AMD Radeon RX 6900 XT specs: Navi 21 GPU with 16GB could cost $499

I’m knee deep in reviewing NVIDIA’s new GeForce RTX 3080 graphics cards, and now the RTX 3090 to have a day one review, but the world hasn’t stopped.

We now have some new rumored specs on the AMD’s next-gen Radeon RX 6900 XT graphics card, where Rogame tweets it will be powered by the Navi 21 GPU and have 16GB of GDDR6 memory. The 16GB of GDDR6 memory is coming from Samsung, and will reportedly be on a 256-bit memory bus.

Not only that, but there is also the Navi 22 GPU that should power a slightly slower card that should materialize into the Radeon RX 6800 XT, or possibly the Radeon RX 6700 XT. This card will reportedly have 12GB of GDDR6 on a 192-bit memory bus.

There are others that think AMD will use 16GB of faster HBM2 memory on its Radeon RX 6900 XT, and then GDDR6 on the replacements for the Radeon RX 5700 XT, RX 5600 XT, and RX 5500 XT. This also makes sense, as it would make the new Big Navi / Radeon RX 6900 XT more like a “TITAN” series graphics card — a competitor for the GeForce RTX 3090 with its ultra-insane 24GB of ultra-fast GDDR6X memory.

NVIDIA’s new GeForce RTX 3090 drops on September 24.

AMD Radeon RX 6900XT leaks in new photos

New photos of alleged AMD Radeon RX 6900XT leak

A YouTuber RedGamingTech has leaked more photographs of the alleged Radeon RX 6900XT graphics card. It seems that the source of the photographs is the same as JayzTwoCents’s who also shared some of these photos yesterday.

It is worth noting that the graphics card name has not been confirmed. It is only alleged that the highest-end model will be called RX 6900XT. AMD is likely to unveil multiple graphics card SKUs based on RDNA2 architecture. The flagship model should feature Navi 21 GPU (known as Big Navi), while the rest of the stack could adopt Navi 22 or Navi 23.

The photos are likely showing a prototype as even the I/O vents are blocked, which is very unusual for a graphics card. Although the render of the graphics card in Fortnite game also lacked the venting holes.

The model is equipped with dual 8-pin power connectors. This configuration is also seen on a dual-fan model, which was revealed by JayzTwoCents yesterday.

AMD Zen3 on October 8th, Radeon RX 6000 coming October 28th.

AMD today announced it will host two separate events in October. The first event will cover Zen3 architecture and most likely Ryzen 4000 series processors codenamed Vermeer. This event is scheduled for October 8th. The name of the event is ‘Next Generation Ryzen Desktop Processors’.

On October 28th, AMD will showcase its Radeon RX 6000 graphics cards (the name has been confirmed in a tweet). These graphics cards are based on RDNA2 architecture. This event is called ‘Next Generation Radeon Graphics’.

The announcement takes place in late October, so we expect the availability of the new graphics cards somewhere in November, right before or shortly after the launch of next-generation gaming consoles, also based on RDNA2 architecture.

The Radeon RX 6000 series will feature graphics cards based on a so-called Big Navi processor. These graphics cards will compete with just announced (and not yet released) NVIDIA GeForce RTX 30 series codenamed Ampere. The performance is a big question, but AMD sounds confident it will deliver twice the performance per watt with the new architecture.

AMD:

Preparing to delight gamers globally with the next horizon of Radeon Graphics, we invite you to learn more about our RDNA 2 architecture, Radeon RX 6000 series graphics cards, and our deep collaboration with game developers and ecosystem partners who will help us bring the best of Radeon to gamers. Tune in for the reveal of the future of Radeon PC gaming at 12 p.m. ET, October 28th.

The company posted an RDNA2 teaser teasing a ‘breakthrough gaming architecture’.

The teaser confirms that Zen3 comes with dual CCD and a single I/O.

