ASRock has announced its new series of ultra-compact form-factor (UCFF) PCs that combine small dimensions with performance of desktop APUs from AMD. ASRock’s Jupiter X300-series is only slightly larger the company’s Mars 4000U-series machines launched last November, but it clearly packs considerably more performance and features than its smaller brother.
ASRock’s Jupiter X300 barebones PC uses AMD’s easy-to-find socketed Ryzen 2000/3000/4000 APU with up to eight cores, built-in Radeon Vega graphics as well as an up to 65W TDP cooled using a copper heatsink and a high-performance blower. The APUs can be paired with up to 64GB of DDR4 3200 MHz memory (using two SO-DIMM modules), an M.2-2280 SSD with a PCIe 3.0 x4 or SATA interface, and a 2.5-inch/9.5-mm drive.
The manufacturer says nothing about compatibility of its Jupiter X300 systems with AMD’s not-yet-announced Ryzen 5000-series ‘Cezanne’ APUs, but the machines are based on a rather outdated AMD X300 chipset. Meanwhile, since the motherboard uses a proprietary form-factor, it will prove tricky to upgrade.
(Image credit: ASRock)
The Jupiter X300 chassis measures 178 × 178 × 34mm, so not as compact as a NUC, but not too far away from NUC-like dimensions. The system — which is actually smaller than Apple’s MacMini — can be attached to VESA mounts of a display or used on the desktop.
(Image credit: ASRock)
Connectivity department of ASRock’s Jupiter X300 is quite advanced. The machine can be equipped with Intel’s AX200 Wi-Fi 6 + Bluetooth 5 module (or any other adapter if you buy it separately), it has one Gigabit Ethernet port, three display outputs (a DisplayPort 1.2, an HDMI 2.0, and a D-Sub to support legacy monitors), two 3.5-mm audio connectors, and eight USB ports (two USB 3.2 Gen 1 Type-C, four USB 3.2 Gen 1 Type-A, two USB 2.0), which might be a record for an UCFF PC.
(Image credit: ASRock)
ASRock positions its Jupiter X300-series machines for home, office, and enterprise customers. Since the latter often require remote management and advances security, ASRock offers them its X300-serie DASH barebones with a DASH-enabled LAN, trusted platform module 2.0, an anti-intrusion sensor. Obviously, to use remote management and other advanced features of AMD’s business platform, the systems have to be equipped with the company’s Ryzen Pro-series APUs.
ASRock did not announce MSRPs of its Jupiter X300-series barebones PC.
AMD reported its fourth-quarter 2020 and full-year results today, setting new records as it continues to whittle away market share from its archrival Intel despite the challenges associated with the pandemic and US-China trade war. AMD is making strides in all facets of its operation; CPU, GPU, data center, and console sales are all strong, but exceptional demand driven by the pandemic has led to rolling shortages.
AMD CEO Lisa Su said the shortages primarily impact the gaming (both consoles and gaming CPUs and GPUs) and the low end of the PC market, and overall demand has exceeded its planning. That’s particularly challenging given that, in the first quarter of availability, the company’s Ryzen 5000 processors have sold twice the number of units compared to any other Ryzen series in history.
Su expects to see ‘tightness’ throughout the first half of 2021 until added production capacity comes online. That means we could see a limited supply of AMD’s PC and console chips until the middle of the year. Su also explicitly called out the low end of the PC market and consoles as being impacted the most, meaning higher-margin products are more readily available at retail.
Interestingly, AMD’s inventory swelled during the quarter, reaching $1.4 billion in unsold goods (up from $930 million the prior quarter), which seems odd given that the company has unmet demand. An AMD representative tells us that the company’s inventory values can consist of chips in various stages of production, so it’s possible that AMD’s increased wafer starts at TSMC has resulted in more chips in various stages of production. AMD also confirms that packaging shortages continue to be a pinch point in the supply chain. That implies that there could be plenty of chip dies also awaiting packaging, meaning lithographic capacity at TSMC (i.e., wafer starts) aren’t the only factor impacting the company’s ability to deliver chips.
