Things You Have to Consider When It Comes to Adopting New Platforms

===========================================================

One crucial factor to consider when adopting new platforms is where you're coming from. Are you already using DDR4 RAM? If so, you may not be able to take advantage of the latest and greatest technologies that come with newer systems. For instance, AMD's 7000 series requires DDR5 RAM, which means you can't use it if you only have DDR4 RAM. This is a significant consideration for anyone looking to upgrade their system.

In contrast, Intel has been notorious for forcing full platform upgrades whenever you want to use its latest and greatest CPUs. However, this doesn't mean that Intel systems are completely incompatible with older motherboards. The Z690 and Z790 series, for example, support both DDR4 and DDR5 RAM, making it easier to upgrade or switch between different platforms.

AMD, on the other hand, has been pushing the boundaries of what's possible with its latest and greatest CPUs. The 12th generation AMD Ryzen 7000 series CPU is a prime example of this. With its forward-facing socket design, it requires DDR5 RAM to function properly. This means that if you're looking to use an AMD system, you'll need to make sure you have the correct type and amount of RAM.

The Value Proposition

--------------------

When it comes to choosing between Intel and AMD, there are significant differences in terms of value proposition. Historically, AMD has been known for its amazing value, offering powerful performance at a lower price point than its competitors. However, with the release of the 12th generation Ryzen 7000 series, Intel has finally caught up.

In our testing, we've seen that the 12th generation Ryzen 7000 series CPU can trade blows with Intel's latest offerings in terms of gaming performance. However, it falls behind in certain areas like Blender and Geekbench. This means that if you're looking for a system that can handle both gaming and content creation tasks, Intel may be the better choice.

On the other hand, Intel's value proposition is hard to beat when it comes to budget-conscious buyers. With CPUs starting at $589, which includes the KF model without an integrated GPU, it's clear that Intel is offering a more affordable option for those on a tight budget.

The Bottom Line

----------------

For the first time in history, we're seeing a situation where the top-tier offerings from both AMD and Intel are priced so closely together that it's almost impossible to choose between them. The 12th generation Ryzen 7000 series CPU is priced at $599, while the 13th generation Intel Core i9-13900K is available for $589.

However, our testing has shown that there's still a difference in terms of performance and features between the two platforms. If you're looking to future-proof your system or take advantage of newer technologies like PCIe Gen 5 lanes, Intel may be the better choice.

Ultimately, the decision comes down to personal preference and what features are most important to you. Whether you choose AMD or Intel, one thing is clear: the competition between these two giants has pushed both companies to offer more value and performance than ever before.

In-Depth Comparison

--------------------

One of the most interesting aspects of this comparison is the fact that we have access to a 13th generation Intel Core i9-13900K CPU for testing. This CPU is a beast, with 14 cores and 20 threads at 3.2 GHz base clock speed.

However, it's not just the CPU that matters. The motherboard also plays a critical role in determining the overall performance of the system. In our testing, we're using a lower-end Z600 series motherboard with DDR4 RAM to see how well it can handle the demands of modern gaming and content creation tasks.

The results are telling: while the 13th generation Intel Core i9-13900K is able to deliver impressive performance in most games and applications, it falls behind slightly compared to the 12th generation Ryzen 7000 series CPU. This is a significant development for AMD fans, who may have been written off as being "behind" Intel in terms of overall value.

However, our testing also showed that when it comes to load times, the 13th generation Intel Core i9-13900K is able to deliver significantly faster performance than the 12th generation Ryzen 7000 series CPU. This could be a major factor for content creators who need fast load times and quick rendering.

Sound Off: What Do You Think?

------------------------------

So, what do you think about this launch? Are you leaning towards AMD or Intel? Let us know in the comments below! We'd love to hear your thoughts on the latest developments in the world of PC hardware.

