**AMD Ryzen 2 Processors: A Comprehensive Guide**

When it comes to AMD Ryzen 2 processors, there are several key factors to consider for optimal performance and efficiency. In this article, we'll delve into the details of these processors and provide recommendations for users.

**Zen Architecture vs. Zen+ Architecture**

The Ryzen 2 series is built on AMD's Zen architecture, which was introduced in 2017. The Zen+ architecture, also known as Ryzen 5 and Ryzen 7, offers improved performance and efficiency compared to the original Zen architecture. However, it's essential to note that the Ryzen 1.5 series (Zen+) has undergone significant improvements since its release.

**Zen 2 Architecture**

AMD has recently released the Zen 2 architecture, which is used in the Ryzen 3000 series. This new architecture offers further improvements in performance and power efficiency compared to the Zen+ architecture. Additionally, AMD's upcoming Ryzen XT series will also utilize this architecture, offering even better performance and efficiency.

**Recommendations for Ryzen 2 Processors**

When it comes to choosing a Ryzen 2 processor, there are several key factors to consider. First and foremost, memory speed is crucial for optimal performance. AMD recommends using at least 3600 MHz memory, as this provides the best balance between performance and cost. However, if you're running slower memory speeds (3600 MHz or lower), it's essential to adjust your CPU settings accordingly.

In particular, we recommend setting the F clock to exactly half of your core speed or memory speed. This can be achieved by adjusting the F clock ratio in your BIOS settings. While experimenting with different ratios may provide some benefits, it's unlikely to have a significant impact on performance beyond 0.5x. Additionally, over-tightening the F clock ratio can lead to instability issues.

Another essential factor is motherboard support for core enhancement features, such as Precision OverBoost. If your motherboard has this feature, ensure that it's enabled and set to auto or manual mode. Disabling this feature will disable Precision OverBoost, which may result in reduced performance.

**Intel Comparison**

When comparing AMD Ryzen 2 processors to Intel CPUs, it's essential to note that the latter often offer better performance due to their higher clock speeds. In our testing, we found that even with the same GPU and settings, Intel CPUs were able to outperform Ryzen 2 processors by a significant margin.

However, it's crucial to remember that this is largely due to Intel's ability to run at faster clock speeds. The difference between 142 FPS and 150 FPS on an Intel system may be noticeable, but in reality, the performance difference may not be as significant when running games at lower frame rates (e.g., 60-80 FPS).

**Conclusion**

In conclusion, AMD Ryzen 2 processors offer excellent performance and efficiency, especially with proper memory settings. By following our recommendations and understanding the limitations of these processors, users can optimize their system for peak performance. Additionally, it's essential to consider motherboard support for core enhancement features and Intel CPUs when comparing AMD and Intel options.

**Additional Tips**

* For optimal performance, ensure that your F clock is set to exactly half of your core speed or memory speed.

* Enable Precision OverBoost on your motherboard if supported, but be aware that disabling this feature will disable the feature altogether.

* Consider using at least 3600 MHz memory for optimal performance and efficiency.

* When comparing AMD Ryzen 2 processors to Intel CPUs, keep in mind that Intel's higher clock speeds may result in better performance.

**AMD and Motherboard Meeting**

We recently had the opportunity to meet with AMD representatives, who informed us that there have been false rumors circulating about the 4000 series being bumped or pushed back. According to AMD, this is not the case, and the 4000 series is still on track for release this year.

In other news, we've seen significant improvements in multi-threaded performance across various benchmarks when transitioning from Ryzen 1.5 to Zen+ architecture. However, it's essential to note that individual results may vary depending on system configuration and workload.

**Recommendations Summary**

* Get at least 3600 MHz memory.

* Set F clock to exactly half of your core speed or memory speed (e.g., 1500/3000).

* Enable Precision OverBoost on motherboard if supported.

* Consider using Intel CPUs for better performance, but be aware that this may not always result in noticeable differences.

