Does RAM Speed REALLY Matter?? (DDR5 Edition)

hey everyone it's me the tech tips man now that ddr5 has reached insane speeds of over 10 000 Mega transfers per second we've gotta know which sticks should you spend your hard-earned money on a single kit can cost you anywhere from 160 dollars to over 400 so making the right choice can be the difference between wallet pain and wallet death to help you out we put the screws to 10 kits ranging from stock GX speeds all the way up to 7 800 Mega transfers per second on both Intel and AMD platforms does it matter a little does it matter a lot has the lab managed to find the tdr5 price to Performance Sweet Spot there's only one way to find out lenode cloud computing from Akamai lenode is a powerful Linux based cloud computing service that's affordable and easy to use whether for a small project check or a business critical application see why developers choose lenode sign up today at lino.com LTT to get a free 100 60 day credit right now Intel has the only platform where we can test both ddr5 and ddr4 on the exact same CPU so that's where we started and Intel also had the benefit of being more stable throughout our testing we'll get into that a little bit more later for now all you need to know is that we took some of the beefiest hardware commercially available plugged in some memory kits enabled XMP and went to town we're gonna have all the gear that we used linked down below if you want to check it out for yourself right out of the gate we can clearly see that in most games fast ddr4 is still a valid option but that price to Performance advantage that it enjoyed in the past seems to have pretty much disappeared this is a 3600 Mega transfer per second kit with pretty tight timings and while it beats our ddr5 jdeck kit by any anywhere from 1 to 10 FPS on average it does so at a considerably higher price and once for spending the same amount on our ddr5 our ddr4 not only loses its Advantage but falls measurably behind especially in the all-important one percent lows the thing is average frame rates are interesting but the difference between 311 and 315 FPS is basically impossible to discern so this bar usually matters a lot more because a 10 or even 5 FPS difference is much more likely to be noticeable and it also represents how your system will perform in the most critical moments when the action is most intense for gamers then ddr5 is looking like an obvious choice and and more must be better right not necessarily adding our 7 600 and 7800 Mega transfer per second kits to the graphs I mean yeah they performed admirably but just a few more FPS and sometimes less for a hundred plus dollars more than our other options what's going on here I mean look at these numbers in Ida 64. the more we spend the more memory bandwidth we get so our performance should go up proportionally right well here's the thing Ida 64's read write and copy benchmarks are super useful for uncovering system bottlenecks and diagnostic purposes but they are purely synthetic that is to say that they are not representative of any real world workload other than copying data directly into or out of memory and not every application is actually hungry for more bandwidth for some applications latency is actually much more important and that's where our ddr4 kit shines with only the fastest ddr5 kits closing the Gap most of the ddr5 kits that we tested today hit first word latency figures of 10 nanoseconds or faster but our ddr4 kit managed less than eight nanoseconds you can roughly calculate these values for your own kit by the way just take your modules cast latency or how long it takes to complete a clock cycle then multiply it by 2000 and divide by the trend answer speed the lower the first word latency generally the snappier your memory will be but as evidenced by the performance improvements we see with our faster kits it is both this latency and transfer speeds that are crucial when it comes to getting the most out of your system's memory for applications and games today look at these two 6000 Mega transfer per second kits there's a pretty Stark difference in cast timings here resulting in a 3.3 nanosecond difference in latency the faster one is a little over twenty dollars more for about a one to five FPS Improvement but if we bump up the speed to this 6400 kit with timings that keep us at 10 nanoseconds we get more performance but start to hit our point of diminishing value returns and then anything past this point sees such a stark drop in performance Improvement that it doesn't really warrant spending tens or hundreds of extra dollars to try to make the line go up at least at current pricing but what about AMD well for team red our biggest hurdle was just getting kits faster than 6400 Mega transfers per second to even run long story short they didn't and that's unlikely to change anytime soon our testing started mid last month so that's December of 2022 and we were running on a gisa version 1.0.0.3 patch d right before we filmed