{"video_id":"fp_GgNNhpklDF","title":"MA: Comparing PC Parts to Apple’s M3 - Part 1","channel":"Mac Address","show":"Mac Address","published_at":"2024-01-27T10:24:00.018Z","duration_s":599,"segments":[{"start_s":0.0,"end_s":7.64,"text":"I've always had a problem comparing Apple Silicon to PC parts because gaming wasn't","speaker":null,"is_sponsor":0},{"start_s":7.64,"end_s":15.24,"text":"ready. I've always wanted to know how Apple Silicon compares to off-the-shelf PC GPUs and CPUs.","speaker":null,"is_sponsor":0},{"start_s":15.24,"end_s":19.36,"text":"And that's not something we've been able to test because Macs and PCs now use completely","speaker":null,"is_sponsor":0},{"start_s":19.36,"end_s":25.48,"text":"different architectures, and there were no games that ran natively on both.","speaker":null,"is_sponsor":0},{"start_s":25.48,"end_s":33.64,"text":"But that's changed. Check this out. I'm playing Baldur's Gate 3, one of the hottest video games right now, on a Mac.","speaker":null,"is_sponsor":0},{"start_s":33.64,"end_s":40.36,"text":"And it's running natively on Apple Silicon. There's now a growing list of games that support both Apple's graphics hardware and","speaker":null,"is_sponsor":0},{"start_s":40.36,"end_s":44.2,"text":"metal graphics APIs without any sort of translation layer.","speaker":null,"is_sponsor":0},{"start_s":44.2,"end_s":48.76,"text":"So with the help of LMG Labs, we're going to finally compare Apple chips with PC graphics","speaker":null,"is_sponsor":0},{"start_s":48.76,"end_s":54.68,"text":"cards and see what GPUs from NVIDIA and AMD compete with Apple's M series.","speaker":null,"is_sponsor":0},{"start_s":54.68,"end_s":60.16,"text":"Though not in this video. That's part two because in part one, we need to build the test benches that the GPUs will","speaker":null,"is_sponsor":0},{"start_s":60.16,"end_s":65.4,"text":"be on. And those test benches have CPUs, so we're going to figure out what CPUs are the closest","speaker":null,"is_sponsor":0},{"start_s":65.4,"end_s":73.84,"text":"to Apple Silicon. And I need to ask Labs if they'll help me.","speaker":null,"is_sponsor":0},{"start_s":73.84,"end_s":76.84,"text":"What did I think? Oh my god. Well, it's going to be interesting.","speaker":null,"is_sponsor":0},{"start_s":76.84,"end_s":82.8,"text":"It's going to be a learning experience. We haven't tested Macs really all that much, so this is going to be the first time Labs","speaker":null,"is_sponsor":0},{"start_s":82.8,"end_s":87.04,"text":"really tackles Macs. Nicholas, hey.","speaker":null,"is_sponsor":0},{"start_s":87.04,"end_s":90.52,"text":"This is Nicholas Harris. He's LTT Labs software developer.","speaker":null,"is_sponsor":0},{"start_s":90.52,"end_s":96.88,"text":"Part of his job has been developing and automating the tests for PCs and other parts here.","speaker":null,"is_sponsor":0},{"start_s":96.88,"end_s":102.96,"text":"The first step is figuring out the tests. And so Nicholas worked with test technician John Deuren to figure out how to measure these","speaker":null,"is_sponsor":0},{"start_s":102.96,"end_s":111.32,"text":"two completely different computers. Our primary goal was to find tests that can natively work on either system to avoid the","speaker":null,"is_sponsor":0},{"start_s":111.32,"end_s":115.84,"text":"Rosetta layer, because that's another variable that we want to isolate for.","speaker":null,"is_sponsor":0},{"start_s":115.84,"end_s":120.28,"text":"But we're also isolating for the CPU that also limits us, because we do have tests that","speaker":null,"is_sponsor":0},{"start_s":120.28,"end_s":125.44,"text":"test the whole system, but we're not looking to test the memory and the SSD and the graphics","speaker":null,"is_sponsor":0},{"start_s":125.44,"end_s":132.0,"text":"card yet. But our current test suite for Markbench is very Windows focused, so there wasn't really","speaker":null,"is_sponsor":0},{"start_s":132.0,"end_s":138.52,"text":"anything we could just