"WEBVTTKind: captionsLanguage: enso it's been a roughly a month since AMD launched at 7950x processor downward and just made a huge Splash with their IPC uptick their core count uh uptick and just the amazing performance that 7000 series is and we all said Intel better pull out one hell of a rabbit from its hat to have any chance of competing well we've got the 13900k review results and Benchmark results today I'm gonna tell you right now this might be the hardest decision for consumers to make regarding CPU in terms of performance that we have seen in like 20 years today's video is brought to you by the Jason's merch store we've got t-shirts gaming mats and mugs and all that kind of stuff so whenever you go buy our stuff we don't have to put other ads here and other annoying crap so go buy our stuff all right so up first today we are talking about the ryzen 9 uh 7950x Flagship mainstream processor for AMD versus Intel's 13900k uh 13th Gen processor from Intel it's interesting when you talk about CPU tests um especially for gamers it's when we do all this computational testing for the most part for most consumers it's not really that important to people especially if you're a gamer you're like I just what's what's the max FPS I can get with this CPU if I introduce a bottlenecking situation which is why the 5800x3d from AMD became such an amazing processor because of the uplift they were able to get in performance just with the 3D V cache available on the CPU now we're talking about 3 900k they're two very different approaches to how these CPUs are achieving the performance that they are for instance when Zen first came out AMD really adopted the whole idea of chiplet scalable chiplet design so you have your memory controller and your cache so you have your memory controller as a a separate i o controller die on the substrate and then you have scalable chiplets one or two depending on how many cores there are and that's how with the AMD 7950x you're able to get 16 cores and 32 threads of smt or simultaneous multi-threading performance on the CPU on Intel however instead of the irony though until now somewhat also using a chip lit design although it's not they're not separated on the substrate like they are with AMD they called this a glued CPU it's funny because now Intel's gluing together their own Concepts as well so we've got the uh built upon the same concept we found the 12th gen CPU which is basically having a p chord and an e-core P chores are performance they have hyper threading enabled they are full scale full size cores that can then hand off and and deal with hyper threading as they have for for Generations but now we have the e-core which is a smaller less boosting less powerful core that's able to handle a lot of the lower demanding tasks or background tasks or whatever to allow your CPU to be able to be running full speed with P cores handling tasks while e-cores are doing other things in the background for 13th gen they chose to basically just up the e-core count kind of across the board so now with 13th gen we have a 24 core but it's an 8p core 16 e-core 32 thread CPU very interesting design what's nice about it it's got 32 or 36 megabytes of uh L3 cache it's got 32 megabytes of L2 cache that's a lot remember the more cash you can get on the CPU the less CPU has to rely on RAM to get its critical task performed which means faster response lower latency although it is using ddr4 and ddr5 because 13th gen is backwards compatible with z690 motherboards for this test we chose the z790 motherboard to make sure that we were not somehow not allowing the CPU to have access to any of its technology or any of its uh chipset features that would maybe somehow hinder its performance we are also running it on a crucial ddr5 so it's important to note that to get the most out of both 12th gen and 13th gen running it on its native chipsets or chipsets designed for them so that being obviously z690 for 12th gen z790 for Intel although a no a more expensive platform to a DOT you are going to get full feature sets by doing so now for Ram ended up using the crucial fairly basic set of ram at 64 gigabytes of gddr or not gdd I'm so so much graphics on the brain ddr5 this is a 64 gigabyte kit 2 times 32 doesn't have a heatsink on it I really wanted to test this Ram because I've seen it uh really sort of talked about and I was curious as to how hot the ram modules would get with no cooling on them and to my surprise I didn't use a thermometer because I don't have a laser thermometer I've lost it I lost it um I Could Touch the chips with it under load and they weren't hot at all but anyway this is ddr5 4800 megahertz I pushed it to 5000 I was curious if I could overclock it ddr5 is where you're going to get the most performance out of um this particular setup but ddr5 is also one of the more expensive things to adopt right now when it comes to memory so that's why crucial um kind of partnering with us on this launch made this a I don't want to say a no-brainer but if you're the kind of person that wants to buy the latest and greatest but you're going to leave performance on the table by running it with ddr4 to me that doesn't make a whole lot of sense but because there are Ram options out there that aren't going to completely break the bank and was still able to overclock it I just went to 5000 to match what we did on 12th gen it could probably even go farther I just haven't personally tested it anyway it is running Micron chips on there it's a cl40 that's 1.1 volt so that's probably one of the reasons why it's also not that hot because it's at 1.1 volts so some of the RAM for ddr5 goes up to 1.3 which could make it very warm but anyway we'll put a link to the ram down below if you guys are looking for ddr5 setups and you aren't looking to break the bank with like crazy RGB setups and stuff when it comes to Ram but you guys can find that there's actually quite a value when it comes to not going with super flashy on the ramp but I do recommend ddr5 if you're adopting 13th January even 12th Jan if you can afford it AMD on the other hand requires ddr5 there's no backwards compatibility there the am5 socket all that sort of stuff but a 13th gen will run in a z690 motherboard so let's say you are running 12th gen and you want to step up to 13th gen you can do that so I just wanted to sort of point that out also too there are two versions of the 3900k there's the K and the KF that's a that's a SKU that has existed now for quite a few Generations since 9900k 900 KF I believe the KF basically just removes the Intel UHD Graphics the 770 Graphics from the die so you save 25 bucks by taking that out I personally feel it's worth it to have it if you aren't extremely tight on your budget because it does come in handy having an igpu to be able to handle some of your troubleshooting tasks if you're looking at having a graphics card that's maybe not turning on or something you can troubleshoot the issue or if you're doing any sort of