If there’s one certainty, it’s that AMD is selling every chip it can put on shelves. AMD raked in a record $3.24 billion in revenue in the fourth quarter, up 53% over the year prior, and net income weighed in at $1.78 billion, a staggering 948% increase over the prior year, albeit buoyed by an income tax benefit of $1.3 billion.
Image 1 of 2
(Image credit: AMD )
Image 2 of 2
(Image credit: AMD )
On the client group side of the house that produces both consumer CPUs and GPUs, AMD’s $1.96 billion in Q4 revenue marked an 18% YoY and quarterly increase. AMD says higher sales of Ryzen processors drove this performance and average selling prices (ASPs) increased during the quarter, though ASPs lagged the prior year due to a higher mix of Ryzen Mobile sales.
That higher mix of laptop chips isn’t surprising – AMD recently recorded its highest laptop market share in history. AMD also says that its Ryzen 5000 processors doubled the launch sales of any prior-gen Ryzen processor. As a result, annual processor revenue grew 50% even though the PC market only grew 13%, meaning the company gained more market share during the year.
On the graphics side of the house, Su said that demand for Radeon 6000 GPUs is strong, marking the fastest-selling GPU over $549 in AMD’s history, and ASPs are up for both the quarter and the year. The company will launch RDNA 2 GPUs in the first half of 2021.
For the company’s EESC unit, which comprises its data center and semi-custom business, AMD raked in $1.38 billion in revenue, a 176% year-over-year (YoY) increase. Su said that the Sony PS5 and Microsoft Xbox ramp is faster than the previous cycle, which seems to be a given in light of the constant shortages. AMD says that it expects sales of its console chips to stay strong in the first half of the year, bucking the normal historical trend of reduced console sales in Q1.
AMD lumps both data center and game console processors under this same unit, making it hard to determine how much of the revenue gain can be attributed to its server chip business. However, during the earnings call, Su noted that server processor revenue reached an all-time high in the fourth quarter and now comprises “high mid-teens” percentage of the company’s overall revenue (for both the quarter and the year), implying AMD raked in ~$550 to ~$600 million in revenue for EPYC server chips for the quarter, and ~$1.4 billion for the year. That’s a big step forward but still lags behind Intel’s $6.1 billion in data center revenue in 2020.
AMD did note that its EPYC ASPs were higher sequentially, meaning the company is making more money per chip. Meanwhile, Intel has taken a haircut on ASPs, and thus margins, as it has cut pricing to fend off AMD’s EPYC.
AMD began production of its next-gen EPYC Milan in the fourth quarter of last year and sampled to HPC and cloud providers. Su said the company is on track for the official launch in March with very strong ecosystem support.
(Image credit: AMD )
AMD’s Q4 gross margins were flat YoY at 45%. For the next quarter, AMD guides for $3.2 billion in revenue, up 79% YoY and down 1% on the quarter.
AMD’s full-year results were exceptional, as well. The company raked in a record $9.76 billion in revenue, a 45% YoY jump, and net income hit $2.5 billion, a staggering 630% increase over the prior year.