By following these recommendations and understanding the intricacies of AMD Ryzen 2 processors, users can optimize their system for peak performance and efficiency.

"WEBVTTKind: captionsLanguage: entoday we're going to be talking about the pinnacle of both performance and sound many of us are working and learning from home during these challenging times and micro center has what you need to create a seamless work or learn from home setup including laptops desktops monitors and more for those of you practicing social distancing but need to get your daily dose of tech microcenter has a community forum where you can discuss and share your favorite tech setups and shared bias and even get help with troubleshooting from fellow enthusiasts to learn more about how to get your essential techniques from micro center and visit the community click the links in the description below all right so a while back we did a video where we specifically showed how uh performance can be impacted with memory speeds on intel and i asked you guys at the end of that video if you want me to do one with amd since verizon was known to be a bit more picky when it came to ram speeds specifically with you know first edition generation ryzen we're in ryzen 2 now uh and a lot of you guys were like yes please do that so i think we're going to do that today i think i hope that's what we're going to do today that's what i'm set up for so i think we're going to we're going to be using windows 95 for these tests though because we want to use a a relevant operating system so first things that we're going to talk about here is just our test we're just going to use geekbench and cinebench reason for that is there are free tests again that you guys can download and try these tests yourself and we're just going to be looking for any of these score differences we're not comparing them to other cpus or any of that we just want to see with out of the box ram settings which on ddr ddr4 is always going to be 2133 megahertz for even if you buy a dim that says 4200 megahertz it's gonna run 2133 if you don't go in and enable docp profile on amd um the other thing is going to be i think we'll play with the fabric clock a little bit or the f clock because that's the other half of the equation when it comes to memory speeds speeds with amd just a quick high level here and phil can put a little graphic up to show if you guys aren't aware the amd cpu is made up of chiplets there's a memory controller on the substrate as well as the cpu chiplets now when the cpu is separated like it is which is why intel was always like glued together cpus which is kind of funny because that's kind of what intel does with their like big xeon stuff but anyway moving on when you have the cpu dies like that that are separate they've got to share the memory as well but the f clock is the speed at which the memory communicates with each other on the substrate so a very bad explanation of it but that's why if you go fast memory but if you don't have your f clock set to auto or at least proportionate to your memory speeds you can actually be losing performance because of the communication between the chiplets but moving on i need to actually install geekbench hold on i thought you're prepared for this man be prepared i am not scar i won't be prepared quit pro quo so we're gonna do testing methodology in basically three different phases if you will i'm just running out of the box settings right now with the cpu kind of fluctuating and doing its thing with its uh precision boost kind of controlling things the memory is out of the box 2133 the fabric clock is at auto and then we're going to do is we're going to lock everything down we're going to lock the cpu to 4.2 i'm just curious how much fluctuation there's going to be just with locking the cpu um and then we're going to do um obviously our memory speed at our slowest which is 2133 and then we're going to do a 3000 and then we're going to do a 3 800. this memory is 3800 modules cl15 i'm going to leave the stock timings i'm still debating whether or not i want to do a timing video because you could actually go in there and cause yourself some headaches by playing with the timings and trying to get them back to factory if sometimes clear cmos doesn't work like we were kind of dealing with before starting this video clear cmos wasn't doing crap when this thing didn't want to boop we found out it was our gpu and that's a whole different video subject yeah i find timings to be not that important unless you're going for like record overclocks and stuff but the reason why i chose 3000 is that it seems to be where a lot of the memory that's affordable today is kind of that for the 16 gigabyte dual 8 gig sticks they seem to be right around 3200 megahertz so you guys can see and then we can see what the difference is between low speed 2133 and nearly 4 000 megahertz um prediction time though because we haven't actually done this test i predict with ryzen 2 which is what the 3950x is is going to see marginal at best score differences between 3000 and 3800 