AMD Ryzen 5800X3D

amazon AMD Ryzen 5800X3D reviews

The battle for the upper hand in gaming CPUs between Intel and AMD is inherently intense, but when the 5800X3D was released by AMD, it created an earthquake when gaming performance could be completely on par with that of Core i9-12900K.

This gaming-only CPU from AMD uses a new advanced 3D stacked SRAM technology, called 3D V-Cache, which enables a total of 96MB of L3 cache to unleash monstrous gaming performance. Consider Intel’s Core i9-12900KS as the worst-performing CPU among the Best CPUs for gaming – but at a more affordable price.

AMD has succeeded with an eight-core 16-thread chip based on the same 7nm process and Zen 3 architecture as the original Ryzen 5000 chip launched in 2020, but using innovative hybrid bonding technology to integrate an additional caching section topped the processor cores, a first for the desktop.

The Ryzen 7 5800X3D represents the company’s final foray into its long-lived Socket AM4 platform, which replaced Ryzen chips from their infancy with the Ryzen 7 1800X in 2017 to achieve a once dominant position at the top of our CPU benchmark gaming hierarchy last year with the Ryzen 9 5900X.

AMD’s chips held the lead in every metric until Intel released the Alder Lake lineup last year, with Intel’s Core i9-12900K being the fastest gaming CPU we’ve ever tested. However, with AMD getting ready to launch the 5800X3D, Intel managed to rise to the top of the gaming performance charts with its new Core i9-12900KS special edition.

Intel’s short-lived advantage in gaming comes at the expense of additional power: the Core i9-12900KS has a 150W processor base power (PBP), a record for a processor mainstream desktop processors, and we’ve measured up to 300W of power consumption all site. In contrast, the Ryzen 7 5800X3D has a 105W TDP rating and maxed out at 130W in our tests, suggesting that it’s a much cooler processor and won’t require expensive amenities, such as powerful coolers, motherboards, and power supplies, like the Core i9-12900KS.

The Ryzen 7 5800X3D’s 96MB L3 cache is OS-transparent, meaning it doesn’t need special support from the operating system or software, but it doesn’t benefit all games. However, we did notice a big increase in most of the titles we tested.

Here’s a snapshot of the 5800X3D’s average performance in our gaming test suite and critical single and multithreaded applications. You’ll find more extensive tests below, but this gives you a good general feel for how the Ryzen 7 5800X3D stacks up when it hits the market on April 20.

As you can see, the Ryzen 7 5800X3D takes the crown as the fastest gaming chip in our test suite and fulfills AMD’s claim that 3D V-Cache delivers an increase in gaming performance comparable to those of the previous generation. What we usually only see from a new microarchitecture. However, the 58000X3D is not as fast as comparable chips in other types of single and multi-threaded work outside of gaming. That’s because other models have advantages in core count and frequency. In fact, due to lower clock speeds than its most directly comparable counterpart, the Ryzen 7 5800X, the 5800X3D is slower in some single-threaded applications.

While Intel’s 12900KS still delivers class-leading performance in applications, its hefty $739 premium isn’t worth buying as much as AMD’s $449 Ryzen 7 5800X3D if you’re only interested in playing games. The same applies to the standard 12900K and 12700K, though the Core i7-12700K is a contender if you’re looking for a more balanced combination of gaming and application performance at the $410 price point.

Of course, the Ryzen 7 5800X3D is a big win if you already own a Ryzen system – this chip will land on almost any AM4 motherboard, save some cash if you have the right supporting components. Overall, the Ryzen 7 5800X3D is what AMD says it is – a chip optimized specifically for gaming that leads the way overall.

Specifications

The Ryzen 7 5800X3D is the first consumer processor with a 3D V-Cache, but the company also uses the technology for its Milan-X processors for the data center. 3D V-Cache leverages a new technique that uses a hybrid link to combine an additional 64MB of 7nm SRAM cache vertically on top of the Ryzen computer chiplet, thus tripling the number of L3 caches per Ryzen mold.

The Ryzen 7 5800X3D comes with the same eight Zen 3 cores and 16 threads as the standard Ryzen 7 5800X but with a lower 3.4GHz base frequency and 4.5GHz boost frequency in its 105W case. AMD has cut 400 MHz of a base clock and boosted frequency by 200 MHz, but in return, you get 64 MB of L3 cache, for a total of 96 MB of L3.

Naturally, 3D V-Cache technology has trade-offs, the most obvious of which is the $449 price tag – you’ll pay an extra $100 for the same number of cores as you get in the vanilla Ryzen 7 5800X.