like Premiere um edit video editing or whatever because having quick sync available to you is also nice to have which in my opinion is worth the 25 bucks alone all right enough chatting let's just go ahead let's put the benchmarks up there let's just put it all out on the table and see how these two now that they have finally met Head to Head how they really compare foreign thank you okay so starting off with cinebench R23 multi-core um 38 714. this is R23 like I mentioned you I want to point out all these tests that we use with the exception of Forza Horizon 5 for bottlenecking is free tests that you can download and compare yourself there's a million different ways you can do CPU tests so I chose to use these free synthetic benchmarks that give us very very solid a b testing they're also free like I said you can go and download your own if you want to see how much uplift you would have with your current system versus this download it run it compare it also two of these are obviously fresh os's with no background tasks all updates performed to make sure that all of the overhead to the CPU is available without OS involved getting involved and slowing anything down so a 38714 and R23 multi-core versus a 38 051 on AMD now the reason for this this test scales very heavily with core speed also none of these tests are very Ram intensive at all so you'll notice as a faster ddr5 available to the X program that we were sent am5 we're only testing expo at this moment so all the Expo RAM available to it is is 6 000 megahertz that we've received so that's why you're seeing a one gigahertz differential between Intel and AMD I can tell you right now I did also test it with 6000 megahertz RAM on here and there was Zero difference whatsoever in any of these tests even the bottlenecking 10 us to the CPU by running that level of overhead that's because these particular tests are not they're not Ram intensive they are straight up CPU core intensive but R23 scales very well with clock speed now the all-core overclock with turbo velocity boost and all limits enabled you'll notice this is the board we used right here this is Rog strix z790 Asus is still sending their motherboard's default Ram bio settings to Auto enhance we don't want Auto enhance features on when we're doing a b testing because that means ASU starts using its Logic on how the CPU should perform not Intel's so we go in there and we disable all of the auto overclocking all of the Auto Limit removal and all that we just go enforce all Intel limits all turbo timers all terminal velocity boost algorithms all of that is enabled by the way so important to note that note that if you're doing your own comparison on Easter's motherboards turn those features off if you want to see the way Intel intended for the CPU to perform so because of that the intel 13900k was boosting up to 5.5 gigahertz all core for about 5 to 10 seconds and then starts to come down and will re-land is a 5.3 gigahertz all core on the p-core and it goes all the way up to 4.3 gigahertz on the e-cores dropping down to a 4.1 gigahertz e-core once turbo velocity boost or thermal velocity boost has subsided and it's gone back to its limits because of the fact that when they boost up as high as they did the up to 5.5 and up to 5.3 which we got for a moment uh the temps go up with it so 13th gen without the limits gets very hot we'll talk about that in a moment but those core clocks are higher than you get with AMD 7950x so AMD 7950x lands at about a 5.1 gigahertz all-core average because remember these are all P cores essentially inside the AMD CPU and so up to 5.3 or 5.2 all core you'll find one CCD lands at about 5.05 or 5 000 50 megahertz and then the other CCD like ccd1 or the better CCD will be like 51.50 so then the average is about 5.1 you're not getting the full 5.2 just like with Intel it comes down slightly uh but the huge uplift of the 3.7 3.8 e-core frequency up to 4.2 and then having double that's where you're getting this performance uplift versus 12th gen 12th gen 27 566. it's like over 10 000 points higher again eight more e-cores and 500 megahertz higher on average with those e-cors that's where we're getting the major uplift here in performance so cinebench R23 the win obviously goes to Intel on that one if we talk about single core though 2259 on the 13900 versus a 2013 on the 7950x the reason for that is just clock speed the Intel went to 5.8 gigahertz and stayed there the AMD goes up to 5.8 but then we see it sort of fluctuate five six five eight five six five seven five six five eight and it hands off the CPU a lot the task whereas Intel kind of just went back and forth between two cores and sort of stayed there so having an average higher frequency and much better I don't know how but much better IPC on 13th gen versus 12th gen is how we were able to get this clock uh speed lift now cinebench has always been pretty good at just really utilizing the ABX instruction sets available on Intel remember what we found with the 7950x is instead of running an AVX 512 it runs AVX 256 twice so it's basically running avx256 per CCD whereas Intel is just running the true ATX 512 so that's where you're noticing a lot of that performance differential on cinebench so that's just the way the test is designed geekbench uh that single core Performance Wing actually went to the 7950x by about 40 points really what you should take away from this is they are trading blows depending on the test 12th gen it's just look it's like hey guys wait for me and 12th gen was actually pretty fast the multi-core though once again went to Intel and that's because of the overall higher clocks um on average now blender Benchmark is essentially a ray tracing test to be honest as most of these type of renderings are though but it's just it's just Ray tracing light Ray uh light samples and so the more samples the better the 7950x just handedly beat the 13th gen um 13th gen as you can compare versus 12th gen though quite a ways ahead 297 uh versus 277. so about 10 up on the AMD system same thing with junk shop uh 178 versus a 163 in terms of uh scene samples classroom once again 139 versus 128. so on average it's just faster Port 1210 down there going don't forget about me I'm only a half gen old I came out within a year don't forget about me okay 3D Mark time spy extreme this is just a CPT best um very close although Intel edged it out once again it's about it being able to just overall on average keep higher clock speeds now this is because IPC on these two CPUs are so close it's just about who can keep the average clock speed higher longer and you would think those e chords would be sort of a detriment to Intel but if the application is very multi-threaded it can definitely make use of those e-cores so a 13258 versus a 13029 12th gen's 9272 going wait a minute remember the 12th gen though is a 24 thread CPU so it is it is eight threads behind again E chord three is we basically added a whole e-core CPU to the 13th chance