffer from this problem. I rarely if ever see AMD rigs with memory over 6000MHz, and even my 9600KF can clock the memory higher than most of the Ryzen processors.

What the hell is AMD doing? Why do they struggle so much in this area? I mean it's a simple multiplier... lol
 
I have personally suffered the Ryzen memory controller for many generations and I do agree this is an area they need to focus on more. With the infinity fabric the Ryzen memory system is more complex than the Intel one but I do not think that is why they have more issues. Intels memory controller has always been excellent and practically idiot proof in most cases and that is because they are the market leader. With the vast majority of systems being Intel the memory manufacturers produce the memory modules with the Intel cpus as the baseline and AMD cpus are an afterthought.
 
AMD while they have good products are still behind mainly because all of this costs a LOT of R&D money and their budget is quite small especially when you compare their R&D budget to the likes of Intel and Nvidia whose budgets are eye wateringly astronomical.

As they gain market share and their R&D budget goes up we'll see these things such as memory compatibility get better and better.
 
Outside of a few niche pieces of software and 1 or 2 games it makes next to no difference, especially if you have one of their new 3D CPUs.

As long as the bandwidth and latency are good enough it's not going to matter for 99% of users on their mainstream platforms.

I enjoy RAM tuning and it does make a system feel more snappy and can help with the 1% lows in competitive titles if you're still playing at 1080p. However, it's time-consuming to get it dialed in and you can easily corrupt your OS.

IIRC you're heavily into benchmarking so I can understand the frustration if you like playing with different platforms but it's about the broader perspective.
 
Last edited:
storm-chaser said:
I've always wondered why this issue continues to plague AMD... ever since Ryzen's market release AMD has struggled in keeping up with Intel on the memory performance front. I can excuse a couple generations of Ryzen as working out the bugs and developing the platform, but this has now been going on for years and years, with no end in sight, because AMDs latest releases also suffer from this problem. I rarely if ever see AMD rigs with memory over 6000MHz, and even my 9600KF can clock the memory higher than most of the Ryzen processors.

What the hell is AMD doing? Why do they struggle so much in this area? I mean it's a simple multiplier... lol
Click to expand...
You don't actually know what you're talking about. More speed doesn't mean more betterer. There are many factors including timing, latency, rank, etc.

AMD uses Infinity Fabric and it prefers to run at a 1:1 ratio with the RAM. AMD have said that DDR5-6000 is the sweet spot for AM5.

Here's an Infinity Fabric explained here and here
 
You must log in or register to reply here.
Back
Top Bottom