reuse from that. There is a test framework out there called Pheronix, which has been there out there for","speaker":null,"is_sponsor":0},{"start_s":138.52,"end_s":146.2,"text":"a long time. So we tried to find some that did compression, stuff that did maybe some encoding, things","speaker":null,"is_sponsor":0},{"start_s":146.2,"end_s":150.0,"text":"that were just pure computational. Okay, so I have to confess something.","speaker":null,"is_sponsor":0},{"start_s":150.0,"end_s":155.16,"text":"I had this idea, fantasy really, that Labs would test a bunch of Macs and then compare","speaker":null,"is_sponsor":0},{"start_s":155.16,"end_s":159.96,"text":"it to a matrix of CPU data that would show us what desktop and laptop CPUs perfectly","speaker":null,"is_sponsor":0},{"start_s":159.96,"end_s":166.48,"text":"match their M-series counterparts. But we have to use new tests, so we're starting from scratch.","speaker":null,"is_sponsor":0},{"start_s":166.48,"end_s":169.72,"text":"While we would have loved to investigate every CPU, it is wholly unreasonable to get","speaker":null,"is_sponsor":0},{"start_s":169.72,"end_s":176.36,"text":"Labs to test them all. For instance, here in Logistics, there are about 150 different CPUs available to test,","speaker":null,"is_sponsor":0},{"start_s":176.36,"end_s":180.12,"text":"and that doesn't even include the laptop CPUs and all the shapes and sizes they're","speaker":null,"is_sponsor":0},{"start_s":180.12,"end_s":186.44,"text":"cooled in, either. So we have to make tradeoffs, and this is about graphics cards, and so that's why we're","speaker":null,"is_sponsor":0},{"start_s":186.44,"end_s":194.2,"text":"only including desktop components. If you're after gaming performance, the question then becomes like, okay, cool, we have three","speaker":null,"is_sponsor":0},{"start_s":194.2,"end_s":199.84,"text":"contenders, right? Apple, Intel, and AMD. Intel changes their socket all the time.","speaker":null,"is_sponsor":0},{"start_s":199.84,"end_s":204.76,"text":"But AMD, you can go back three generations on AM4 on the same platform.","speaker":null,"is_sponsor":0},{"start_s":204.76,"end_s":208.76,"text":"And so that's why we're going to be sticking with AM4 chips in our conclusion, though we","speaker":null,"is_sponsor":0},{"start_s":208.76,"end_s":212.28,"text":"did run the tests on a few Intel chips earlier in this project.","speaker":null,"is_sponsor":0},{"start_s":212.28,"end_s":216.24,"text":"Alright, so let's go through the tests and the results.","speaker":null,"is_sponsor":0},{"start_s":216.24,"end_s":220.16,"text":"The first we did is, of course, Cinebench, it's widely used in the tech media space","speaker":null,"is_sponsor":0},{"start_s":220.16,"end_s":223.96,"text":"and they just came out with an update for it, though we did R23.","speaker":null,"is_sponsor":0},{"start_s":223.96,"end_s":227.4,"text":"It includes both a multi-core and single-core score.","speaker":null,"is_sponsor":0},{"start_s":227.4,"end_s":233.2,"text":"We chose Cinebench because it kind of chooses itself as it's the prolific go-to processor","speaker":null,"is_sponsor":0},{"start_s":233.2,"end_s":239.4,"text":"benchmark, and it's really good that it supports Apple natively as well as Windows.","speaker":null,"is_sponsor":0},{"start_s":239.4,"end_s":243.2,"text":"Looking at the single-core results, you can see how the newest chips rise to the top,","speaker":null,"is_sponsor":0},{"start_s":243.2,"end_s":248.2,"text":"but once all the cores get involved, you can see just how the 24 and the M2 Ultra push","speaker":null,"is_sponsor":0},{"start_s":248.2,"end_s":254.08,"text":"it to the top. Along with single-core tests, Lab did a Flak encode test where they encoded a bunch of","speaker":null,"is_sponsor":0},{"start_s":254.08,"end_s":257.6,"text":"copies of a 9-inch nail song from Wave to Flak.","speaker":null,"is_sponsor":0},{"start_s":257.6,"end_s":261.92,"text":"We actually struggled to find single-core tests