essentially what happened um IPC test though a 1978 versus 1806 on the 7950x I know I said a moment ago that they were higher that they were closer together I was actually referring to the 12th gen in my brain going they were close but it's actually quite a bit ahead of AMD 1806 versus a 1978 of the CPUs locked at five gigahertz so what we do to do our IPC test if you're wondering how we do this is we go in and make sure all limits are enabled but we lock the CPUs to five gigahertz this is the first time I've ever been able to do that with AMD because even single core we were having a hard time getting five gigahertz stable on ryzen 5000 series was able to do it but three thousand two thousand one thousand was nowhere near it um they were barely like four four two and then we run the IP the the single core thread in uh send a bench and then we compare the score so at five gigahertz locked the Intel cpu's IPC is fairly significantly higher than amds so by able being able to keep the average clock speed higher on Intel plus having better IPC is why you're seeing in those very multi-threaded tests um a win for Intel handbrake transcode that's where that really just shows 79 seconds versus 84 and then the 12th gen as you can see is is the slowest at 97. if we take a look now at our bottleneck test this is where gamers are gonna care average GPU this is Forza Horizon 5 which gives us an average GPU an average CPU render and an average CPU simulation so the CPU has to handle multiple things it has to run the simulator the simulation it has to also have CPU rendering when it comes to the things that it's rendering in the scene and then the GPU is actually handling all of the post-processing and all that stuff so the higher the FPS from the GPU the harder it is on the CPU to draw those frames in a consistent frame rate or consistent frame pacing I should say the amount of milliseconds between each frame draw now we use the 30 30 90 TI I would have used the 490 for the initial test for AMD but the problem was that 4D series wasn't launched yet so doing this with a 40 90 would show even bigger Gap probably but if we take a look at all all of it CPU render and GPU render it is slightly faster on Intel than it is on AMD and that's again the IPC and the clock speed average they're not different enough in this particular test though for me to say oh if you don't go with an Intel CPU you're totally bottlenecking your GPU which is something I would have said in the past like 3000 Series 2000 Series ryzen today at least this particular bottlenecking test which we run a 1080p very low all settings down to low to run as much frames as we can I would tell you that you could go with either of these generations of CPU 7000 series or 13th gen and have an amazing gaming experience in fact if you're on a 5800 x3d right now and all you care about is gaming then 7000 series wouldn't even be that interesting to you yet until the new v3dc uh V3 v3d cache CPUs come out then it would matter but fun fact I have the 522.25 driver on there which is the latest driver which gives huge dx12 uplift for NVIDIA like across the board so I had to go back and uninstall it because I forgot I did Auto install I had to run back to 517.43 which was the same driver I ran on the AMD test and the 12th gen test and everything came down on the CPU or the the GPU average in fact my GPU average was like 409 from 365 to 409. so the big story here that we should be talking about is if you're not running 522.25 and you have an in a Nvidia GPU you're leaving pre-performance on the table now there's another thing we have to talk about here and that's going to be power consumption and temperatures one of the things that we obviously talked about with the 7 thousand series is the fact that 95 Celsius is The New Normal we also have a video that we talked about where we show that you could actually do a lot of things to bring that temperature down but it revolves a boarding your warranty and filing down the IHS and a whole bunch of things with the out of the box settings uh not touching anything leaving the limits enabled remember I talked about the fact that the motherboards with Asus have you know enhanced optimizations and stuff that if you enable those they allow the thermal velocity boost to go longer and such which will affect the temperatures but with all of the Intel logic in place all of the Intel limits and controls the hottest I saw the CPU get was 82c now that's on a 360 millimeter AIO the exact same 360 millimeter AIO I used on a 7000 series so I moved it from one platform to the other so I had the identical cooling solution to see how well it was able to handle it now what's interesting about that is the fact that it's a I saw a max power draw of 253 Watts which is exactly what it shows on the spec sheet the average wattage on the 7950x was hovering right around 200-ish Watts with a 225 watt Max TDP it wasn't pulling the full 225 available to it whereas the intel was going to the full 253 Watts but staying at around 82c however the moment you lift those limits it went right to 96 C so I didn't see any throttling which is interesting about that because of the fact just like on the 7950x 95c it would still run you know nearly full clock speeds on the Intel system it was at 95 C and it would Peak 96 come back down to 95 I was still getting the full 5.5 gigahertz all core and 4.3 gigahertz on the e-course now that's with the AI optimized etc etc happening on the motherboard so it is a slight overclock uh well actually it's not a slide overclock with a slide over clock I was able to get that even higher but yeah with the 360 AIO we were still seeing deep into the 90s once all those limits are lifted so to kind of put it in perspective with the thermal velocity boost and such it goes to those clock speeds but then it comes down like I said to like four one all core e-core and five three all core P core so cooling this time around on both sides Winter's coming up so hey your space heaters are already there this is the part where I talk about my recommendations on CPU there's a couple of things you have to consider when it comes to adopting these platforms one where are you coming from do you already have ddr4 Ram you want to utilize well then you can't go AMD period AMD 7000 series it's a forward-facing socket which that what that means is it's the first generation of am5 that was requiring ddr5 so there's no other CPUs no other Ram you can use it must be a 7000 series and it must be ddr5 Intel on the other hand which is notorious for forcing full platform upgrades whenever you want to use it is still backwards compatible with a previous generation motherboard the z690 and both of those and the Z6 790 available now as well with just basically uh pcie Gen 5 Lanes direct to the CPU for faster storage and stuff you can run it on a z690 motherboard which you can find for cheaper than the z790 and ddr4 so you're not going to get the full potential however you could upgrade