EVGA Launches Z490 Dark K|NGP|N Edition: Built for Overclockers

One of the biggest names on the extreme overclocking scene is Vince ‘K|NGP|N’ Lucido, who is highly regarded as one of the best sub-zero overclockers. He is a longtime employee of EVGA, and collaborates exclusively with its hardware team to develop unique overclocking-centric hardware. The latest collaboration is the EVGA Z490 Dark K|NGP|N – a limited edition motherboard with an 18-phase power delivery designed for overclocking. There is also more conventional features such as two PCIe 3.0 x4 M.2 slots, eight SATA ports, U.2 port, with a 2.5 G Ethernet controller and Wi-Fi 6.

One of the most interesting elements on the EVGA Z490 Dark is its unconventional desktop design, with just two memory slots mounted horizontally across the top, just above the transposed LGA1200 socket. The 90-degree angle is usually done for airflow reasons with specific chassis in mind, and only two memory slots assists in memory-latency driven workloads.

The memory slots are qualified up to 64 GB of DDR4-5000, although for this board it wouldn’t be a surprise to see overclockers push speeds further. The socket is also better angled for easier mounting of the large copper pots that sub-zero overclockers use with liquid nitrogen. Overclocking features for this sort of board usually extend to driving extra power, sometimes even external voltage control, as well as monitoring tools for accurate thermal and voltage measurements. 

For storage there is a pair of PCIe 3.0 x4 M.2 slots, with a single U.2 port and eight SATA ports with six of these supporting RAID 0, 1, 5, and 10 arrays, while an ASMedia SATA controller powers the other two ports. Three full-length PCIe 3.0 slots operate at x16, x8/x8, and x8/x8/+4, with a secondary half-length PCIe 3.0 x4 slot.

The top right-hand corner features two 2-digit LED panels that monitor CPU voltage and temperature, with a power and reset button also added. Focusing on networking, the Z490 Dark is using a pair of Intel-based Ethernet controllers, one I225-V 2.5 GbE and one I219-V Gigabit, with an Intel AX201 providing both Wi-Fi 6 and BT 5.1 device support. The audio codec looks ‘amped’ as well.

The limited-edition EVGA Z490 Dark K|NGP|N model is currently available for $600 direct from EVGA, with a limit of two per household. This is $100 more expensive than the standard EVGA Z490 Dark model ($500), although the K|NGP|N variant is likely to use perhaps better binned components to ensure extreme overclocking consistency. Both models come with a 3-year manufacturer warranty.

NVIDIA GeForce RTX 3080 Delivers Up To 100 FPS at 4K In Several AAA Gaming Titles With Max Quality & RTX On

A couple of hours ago, we shared how NVIDIA’s upcoming GeForce RTX 3080 absolutely demolishes the GeForce RTX 2080 Ti in Doom Eternal, delivering over 150 FPS at 4K resolution with all quality and image settings maxed out. Well, it looks like NVIDIA has shared even more performance metrics for its next-generation flagship and it looks like it’s a whole different beast compared to the RTX 2080 Ti.

NVIDIA’s GeForce RTX 3080 Is Going To Make Us All Forget The RTX 2080 Ti – Over 100 FPS In AAA Titles With Max Quality & RTX On at 4K

During the unveil, Jensen only shared the performance metrics in percentages versus the previous generation of cards but during the same Reddit Q&A session which we had a whole article on a few hours ago, NVIDIA’s director of GeForce product management, Justin Walker, shared additional FPS numbers for the GeForce RTX 3080 graphics card.

Answering to whether the NVIDIA GeForce RTX 3080 with its 10 GB memory would be sufficient to run next-generation AAA titles, Justin replied that the goal of their next flagship is to provide great performance at up to 4K resolution with all settings cranked to the max.

Justin even shared some game examples where he mentions that the GeForce RTX 3080 is able to deliver 60-100 FPS in several AAA titles at 4K resolution with everything maxed out (high-resolution textures and RTX On where applicable). The titles mentioned include Shadow of The Tomb Raider, Assassins Creed: Odyssey, Wolfenstein: Youngblood, Gears of War 5, Borderlands 3, Metro Exodus & Red Dead Redemption 2.