i think we might see a little bit of an improvement from 2133 to 3000 but i think what we're kind of going to show you here too is that you don't need to spend all kinds of money on expensive ram which was our whole point with the intel video now the other thing we did with intel video was some gaming benchmarks which showed fps improvement but anyway here's our first test with cinebench let's see how this does so we got a 93 40 and then before that we had a 93 30 and then another one we had a 93 53 so you can see it's fairly consistent um and it scores and i mean 10 f 10 score difference here is really not something you're gonna notice it's we can run it 10 times to get 10 different numbers but they'll all be within like half a percent or a quarter of a percent of each other which is totally normal remember there are background processes happening here um all right so that was a 93 40 or 93 40. here's our baseline geekbench scores 1312 single core and uh 12 163 multi core now we're going to do only one gaming title because what we what we didn't expect with the intel system was that we got like a 30 fps improvement in like shadow of the tomb raider and far cry 5 but we saw no difference really in wildlands like one or two fps at the most we just realized though our 5700 that's sitting on here this is not an xt just 5700 is not going to bottleneck our 3950x so we're going to use the same graphics card we did in those tests which is actually the msi gaming x trio 2080 ti and we're going to set it at 1080p because obviously we want to see how memory speed is going to help the potential bottlenecking you're going to do with the cpu um so i guess i need to swap out the gpu now but let me say this fanboys this is not a 3950x gaming performance versus a what was that a 9900 i think we did with that this is not about that i think the 9900k is going to give you better gaming performance than a 3950x because we were running that thing at 5 gigahertz we all know gigahertz reigns supreme when it comes to gaming we are just looking at what memory speeds do for this particular cpu not we don't care how it performs to other cpus that's not the point of this video they were tight all right so here we go graphic settings were set to highest 1080p uh our trio is set to plus 75 megahertz on the core plus 500 on the memory fan set to 100 per 100 per spent it's only wednesday and it's been a long week okay you know what i'm trying to say all right well you know obviously we need to uh do all the tests now so i already explained to you how we're gonna do it we're gonna chart it we'll come back we'll talk about the differences and where the real sweet spot is gonna be in terms of memory speed and where to spend your money and such on on memory so it has been hours upon hours later and we have run so many different iterations of these tests now what you're gonna see on the charts um like i said we locked the core at 4.2 gigahertz and then we did 2133 and thirty eight hundred but what we saw was a a decent jump i say decent i mean we're talking not a lot of percentage-wise numbers like what four percent or something like that i think it was at the most in terms of cpu usage or cpu scores but once we went into gaming with shadow of the tomb raider we saw pretty significant increases across the board from 120 fps all the way up to like 140 fps with the 2080 ti that is overclocked so that tells us that by speeding up the memory just like we saw in the intel test increased our fps as well which is obviously important but we started thinking about this we're like okay ram and the way it works with ryzen is totally different than intel and we already explained the fabric clock now i do know for a fact that the fabric clock running at exactly half the rate of the memory speed is where you get the most performance out of it now this being a docp of 38 68 or 66 it says 3868 when you go into the bios but that means that in order to actually get your f clock at half that speed it would have to be running at 1933. now we also know that trying to go above 1800 megahertz on the f clock leads to some significant instabilities in our case a no post now most people aren't going to go in there and start tweaking the f clock and all that sort of stuff so what we did just to kind of see what would happen here because you should also never go in there and manually set your core clock for your cpu because when you do that on amd unlike intel where the boost speeds for both all core and single core are so much higher than the base clocks and such when you lock the core to 4.2 on any ryzen processor at this point although we are talking about the 3950x the single core going all the way up to 4.7 gigahertz is now 500 megahertz limited because it can't go past the 4.2 now i don't personally know if there's a way to go in there into precision boost and set the clocks individually for single core 2 core 3 core whatever like intel you can do that i'm not sure if you can do it with ryzen if you can please link down below a resource we could use so we could actually