Image 1 of 32
(Image credit: AMD)
Image 2 of 32
(Image credit: AMD)
Image 3 of 32
(Image credit: AMD)
Image 4 of 32
(Image credit: AMD)
Image 5 of 32
(Image credit: AMD)
Image 6 of 32
(Image credit: AMD)
Image 7 of 32
(Image credit: AMD)
Image 8 of 32
(Image credit: AMD)
Image 9 of 32
(Image credit: AMD)
Image 10 of 32
(Image credit: AMD)
Image 11 of 32
(Image credit: AMD)
Image 12 of 32
(Image credit: AMD)
Image 13 of 32
(Image credit: AMD)
Image 14 of 32
(Image credit: AMD)
Image 15 of 32
(Image credit: AMD)
Image 16 of 32
(Image credit: AMD)
Image 17 of 32
(Image credit: AMD)
Image 18 of 32
(Image credit: AMD)
Image 19 of 32
(Image credit: AMD)
Image 20 of 32
(Image credit: AMD)
Image 21 of 32
(Image credit: AMD)
Image 22 of 32
(Image credit: AMD)
Image 23 of 32
(Image credit: AMD)
Image 24 of 32
(Image credit: AMD)
Image 25 of 32
(Image credit: AMD)
Image 26 of 32
(Image credit: AMD)
Image 27 of 32
(Image credit: AMD)
Image 28 of 32
(Image credit: AMD)
Image 29 of 32
(Image credit: AMD)
Image 30 of 32
(Image credit: AMD)
Image 31 of 32
(Image credit: AMD)
Image 32 of 32
(Image credit: AMD)
For the full-year 2021 guidance, AMD pegs revenue to grow ~37% (~$13B) and non-GAAP gross margins to weigh in at 47%, driven by growth across all of AMD’s business segments.
AMD also says it is on track to complete its Xilinx acquisition, which should be complete by the end of 2021.
Custom Radeon RX 6900 XT graphics cards are slowly invading the hardware market, and Gigabyte has just announced a couple of its own. The Aorus Radeon RX 6900 XT Master 16G and Radeon RX 6900 XT Gaming OC 16G are designed to release Big Navi’s true power.
The Aorus Radeon RX 6900 XT Master 16G comes with a robust cooling system to keep Big Navi in check. According to Gigabyte, the heat dissipation area on this generation has increased in comparison to the previous generation. A total of seven composite copper heat pipes carry the heat away from the graphics card’s Navi 21 die and surrounding memory chips to the huge heatsink. Given the size of the graphics card, the Aorus Radeon RX 6900 XT Master 16G employs a triple-fan setup to ensure maximum airflow.
As a bit of eye candy, the Aorus Radeon RX 6900 XT Master 16G is equipped with a cool LCD screen on one side of the shroud, which allows consumers to display things like text, images or GIFs. The graphics card’s product page hasn’t gone live yet, so its full specifications and dimensions are unknown.
The Radeon RX 6900 XT Gaming OC 16G, which sports dimensions of 286 x 118 x 58mm, utilizes Gigabyte’s WindForce 3X cooler. Unlike the Aorus model, the Gaming OC only has six composite copper heat pipes for heat dissipation. The trio of semi-passive 80mm cooling fans spin in opposite directions that, according to Gigabyte, helps reduce the turbulence generated between adjacent fans and increase air pressure.
As far as specifications go, the Radeon RX 6900 XT Gaming OC 16G’s game and boost clock peaks at 2,050 MHz and 2,285 MHz, respectively. In comparison to the reference edition, Gigabyte’s iteration has a 1.7% higher game clock and 1.6% higher boost clock.
The Radeon RX 6900 XT Gaming OC 16G sips its juice from three 8-pin PCIe power connectors and the Aorus Radeon RX 6900 XT Master 16G in all likelihood features a similar configuration. Although we can’t speak for the latter yet, the first offers two DisplayPort 1.4a outputs and two HDMI 2.1 ports.
Both the Aorus Radeon RX 6900 XT Master 16G and Radeon RX 6900 XT Gaming OC 16G have two BIOS profiles. The switch to alternate bewteen the silent and OC profile is conveniently placed on the side of the graphics card. A reboot is all that’s necessary to get the profile up and working.
Intel’s Xe Graphics has been out for a while now, in the form of 11th Gen Tiger Lake laptops, but the discrete variant (Iris Xe Max) has only been in a few notebooks. Now, Intel is finally releasing its Iris Xe for desktops, but they’re not what you would expect, nor can you just go to the shop and buy them.
The GPU in question is the renowned DG1, similar to what we found in the Tiger Lake laptop we tested last year. Except, for the discrete graphics cards for desktops, Intel has seemingly cut the GPU down to 80 Execution Units for 640 cores, and it’s wired to 4 GB of memory. It’s not clear what kind of memory the GPU is wired to (Iris Xe Max used LPDDR4), but it does have its own dedicated pool, unlike the Xe Graphics integrated into Tiger Lake CPUs.