[Justin Walker] We’re constantly analyzing memory requirements of the latest games and regularly review with game developers to understand their memory needs for current and upcoming games. The goal of 3080 is to give you great performance at up to 4k resolution with all the settings maxed out at the best possible price.

In order to do this, you need a very powerful GPU with high speed memory and enough memory to meet the needs of the games. A few examples – if you look at Shadow of the Tomb Raider, Assassin’s Creed Odyssey, Metro Exodus, Wolfenstein Youngblood, Gears of War 5, Borderlands 3 and Red Dead Redemption 2 running on a 3080 at 4k with Max settings (including any applicable high res texture packs) and RTX On, when the game supports it, you get in the range of 60-100fps and use anywhere from 4GB to 6GB of memory.

Extra memory is always nice to have but it would increase the price of the graphics card, so we need to find the right balance. via Reddit

At least three of the titles mentioned here utilize NVIDIA’s RTX feature set. These titles are Metro Exodus, Shadow of The Tomb Raider, and Wolfenstein: Youngblood. Starting with Metro Exodus, our own testing delivered around 47 FPS with an RTX 2080 Ti with RTX enabled at the Ultra preset without any Gameworks features enabled. Considering that the NVIDIA says they have everything maxed out (including Gameworks effects), a 60 FPS+ performance is a stellar increase. NVIDIA does not mention if they used DLSS but our testing was without DLSS enabled and at native 4K res.

Moving on to Shadow of Tomb Raider, the game was once again tested at Ultra settings. Here, our RTX 2080 Ti delivers around 55 FPS on average with the Ultra preset and DLSS enabled. Percentile (99%) performance is around 44 FPS on average. We don’t know the exact FPS for the RTX 3080 but they could hover around 70-80 FPS which would mark a 45% increase in performance over the RTX 2080 Ti or a 90% jump over the RTX 2080 which would be incredible.

For Wolfenstein Youngblood, the performance at max settings is around 55 FPS with RTX On and about 80-85 FPS with DLSS enabled. An RTX 3080 given we get the same 40-50% performance jump over the RTX 2080 Ti which we are seeing in the other titles should get at least 100+ FPS with the same settings. Overall, the NVIDIA GeForce RTX 3080 is looking like a fantastic graphics card for 4K gaming with an even more fantastic price of just $699 US.

The RTX 3080 features some insane specifications such as 68 SM units which results in a total of 8704 CUDA cores. In addition to the CUDA cores, the card also comes packed with next-gen RT (Ray-Tracing) cores, Tensor cores, and brand new SM or streaming multi-processor units.

In terms of memory, the NVIDIA GeForce RTX 3080 comes packed with 10 GB of memory and that too the next-generation GDDR6X design. With Micron’s latest and greatest graphics memory dies, the RTX 3080 can deliver GDDR6X memory speeds of 19.0 Gbps. That along with a bus interface of 320-bit will deliver a cumulative bandwidth of 760 Gbps.

The NVIDIA GeForce RTX 3080 will be launching to consumers on the 17th of September, 2020. The reference Founders Edition variant will cost $699 US. The NVIDIA GeForce RTX 3080 will feature a price of $699 (MSRP) however custom models will vary depending on their design and the extra horse-power that they have to offer.

There aren’t any detailed performance numbers that NVIDIA is sharing right now but from what has been showcased, the GeForce RTX 3070 is faster than an RTX 2080 Ti, the RTX 3080 is a good bit ahead of the RTX 2080 Ti and the RTX 3090 is about as much as 50% faster than the RTX 2080 Ti which is very impressive for the full lineup stack. The NVIDIA GeForce RTX 3080 itself is twice as fast as the RTX 2080 and is considerably faster than the RTX 2080 Ti making it a perfect 60 FPS 4K gaming graphics card.