maybe get better at tuning these overclocks but when you lock into 4.2 you lose a crap ton of single core performance now you might be thinking but jay everything's multi-threaded these days but you don't understand when there's bursty single load instructions that happen all the time on various programs those are definitely benefited by that single core boost going all the way up to 4.7 in this case now what we did see was by reducing our ram or enabling docp for the timings which it's funny because the timings on this particular set of ram that we have are not the fastest it's like cl 18 19 19 39 or 38 i think when we enable docp for the timings because if you don't do that the timings are even looser when they're just out of the box but then we set the memory to 3600 and then manually set the f clock to 1800 is where we finally saw our most beneficial across-the-board user experience sure our all core was automatically boosting to 4.1 and by setting it to 4.2 we saw an increase in multi-threaded performance across the board because an extra 100 megahertz across 16 cores and 32 threads adds up but the the amount of loss you take everywhere else just isn't worth it so at the end of the day our recommendation regarding ryzen 2 processors now remember this is different than ryzen 1.5 we'll call it or zen plus architecture is what they actually refer to it as so you have zen architecture which is like the 1800x zen plus which is all the 2000 series stuff now zen 2 is the 3000 and then now the new 3000 uh xt series that's coming out plus i guess the 4000 series is still slated for this year amd and and and us we'll just say us all of our content creators we did have a meeting with amd and they wanted us to basically mention that i guess there's false rumors saying that the 4000 series has been bumped or pushed they say that's not true it's still coming this year back to what i started to say a couple minutes ago though our recommendation quite honestly is just get 3600 megahertz memory and call it a day you can still get it for a pretty decent price but if you're gonna also be running any memory 3600 or less i highly recommend going into your bios and setting your f clock to be exactly half of whatever your core speed is or your memory speed is now you can play with it and and increase it if you want but it seems to be if you keep it at a ratio of you know 0.5 of your actual ram speed it seems to give you the best performance so if you have 3000 megahertz dimms set it to 1500 you can play with it set it to 16 set it to 18 and see how it does but i believe there's diminishing returns there where it's not going to really do anything and you could be you know potentially over stressing the f clock voltage and all that stuff we're trying to push those faster clocks for no reason whatsoever um the other recommendation too to me is if your motherboard does have a core enhancement feature like all the asus boards do and i think they all do these days make sure you leave that to auto or enabled if you disable it what that's going to do is it's going to also disable the precision overboost stuff so over boost precision boost turbo boost no leave that enabled or auto because if you don't then it's not gonna boost at all it'll just sit at its base clock of like three five and nothing you'll have a bad time because it's three five is slow these days that's slow just like intel lots of gaming performance to be had 22 fps improvement in our case from 120 fps average at 1080p remember we're intentionally pushing a lot of frame rate to try and see where the bottlenecks are on the cpu um to the point to where as you can see right now we left this up on the screen an average of one oh i tried oh that's a power button the power button shouldn't be right where i'm trying to grab it there we go so up to 142 fps but check this out we finally got to the point to where our gpu was our limiting factor we couldn't get the number higher than 142 fps at our current overclock of only plus 75 on the core and plus 500 on the memory and then power limit increase all the way as far as it'll go 17 gpu bound that tells us our gpu was doing the best that it could now the weirdness here is that if we took the same gpu with the same settings and put it on an intel cpu we would get more fps and simply because of the five gigahertz that you could run on like intel cpus these days it does make a difference however i challenge anyone to tell me that they can truly see a difference in a game running 142 fps versus 150 because i think about 150 155 or something that is where we were running this on an intel system so good news you have to spend a lot of money on memory big surprises there you'll see people advertising you know 4 200 megahertz you're not gonna get there on your ryzen system out of the box you've got to do a lot of tweaking to get the f clock there and with that comes a lot of instability unless you want to be an overclocker and then that's always fun as well so recommendation out of the box turn on core enhancement leave it auto or just turn it