because most tests are all about loading","speaker":null,"is_sponsor":0},{"start_s":261.92,"end_s":266.38,"text":"the CPU and trying to, you know, how fast you compute this thing.","speaker":null,"is_sponsor":0},{"start_s":266.38,"end_s":271.64,"text":"In this test, and another you'll see, Apple Silicon is so far ahead of the other chips,","speaker":null,"is_sponsor":0},{"start_s":271.64,"end_s":277.92,"text":"and they're all grouped together. That's why we're going to be weighing this test less when we figure out our matches.","speaker":null,"is_sponsor":0},{"start_s":277.92,"end_s":283.16,"text":"The last single-core tests are the XZ and LZ4 compression tests, with both compressing","speaker":null,"is_sponsor":0},{"start_s":283.16,"end_s":289.64,"text":"an Ubuntu image. We actually tried like four different compression algorithms, or compression tests, but not","speaker":null,"is_sponsor":0},{"start_s":289.64,"end_s":295.38,"text":"all of them worked. Sometimes they worked on one, but not the other, even though they're advertised for cross-platform.","speaker":null,"is_sponsor":0},{"start_s":295.38,"end_s":300.96,"text":"So we did find that XZ and LZ4, we were able to compile for both natively.","speaker":null,"is_sponsor":0},{"start_s":300.96,"end_s":306.08,"text":"LZ4 single-core shakes out slightly differently from Cinebench, with most of the Ryzen 5,000s","speaker":null,"is_sponsor":0},{"start_s":306.08,"end_s":313.04,"text":"closer to the M3 generation. But it appears that with the XZ compression test, Ryzen has a bit more strength than it","speaker":null,"is_sponsor":0},{"start_s":313.04,"end_s":318.36,"text":"does in Cinebench. What was it like to do all the testing?","speaker":null,"is_sponsor":0},{"start_s":318.36,"end_s":325.28,"text":"Illuminating. Testing's pretty straightforward. Once you identify the tests and you come up with your test suite, the execution is just","speaker":null,"is_sponsor":0},{"start_s":325.28,"end_s":331.2,"text":"while you do your testing. And now as results come in, it's like putting your puzzle together, right?","speaker":null,"is_sponsor":0},{"start_s":331.2,"end_s":334.64,"text":"As the pieces slowly fit in more, you get more of the picture.","speaker":null,"is_sponsor":0},{"start_s":334.76,"end_s":338.48,"text":"However, the difference is you're completing a puzzle that doesn't, you don't know what","speaker":null,"is_sponsor":0},{"start_s":338.48,"end_s":343.92,"text":"the end picture is. So it's interesting that way to kind of see the story reveal itself to you.","speaker":null,"is_sponsor":0},{"start_s":343.92,"end_s":348.36,"text":"Alright, so how about the puzzle pieces that tested multiple CPU cores?","speaker":null,"is_sponsor":0},{"start_s":348.36,"end_s":353.12,"text":"Blender is a popular 3D modeling program, and in it, Labs rendered the barbershop scene.","speaker":null,"is_sponsor":0},{"start_s":353.12,"end_s":356.24,"text":"We might need to find a new scene as it can render fairly quickly.","speaker":null,"is_sponsor":0},{"start_s":356.24,"end_s":360.24,"text":"It's a popular scene used to benchmark rendering performance in Blender.","speaker":null,"is_sponsor":0},{"start_s":360.24,"end_s":364.96,"text":"In Blender, the mid-range AMDs provide a transition between the M2 and M3s.","speaker":null,"is_sponsor":0},{"start_s":364.96,"end_s":368.72,"text":"The 5600X and G are surprisingly weaker here.","speaker":null,"is_sponsor":0},{"start_s":368.72,"end_s":371.92,"text":"Another render type test that I'd never heard of is C-Ray.","speaker":null,"is_sponsor":0},{"start_s":371.92,"end_s":375.72,"text":"It's a simple ray tracer that outputs this 90s looking image.","speaker":null,"is_sponsor":0},{"start_s":375.72,"end_s":383.08,"text":"C-Ray gives us another render type of test, but mostly we chose it because it works on","speaker":null,"is_sponsor":0},{"start_s":383.08,"end_s":387.04,"text":"both. That's definitely got for some of