Intel Tiger Lake SuperFin-ishes AMD Ryzen 4000 In Latest Benchmarks

Shortly after leaking several Tiger Lake promotional videos, Twitter user WalkingCat is making headlines again. This time, the leaker has shared two new videos where the 10nm SuperFin-powered Core i7-1185G7 goes up against AMD’s Ryzen 7 4800U in several scenarios. The videos appear to be conducted by Intel, and the exact conditions of the test environment and the specifications of the two systems are unknown. For now, we recommend you exercise a bit of caution with the results and wait for our testing before you make a choice of which is the Best CPU. 

Intel Core i7-1185G7 vs. AMD Ryzen 7 4800U

The first video has the Core i7-1185G7 facing the Ryzen 7 4800U in multiple games. Apparently, Intel configured the Core i7-1185G7 to 28W (PL1), while the Ryzen 7 4800U was running on Extreme Performance Mode. We’ve seen the Extreme Performance Mode feature before in a Lenovo laptops before. The functionality essentially unlocks a higher thermal limit for the processor. Giving Intel the benefit of the doubt, the Ryzen 7 4800U in the chipmaker’s comparison should have had a 25W ceiling.

The results showed the Core i7-1185G7 surpassing the Ryzen 7 4800U in iGPU gaming, which is something we’re not accustomed to see from Intel. Back in June, Intel employee Ryan Shrout demoed a Tiger Lake-powered laptop running Battlefield V on solely integrated graphics. These latest benchmarks seemingly confirms Xe LP’s graphical prowess.

The general gist of the second video is that the Core i7-1185G7 practically outperformed the Ryzen 7 4800U in all the tasks that Intel threw at it. The workloads included video edition, 4K video playback, export times, among others. According to the videos, the margins were pretty substantial – the Tiger Lake chip completed certain activities up to twice as fast.

To put things into context, the Ryzen 7 4800U (codename Renoir) is an eight-core, 16-thread processor with a 8MB L3 cache. The Zen 2 APU, which is a product of TSMC’s 7nm FinFET manufacturing process, runs with a 1.8 GHz base clock and 4.2 GHz boost clock. On the graphics side of things, the Ryzen 7 4800U sports eight Vega Compute Units (CUs) at 1,750 MHz. The Ryzen 7 4800U comes with a cTDP (configurable thermal desin power) rating that spans between 10W and 25W.

The Core i7-1185G7 (codename Tiger Lake) is unreleased, however, the processor has surfaced multiple times over the past months. Thus far, we know that the Core i7-1185G7 has a four-core, eight-thread configuration that’s complemented with 12MB of L3 cache. The quad-core chip comes out of Intel’s 10nm SuperFin furnace and packs Willow Cove cores and Xe LP graphics. The most recent Geekbench 5 submission detected the Core i7-1185G7 with a 3 GHz base clock and 4.8 GHz boost clock. The processor’s iGPU is reportedly comprised of 96 Execution Units (EUs) at 1,550 MHz. The cTDP for the Core i7-1185G7 ranges between 15W and 28W.

It’s too early to proclaim Intel as the undisputed winner, though. As we’ve learned in the past, manufacturer benchmarks aren’t exactly the most accurate. From what we’ve seen so far, Tiger Lake is absolutely a game changer. Don’t forget to tune in tomorrow for the official launch.

Sabrent’s new PCIe 4 SSD is absurdly fast, beats out Samsung’s 980 Pro

The world of PCIe 4 SSDs already has plenty of speed demons, but Sabrent just took the crown. The company’s new Rocket 4 Plus SSD sports the fastest combination of read and write speeds currently available, beating out even Samsung’s latest 980 Pro PCIe 4 NVMe drive.

Both drives would probably make any list of the best SSDs, but Sabrent manages to just squeeze out the edge. The Rocket 4 Plus matches the Samsung 980 Pro’s 7,000 MB/s read time, but it surpasses Samsung’s drive with a 6,850 MB/s write speed. The Samsung drive, for comparison, reaches maximum write speeds of 5,000 MB/s.

According to Sabrent’s Rocket 4 Plus product page, it will be available in 2TB, 1TB, and 500GB capacities. That’s a far cry from the largest PCIe 4 SSD out there, a 4TB monster also made by Sabrent. However, in exchange for the lower capacities, you’re getting vastly superior maximum read and write speeds. Sabrent also shrouds the Rocket 4 Plus in a beefy custom heatsink to keep the drive running cool.