With this design handed over to Asus and Colorful, the following GPUs were conjured:
Image 1 of 2
(Image credit: Intel)
Image 2 of 2
(Image credit: Intel)
Asus’s flavor of the DG1 comes passively cooled, and Colorful’s card comes with a cheap-looking dual-fan cooler. Neither of the cards have PCIe power connectors, so it’s clear these are low-power chips. We haven’t found exact TDPs yet, but with cooling solutions like these and a cut down chip design in the desktop parts, expect a sub-30W TDP. The cards both have three display outputs: DVI, HDMI, and DisplayPort.
The chips are manufactured on Intel’s 10nm SuperFIN fabrication process, and performance is supposed to be pretty decent at 1080p. Our tests on the mobile platform confirmed this, and chances are that the desktop parts run at higher clock speeds to make up for the deficit in EUs — but exact details aren’t available and neither Asus nor Colorful have working product pages yet.
(Image credit: Intel)
OEM Only — How Bad Is It Really?
However, as much as everyone has been jumping at the thought of Intel joining the discrete high-end GPU market for gamers, that isn’t happening yet. That’s no surprise given what we know about Xe Graphics, as the high-end variants are mostly aimed at HPC and data centers and have not yet launched.
At this level, the DG1 level, Intel is only selling the GPUs directly to system integrators to put in prebuilt desktops — think parties such as Dell and HP. That might seem like a bummer, but let’s be honest: If you’re a gamer, would you really be jumping at the performance offered by the DG1 GPU? Probably not, so we doubt the DIY space will miss the DG1. We’ll keep our fingers crossed for the HPG (High Performance Gaming) Xe solutions in the future.
As a result of this OEM-only strategy, we also don’t know what pricing will look like. Availability is slated for Q1, and this looks like a half-hearted move by Intel at best. Iris Xe Max hasn’t been particularly impressive, and unless Intel has some ace up its sleeve, the desktop DG1 will be pretty anemic compared to competing solutions from AMD and Nvidia.
Then again, even budget graphics cards are currently sold out or selling at inflated prices. Nvidia’s GTX 1050 Ti from over four years back starts at around $170, and the newer GTX 1650 is selling at $300+ on Newegg. Similarly, the Radeon RX 570 that was selling for $120-$130 for over a year now starts at $290. If Intel’s DG1 can keep up with a GTX 1050 for $150, maybe it has a chance?
The first alleged independent benchmark results of AMD’s recently introduced eight-core Ryzen 9 5980HS “Cezanne” laptop processors have been published. AMD’s Zen 3-based chip uses integrated Radeon graphics, and, according to the new numbers, beats its predecessor and Intel 10th Gen Comet Lake in single- and multi-core workloads, as well as 11th Gen intel Tiger Lake in single-core. However, there is a processor that still beats AMD’s Cezanne.
Hardware enthusiast @Tum_Apisak found two Geekbench 5 results from the Asus ROG Flow X13. The gaming notebook runs the eight-core Ryzen 9 5980HS at a 3.30 GHz default clock speed and can boost it all the way to a 4.53 GHz. In one case, AMD’s Cezanne APU hit a 1,532 single-core score and 8,219 multi-core score. In another case, the processor finished with 1,541 single-core points and 8,224 multi-core points.
