Stop wasting money on fast Ram!! 7200MHz vs 4800MHz...

too bad today's video isn't sponsored by crucial because we're going to be talking about how crucial DDR Ram is to your system but how much Fury you can feel if you overspend on that although that's not a crucial brand but you get it dad jokes and puns aside today we're going to talk about ddr5 very fast speeds versus base speeds on gaming PCs and whether or not you're just throwing money away by getting really fast Ram the vonic x24k LED projector is the world's first projector designed for Xbox consoles featuring screen sizes up to 120 in Native refresh rates up to 240 HZ built-in Harmon Card and speakers and low latency gaming modes the x24k gaming projector from vsonic delivers a truly customizable experience for both gaming and HDR Cinema to see our full length video covering features and user experiences follow the links in the description below it's not a new topic it's a topic that tends to come up all the time especially as RAM sort of progresses i' I've done this topic on DDR3 ddr4 and we're doing it now ddr5 and I already said in the intro we're talking about for gaming PCs specifically gaming PCs whether or not having very fast Ram gives you any sort of benefit compared to slower Ram so what we did uh well what I did anyway for the comparisons is I compared 7200 MHz 32 GB uh sets of ram versus base clock speeds at 4,800 so if you guys don't know base clocks ddr5 J deck specifications are 4800 MHz cl40 versus 7200 MHz at CL 34 but we also threw in their 6,400 MHz Ram at CL 38 so what you're going to see right here are a couple of different things compared to sort of get like a slow medium and fast now one thing to point out too is the fact that Ram pricing is all over the place and this video is not specifically talking about the capacity of RAM and how that affects your gaming the capacity of ram should have no impact on your gaming whatsoever unless your graphics card Ram is so low in terms of GD gddr RAM available to the graphics chip that it's doing Swap all the time with the system Ram then you're going to notice terrible performance because system Ram is significantly slower than the ram that you find on your graphics card and not to mention that interface between graphics card going through the PCI all the way through the chipset into the RAM and then back to the graphics card is extremely slow um but you can see right here prices all over the place 7200 MHz dims can cost you as much as300 or $400 although they've started coming down a bit because now they're up to like what 8400 MHz Ram which isn't even guaranteed to run on your CPU by the way this probably most guaranteed to run on Intel CPU AMD is still a bit picky when it comes to Ram speeds even things that are Expo but um it is an overclock because the printed speeds on your RAM are not the speeds that are going to load when you put it in your system and turn it on you've got to go and enable either the XMP profile the docp profile for the ryzen processors that are older than than uh am5 or on the am5 processors what they refer to now as in AMD Expo profiles but you can see pricing here is $230 for a 5200 MHz dim on that Fury that's that's a complete ripoff but you can find Ram also extremely cheap as much as like 100 bucks for 32 GB of like slower Ram like 5200 MHz or whatever but Ram pricing is all over the place which is why people get extremely confused they're like why is this one $5 more this one's $20 more and it's slower it's just stuff is all over the place so what I did was I took what I consider to be a a high-end GPU but not so high-end that we're going to bottleneck everything so I took a 4080 super yeah I know that's that's almost a 4090 but not quite to be honest I just 490s are so unobtanium for most people at that point it's not even worth talking about even though the 480 super is still pretty unobtanium for most people and I ran it through our a few of our benchmarks both synthetics as well as track benchmarks to see whether or not we get any sort of uplift in performance in either 1080P 1440p and 4K so let's go ahead and get started with some of our synthetics so 3D Mark Port Royal is a um it's a a benchmarking tool that's designed specifically for like GPU drag racing I mean sure it gives you a number but it it's basically just about comparing one GPU to another comparing overclocks and it's extremely efficient what I mean by that is it's not loading it's not waiting on the CPU to have to really do much the entire process is is stored entirely in Ram to make it as fast as possible so the CPU is pretty much doing nothing during this test and that's pretty obvious that it's not uh the ram is having no effect on the performance either as you can see we've got the 7200 MHz RAM on the top at 18676 the 40 the 4,800 in the middle at 18598 and then the 6,400 in the uh bottom at 18545 now all three of these scores are actually margin of error because when you start dealing with very fast graphics cards very not so much in Port Royal and times spy but as the frame rates get really high even though it's stored entirely in system Ram background tasks can have a huge effect on some of these these numbers so I could have ran this test 10 times and probably gotten three or different orders of all three cards all 10 times because they're so close uh in this type of test I mean the difference between 18676 and 18545 would be absolutely impossible to determine when watching the test I mean talking down to the tth of a frame per second average change that is giving these score differences if we move over to a Time spy extreme which is just a rasterization test there's no P there's no RT in there you can see once again we've got which is really weird and you're going to see this trend as we go forward here the 7200s on top at 14730 4,800 MHz in the middle