AMD to Push DDR5 Memory Overclocking With EXPO Technology
Back in January, we reported that AMD was ramping up its memory overclocking marketing. The new system is called RAMP, and stood for Ryzen Accelerated Memory Profiles. Though we’re not Professional Acronym Judges (PAJ), it seemed like a good effort at branding for AMD. It was concise, conveyed forward movement, and communicated the company’s goals; to “ramp up” its memory speeds. However, the company seems to have fallen out of favor with it, and has changed the name to EXPO. This new branding apparently stands for EXtended Profiles for Overclocking.
AMD filed for a trademark for the name EXPO back in February. Like any trademark filing there’s not much information other than it relates to memory and integrated circuits. However, more information has now come to light via Videocardz. According to the site’s sources, EXPO will allow DDR5 memory modules to store two overclocking profiles. One will offer high bandwidth, and the other will allow for lower latency. The site says the second profile will be optional. The technology will also work with all DDR5 modules, so it won’t be limited to just desktop users looking to overclock their RAM sticks. If it works like Intel’s version, you’ll be able to overclock your memory by enabling it in your computer’s UEFI/BIOS. The memory modules are pre-tested by the manufacturer and validated to run on AMD’s platforms at specific speeds. That data is stored in the EEPROM of the memory sticks, and enabled by the user if they wish.
AMD is seemingly trying to offer something that sounds similar to Intel’s own memory overclocking branding. Intel’s is called Extended Memory Profiles, known as XMP, and for DDR5 it’s called XMP 3.0. Perhaps it’s trying to ease the transition for folks switching teams, but in our minds it just sounds confusing. Either way it seems AMD is serious about at least one type of overclocking for its upcoming architecture.
The company recently made headlines saying overclocking would be a big focus for Zen 4. As a reminder, Joseph Tao, Memory Enabling Manager at AMD, made the following remarks. “Our first DDR5 platform for gaming is our Raphael platform and one of the awesome things about Raphael is that we are really gonna try to make a big splash with overclocking and I’ll just kinda leave it there but speeds that you maybe thought couldn’t be possible, maybe possible with this overclocking spec.” We noted our skepticism at the time, as the company hasn’t offered a good overclocking CPU in many moons. Upon reflection, it’s entire possible the company was talking about overclocking memory, not the CPU. As you recall, its newest V-Cache CPU doesn’t allow overclocking, and it’s plausible Zen 4 CPUs will also feature additional L3 cache. If they can’t be overclocked, then it makes sense to focus on memory overclocking. We doubt that will be the case, but we’ll find out at the Zen 4 launch later this year.
Continue reading
Nvidia Will Mimic AMD’s Smart Access Memory on Ampere: Report
AMD's Smart Access Memory hasn't even shipped yet, but Nvidia claims it can duplicate the feature.
Nvidia Unveils Ampere A100 80GB GPU With 2TB/s of Memory Bandwidth
Nvidia announced an 80GB Ampere A100 GPU this week, for AI software developers who really need some room to stretch their legs.
AMD Will Bring Smart Access Memory Support to Intel, Nvidia Hardware
AMD is reportedly working with Nvidia and Intel to bring hardware support for Smart Access Memory to other GPU and CPU platforms.
Tesla Ordered to Recall 150K+ Vehicles to Repair Memory Failures
Tesla has been asked — or "asked" — to recall some 159,000 vehicles to repair a NAND memory issue that will eventually cause failures on every affected vehicle.