Sabrent hasn’t provided a release date or pricing information yet, but we’ll likely hear more soon.

Hyper X Predator DDR4 3600MHz 2x16GB Memory Review

Kingston and the Hyper X Predator family have a long-running history of uncompromising high performance. Their 2020 lineup of the Hyper X Predator DDR4 kits is no exception. Aesthetically speaking not much has changed since this is more of a refresh than a new launch series and new XMP rated kits have been added running all the way up to DDR4 4800.

For this review, Hyper X sent over their Hyper X Predator DDR4 3600 CL17 32GB (2x16GB) kit. We will be testing it out on the X570 platform using the Ryzen 9 3900X CPU as the new lineup of Ryzen 3000 Series CPUs are handling high clock speeds so well, and we changed out our test bench.

A Closer Look

The kit ditches the standard blister pack, but instead, quite the fancy box is accompanying the memory kit. The box is light on actual information for the memory kit with the front displaying the XMP rating as well as the capacity and DIMM count. The back of the box has a single sentence repeated in various languages with the model sticker displaying the capacity, XMP rating, and latency rating. Inside includes a case badge for those who like to use those kinds of things.

Sporting a thick black heatsink shroud with a two-toned aluminum finish for the heatsink with and brushed aluminum lettering and branding of the DIMMs. Sitting on top of the modules is the diffuser for the RGB modules that have the Hyper X logo printed onto it, so if you disable the lighting the module still looks attractive because sometimes you just have days where you want to disable the lights.

RGB is supported by their own Hyper X Ngenuity software as well as the more popular suites from vendors including; ASUS Aura Sync, ASRock’s, Gigabyte’s RGB Fusion 2.0, and MSI Mystic Light Sync. I found it integrates nicely with the ASUS Aura Sync on the ASUS TUF Gaming X570 tesh bench and the Hyper X Ngenuity software handled the RGB settings nicely on the EVGA Z370 motherboard as well.

Testing Setup

Our Memory test bench is based on the ASUS TUF Gaming X570 Pro-Plus and a Ryzen 9 3900X with a T-Force Cardea PCIe Gen 4.0 NVMe storage drive. Our tests are primarily synthetic tests based on real-world memory-intensive programs from Aida64 for measurements to compression/decompression workloads. One of the things we wanted to touch on with the Ryzen 3000 Series CPUs was to not just show dual CCD Ryzen 9 3900X performance but also show single CCD Ryzen 5 3600X performance as well. The biggest reason for this is that the single CDD processors will behave differently in the latency and write tests. So to ‘simulate’ this we disabled one fo the CCDs on the Ryzen 9 3900X and kept everything else the same, just something to note through the results

ComponentModel
CPUAMD Ryzen 9 3900X @ 4.3GHz All Core
MotherboardASUS TUF Gaming X570 Pro-WiFi
PSUCooler Master V1200P
OSWindows 10-64 Bit
GPUNVIDIA GeForce RTX 2080 Ti FE
CaseLian Li T70X

The Hyper X Predator kits are making use of the latest Micron J-die memory that was recently released. As of our last checking, you won’t find these in the famous Ryzen DRAM Calculator as they’re just that new. Compared to the Viper Black RGB’s Samsung B-die modules it’s clear that the Micron J-die is quite Ryzen friendly.

Hyper X Predator DDR 4 3600MHz 2x16GB

Aida 64 Bandwidth

Using Aida 64 we are able to get a good idea of the overall bandwidth from the memory kits in Read, Write, and Copy performance. I separated them out into their own charts per each measurement so they would be easier to read and digest. A great aspect of this benchmark is that it is readily available to everyone to try out and see where your system’s performance lands for a point of comparison.

Aida 64 Latency

Latency is the name of the game in this benchmark. The lower the latency the faster the response from the memory.