these.","speaker":null,"is_sponsor":0},{"start_s":387.76,"end_s":392.32,"text":"It's a very simple, efficient load to multi-core.","speaker":null,"is_sponsor":0},{"start_s":392.32,"end_s":397.48,"text":"AMD is relatively weaker in this test with the 3600 sitting between the M1s and their","speaker":null,"is_sponsor":0},{"start_s":397.48,"end_s":404.56,"text":"different cooling. The next test Labs did is LibRAW, which tests how well CPUs handle raw photographs.","speaker":null,"is_sponsor":0},{"start_s":404.56,"end_s":409.68,"text":"LibRAW is also nice that it has a built-in post-processing benchmark, which we run 30","speaker":null,"is_sponsor":0},{"start_s":409.68,"end_s":414.48,"text":"times on the test image that comes with Fronix, and then it spits out like a megapixels per","speaker":null,"is_sponsor":0},{"start_s":414.48,"end_s":419.08,"text":"second. LibRAW is the other test we're going to have to weigh less because of Apple Silicon's apparent","speaker":null,"is_sponsor":0},{"start_s":419.08,"end_s":424.76,"text":"supremacy. It does feel like the Macs are especially tailored to calculate audio and visual codecs.","speaker":null,"is_sponsor":0},{"start_s":424.76,"end_s":428.44,"text":"And lastly, if you're into numbers, there's PrimeSiv.","speaker":null,"is_sponsor":0},{"start_s":428.44,"end_s":432.92,"text":"We chose it because Y Cruncher doesn't work natively on Mac, because we do favor Y Cruncher.","speaker":null,"is_sponsor":0},{"start_s":432.92,"end_s":439.16,"text":"It's a very popular benchmarking one, but we found PrimeSiv, PrimeSiv, PrimeSiv, PrimeCV,","speaker":null,"is_sponsor":0},{"start_s":439.16,"end_s":443.2,"text":"PrimeSciEV, it calculates Prime numbers up to a certain length.","speaker":null,"is_sponsor":0},{"start_s":443.2,"end_s":449.44,"text":"So we consider that it's our stand-in for Y Cruncher and that it's something computational,","speaker":null,"is_sponsor":0},{"start_s":449.44,"end_s":452.88,"text":"generating number over a long period.","speaker":null,"is_sponsor":0},{"start_s":452.88,"end_s":458.04,"text":"It's multi-core as well. We learned a lot by doing this, and that is that this is hard.","speaker":null,"is_sponsor":0},{"start_s":458.04,"end_s":462.72,"text":"For one, picking AM4 means that we've got an array of chips that don't quite fit with","speaker":null,"is_sponsor":0},{"start_s":462.72,"end_s":467.72,"text":"single-core performance, as that's where Apple Silicon shines, and these are old.","speaker":null,"is_sponsor":0},{"start_s":467.72,"end_s":472.08,"text":"But then, with AM5, there aren't any low-end chips to compare with the lower-end Mac chips","speaker":null,"is_sponsor":0},{"start_s":472.08,"end_s":477.28,"text":"either. Single-core is more important for gaming, so we waited higher, but we waited Flak and","speaker":null,"is_sponsor":0},{"start_s":477.28,"end_s":481.8,"text":"Libra less because they favor Apple Silicon egregiously in a way that's not related to","speaker":null,"is_sponsor":0},{"start_s":481.8,"end_s":489.08,"text":"gaming. Alright, our picks. These choices for CPUs are still even a best guess, because we don't know what's the bottleneck","speaker":null,"is_sponsor":0},{"start_s":489.08,"end_s":493.32,"text":"CPU-wise once the GPUs are installed in running games.","speaker":null,"is_sponsor":0},{"start_s":493.32,"end_s":498.48,"text":"So these are not exact matches. I really was in Fantasyland thinking this was possible.","speaker":null,"is_sponsor":0},{"start_s":498.48,"end_s":505.08,"text":"But we've learned a lot. We're going to use the AMD 5800X3D as a control and to match the M2 Ultra, because","speaker":null,"is_sponsor":0},{"start_s":505.08,"end_s":510.28,"text":"its 3D cache really helps in gaming, and the M2 Ultra screamed well ahead in every test","speaker":null,"is_sponsor":0},{"start_s":510.28,"end_s":516.84,"text":"we threw at it. The M2 Pro and Max, as well as the M3 Pro chips, will be matched against the 