this gigabyte dropped a new bios for our motherboard that included an update to 1.0.0.4 supposedly improving memory support and it does but the difference wasn't enough to boot any of our previously incompatible kits I'm certain that professional overclockers will be able to use this extra Headroom to push the platform to new heights but our goal was to measure the out of box experience so we didn't do any tuning Beyond plugging in the CPU plugging in the RAM and enabling XMP or Expo as it were the good news is that even though it's relatively slow our 6400 memory absolutely slayed on AMD going from a basic ddr5 kit to a premium but still not outlandishly priced kit saw almost every game gain 10 average FPS or more this is especially notable because these are similar gains to what we saw on Intel but we aren't spending as much on our top performing memory since we topped out at a lower speed and it's safe to assume that if future bios updates can get the jello to stop shaking amd's performance should only improve as we get stable access to significantly faster memory right hmm well not exactly we are still hitting the point of diminishing returns here especially when you consider the price amd's chips really like tighter timings and that seems to be the biggest Factor when it comes to games I mean sure the jadex spec 4800 cl40 kit is way slower than our top end but look at this 5600 cl28 result it's on par or just behind its more expensive counterpart and for about twenty dollars less and the reason lies here while the transfer speed might be higher on that 6400 kit it actually has the same first word latency as the 5600 configuration I mean if you were to take that faster kit and sit around for hours tuning the timings chances are that you would end up with a better result during our quick sanity check with the latest bios we found that a GSA 1.0.0.4 does improve stability at 6400 with manually tightened timings but but the reality is that many of you don't even overclock your CPU these days let alone your memory and realistically your time is probably better spent buying our new underwear from lttstore.com we've got new colors available check them out today at least that's true for gaming for productivity where we'll have a lot more CPU cores to feed with delicious data it's a whole different Beast similar to what we saw with Intel faster modules seem to make a huge difference in some workloads and then basically none and others handbrake for example sees an impressive boost of 22 percent over jdac wow but then just about everything else sees either no improvement a clearly linear progression that is so small you might as well save your money or what could best be described as run to run variants one major issue to note is that we have more productivity results but our 6400 kit didn't manage to finish all of them so hence the did not finish results we do expect this to continue to improve over time as am5 matures as a platform and new bios updates roll out but it was a major problem for us at the time of testing and the improvements that were available by the time we're actually filming it have been pretty marginal so nobody knows exactly what the future will hold and it's still not 100 clear where the ddr5 bang for your buck sweet spot is going to ultimately end up I mean considering that ddr4 launched in 2014 and higher speed kits took years to become the norm it's still early days for this new generation but what is clear is that you don't want to buy something with significantly slower first word latency if you can avoid it particularly on the lower transfer speed modules oh and also that there is no point buying faster than 6000 or at most 6400 speed memory for AMD at least for this generation you know what else is clear this piece of glass wait that's not clear it's got a Segway to our sponsor Samsung you know what you can never have enough of storage Samsung aims to deliver with the 970 Evo plus nvme m.2 SSD powered by the latest v-nan technology and firmware optimization it can reach read and write speeds of up to 3 500 and 3300 megabytes per second respectfully Samsung's Advanced nickel coated controller and heat spreader enables Superior heat dissipation keeping a cool even under heavy load and now Drive management has never been simpler the Samsung magician software provides a suite of user-friendly tools that help keep your drive up to date monitor drive speed and health and even increase performance so upgrade your storage now visit the link below and check out the Samsung 970 Evo plus if you enjoyed this video we haven't tried it with ddr5 yet but you might want to check out our just how bad is mixing memory video it's not as bad as you might think or at least wasn't

Channel: Linus Tech Tips

Views: 2,166,895

Keywords: RAM, DDR5, DDR4, Memory, PC, RTX 4090, 13900K, 7950X, Intel, AMD, NVIDIA, Crucial, G.Skill, 6000MT/s, CAS Latency, First Word Latency

Id: b-WFetQjifc

Length: 10min 45sec (645 seconds)

Published: Sun Feb 12 2023