The big draw of the 5800X3D is AMD’s claim of an average 15% increase in gaming performance over AMD’s fastest gaming chip, the Ryzen 9 5900X which also retails for $450 now. 3D V-Cache doesn’t boost performance in other types of work than gaming, so compared to the 5900X you’re sacrificing four cores and eight threads in exchange for additional caching, thus reducing performance in a short period number of productivity applications. That means the Ryzen 9 5900X will be a better choice for productivity-focused work, but you should also look to the Alder Lake alternatives if you’re after a more balanced performance profile.

The 5800X3D fully supports memory overclocking and Infinity Fabric, but you cannot overclock CPU cores or use the Precision Boost Overdrive feature that automatically overclocks. The company cites a voltage limit, but our thermal testing below certainly implies that heat dissipation is a more aggravating problem. AMD says this is the first iteration of this technology, and there’s the possibility that overclocking could be enabled on potential 3D V-Cache processors in the future. However, the company has yet to officially commit to releasing other models in the future. Given the performance we’ve seen, it wouldn’t be surprising to see this technology carried over to the Zen 4 era.

However, that won’t stop enthusiasts from taking the challenge. We’ve seen reports of a limited BCLK overclock that could generate a few hundred megahertz more (perhaps more on motherboards with an external clock) and there seems to be a solution to varying the voltage motherboard’s output to the CPU, thus giving the chip more voltage than AMD intended. Of course, the latter can pose significant risks.

Like all other Ryzen 5000 105W chips, the Ryzen 7 5800X3D does not come with a cooler. This chip has the same thickness (Z-height) as all other Ryzen 5000 models, so it’s compatible with the vast ecosystem of standard coolers for the AM4 socket. The 5800X3D will fall under the existing 400 and 500 series motherboards (Socket AM4) and AMD’s upcoming BIOS updates will also enable support on older 300 series platforms. You will need a BIOS with AGESA 1.2.0.6b (or later) for the Ryzen 7 5800X3D.

AMD says Ryzen 5000 support will vary by vendor, as well as the schedule for new BIOS revisions. However, we should see them in the April to May timeframe. Notably, these BIOS revisions will also include a fix for AMD’s fTPM stuttering issues.

The 5800X3D also doesn’t support the top connectivity options, like DDR5 and PCIe 5.0, that you’ll find with Alder Lake, but it does support up to DDR4-3200 and PCIe 4.0. AMD won’t be able to match intel’s connectivity technology until their Ryzen 7000 ‘Raphael’ Zen 4 5nm CPU launches later this year.

3D V-Cache Technology

The idea behind 3D V-Cache is relatively simple, but the implementation is complex. The basic idea behind any on-chip cache is to keep frequently accessed data as close to the execution cores as possible, thus eliminating high-latency trips to main memory. As a result, the cores don’t have to wait for data, hence busier operation and increased performance. L3 cache is slower than other caches (like L1 and L2), but higher capacity means it can store more data, improving the access rate (number of times useful data is available). the utility is kept in the cache). There’s a reason AMD calls it “Game Cache” – L3 cache is critical to performance and games, in particular, can suffer from high L3 latency or reduced cache space/billion access rate.

But the best is a large cache. As you can see in the album above, AMD stacks an additional SRAM chiplet, connected via TSV to the bottom die, directly in the center of the computer die (CCD) to isolate it from the heat-generating cores at the cores face of the chiplet. However, AMD had to use silicon pads on top of the cores to create an even surface for the heatsink to rest on top of the chiplet. Contrary to popular belief, this is a single pad that wraps around the chiplet on three sides. Silicon is an excellent conductor of heat, but the gasket and the extra SRAM die will certainly reduce heat dissipation from the bottom die, thus resulting in less heat dissipation headroom. We showed that effect in our boost frequency and thermal load test. Extra memory also consumes more power.

AMD says it can’t overclock because the cache chiplet and CCD share the same power plane and limit the effective voltage for the heavy SRAM chiplet at 1.35V. Since the core voltage cannot be changed separately, it prevents overclocking the core frequencies of the CPU. Unfortunately, this also interferes with the chip’s peak frequency during normal operation, so 3D V-Cache technology contributes to the 5800X3D’s lower clock speed. In terms of perspective, the Ryzen 7 5800X has a 1.5V limit, so it can hit higher clock speeds.

AMD’s 3D chip stacking technology is based on TSMC’s SoIC technology. TSMC’s SoIC is an easy chip-stacking technology, which means it doesn’t use micro bumps or solder to connect two dies. Instead, the two dies are ground into such a perfectly flat surface that the TSV channels can mate without any kind of bonding material, reducing the distance between the cache and the core by 1000X. That reduces heat and power consumption while increasing bandwidth. AMD says the technique uses a silicon-like manufacturing process with a backing like TSV, which means the manufacturing process is similar to that of conventional chips.