5800X.","speaker":null,"is_sponsor":0},{"start_s":516.84,"end_s":522.96,"text":"The 5700X were pitting alongside the basic M3, and the basic M2 and M1 chips are matched","speaker":null,"is_sponsor":0},{"start_s":522.96,"end_s":527.2,"text":"against a 5600G. I'm feeling as well as I could.","speaker":null,"is_sponsor":0},{"start_s":527.92,"end_s":542.24,"text":"And that, I guess, is the biggest lesson on this journey.","speaker":null,"is_sponsor":0},{"start_s":542.24,"end_s":548.68,"text":"And it's that we're always learning. But now that we have our CPUs figured out, the next step will be to test the GPUs.","speaker":null,"is_sponsor":0},{"start_s":548.68,"end_s":552.96,"text":"Have NVIDIA and AMD met their match? We're going to have to see where they line up.","speaker":null,"is_sponsor":0},{"start_s":552.96,"end_s":561.08,"text":"But things are looking good in their own little way. Personally, I wasn't expecting the Mac, the Apples, to be as strong as they were.","speaker":null,"is_sponsor":0},{"start_s":561.08,"end_s":565.36,"text":"I knew they were super efficient, so this is my first really experience.","speaker":null,"is_sponsor":0},{"start_s":565.36,"end_s":571.8,"text":"Now during this project, I started daily driving the 15-inch MacBook Air, and I loved it.","speaker":null,"is_sponsor":0},{"start_s":571.8,"end_s":577.64,"text":"The fact that I could close it, neglect it for three days, and it still had power.","speaker":null,"is_sponsor":0},{"start_s":577.64,"end_s":580.8,"text":"I mean, I ended up buying one. What?","speaker":null,"is_sponsor":0},{"start_s":580.8,"end_s":584.24,"text":"You bought a Mac from this project? I mean, I did immediately stick or bomb it.","speaker":null,"is_sponsor":0},{"start_s":584.24,"end_s":588.84,"text":"How dare you? Thanks for testing this Mac Address, Labs.","speaker":null,"is_sponsor":0},{"start_s":588.84,"end_s":593.52,"text":"If you want to check out another video we did, check out the iPad tier list video.","speaker":null,"is_sponsor":0},{"start_s":593.52,"end_s":597.68,"text":"And I'm curious in the comments below, who of you are like Nicholas and bought a Mac","speaker":null,"is_sponsor":0},{"start_s":597.68,"end_s":598.4,"text":"for gaming?","speaker":null,"is_sponsor":0}],"full_text":"I've always had a problem comparing Apple Silicon to PC parts because gaming wasn't ready. I've always wanted to know how Apple Silicon compares to off-the-shelf PC GPUs and CPUs. And that's not something we've been able to test because Macs and PCs now use completely different architectures, and there were no games that ran natively on both. But that's changed. Check this out. I'm playing Baldur's Gate 3, one of the hottest video games right now, on a Mac. And it's running natively on Apple Silicon. There's now a growing list of games that support both Apple's graphics hardware and metal graphics APIs without any sort of translation layer. So with the help of LMG Labs, we're going to finally compare Apple chips with PC graphics cards and see what GPUs from NVIDIA and AMD compete with Apple's M series. Though not in this video. That's part two because in part one, we need to build the test benches that the GPUs will be on. And those test benches have CPUs, so we're going to figure out what CPUs are the closest to Apple Silicon. And I need to ask Labs if they'll help me. What did I think? Oh my god. Well, it's going to be interesting. It's going to be a learning experience. We haven't tested Macs really all that much, so this is going to be the first time Labs really tackles Macs. Nicholas, hey. This is Nicholas Harris. He's LTT Labs software developer. Part of his job has been developing and automating the tests for PCs and other parts here. The first step is figuring out the tests. And so Nicholas worked with test technician John Deuren to figure out how to measure these two completely different computers. Our primary goal was to find tests that can natively work on either system to avoid the Rosetta