that platform later if you get the CPU now or if you already have that platform and you just want the CPU upgrade after seeing what 12 900k versus 13900 is the other thing you have to consider the fact that the 12 or the 7950x is a 700 CPU right now at the time of filming this 699.99 on Amazon Micro Center all that but on Intel the 13900k is 589 and the KF which takes away the igpu is 25 left at five less at 564 dollars so as you can see on these tests this is sort of a problem for AMD at the moment you have a CPU that in our particular benchmarks and please watch as many Benchmark reviews as you can because they're all going to be different methodologies different approaches different tests that you can use to extrapolate all the data and not knock over your motherboard extrapolate all your data and determine which GPU is right for you if we are looking at this purely from a budget standpoint the Intel is pretty much the no-brainer because it's trading blows it's beating the the 7950x in this particular Suite enter of games and and stuff and loses to blender and geekbench that's it it's cheaper by 130 minimum or 120 minimum it's cheaper when it comes to Ram if you're truly on a budget and it's cheaper when it comes to motherboards because it's backwards compatible so for the first time the value is Intel what a weird weird world we're living in right now I'm super conflicted because I I'm the kind of guy that goes you know if I'm going to build a just a badass rig I go with what's the top out there I guess I'd have to wait for the the newest threadripper but that thread repair's never been great at gaming and that's pretty much all I do with my systems at home and stuff but you can see now the 12th gen uh it's just it's Intel made it so obsolete so fast in terms of performance but the budget value goes to Intel because their top CPU right now is 589 and we'll be testing the 13 600k because we have one of those as well in fact it is you know great there that's 319 and that is a 14 core 20 thread CPU 6p cores 80 cores that's where we're gonna have fun checking out to see what happens we put one of those on a lower end z600 series motherboard with ddr4 to see what happens there so anyway the bottom line which you should take away from this right now is for the first time if you take the 120 differential until being cheaper by that much for the top tier offerings for the first time you can really go with whichever brand you prefer it comes out of business practices or just trust in the stability of the platform and adoption rate and just things working you can go with whichever brand your happiest with and not feel like you're leaving something on the table massive back in the day it was like oh especially with FX AMD is an amazing value but you're not going to get the greatest you know load times you're gonna get the greatest performance with your games and stuff especially if the game uses you know a lot of CPU for World rendering and stuff now let's go with whichever platform you prefer and you're gonna have a great experience on it no matter what I'm curious how many of you would pick Intel because of the fact that it's cheaper overall it has cheaper options available with random motherboard or AMD because of the fact that it's not Intel I find more people buy AMD because it's not Intel than actually being huge fans of AMD anyway sound off down below how you feel about this launch what you would buy and why thanks for watching guys and as always we'll see in the next oneso it's been a roughly a month since AMD launched at 7950x processor downward and just made a huge Splash with their IPC uptick their core count uh uptick and just the amazing performance that 7000 series is and we all said Intel better pull out one hell of a rabbit from its hat to have any chance of competing well we've got the 13900k review results and Benchmark results today I'm gonna tell you right now this might be the hardest decision for consumers to make regarding CPU in terms of performance that we have seen in like 20 years today's video is brought to you by the Jason's merch store we've got t-shirts gaming mats and mugs and all that kind of stuff so whenever you go buy our stuff we don't have to put other ads here and other annoying crap so go buy our stuff all right so up first today we are talking about the ryzen 9 uh 7950x Flagship mainstream processor for AMD versus Intel's 13900k uh 13th Gen processor from Intel it's interesting when you talk about CPU tests um especially for gamers it's when we do all this computational testing for the most part for most consumers it's not really that important to people especially if you're a gamer you're like I just what's what's the max FPS I can get with this CPU if I introduce a bottlenecking situation which is why the 5800x3d from AMD became such an amazing processor because of the uplift they were able to get in performance just with the 3D V cache available on the CPU now we're talking about 3 900k they're two very different approaches to how these CPUs are achieving the performance that they are for instance when Zen first came out AMD really adopted the whole idea of chiplet scalable chiplet design so you have your memory controller and your cache so you have your memory controller as a a separate i o controller die on the substrate and then you have scalable chiplets one or two depending on how many cores there are and that's how with the AMD 7950x you're able to get 16 cores and 32 threads of smt or simultaneous multi-threading performance on the CPU on Intel however instead of the irony though until now somewhat also using a chip lit design although it's not they're not separated on the substrate like they are with AMD they called this a glued CPU it's funny because now Intel's gluing together their own Concepts as well so we've got the uh built upon the same concept we found the 12th gen CPU which is basically having a p chord and an e-core P chores are performance they have hyper threading enabled they are full scale full size cores that can then hand off and and deal with hyper threading as they have for for Generations but now we have the e-core which is a smaller less boosting less powerful core that's able to handle a lot of the lower demanding tasks or background tasks or whatever to allow your CPU to be able to be running full speed with P cores handling tasks while e-cores are doing other things in the background for 13th gen they chose to basically just up the e-core count kind of across the board so now with 13th gen we have a 24 core but it's an 8p core 16 e-core 32 thread CPU very interesting design what's nice about it it's got 32 or 36 megabytes of uh L3 cache it's got 32 megabytes of L2 cache that's a lot remember the more cash you can get on the CPU the less CPU has to rely on RAM to get its critical task performed which means faster response lower latency although it is using ddr4 and ddr5 because 13th gen is backwards compatible with z690 motherboards for this