As before, the 7nm Core Complex Die (CCD) has 4.15 billion transistors spread across 80.7mm^2 silicon. Meanwhile, the new 7nm 3D V-Cache die is smaller at only 41mm^2 but has 4.7 billion transistors. As you can see in the table, that means it has a little more than double the transistor density, which is because AMD uses a density-optimized version specifically for SRAM. It is important to remember that a standard compute die includes several types of transistors (library, standard cells) for different purposes, so the density varies across the die. In contrast, the V-Cache template uses a uniform layout.

The L3 cache chiplet spans the same amount of space as the L3 cache on the underlying CCD, but it also has twice the capacity. That’s because the process is optimized, but also partly because the additional L3 cache slice is a bit ‘strange’ – all control circuitry is on the base die, which helps to reduce the inevitable latency costs associated with involves fetching data from a separate die.

The compilation, Compression, AVX Benchmarks

This bag of various tests shows that the Ryzen 7 5800X3D is largely behind its similarly priced competition, although the performance boost in the threaded y-cruncher test is almost certainly a result of the 3D V-Cache. However, the 5800X3D’s lower clock speed means that the same trend does not carry over to the lightly threaded y-cruncher benchmark.

where can you get a Lian li SL120 Infinity online

AMD Ryzen™ 7 5800X3D 8-core, 16-Thread Desktop Processor with AMD 3D V-Cache™ Technology: Buy it now

3D V-Cache Design and Latency

Several factors have influenced AMD’s decision to use 3D stacked SRAM, but the key among them is that SRAM density doesn’t scale as quickly as logic density. As a result, the cache now occupies a higher percentage of the die area than before but does not provide a meaningful increase in capacity. Furthermore, the lateral caching expansion will result in higher latency due to longer wire lengths and eat into the available die area that AMD can use for cores. Also, adding another SRAM chiplet in a 2D layout is not feasible due to latency and bandwidth impacts.

To address those issues, AMD stacks additional SRAM directly on top of the core of the chassis where the existing L3 resides. This L3-on-L3 stacking allows the underlying die to power and communicates via two rows of TSV connections that extend onto the bottom of the L3 cache chiplet. These connections go vertically into the top die and fan out, which reduces the amount of distance data that has to travel, thus reducing the number of cycles required to transfer compared to expanding the flat cache (2D) standard. Thus, the L3 chiplet offers the same 2 TB/s peak throughput as the on-die L3 cache, but it only comes with the four-cycle latency penalty.

These tests measure cache latency with different sizes of data blocks, and the first slide zooms in on the L3 portion of the cache. Here we can see that the Ryzen 7 5800X3D’s L3 latency gauges at 12-13ns, while the 5800X measures at 10-11ns. We also use AIDA to record the latency measurements, which we have listed. Overall, 3D V-Cache triples the amount of L3 cache but suffers a rather insignificant latency impact of ~2ns and a four-cycle penalty.

As mentioned before, the L3 cache chiplet has the same area as the L3 cache on the underlying CCD, but it has twice the capacity. That’s partly because the extra L3 cache slice is a bit ‘freezing’ – all of the control circuitry is on the base die, which minimizes latency. AMD also uses a density-optimized 7nm version specifically for SRAM. The L3 chiplet is also thinner than the base die (13 layers of metal).

AMD manufactures all Zen 3 silicon with TSV, so all of their Zen 3 silicon support 3D V-Cache profiles. However, TSVs are not disclosed unless they are necessary. For the 3D V-Cache models, AMD also thinned the base slightly to both expose the TSV connections and to maintain the same overall package thickness (Z-Height) as the existing models.

The lack of control circuitry in the L3 chiplet also maximizes capacity and allows AMD to selectively ‘light’ only the parts of the cache that are being accessed, thereby reducing (and even eliminating) overhead power cost triples L3 cache capacity. In addition, because a larger cache reduces trips to main memory due to higher L3 cache access speeds, the additional capacity reduces bandwidth pressure on main memory, which reduces latency and thus improves application performance from multiple axes. Fewer trips to the main memory also reduce overall power consumption.

The L3 cache chiplet consumes significantly less power per square millimeter than CPU cores. However, stacking vertically increases power density, so it’s best to isolate it from the heat-generating cores on the sides of the chiplet. However, this would leave a protruding die on top of the CCD, so AMD used a silicon gasket around three sides of the L3 chiplet to create an even surface for the heatsink to rest on top of the chiplet. Silicon is an excellent conductor of heat and serves as a spacer to allow heat to flow from the cores to the heat exchanger.