layer, because that's another variable that we want to isolate for. But we're also isolating for the CPU that also limits us, because we do have tests that test the whole system, but we're not looking to test the memory and the SSD and the graphics card yet. But our current test suite for Markbench is very Windows focused, so there wasn't really anything we could just reuse from that. There is a test framework out there called Pheronix, which has been there out there for a long time. So we tried to find some that did compression, stuff that did maybe some encoding, things that were just pure computational. Okay, so I have to confess something. I had this idea, fantasy really, that Labs would test a bunch of Macs and then compare it to a matrix of CPU data that would show us what desktop and laptop CPUs perfectly match their M-series counterparts. But we have to use new tests, so we're starting from scratch. While we would have loved to investigate every CPU, it is wholly unreasonable to get Labs to test them all. For instance, here in Logistics, there are about 150 different CPUs available to test, and that doesn't even include the laptop CPUs and all the shapes and sizes they're cooled in, either. So we have to make tradeoffs, and this is about graphics cards, and so that's why we're only including desktop components. If you're after gaming performance, the question then becomes like, okay, cool, we have three contenders, right? Apple, Intel, and AMD. Intel changes their socket all the time. But AMD, you can go back three generations on AM4 on the same platform. And so that's why we're going to be sticking with AM4 chips in our conclusion, though we did run the tests on a few Intel chips earlier in this project. Alright, so let's go through the tests and the results. The first we did is, of course, Cinebench, it's widely used in the tech media space and they just came out with an update for it, though we did R23. It includes both a multi-core and single-core score. We chose Cinebench because it kind of chooses itself as it's the prolific go-to processor benchmark, and it's really good that it supports Apple natively as well as Windows. Looking at the single-core results, you can see how the newest chips rise to the top, but once all the cores get involved, you can see just how the 24 and the M2 Ultra push it to the top. Along with single-core tests, Lab did a Flak encode test where they encoded a bunch of copies of a 9-inch nail song from Wave to Flak. We actually struggled to find single-core tests because most tests are all about loading the CPU and trying to, you know, how fast you compute this thing. In this test, and another you'll see, Apple Silicon is so far ahead of the other chips, and they're all grouped together. That's why we're going to be weighing this test less when we figure out our matches. The last single-core tests are the XZ and LZ4 compression tests, with both compressing an Ubuntu image. We actually tried like four different compression algorithms, or compression tests, but not all of them worked. Sometimes they worked on one, but not the other, even though they're advertised for cross-platform. So we did find that XZ and LZ4, we were able to compile for both natively. LZ4 single-core shakes out slightly differently from Cinebench, with most of the Ryzen 5,000s closer to the M3 generation. But it appears that with the XZ compression test, Ryzen has a bit more strength than it does in Cinebench. What was it like to do all the testing? Illuminating. Testing's pretty straightforward. Once you identify the tests and you come up with your test suite, the execution is just while you do your testing. And now as results come in, it's like putting your puzzle together, right? As the pieces slowly fit in more, you get more of the picture. However, the difference is you're completing a puzzle that doesn't, you don't know what the end picture is. So it's interesting that way to kind of see the story