AMD Ryzen 7 5800X3D review: cache money
ORIGINAL SOURCE LINK
The $449/£429 Ryzen 7 5800X3D is something a bit different from AMD, a processor that exists to demonstrate the power of the firm’s 3D V-Cache design for its upcoming Ryzen CPUs and fight off Intel’s 12900KS for the title of ‘fastest gaming processor’. It’s also something of a last hurrah for the surprisingly long-lived AM4 platform, which debuted back in 2017 and outlasted half a dozen Intel generations as Ryzen CPUs improved in leaps and bounds.
So what exactly is a 3D V-Cache anyway? Let’s start with the basics. You can think of a processor’s cache as a place to store data it’s currently working on – a bit like RAM, but because it’s inside the CPU it is an order of magnitude faster to access and an order of magnitude smaller in terms of the amount of data it can store. Modern processors typically use three levels of cache – L1, L2, L3 – with L1 cache being the fastest to access but the smallest, L2 being slower but larger, and L3 being slower and larger again. It’s this third level of cache that AMD has changed, moving from a traditional 2D design to a 3D design, a stack of cache that takes up more vertical space. This allows for way more data to be stored inside the CPU at once, thus increasing the chances that the data needed is already inside and speeding up any subsequent processing.
AMD is slated to use this technology for its future Zen 4 processors, but in the here and now it’s just this one special 5800X3D, an upgraded version of the Ryzen 7 5800X that launched back in 2020. Compared to the 5800X, the 5800X3D trades a bit of frequency and some overclocking controls for a significantly larger 96MB L3 cache – triple the size of the 5800X’s.
|CPU design||Boost||Base||L3 cache||TDP||RRP|
|Ryzen 5950X||Zen 3 16C/32T||4.9GHz||3.4GHz||64MB||105W||$799|
|Ryzen 5900X||Zen 3 12C/24T||4.8GHz||3.7GHz||64MB||105W||$549|
|Ryzen 5800X3D||Zen 3 8C/16T||4.5GHz||3.4GHz||96MB||105W||$449|
|Ryzen 5800X||Zen 3 8C/16T||4.7GHz||3.8GHz||32MB||105W||$449|
|Ryzen 5700G||Zen 3 8C/16T||4.6GHz||3.8GHz||16MB||65W||$359|
|Ryzen 5600X||Zen 3 6C/12T||4.6GHz||3.7GHz||32MB||65W||$299|
|Ryzen 5600G||Zen 3 6C/12T||4.4GHz||3.9GHz||16MB||65W||$259|
Before we get into the first test results, let’s briefly cover the rig we’re using. For the AMD side, we’re using an Asus ROG Crosshair 8 Hero, while 11th-gen Intel gets an Asus ROG Maximus Z590 Hero and 12th-gen gets the Asus ROG Z690 Maximus Hero – all high-end boards for their respective platforms. DDR4 motherboards used G.Skill 3600MT/s CL16 memory, while 12th-gen Intel got the benefit of faster but higher-latency Corsair 5200MT/s CL38 RAM.
The AMD and 11th-gen Intel CPUs were cooled with an Eisbaer Aurora 240mm AiO, while the 12th-gen testing was conducted with an Asus ROG Ryujin 2 360mm AiO. (And to answer the obvious question: 240mm and 360mm AiOs tend to provide equivalent performance based on our testing – especially for an open air test bench in cool (21C) ambient conditions. The only difference tends to be fan speed, which is higher on the 240mm than the 360mm.) Our rig was completed with a 1000W Corsair RM1000x power supply from Infinite Computing.
In order to reduce run-to-run variance and ensure we’re CPU-limited as much as possible, we’re using the Asus ROG Strix 3090 OC Edition. This is a massive three-slot, triple-fan design that keeps the card surprisingly cool and quiet.
One of the biggest questions over the 5800X3D is exactly where that upgraded cache will come in useful – because if a game or other application doesn’t fit a specific performance profile, it may see no performance advantage at all running on the 5800X3D – and indeed, it may even run worse due to the clock speed that AMD has sacrificed to make the design work.
To find out, we’ve tested the 5800X3D in a range of content creation and gaming scenarios – against the original 5800X and a number of other recent AMD and Intel processors. We’re hoping to see some big performance increases, especially in video games, but we’ll start with a couple of quick content creation benchmarks: a Cinebench R20 3D render and a Handbrake video transcode.
|CB R20 1T||CB R20 MT||HB h.264||HB HEVC||HEVC Power Use|
|Core i9 12900K||760||10416||70.82fps||29.26fps||373W|
|Core i7 12700K||729||8683||57.64fps||25.67fps||318W|
|Core i5 12600K||716||6598||44.27fps||19.99fps||223W|
|Core i5 12400F||652||4736||31.77fps||14.70fps||190W|
|Core i9 11900K||588||5902||41.01fps||18.46fps||321W|
|Core i5 11600K||541||4086||29.00fps||13.12fps||250W|
|Ryzen 9 5950X||637||10165||70.28fps||30.14fps||237W|
|Ryzen 7 5800X3D||546||5746||42.71fps||19.10fps||221W|
|Ryzen 7 5800X||596||6118||44.18fps||19.50fps||229W|
|Ryzen 5 5600X||601||4502||31.75fps||14.43fps||160W|
Neither content creation result is particularly impressive for the 5800X3D, which outperforms the 5600X and Intel 12400F but falls behind its erstwhile competitors like the 12900K, 12700K and 5800X (the latter by between two to six percent). This isn’t a massive surprise – neither task would logically benefit from having larger cache, so you only see the effect of the new CPU’s lowered core clocks compared to the standard 5800X. However, the results aren’t disastrous either; this is still a perfectly capable CPU for these tasks that handily outperforms prior generations, just not a class-leading one.
With those out of the way, let’s move onto the fun stuff: checking out how the 5800X3D performs in a range of games. Click the quick links below to move onto the titles you’re most interested in, or hit the ‘next page’ button to take it all in!