at 14714 so we're talking 16 points difference between these two and then 300 points lower almost 3 00 points is the 6400 but one thing I want to point out before we move on to these tests cuz you're going to notice some weird anomalies here is the fact that the 6400 MHz Ram is being uh run by our tforce team group tforce RGB Ram this is a CL 38 I believe yes it's a CL 38 so it's running 6,400 MHz at CL 38 which is the timing versus our Trident Z5 7200 MHz uh kit for the 7200 and the 4800 so we took the max speed on this one ran the test and then we dropped it to base clocks for the jde specifications which is 4,800 MHz at cl40 um for the rest of the test so the 6400 keep in mind is the team group where I guess technically if we should have we should have just run this in the middle setting we wanted to compare it to our normal bench marking Hardware that we use to see how much does timing really take effect because the Trident is CL 34 and these are CL 38 so now you're kind of seeing where memory timings can start to really sort of play in versus the actual um memory like Max Speed itself but it's important to keep in mind though at the 4800 it's at cl40 not cl34 so if we had run 4800 cl34 we probably would have seen it be even farther ahead but anyway this this 14486 versus the 14714 um again there's a myate of things that it could have been to give that sort of a differential and that thing is about synthetic benchmarks they're intended to show how little things can give big effects in the numbers that's part of the reason why it's like a drag racing tool but if we start moving into actual gaming and we look at resolutions this is what we get and I don't know why this is the case but the RTX 480 super at 4800 MHz stock speeds on the Trident was fastest in 1080p by 5 FPS like a significant margin which is really odd the fact that at 4,800 uh or the 480 at 7200 was 5 FPS lower if you look at our 1080 our 1440p numbers though margin of error 2 FPS that's like not even 1% it's barely 1% of of what it's actually just barely over 1% of the overall score a million different things that could be that could be Windows background updates might have checked is there an update that little check could have affected that but if you look at 4K entirely untouched the 4800 or the 480 at 6400 257 a significant difference right there and this right here between these two could quite honestly be the cl34 versus CL 38 to be honest but if we look at our 1440p numbers again 191 versus 194 3 FPS and then 103 and 4K so slightly slower overall on the team group Ram which I thought was interesting but Borderlands is also a very heavily aimed the optimized title Nvidia is definitely gotten faster at it but it's it's always been weird on Nvidia titles if we move on to cyberpunk no RT the 4080 at 6,400 MHz was our fastest 1080p at 204 but if we look at our 1440p and our 4K numbers they're actually slower than the 7200 MHz Ram at 126 and 58 cyberpunk is another strange title um it's we can run the test five times in a row and get five different variants in one or two FPS performance changes it's just the way that title is it's frustrating but it is just the way it is and then you can see down here our 4800 MHz um 1 FPS better in 1440 versus 7200 and then exactly the same in 4k but you can see a couple of FPS Improvement when it comes to the 7200 here can explain why the 6400 beat it with slower timings or I should say looser timings but it is what it is but as soon as you turn on RT now we're GPU bound right we're very GPU bound because the GPU is having to do all the all that um that post-processing which means the CPU is not trying to deal with 200 frames per second it's dealing with half that at 103 frames per second 103 103 and 102 again the 4800 on top margin of error this is 100% margin of error here 65 versus 66 and 1440 32 32 and then 31 down here in 4k at the bottom 64 versus 66 and 65 102 103 103 so all three of those tests are a wash it's just margin of error whatever one process in the background might have triggered and started in the middle of this test um is probably what led to these couple of differences Shadow the Tomb Raider is a interesting one to look at here because it actually shows an proper linear change with frequency and it's fun because it's an older title and it's an older title that has actually increased its engine limits over time very rarely do game developers ever increase the engine capacity it used to be 200 FPS now it's over 300 as you can see but our 4080 super with 7200 MHz at 308 FPS we saw an improvement of 10 FPS versus the 6400 probably again the timings as well as the just super fast transfer rates um and then n 1080p down here all the way down to 287 for the 4,800 and if we look we see that same linear Improvement in 1440p 248 257 261 then in 4k we see it doesn't matter because we're not waiting on the CPU at all there so you're seeing now in games that are achievable at very high frame rates is where you might start to see some measurable difference in Ram speed versus um resolution at that point so as the resolution gets higher and there's more load on the GPU and it's more GPU bound the less benefit to having faster Ram is going to be we turn on rate tracing exact same story as you can see although at 1440p it's pretty much a wash 189 188 188 because again GPU bound RT is on much lower frame rates but you can see the 7200 pulled out quite ahead 255 versus 246 again no one is going to notice that with their eyeballs period you are going to notice that with your wallet though in the difference of price between the RAM and then even the slowest 4,800 MHz Ram is running 239 and 1080p but look at the 4K and 1440p numbers absolutely dead even moving on to Forza which is my last title here um you can see once again a nominal Improvement of 7200 MHz Ram over the 4,800 uh the