test we chose the z790 motherboard to make sure that we were not somehow not allowing the CPU to have access to any of its technology or any of its uh chipset features that would maybe somehow hinder its performance we are also running it on a crucial ddr5 so it's important to note that to get the most out of both 12th gen and 13th gen running it on its native chipsets or chipsets designed for them so that being obviously z690 for 12th gen z790 for Intel although a no a more expensive platform to a DOT you are going to get full feature sets by doing so now for Ram ended up using the crucial fairly basic set of ram at 64 gigabytes of gddr or not gdd I'm so so much graphics on the brain ddr5 this is a 64 gigabyte kit 2 times 32 doesn't have a heatsink on it I really wanted to test this Ram because I've seen it uh really sort of talked about and I was curious as to how hot the ram modules would get with no cooling on them and to my surprise I didn't use a thermometer because I don't have a laser thermometer I've lost it I lost it um I Could Touch the chips with it under load and they weren't hot at all but anyway this is ddr5 4800 megahertz I pushed it to 5000 I was curious if I could overclock it ddr5 is where you're going to get the most performance out of um this particular setup but ddr5 is also one of the more expensive things to adopt right now when it comes to memory so that's why crucial um kind of partnering with us on this launch made this a I don't want to say a no-brainer but if you're the kind of person that wants to buy the latest and greatest but you're going to leave performance on the table by running it with ddr4 to me that doesn't make a whole lot of sense but because there are Ram options out there that aren't going to completely break the bank and was still able to overclock it I just went to 5000 to match what we did on 12th gen it could probably even go farther I just haven't personally tested it anyway it is running Micron chips on there it's a cl40 that's 1.1 volt so that's probably one of the reasons why it's also not that hot because it's at 1.1 volts so some of the RAM for ddr5 goes up to 1.3 which could make it very warm but anyway we'll put a link to the ram down below if you guys are looking for ddr5 setups and you aren't looking to break the bank with like crazy RGB setups and stuff when it comes to Ram but you guys can find that there's actually quite a value when it comes to not going with super flashy on the ramp but I do recommend ddr5 if you're adopting 13th January even 12th Jan if you can afford it AMD on the other hand requires ddr5 there's no backwards compatibility there the am5 socket all that sort of stuff but a 13th gen will run in a z690 motherboard so let's say you are running 12th gen and you want to step up to 13th gen you can do that so I just wanted to sort of point that out also too there are two versions of the 3900k there's the K and the KF that's a that's a SKU that has existed now for quite a few Generations since 9900k 900 KF I believe the KF basically just removes the Intel UHD Graphics the 770 Graphics from the die so you save 25 bucks by taking that out I personally feel it's worth it to have it if you aren't extremely tight on your budget because it does come in handy having an igpu to be able to handle some of your troubleshooting tasks if you're looking at having a graphics card that's maybe not turning on or something you can troubleshoot the issue or if you're doing any sort of like Premiere um edit video editing or whatever because having quick sync available to you is also nice to have which in my opinion is worth the 25 bucks alone all right enough chatting let's just go ahead let's put the benchmarks up there let's just put it all out on the table and see how these two now that they have finally met Head to Head how they really compare foreign thank you okay so starting off with cinebench R23 multi-core um 38 714. this is R23 like I mentioned you I want to point out all these tests that we use with the exception of Forza Horizon 5 for bottlenecking is free tests that you can download and compare yourself there's a million different ways you can do CPU tests so I chose to use these free synthetic benchmarks that give us very very solid a b testing they're also free like I said you can go and download your own if you want to see how much uplift you would have with your current system versus this download it run it compare it also two of these are obviously fresh os's with no background tasks all updates performed to make sure that all of the overhead to the CPU is available without OS involved getting involved and slowing anything down so a 38714 and R23 multi-core versus a 38 051 on AMD now the reason for this this test scales very heavily with core speed also none of these tests are very Ram intensive at all so you'll notice as a faster ddr5 available to the X program that we were sent am5 we're only testing expo at this moment so all the Expo RAM available to it is is 6 000 megahertz that we've received so that's why you're seeing a one gigahertz differential between Intel and AMD I can tell you right now I did also test it with 6000 megahertz RAM on here and there was Zero difference whatsoever in any of these tests even the bottlenecking 10 us to the CPU by running that level of overhead that's because these particular tests are not they're not Ram intensive they are straight up CPU core intensive but R23 scales very well with clock speed now the all-core overclock with turbo velocity boost and all limits enabled you'll notice this is the board we used right here this is Rog strix z790 Asus is still sending their motherboard's default Ram bio settings to Auto enhance we don't want Auto enhance features on when we're doing a b testing because that means ASU starts using its Logic on how the CPU should perform not Intel's so we go in there and we disable all of the auto overclocking all of the Auto Limit removal and all that we just go enforce all Intel limits all turbo timers all terminal velocity boost algorithms all of that is enabled by the way so important to note that note that if you're doing your own comparison on Easter's motherboards turn those features off if you want to see the way Intel intended for the CPU to perform so because of that the intel 13900k was boosting up to 5.5 gigahertz all core for about 5 to 10 seconds and then starts to come down and will re-land is a 5.3 gigahertz all core on the p-core and it goes all the way up to 4.3 gigahertz on the e-cores dropping down to a 4.1 gigahertz e-core once turbo velocity boost or thermal velocity boost has subsided and it's gone back to its limits because of the fact that when they boost up as high as they did the up to 5.5 and up to 5.3 which we got for a moment uh the temps go up with it so 13th gen without the limits gets very hot we'll talk about that in a moment but those core clocks are higher than you get with AMD 7950x so AMD 7950x lands at about a 5.1 gigahertz all-core