on to verify that the guarantee that it goes 3 600 megahertz memory 1800 megahertz f clock profittoday we're going to be talking about the pinnacle of both performance and sound many of us are working and learning from home during these challenging times and micro center has what you need to create a seamless work or learn from home setup including laptops desktops monitors and more for those of you practicing social distancing but need to get your daily dose of tech microcenter has a community forum where you can discuss and share your favorite tech setups and shared bias and even get help with troubleshooting from fellow enthusiasts to learn more about how to get your essential techniques from micro center and visit the community click the links in the description below all right so a while back we did a video where we specifically showed how uh performance can be impacted with memory speeds on intel and i asked you guys at the end of that video if you want me to do one with amd since verizon was known to be a bit more picky when it came to ram speeds specifically with you know first edition generation ryzen we're in ryzen 2 now uh and a lot of you guys were like yes please do that so i think we're going to do that today i think i hope that's what we're going to do today that's what i'm set up for so i think we're going to we're going to be using windows 95 for these tests though because we want to use a a relevant operating system so first things that we're going to talk about here is just our test we're just going to use geekbench and cinebench reason for that is there are free tests again that you guys can download and try these tests yourself and we're just going to be looking for any of these score differences we're not comparing them to other cpus or any of that we just want to see with out of the box ram settings which on ddr ddr4 is always going to be 2133 megahertz for even if you buy a dim that says 4200 megahertz it's gonna run 2133 if you don't go in and enable docp profile on amd um the other thing is going to be i think we'll play with the fabric clock a little bit or the f clock because that's the other half of the equation when it comes to memory speeds speeds with amd just a quick high level here and phil can put a little graphic up to show if you guys aren't aware the amd cpu is made up of chiplets there's a memory controller on the substrate as well as the cpu chiplets now when the cpu is separated like it is which is why intel was always like glued together cpus which is kind of funny because that's kind of what intel does with their like big xeon stuff but anyway moving on when you have the cpu dies like that that are separate they've got to share the memory as well but the f clock is the speed at which the memory communicates with each other on the substrate so a very bad explanation of it but that's why if you go fast memory but if you don't have your f clock set to auto or at least proportionate to your memory speeds you can actually be losing performance because of the communication between the chiplets but moving on i need to actually install geekbench hold on i thought you're prepared for this man be prepared i am not scar i won't be prepared quit pro quo so we're gonna do testing methodology in basically three different phases if you will i'm just running out of the box settings right now with the cpu kind of fluctuating and doing its thing with its uh precision boost kind of controlling things the memory is out of the box 2133 the fabric clock is at auto and then we're going to do is we're going to lock everything down we're going to lock the cpu to 4.2 i'm just curious how much fluctuation there's going to be just with locking the cpu um and then we're going to do um obviously our memory speed at our slowest which is 2133 and then we're going to do a 3000 and then we're going to do a 3 800. this memory is 3800 modules cl15 i'm going to leave the stock timings i'm still debating whether or not i want to do a timing video because you could actually go in there and cause yourself some headaches by playing with the timings and trying to get them back to factory if sometimes clear cmos doesn't work like we were kind of dealing with before starting this video clear cmos wasn't doing crap when this thing didn't want to boop we found out it was our gpu and that's a whole different video subject yeah i find timings to be not that important unless you're going for like record overclocks and stuff but the reason why i chose 3000 is that it seems to be where a lot of the memory that's affordable today is kind of that for the 16 gigabyte dual 8 gig sticks they seem to be right around 3200 megahertz so you guys can see and then we can see what the difference is between low speed 2133 and nearly 4 000 megahertz um prediction time though because we haven't actually done this test i predict with ryzen 2 which is what the 3950x is is going to see marginal at best score differences between 3000 and 3800 i think we might see a little bit of an improvement from 2133 to 3000 but i think what we're kind