same Ram kit at base JX speads 1440p is a is a dead heat 4K exactly the same same thing moving on here with the 6400 and slightly slower timing or looser timings at CL 38 197 178 and 136 exactly the same so that's where we kind of stopped the testing CU we we had our patterns really start to to show there and this is not to say fast Ram is not beneficial to you it's just if we're talking specifically in the discussion of Gaming it comes entirely down to multiple factors one what resolution are you running if you're running 1440p or higher it is very unlikely that you are going to notice any sort of performance difference with 7200 MHz or higher versus base clock speeds of ddr5 at 4800 now ddr5 is significantly faster than ddr4 if we did these tests again with ddr4 you might notice a little bit more of an improvement of the faster ddr4 speeds as we go up the stack in a faster graphics card because of just the ddr5 base clocks are so fast versus ddr4 maybe we'll do this test again if you want me to with ddr4 s off in the comments below maybe we'll try and make that one even more exaggerated by putting like a 4090 that's overclocked on like a ryzen CPU or an Intel 12th gen or something with ddr4 and see what happens there um this is also a 139k I forgot to mention at the start of this video so it's already a very fast multi-threaded CPU that is running AI overclock as well um if you are doing multiple things with your system if you are doing like any sort of video editing or uh photo editing and such you might find that capacity is more important to you than speed of the ram Because unless you're doing tasks that are stored entirely in the ram we're having how fast that data can be swapped out of the ram is more important than how much data the ram has to hold then faster Ram speeds at lower capacity might be important to you you might have you might say okay I'm dealing with something that is this program is very notorious for needing fast Ram go SE with 32 gabt of 7200 or 7,400 or something but if you're dealing with high amounts of data that's stored in RAM for editing purposes might make more sense to go with 64 GB of RAM and drop it down to like 6,000 MHz because the capacity at that point is more important than the transfer speed but when it comes to gaming this is not this is the same story being retold over and over and over fast Ram is very difficult to even notice without it being in a chart form unless you're dealing with like lower resolutions like 1080p with very fast graphics cards which are already a terrible value proposition to go with low resolution and high-end Graphics unless you're again running a very specific type of game that needs those super ridiculous frame rates um that there's that's such a niche group versus the general consumer that that's not even worth going into as as a reason why I would ever pair a 40 90 with a 1080p monitor or something stupid like that and then pairing it with very fast Ram but that's the only time you're going to notice any sort of difference so my recommendation to you would honestly be save the money if we go back to like new egg or something here let's just see cuz I'm curious now what kind of deals we can find on RAM if we look up ddr5 and again this is very specific to ddr5 uh for this particular test I'll do ddr4 if you guys want to see it if we look at 32 GB capacities which is quite honestly the most common kit you're going to find because um ddr5 16 gab dims are like the standard right finding 8 gigabyte dims are actually harder than finding 16 gab dims so if we look now at um I don't know if we just sort by Price lowest price um here's team group V Vulcan 32 GB $ 5687 199 that would be a great set of ram to put in your system that's cost effective fast fast enough that you're never going to notice a difference enough capacity to where you could still do video editing photo editing and such and have 32 gigabytes available to you you're just not getting the the RGB aspect or anything to it which I think most people are truly not going to care about but if you did want RGB spend $2 more and here it is right here the tforce Delta which has RGB it's 5600 and still 32 GB of RAM so the point is you could easily easily overspend you could probably overspend enough on your RAM that the price difference of going to more reasonable Ram might actually bump your CPU up one spec maybe from like a 13400 F to like a 13600 K which would be arguably a bigger Improvement in your system overall than going with fast ramp the old days of I'm upgrading my memory and my system so I can have a faster system are long gone unless You Were Somehow running so far below the standard like like let's say you're running 8 GB of system Ram then you're going to have bigger problems in gaming at that point your whole system is probably going to feel sluggish on that um but yeah the the old the old adage of I added Ram to my system to make it faster is really not a thing anymore so don't overspend on your RAM and don't let these Ram companies that put out 82 different kits with varying prices between 80 bucks to $800 fool you don't overspend on your RAM it's the easiest place to waste money all right guys thanks for watching like I said if you want me to do the DDR video comparison so you can see how that looks which might have a little bit more tangible difference because of the slower speeds of ddr4 then we'll do that just sound off down below give this video a like And subscribe if you're new around here and as always we'll see you in the next one

Channel: JayzTwoCents

Views: 392,062

Keywords: pc ram, ram speed, ram speed for gaming, is ddr5 worth it, is fast ram worth it, is high speed ram worth it, what speed ram, what is the best ram speed, what ram for gaming, gaming ram

Id: W_lbsSFYVvc

Length: 18min 9sec (1089 seconds)

Published: Wed Feb 21 2024