Save Money on RAM.. with Optane??

Alright, time for a build. 8700k? Check. Z370 motherboard? Check too. RAM? One stick?! Hold on. Where the devil is the other RAM? Aaand… That’s actually a terrific question. DRAM has been in a global shortage allegedly due to collusion between the major producers, for months! So when Intel reached out asking for some creative ideas for a sponsored video, we pitched using their Potane caching modules as a way of… topping up your system memory. And for some reason, they greenlit this, so… Let’s give it a shot. [intro music] So obviously we’re still gonna need some system memory but the plan really is to take this and use it to replace some of this, and the rationale behind this is actually a lot less crazy than you’d think.

In our testing we’ve seen that Optane performs more like RAM than it does like an SSD which is to say that it’s got ridiculously low response times compared to this all while being much cheaper than this, and conveniently, even though our Optane module here plugs into an M.2 slot, so this is something that is designed for an SSD Windows already has functionality that lets us use a regular old …storage device as an extension of the system memory.

So we will give up quite a lot of our throughput compared to actual RAM but for small transactions that might not actually hurt us that much and it will certainly be faster than using a traditional NVMe or SATA SSD or heaven forbid, a hard-drive in this role. To put this theory to the test we’re gonna do something a little bit strange. So we’ve actually installed two 8 gigs sticks of memory in our bench to give us the benefit of dual channel but then we’re gonna use Windows as built in MS config utility to artificially limit our memory to 4 gigs. Without a plus sign in there. So our 4 gigs entry actually gives us just shy of 3 gigs of usable memory All right.

Let’s fire up some gaming benchmarks. These numbers actually start out showing little concern for the lack of available RAM. However, in the open world Far Cry 5, we see a significant drop in 97th percentile frame rates as assets loaded on-the-fly. As for Unigine Superposition, we did see a small dip, but that could just be due to run to run variation. Moving on to productivity, things get painful. Adobe Premiere took a whopping 3 times longer than our native 16 gig tests and Blender is much the same story taking over two hours to complete where our non-limited testing gave us a result under 45 minutes. Let’s hope for Intel’s sake that things go a little bit better with Optane. Unsurprisingly, our gaming results don’t really change much. Though we do gain back most of our 97th percentile frames on Far Cry 5 compared to the 3 gigs test, and Unigine Superposition did pick up a cool 33 points of probably run to run variants. Moving on to productivity, things look… well, actually a lot better.

We shave about a minute off of Adobe Premiere, though we didn’t get that close to true Ram, but as for Blender, wow. Blender took a whole hour off the render time. I mean, to be clear, that’s still 25 and a half minutes longer than if we had all just RAM, but considering that this render calls for a minimum of 12 gigs of RAM and takes about 12 and a half, we know that a considerable amount of our working data was sitting in our Optane module. But that’s only part of the story. Let’s take an even closelier look at these results. What you’re looking at now is a scatterplot of frame-times and the corresponding disk activity, where we can see that most of our spikes in instantaneous frame times, so these would be perceived as hitches, are actually caused by spikes in activity and this is especially evident when we look at Far Cry 5.

We can also look at how the relationship works between drive activity and CPU usage during our Premiere and Blender tests, where we can see… well, mostly a mess. But- but squint harder. Look at this. The dips and drive activity actually correspond to spikes in CPU usage meaning our CPU is being kept fed. So the less a drive thrashes, the more our CPU gets to work. Nowhere is this more apparent than our Blender test, where our non-Optane result had the CPU working at below 50% capacity most of the time. So what can we take away here? Well, for one, Optane memory modules aren’t going to be replacing system RAM anytime soon, but think of the potential. Out of necessity it’s common practice for computing workloads to be coded around the availability of modern hardware, so we needed to create a pretty contrived experiment to show such a clear benefit today.

But imagine if developers, or creative types, or researchers were free to create much larger data sets knowing that they’d have a cost-effective way to work around bottlenecking their CPU. We actually saw this in action in our Optane 900p video. So for today, we’re still recommending Optane modules for accelerating slower storage devices, like hard drives. But this was a really fun exercise in future crafting and it really shows off the potential of the underlying Optane memory. So thanks for watching guys. If you disliked this video you can hit that button but if you liked it hit like, get subscribed, maybe consider checking out where to buy the stuff we featured at the link in the video description.

Again, to accelerate other storage, not to replace your system memory. While you’re down there you can check out our merch store which has cool shirts like this one and our community forum. It’s kind of a cool place to hang out and talk tech..

As found on Youtube