of going to show you here too is that you don't need to spend all kinds of money on expensive ram which was our whole point with the intel video now the other thing we did with intel video was some gaming benchmarks which showed fps improvement but anyway here's our first test with cinebench let's see how this does so we got a 93 40 and then before that we had a 93 30 and then another one we had a 93 53 so you can see it's fairly consistent um and it scores and i mean 10 f 10 score difference here is really not something you're gonna notice it's we can run it 10 times to get 10 different numbers but they'll all be within like half a percent or a quarter of a percent of each other which is totally normal remember there are background processes happening here um all right so that was a 93 40 or 93 40. here's our baseline geekbench scores 1312 single core and uh 12 163 multi core now we're going to do only one gaming title because what we what we didn't expect with the intel system was that we got like a 30 fps improvement in like shadow of the tomb raider and far cry 5 but we saw no difference really in wildlands like one or two fps at the most we just realized though our 5700 that's sitting on here this is not an xt just 5700 is not going to bottleneck our 3950x so we're going to use the same graphics card we did in those tests which is actually the msi gaming x trio 2080 ti and we're going to set it at 1080p because obviously we want to see how memory speed is going to help the potential bottlenecking you're going to do with the cpu um so i guess i need to swap out the gpu now but let me say this fanboys this is not a 3950x gaming performance versus a what was that a 9900 i think we did with that this is not about that i think the 9900k is going to give you better gaming performance than a 3950x because we were running that thing at 5 gigahertz we all know gigahertz reigns supreme when it comes to gaming we are just looking at what memory speeds do for this particular cpu not we don't care how it performs to other cpus that's not the point of this video they were tight all right so here we go graphic settings were set to highest 1080p uh our trio is set to plus 75 megahertz on the core plus 500 on the memory fan set to 100 per 100 per spent it's only wednesday and it's been a long week okay you know what i'm trying to say all right well you know obviously we need to uh do all the tests now so i already explained to you how we're gonna do it we're gonna chart it we'll come back we'll talk about the differences and where the real sweet spot is gonna be in terms of memory speed and where to spend your money and such on on memory so it has been hours upon hours later and we have run so many different iterations of these tests now what you're gonna see on the charts um like i said we locked the core at 4.2 gigahertz and then we did 2133 and thirty eight hundred but what we saw was a a decent jump i say decent i mean we're talking not a lot of percentage-wise numbers like what four percent or something like that i think it was at the most in terms of cpu usage or cpu scores but once we went into gaming with shadow of the tomb raider we saw pretty significant increases across the board from 120 fps all the way up to like 140 fps with the 2080 ti that is overclocked so that tells us that by speeding up the memory just like we saw in the intel test increased our fps as well which is obviously important but we started thinking about this we're like okay ram and the way it works with ryzen is totally different than intel and we already explained the fabric clock now i do know for a fact that the fabric clock running at exactly half the rate of the memory speed is where you get the most performance out of it now this being a docp of 38 68 or 66 it says 3868 when you go into the bios but that means that in order to actually get your f clock at half that speed it would have to be running at 1933. now we also know that trying to go above 1800 megahertz on the f clock leads to some significant instabilities in our case a no post now most people aren't going to go in there and start tweaking the f clock and all that sort of stuff so what we did just to kind of see what would happen here because you should also never go in there and manually set your core clock for your cpu because when you do that on amd unlike intel where the boost speeds for both all core and single core are so much higher than the base clocks and such when you lock the core to 4.2 on any ryzen processor at this point although we are talking about the 3950x the single core going all the way up to 4.7 gigahertz is now 500 megahertz limited because it can't go past the 4.2 now i don't personally know if there's a way to go in there into precision boost and set the clocks individually for single core 2 core 3 core whatever like intel you can do that i'm not sure if you can do it with ryzen if you can please link down below a resource we could use so we could actually maybe get better at tuning these overclocks but when you lock into 4.2 you lose a crap ton of single