Zip Benchmark

7-Zip is a very popular and free compression/decompression software tool that is used to zip and unzip files. We tested

WinRar

WinRar’s benchmark takes a measurement of the system’s capacity for file compression and translates that to KB/s showing that it really favors overall speed for the highest output. A great aspect of this benchmark is that it is readily available to everyone to try out and see where your system’s performance lands for a point of comparison.

Far Cry New Dawn

Far Cry New Dawn is still a title that is known to put the hurt on the CPU so it should also share a very similar outcome. While the performance delta is staggering, it’s still a big enough gap for people to really consider faster memory when using these higher-performing CPUs. This was tested at 720p Low Preset paired with an RTX 2080 Ti, just a bit of info on the results is that they’re entire CPU bound and this is allowing the impact of the memory speeds show their advantage.

Quick Overclock

Title:Aida 64 Read MB/s

Labels:R9 3900X XMP,R9 3900X 3733,R5 3600X XMP,R5 3600X 3733

HyperX Predator DDR4 3600 CL17:55228,57501,50812,52417

[/chart]

Aida 64 Latency

Latency is the name of the game in this benchmark. The lower the latency the faster the response from the memory.

Conclusion

At the end of it all Kingston’s Hyper X Predator RGB DDR4 3600 32GB (2x16GB) kit is a pretty well-rounded kit of memory. It offers solid and comparable performance at it’s XMP rated speeds but does fall a bit short if you have to run it at lower frequencies like DDR 3200. If that’s the case just tighten up the timings to regain some of the performance lost and you’ll be back in line with competing kits.

The aesthetics and lighting can easily be found complimentary of almost any build. The side profile is a bit garish but not overly done so it blends in well enough. If the RGB isn’t your thing I would either tell you to turn it off or steer you into the direction of the Hyper X Predator Black lineup and go sans RGB altogether.

The pricing is a bit on the higher side, but the Predator lineup has never been known as the ‘budget’ option, if you want that I encourage you to look to the Fury lineup if Hyper X is where you want to land. But, more interestingly here is that I can’t seem to find this particular kit available for sale anywhere at the time of wrapping up this review, I did find the lower 16GB (2x8GB) capacity DDR4 3600 on Amazon for a respectable $125.99 and the 32GB (2x16GB) DDR4 3200MHz for $180.33.

Intel Claims Its Cheaper To Build A Faster Gaming PC With Its 10th Gen Core CPUs Than AMD’s Ryzen 3000 CPUs, Pits The Core i7-10700K Against Ryzen 9 3900XT

Intel has released a new set of ‘real-world’ performance benchmarks that compare its 10th Gen Desktop CPUs to the AMD Ryzen 3000XT lineup. The company specifically compares the gaming performance of its chip against its competitors and claims that they offer the best performance package at a cheaper price.

Intel Claims Their Core i7-10700K Is Not Only Faster in Gaming But Also Better Value Than AMD’s Ryzen 3000XT CPUs

The new ‘real world’ performance slides were showcased during a presentation to the APAC regional market. While Intel’s own CEO has stated that the industry should move on from the generic benchmarks showcase, the recent decline in desktop processor shipments has led the company to win over the minds of consumers in the mainstream and high-performance CPU segments with a new set of comparisons. We won’t start with how much controversy the real performance benchmark slides have been marked with since their first outing but recently, the blue team has started using similar priced and similar spec’d configurations that do lead to better representation of real-world usage.

Intel Core i7-10700K 8 Core vs AMD Ryzen 9 3900XT 12 Core CPUs Compared

The comparison consists of the Intel Core i7-10700K and the AMD Ryzen 9 3900XT. The Intel Core i7-10700K is an 8 core and 16 thread CPUs with a base clock of 3.80 GHz, boost clock of up to 5.10 GHz, 20 MB of cache, and a TDP of 125W (229W PL2). The AMD Ryzen 9 3900XT is a 12 core and 24 thread CPU with a base clock of 3.80 GHz, boost clock of up to 4.70 GHz, 70 MB of cache, and a TDP of 105W.