Previous builds of the design have shown two separate silicon pads and appear to show the L3 cache die stretching from one side to the other. However, AMD’s materials for the Milan-X launch clearly show a long pad covering the computer die and a thin section on the edge of the die not covered by the L3 cache chiplet. This thin extension of the bottom dies covers the I/O functions that the chiplet uses to communicate with the I/O die. AMD confirms that this is the de facto layout on all 3D V-Cache processors, like the Ryzen 7 5800X3D, but not the shared stylized renders that show two separate shims.

AMD Ryzen 7 5800X3D Boost Frequencies, Power, and Thermals

Heat sinks can limit a chip’s peak performance, especially as process nodes become denser. Adding the complexity of the 3D stacked design adds to the thermal challenges, which in this case are exacerbated by the stacked silicon pads on the CPU cores – which transmit heat from the cores to the integrated heatsink (IHS), but can certainly reduce the efficiency of heat transfer from the cores. It can hold a small amount of heat. Signs of those challenges appeared in our thermal and turbocharger testing, prompting us to conduct more in-depth testing to back up our findings.

Ryzen 7 5800X3D and its near-similar counterpart, the Ryzen 7 5800X, without the 3D V-Cache design, run through a series of standard heavy threading applications (Cinebench, HandBrake, AVX-heavy y -cruncher) to measure power and thermoelectricity. We used the Corsair H115i 280mm AIO with the fans turned at 100% to keep the chips as cool as possible during this test run.

The 5800X3D has a 400MHz lower base and 200MHz lower boost clock than the Ryzen 7 5800X and we can see that the 5800X3D runs at 4.35GHz during the heaviest multi-core workloads while the 5800X runs at 4.5 GHz. This is to be expected given the specs, but we also found that the 5800X consumes up to 145W while hitting higher clock speeds, while the 5800X3D only peaks at around 120W. This is despite both chips having the same 105W TDP and 142W PPT.

Both chips hit the same peak of around 80C in the heavy parts of the test, which shows that the 5800X3D runs at the same temperature even though the 5800X is drawing 25W more power and running at a higher clock. AMD tells us that temperature is not the limiting factor preventing higher clock speeds or allowing higher voltages (and therefore more heat) to overclock, but these results certainly imply Note that the 3D stacked design does not dissipate heat as well as the standard design.

We turned to test Prime95 under rigorous conditions for a closer look at heat dissipation. We typically don’t include Prime95 power measurements in our standard CPU reviews, largely because there’s a big disconnect between this extremely rigorous stress test and power consumption and heat load generated by most real-world applications. However, we’re specifically looking to push the chips to their throttle point for this test.

To push the processor limit even harder, we unplugged the fans on the Corsair H115i cooler but kept the pump running (unplugging the pump caused the clocks to drop so quickly that the write Our logs cannot provide any details). Then we started running Prime95 with a small FFT, but with AVX instructions disabled. This type of Prime 95 stress test is brutal and you won’t see this kind of stress during normal use even at the heaviest. So keep in mind that we do this for the sake of science, not as an indicator of how these chips will perform on your PC.

The results clearly show that the 5800X3D peaks at 130W while the 5800X peaks at 145W. Both chips then reach a steady-state temperature of 90C before they start to drastically conserve power, and therefore clock speed, to stay at this temperature threshold. You’ll notice that the 5800X3D experiences this high temperature earlier than the 5800X and it drops to a lower clock speed than the 5800X before the end of the test.

Also, the 5800X3D stayed at 90C while drawing 86W at its lowest point, but the 5800X drew 110W at its lowest point at the same 90C. This shows that the 5800X3D does not dissipate as much heat as the 5800X in the same temperature range. In other words, all else being equal, the 5800X3D will run hotter than the 5800X under the same conditions. Even with less power consumption, the 5800X3D is hotter than the 5800X.

But it should be clear: You should never experience these conditions during normal use, and the Ryzen 7 5800X3D runs perfectly fine within its specifications. The 5800X3D is much cooler than competing Core i9 processors.

Heatsinks are one of the main points preventing high-performance 3D chips from going mainstream, but AMD has done an amazing engineering job in controlling enough temperatures to deliver a chip that delivers excellent performance within the acceptable TDP threshold.

Our results certainly imply that heat dissipation will remain a serious challenge at higher power thresholds, and while AMD considers voltage to be the dominant limiting factor in the 5800X3D’s clock speeds and prohibits overclocking, there may be some room for that statement. Most take AMD’s claims to mean that heat is not an issue, although the company has also cited heat as a factor with 3D V-Cache at other times.

Again, voltage, frequency, and temperature are all interrelated. Higher clock frequencies require more voltage, but more voltage leads to more heat. A higher voltage could simply push the chip out of its comfortable thermal enclosure. Thus, the 3D V-Cache’s 1.35V limit could simply be a product definition designed to test temperatures against the chip’s heatsink design rather than an actual physical limit economics of technology itself.