average because remember these are all P cores essentially inside the AMD CPU and so up to 5.3 or 5.2 all core you'll find one CCD lands at about 5.05 or 5 000 50 megahertz and then the other CCD like ccd1 or the better CCD will be like 51.50 so then the average is about 5.1 you're not getting the full 5.2 just like with Intel it comes down slightly uh but the huge uplift of the 3.7 3.8 e-core frequency up to 4.2 and then having double that's where you're getting this performance uplift versus 12th gen 12th gen 27 566. it's like over 10 000 points higher again eight more e-cores and 500 megahertz higher on average with those e-cors that's where we're getting the major uplift here in performance so cinebench R23 the win obviously goes to Intel on that one if we talk about single core though 2259 on the 13900 versus a 2013 on the 7950x the reason for that is just clock speed the Intel went to 5.8 gigahertz and stayed there the AMD goes up to 5.8 but then we see it sort of fluctuate five six five eight five six five seven five six five eight and it hands off the CPU a lot the task whereas Intel kind of just went back and forth between two cores and sort of stayed there so having an average higher frequency and much better I don't know how but much better IPC on 13th gen versus 12th gen is how we were able to get this clock uh speed lift now cinebench has always been pretty good at just really utilizing the ABX instruction sets available on Intel remember what we found with the 7950x is instead of running an AVX 512 it runs AVX 256 twice so it's basically running avx256 per CCD whereas Intel is just running the true ATX 512 so that's where you're noticing a lot of that performance differential on cinebench so that's just the way the test is designed geekbench uh that single core Performance Wing actually went to the 7950x by about 40 points really what you should take away from this is they are trading blows depending on the test 12th gen it's just look it's like hey guys wait for me and 12th gen was actually pretty fast the multi-core though once again went to Intel and that's because of the overall higher clocks um on average now blender Benchmark is essentially a ray tracing test to be honest as most of these type of renderings are though but it's just it's just Ray tracing light Ray uh light samples and so the more samples the better the 7950x just handedly beat the 13th gen um 13th gen as you can compare versus 12th gen though quite a ways ahead 297 uh versus 277. so about 10 up on the AMD system same thing with junk shop uh 178 versus a 163 in terms of uh scene samples classroom once again 139 versus 128. so on average it's just faster Port 1210 down there going don't forget about me I'm only a half gen old I came out within a year don't forget about me okay 3D Mark time spy extreme this is just a CPT best um very close although Intel edged it out once again it's about it being able to just overall on average keep higher clock speeds now this is because IPC on these two CPUs are so close it's just about who can keep the average clock speed higher longer and you would think those e chords would be sort of a detriment to Intel but if the application is very multi-threaded it can definitely make use of those e-cores so a 13258 versus a 13029 12th gen's 9272 going wait a minute remember the 12th gen though is a 24 thread CPU so it is it is eight threads behind again E chord three is we basically added a whole e-core CPU to the 13th chance essentially what happened um IPC test though a 1978 versus 1806 on the 7950x I know I said a moment ago that they were higher that they were closer together I was actually referring to the 12th gen in my brain going they were close but it's actually quite a bit ahead of AMD 1806 versus a 1978 of the CPUs locked at five gigahertz so what we do to do our IPC test if you're wondering how we do this is we go in and make sure all limits are enabled but we lock the CPUs to five gigahertz this is the first time I've ever been able to do that with AMD because even single core we were having a hard time getting five gigahertz stable on ryzen 5000 series was able to do it but three thousand two thousand one thousand was nowhere near it um they were barely like four four two and then we run the IP the the single core thread in uh send a bench and then we compare the score so at five gigahertz locked the Intel cpu's IPC is fairly significantly higher than amds so by able being able to keep the average clock speed higher on Intel plus having better IPC is why you're seeing in those very multi-threaded tests um a win for Intel handbrake transcode that's where that really just shows 79 seconds versus 84 and then the 12th gen as you can see is is the slowest at 97. if we take a look now at our bottleneck test this is where gamers are gonna care average GPU this is Forza Horizon 5 which gives us an average GPU an average CPU render and an average CPU simulation so the CPU has to handle multiple things it has to run the simulator the simulation it has to also have CPU rendering when it comes to the things that it's rendering in the scene and then the GPU is actually handling all of the post-processing and all that stuff so the higher the FPS from the GPU the harder it is on the CPU to draw those frames in a consistent frame rate or consistent frame pacing I should say the amount of milliseconds between each frame draw now we use the 30 30 90 TI I would have used the 490 for the initial test for AMD but the problem was that 4D series wasn't launched yet so doing this with a 40 90 would show even bigger Gap probably but if we take a look at all all of it CPU render and GPU render it is slightly faster on Intel than it is on AMD and that's again the IPC and the clock speed average they're not different enough in this particular test though for me to say oh if you don't go with an Intel CPU you're totally bottlenecking your GPU which is something I would have said in the past like 3000 Series 2000 Series ryzen today at least this particular bottlenecking test which we run a 1080p very low all settings down to low to run as much frames as we can I would tell you that you could go with either of these generations of CPU 7000 series or 13th gen and have an amazing gaming experience in fact if you're on a 5800 x3d right now and all you care about is gaming then 7000 series wouldn't even be that interesting to you yet until the new v3dc uh V3 v3d cache CPUs come out then it would matter but fun fact I have the 522.25 driver on there which is the latest driver which gives huge dx12 uplift for NVIDIA like across the board so I had to go back and uninstall it because I forgot I did Auto install I had to run back to 517.43 which was the same driver I ran on the AMD test and the 12th gen test and everything came down on the CPU or the the GPU average in fact my GPU average was like 409 from 365 to 409. so the big story here that we should be talking about is if you're not running 522.25 and you have an in a