Intel says that their Core i7-10700K CPU is cheaper with a starting price of $387 US ($409.99 US at Newegg) while the AMD Ryzen 9 3900XT CPU is expensive at $499 US ($479.99 US at Amazon). The standard Ryzen 9 3900X which is not featured in these performance metrics costs $429.99 US that is a far better deal than the Ryzen 9 3900XT which offers up to 5% better performance but for a 10 percent price difference.

Intel used similar test setups which featured an NVIDIA GeForce RTX 2080 Ti graphics card and 16 GB of DDR4 memory. The memory on the Intel platform was running at 2933 MHz speeds while the AMD platform was running 3200 MHz DIMMs. Now there’s no mention of which motherboard both setups were using but considering that X570 and Z490 boards end up costing around the same, the price difference would be close with Intel offering a slightly better value versus the Ryzen 9 3900XT.

The performance benchmarks show that the Intel Core i7-10700K outperformed the AMD Ryzen 9 3900XT in several AAA titles. The Intel CPU was up to

23% faster in Total War: Warhammer 2

17% faster in Leauge of Legends

15% faster in Monster Hunter World: Iceborne

14% faster in Rocket Leauge

12% faster in War Thunder Tank Battle

6% faster in Total War Three Kingdoms

In addition to the games where the Core i7 scored a double-digit win against the Ryzen 9 CPU, Intel also showcased several other titles where their CPU was more than 3% faster than the 3900XT. The lead was seen in 24 out of the 30 titles tested. In the remaining tests, the Ryzen 9 3900XT secured a lead and surprisingly, the AMD CPU scored better than the Intel part in CSGO which is one of the top played eSports titles. All performance tests were performed on a 1080p resolution. The slides were shared by Benchlife however the exact appendix for workload and configurations isn’t shown.

Sure Intel has a solid lead in gaming performance which was the whole purpose of this comparison but a 12 core and 24 thread CPU also has several other benefits than just gaming. The AMD Ryzen 9 3900XT can deliver better performance to game streamers and for general productivity tasks and professional uses which many gamers these days also run simultaneously on their desktop PCs.

Its an added benefit from the several cores that Ryzen CPUs provide and while Intel leads the gaming benchmarks, the gap isn’t huge, especially if we compare the standard Ryzen 9 3900X CPU which costs pretty much identical to the Core i7-10700K in the retail segment. If you’re purely building a gamer-focused machine, then Intel CPUs do retain the advantage but the real question is how long will this advantage reside with Intel with Zen 3 fast approaching the desktop PC market.

AMD Releases the A520 Chipset: For Ryzen on a Budget

AMD has officially lifted the lid on its latest entry-level chipset designed for its Zen 2-based Ryzen 3000 and 4000 processors, the A520 chipset. Following AMD’s mid-tier B550 chipset launch a few months back, A520 brings up the bottom of AMD’s now fully-modernized chipset stack, dialing down on things like PCIe speeds to allow its board partners to build more budget-friendly Ryzen 3000/4000 motherboards.

Overall, A520 offers up to twenty-six available PCIe 3.0 lanes available when paired with a Ryzen 3000 processor, with twenty of these coming directly from the CPU. And, wasting no time, numerous vendors, including ASUS, ASRock, GIGABYTE, MSI, and Biostar have already unveiled some of their entry-level A520 boards.

A520 Chipset, The A-Team For Value

The Zen 2 architecture, which is the foundation of AMD’s latest Ryzen series of processors, has been its most popular series of processors to date. One of the marquee features of the B550 and X570 chipsets is PCIe 4.0, which, although it hasn’t been fully utilized in devices like video cards and storage at present, it is still a popular feature that users hope will future-proof their systems for years to come. The A520 chipset, in turn, drops support for PCIe 4.0 speeds entirely in favor of PCIe 3.0, which is relatively easier to wire for and gives extra scope for the motherboard vendors to save costs compared to B550 and X570. The chipset officially represents AMD’s now current-generation entry-level pathway, and as always, is based around AMD’s AM4 socket.