reveal itself to you. Alright, so how about the puzzle pieces that tested multiple CPU cores? Blender is a popular 3D modeling program, and in it, Labs rendered the barbershop scene. We might need to find a new scene as it can render fairly quickly. It's a popular scene used to benchmark rendering performance in Blender. In Blender, the mid-range AMDs provide a transition between the M2 and M3s. The 5600X and G are surprisingly weaker here. Another render type test that I'd never heard of is C-Ray. It's a simple ray tracer that outputs this 90s looking image. C-Ray gives us another render type of test, but mostly we chose it because it works on both. That's definitely got for some of these. It's a very simple, efficient load to multi-core. AMD is relatively weaker in this test with the 3600 sitting between the M1s and their different cooling. The next test Labs did is LibRAW, which tests how well CPUs handle raw photographs. LibRAW is also nice that it has a built-in post-processing benchmark, which we run 30 times on the test image that comes with Fronix, and then it spits out like a megapixels per second. LibRAW is the other test we're going to have to weigh less because of Apple Silicon's apparent supremacy. It does feel like the Macs are especially tailored to calculate audio and visual codecs. And lastly, if you're into numbers, there's PrimeSiv. We chose it because Y Cruncher doesn't work natively on Mac, because we do favor Y Cruncher. It's a very popular benchmarking one, but we found PrimeSiv, PrimeSiv, PrimeSiv, PrimeCV, PrimeSciEV, it calculates Prime numbers up to a certain length. So we consider that it's our stand-in for Y Cruncher and that it's something computational, generating number over a long period. It's multi-core as well. We learned a lot by doing this, and that is that this is hard. For one, picking AM4 means that we've got an array of chips that don't quite fit with single-core performance, as that's where Apple Silicon shines, and these are old. But then, with AM5, there aren't any low-end chips to compare with the lower-end Mac chips either. Single-core is more important for gaming, so we waited higher, but we waited Flak and Libra less because they favor Apple Silicon egregiously in a way that's not related to gaming. Alright, our picks. These choices for CPUs are still even a best guess, because we don't know what's the bottleneck CPU-wise once the GPUs are installed in running games. So these are not exact matches. I really was in Fantasyland thinking this was possible. But we've learned a lot. We're going to use the AMD 5800X3D as a control and to match the M2 Ultra, because its 3D cache really helps in gaming, and the M2 Ultra screamed well ahead in every test we threw at it. The M2 Pro and Max, as well as the M3 Pro chips, will be matched against the 5800X. The 5700X were pitting alongside the basic M3, and the basic M2 and M1 chips are matched against a 5600G. I'm feeling as well as I could. And that, I guess, is the biggest lesson on this journey. And it's that we're always learning. But now that we have our CPUs figured out, the next step will be to test the GPUs. Have NVIDIA and AMD met their match? We're going to have to see where they line up. But things are looking good in their own little way. Personally, I wasn't expecting the Mac, the Apples, to be as strong as they were. I knew they were super efficient, so this is my first really experience. Now during this project, I started daily driving the 15-inch MacBook Air, and I loved it. The fact that I could close it, neglect it for three days, and it still had power. I mean, I ended up buying one. What? You bought a Mac from this project? I mean, I did immediately stick or bomb it. How dare you? Thanks for testing this Mac Address, Labs. If you want to check out another video we did, check out the iPad tier list video. And I'm curious in the comments below, who of you are like Nicholas and bought a Mac for gaming?"}