Nvidia GPU you're leaving pre-performance on the table now there's another thing we have to talk about here and that's going to be power consumption and temperatures one of the things that we obviously talked about with the 7 thousand series is the fact that 95 Celsius is The New Normal we also have a video that we talked about where we show that you could actually do a lot of things to bring that temperature down but it revolves a boarding your warranty and filing down the IHS and a whole bunch of things with the out of the box settings uh not touching anything leaving the limits enabled remember I talked about the fact that the motherboards with Asus have you know enhanced optimizations and stuff that if you enable those they allow the thermal velocity boost to go longer and such which will affect the temperatures but with all of the Intel logic in place all of the Intel limits and controls the hottest I saw the CPU get was 82c now that's on a 360 millimeter AIO the exact same 360 millimeter AIO I used on a 7000 series so I moved it from one platform to the other so I had the identical cooling solution to see how well it was able to handle it now what's interesting about that is the fact that it's a I saw a max power draw of 253 Watts which is exactly what it shows on the spec sheet the average wattage on the 7950x was hovering right around 200-ish Watts with a 225 watt Max TDP it wasn't pulling the full 225 available to it whereas the intel was going to the full 253 Watts but staying at around 82c however the moment you lift those limits it went right to 96 C so I didn't see any throttling which is interesting about that because of the fact just like on the 7950x 95c it would still run you know nearly full clock speeds on the Intel system it was at 95 C and it would Peak 96 come back down to 95 I was still getting the full 5.5 gigahertz all core and 4.3 gigahertz on the e-course now that's with the AI optimized etc etc happening on the motherboard so it is a slight overclock uh well actually it's not a slide overclock with a slide over clock I was able to get that even higher but yeah with the 360 AIO we were still seeing deep into the 90s once all those limits are lifted so to kind of put it in perspective with the thermal velocity boost and such it goes to those clock speeds but then it comes down like I said to like four one all core e-core and five three all core P core so cooling this time around on both sides Winter's coming up so hey your space heaters are already there this is the part where I talk about my recommendations on CPU there's a couple of things you have to consider when it comes to adopting these platforms one where are you coming from do you already have ddr4 Ram you want to utilize well then you can't go AMD period AMD 7000 series it's a forward-facing socket which that what that means is it's the first generation of am5 that was requiring ddr5 so there's no other CPUs no other Ram you can use it must be a 7000 series and it must be ddr5 Intel on the other hand which is notorious for forcing full platform upgrades whenever you want to use it is still backwards compatible with a previous generation motherboard the z690 and both of those and the Z6 790 available now as well with just basically uh pcie Gen 5 Lanes direct to the CPU for faster storage and stuff you can run it on a z690 motherboard which you can find for cheaper than the z790 and ddr4 so you're not going to get the full potential however you could upgrade that platform later if you get the CPU now or if you already have that platform and you just want the CPU upgrade after seeing what 12 900k versus 13900 is the other thing you have to consider the fact that the 12 or the 7950x is a 700 CPU right now at the time of filming this 699.99 on Amazon Micro Center all that but on Intel the 13900k is 589 and the KF which takes away the igpu is 25 left at five less at 564 dollars so as you can see on these tests this is sort of a problem for AMD at the moment you have a CPU that in our particular benchmarks and please watch as many Benchmark reviews as you can because they're all going to be different methodologies different approaches different tests that you can use to extrapolate all the data and not knock over your motherboard extrapolate all your data and determine which GPU is right for you if we are looking at this purely from a budget standpoint the Intel is pretty much the no-brainer because it's trading blows it's beating the the 7950x in this particular Suite enter of games and and stuff and loses to blender and geekbench that's it it's cheaper by 130 minimum or 120 minimum it's cheaper when it comes to Ram if you're truly on a budget and it's cheaper when it comes to motherboards because it's backwards compatible so for the first time the value is Intel what a weird weird world we're living in right now I'm super conflicted because I I'm the kind of guy that goes you know if I'm going to build a just a badass rig I go with what's the top out there I guess I'd have to wait for the the newest threadripper but that thread repair's never been great at gaming and that's pretty much all I do with my systems at home and stuff but you can see now the 12th gen uh it's just it's Intel made it so obsolete so fast in terms of performance but the budget value goes to Intel because their top CPU right now is 589 and we'll be testing the 13 600k because we have one of those as well in fact it is you know great there that's 319 and that is a 14 core 20 thread CPU 6p cores 80 cores that's where we're gonna have fun checking out to see what happens we put one of those on a lower end z600 series motherboard with ddr4 to see what happens there so anyway the bottom line which you should take away from this right now is for the first time if you take the 120 differential until being cheaper by that much for the top tier offerings for the first time you can really go with whichever brand you prefer it comes out of business practices or just trust in the stability of the platform and adoption rate and just things working you can go with whichever brand your happiest with and not feel like you're leaving something on the table massive back in the day it was like oh especially with FX AMD is an amazing value but you're not going to get the greatest you know load times you're gonna get the greatest performance with your games and stuff especially if the game uses you know a lot of CPU for World rendering and stuff now let's go with whichever platform you prefer and you're gonna have a great experience on it no matter what I'm curious how many of you would pick Intel because of the fact that it's cheaper overall it has cheaper options available with random motherboard or AMD because of the fact that it's not Intel I find more people buy AMD because it's not Intel than actually being huge fans of AMD anyway sound off down below how you feel about this launch what you would buy and why thanks for watching guys and as always we'll see in the next one\n"