Overall, the limitation in PCIe speeds for A520 is board-wide. Along with the chipset itself being limited to PCIe 3.0 support for both downstream and upstream, Ryzen processors will cap their on-chip PCIe lanes to PCIe 3.0 speeds when paired with an A520 motherboard. Further cutting PCIe-related costs, with A520 AMD offers less scope for lane allocation compared to B550 and X570, which allows vendors reduce the number of components required. There are only a couple of avenues in which vendors can configure their boards, such as selecting between using two PCIe 3.0 x1 slots or two SATA ports. 

Meanwhile, A520 also supports a couple of different configuration options with the 4 high-speed storage lanes coming from the host CPU. This includes building a full-fledged PCIe 3.0 x4 M.2 slot, two PCIe 3.0 x2 M.2 slots, or a PCIe 3.0 x2 M.2 slot with two additional SATA ports.

CPU x4 PCIe 3.0 NVMe

x2 PCIe 3.0 NVMe + 2 x SATA

x2 PCIe 3.0 NVMe + x2 PCIe 3.0 NVMe

Chipset 4x PCIe 3.0 + 2x PCIe 3.0

4x PCIe 3.0 + 2 SATA

The remaining PCIe lanes from the CPU, as always, are divided between a full-length PCIe 3.0 x16 slot, and a PCIe 3.0 x4 link between the CPU and the chipset.

As for USB connectivity, A520 offers support for up to four USB 3.2 G2 ports from the CPU. Meanwhile the chipset itself adds another USB 3.2 G2 port, two USB 3.2 G1 ports, and six USB 2.0 ports.

Processor Support

The A520 chipset has identical processor support to that of the mid-range B550 chipset, with official support for AMD’s current Ryzen 3000 desktop processors as well as AMD’s next-generation Ryzen 4000 processors. Like with the B550 chipset, some A520 boards come with video outputs, which all but confirms support for AMD’s Ryzen 4000 APUs – especially seeing as how A520 doesn’t officially support the earlier Ryzen 3000 or 2000 series APUs.

The AMD Ryzen 3000 CPU stack starts at the Ryzen 3 3100 with four cores and eight threads, which is currently available for $99. This would be a very suitable processor to pair with an A520 motherboard, though more aspiring builders can use any of of the Ryzen 3000 desktop CPUs, including the goliath sixteen core Ryzen 9 3950X.

The Current A520 Motherboard Stack

Accompanying AMD’s announcement of the A520 chipset, motherboard vendors have begun rolling out their respective A520 boards. The main form factor of choice for A520 thus far is Micro-ATX, with a couple of Mini-ITX boards including the ASRock A520M ITX/ac and the GIGABYTE A520I AC. Gigabyte also has the only full-sized ATX model announced so far, which is the GIGABYTE A520 Aorus Elite.

A520 boards set to hit the market include some gaming themed models that primarily come from GIGABYTE with its Aorus models, as well as the ASUS TUF Gaming A520M-Plus. The solitary offering from Biostar is aimed at business and casual users with its A520MH V6.0.

The most feature-packed A520 model looks to be the GIGABYTE A520 Aorus Elite with four memory slots that support up to 128 GB of DDR4-4400 memory, a Realtek ALC1200 HD audio codec and an unspecified Realtek Gigabit Ethernet controller. It has two full-length PCIe 3.0 slots that operate at x16/x2, with three PCIe 3.0 x1 slots and a single PCIe 3.0 x4 M.2 slots with four SATA ports.

At present, we don’t have MSRP pricing from any of the vendors, but we have reached out to each vendor individually, and we will update the above chart once we have the official pricing. It is likely that the vast majority of A520 boards will be below $150 due to B550 models starting at that price point. The A520 chipset is its most basic entry-level chipset for Ryzen 3000 processors.

Finally, today’s A520 release is a hard launch, with Newegg and other retailers already offering boards for sale. Not every board vendor has their wares on shelves just yet – as is usually the case for low-end board launches, this is a very relaxed launch – but over the coming days and weeks we should see the remainder of the A520s trickle into retail shops.