core performance now you might be thinking but jay everything's multi-threaded these days but you don't understand when there's bursty single load instructions that happen all the time on various programs those are definitely benefited by that single core boost going all the way up to 4.7 in this case now what we did see was by reducing our ram or enabling docp for the timings which it's funny because the timings on this particular set of ram that we have are not the fastest it's like cl 18 19 19 39 or 38 i think when we enable docp for the timings because if you don't do that the timings are even looser when they're just out of the box but then we set the memory to 3600 and then manually set the f clock to 1800 is where we finally saw our most beneficial across-the-board user experience sure our all core was automatically boosting to 4.1 and by setting it to 4.2 we saw an increase in multi-threaded performance across the board because an extra 100 megahertz across 16 cores and 32 threads adds up but the the amount of loss you take everywhere else just isn't worth it so at the end of the day our recommendation regarding ryzen 2 processors now remember this is different than ryzen 1.5 we'll call it or zen plus architecture is what they actually refer to it as so you have zen architecture which is like the 1800x zen plus which is all the 2000 series stuff now zen 2 is the 3000 and then now the new 3000 uh xt series that's coming out plus i guess the 4000 series is still slated for this year amd and and and us we'll just say us all of our content creators we did have a meeting with amd and they wanted us to basically mention that i guess there's false rumors saying that the 4000 series has been bumped or pushed they say that's not true it's still coming this year back to what i started to say a couple minutes ago though our recommendation quite honestly is just get 3600 megahertz memory and call it a day you can still get it for a pretty decent price but if you're gonna also be running any memory 3600 or less i highly recommend going into your bios and setting your f clock to be exactly half of whatever your core speed is or your memory speed is now you can play with it and and increase it if you want but it seems to be if you keep it at a ratio of you know 0.5 of your actual ram speed it seems to give you the best performance so if you have 3000 megahertz dimms set it to 1500 you can play with it set it to 16 set it to 18 and see how it does but i believe there's diminishing returns there where it's not going to really do anything and you could be you know potentially over stressing the f clock voltage and all that stuff we're trying to push those faster clocks for no reason whatsoever um the other recommendation too to me is if your motherboard does have a core enhancement feature like all the asus boards do and i think they all do these days make sure you leave that to auto or enabled if you disable it what that's going to do is it's going to also disable the precision overboost stuff so over boost precision boost turbo boost no leave that enabled or auto because if you don't then it's not gonna boost at all it'll just sit at its base clock of like three five and nothing you'll have a bad time because it's three five is slow these days that's slow just like intel lots of gaming performance to be had 22 fps improvement in our case from 120 fps average at 1080p remember we're intentionally pushing a lot of frame rate to try and see where the bottlenecks are on the cpu um to the point to where as you can see right now we left this up on the screen an average of one oh i tried oh that's a power button the power button shouldn't be right where i'm trying to grab it there we go so up to 142 fps but check this out we finally got to the point to where our gpu was our limiting factor we couldn't get the number higher than 142 fps at our current overclock of only plus 75 on the core and plus 500 on the memory and then power limit increase all the way as far as it'll go 17 gpu bound that tells us our gpu was doing the best that it could now the weirdness here is that if we took the same gpu with the same settings and put it on an intel cpu we would get more fps and simply because of the five gigahertz that you could run on like intel cpus these days it does make a difference however i challenge anyone to tell me that they can truly see a difference in a game running 142 fps versus 150 because i think about 150 155 or something that is where we were running this on an intel system so good news you have to spend a lot of money on memory big surprises there you'll see people advertising you know 4 200 megahertz you're not gonna get there on your ryzen system out of the box you've got to do a lot of tweaking to get the f clock there and with that comes a lot of instability unless you want to be an overclocker and then that's always fun as well so recommendation out of the box turn on core enhancement leave it auto or just turn it on to verify that the guarantee that it goes 3 600 megahertz memory 1800 megahertz f clock profit\n"