CPU
Single-Core
Multi-Core
Cores/Threads, uArch
Cache
Clocks
TDP
Link
AMD Ryzen 9 5980HS
1540
8,225
8C/16T, Zen 3
16MB
3.30 ~ 4.53 GHz
35W
https://browser.geekbench.com/v5/cpu/6027200
AMD Ryzen 9 4900H
1230
7,125
8C/16T, Zen 2
8MB
3.30 ~ 4.44 GHz
35~54W
https://browser.geekbench.com/v5/cpu/6028856
Intel Core i9-10885H
1335
7,900
8C/16T, Skylake
16MB
2.40 ~ 5.08 GHz
45W
https://browser.geekbench.com/v5/cpu/6006773
Intel Core i7-1185G7
1550
5,600
4C/8T, Willow Cove
12MB
3.0 ~ 4.80 GHz
28W
https://browser.geekbench.com/v5/cpu/5644005
Apple M1
1710
7,660
4C Firestorm + 4C Icestorm
12MB + 4MB
3.20 GHz
20~24W
https://browser.geekbench.com/v5/cpu/6038094
Typically, Cezanne looks very good compared to previous-generation AMD and Intel architectures. The most interesting comparison we can make with a Zen 3 APU is with an Intel Willow Cove processor. Since Intel hasn’t launched its eight-core Tiger Lake-H chips yet, quad-core Core i7 1100-series “Tiger Lake-U” processors are the only available CPUs featuring the Willow Cove microarchitecture. These CPUs are not quite meant for gaming machines and, therefore, come inside notebooks with less sophisticated cooling.
Generally, Intel Core i7-1185G7-based machines score 1,350-1450 single-core points on Geekbench 5. A well-cooled example can hit around 1,550 on a single core and about 5,600 on multi-cores.
Therefore, it looks like mobile CPUs featuring AMD’s Zen 3 and Intel’s Willow Cove cores have comparable single-core performance (assuming that both are cooled properly). Naturally, AMD’s eight-core gaming APU naturally beats Intel’s quad-core CPU in workloads leveraging multiple cores.
As far as Geekbench 5 results go, AMD’s Ryzen 9 5980HS looks like a very potent mobile APU with a 35W TDP. Yet, it’s not unbeatable.
Apple’s tiny M1 system-on-chip (SoC) running at 3.20 GHz scored 11% better than the Ryzen 9 5980HS in single-core workloads and 7% worse in multi-core workloads while consuming about 30% less power, assuming that its TDP is up to 24W.
AMD’s eight-core Ryzen 9 4900H “Renoir” APUs, based on the Zen 2 microarchitecture, scores about 1,230 single-core points and around 7,100 multi-core points when running at 3.30 / 4.44 GHz clocks in Geekbench 5. Therefore, the new Cezanne APU is apparently 25% faster than its Renoir predecessor in single-core tasks and about 15% faster in multi-core workloads.
Cezanne’s noticeably higher performance compared to its predecessor can be explained by microarchitectural improvements, as well as a two times larger L2 cache. The Ryzen 94900H is rated for up to a 54W TDP, whereas the new one has a default TDP of 35W.
A comparison of the new numbers for the Ryzen 9 5980HS to Intel’s eight-core Core i9-10885H, Intel’s fastest mobile Comet Lake CPU with a locked multiplier, suggests the Ryzen 9 5980HS is 15% faster in single-core workloads and 4% faster in multi-core tasks.
It should be noted that the Ryzen 9 5980HS numbers haven’t been confirmed, so you should take them with a grain of salt.
AMD’s Ryzen 3000 and Athlon processors (including the unlocked $49 Athlon 3000G) have shaken up the low-cost landscape and made a splash on our CPU Benchmark Hierarchy, but crushing shortages of chips has gripped the industry, which impacts the low end of the market in a particularly painful way. So even though quad-core models with gaming-capable integrated graphics have an MSRP for a mere $100, and the Athlon lineup now dips below 50 bucks (although it’s not always easy to find the 200GE at the moment), supply is short. Intel’s response to AMD’s challenge has brought Hyper-threading to its low-end Pentium processors and two additional cores to the Core i3 line, which greatly improves performance for its budget chips even though they’re still limited in terms of their graphics.
AMD hasn’t released its Ryzen 5000 chips for the low-end yet; the series bottoms out at the Ryzen 5 family. We expect that Ryzen 3 models will come in due course, shaking up our low-end rankings. Intel also has its Rocket Lake processors incoming next month, but these chips will use the refreshed Comet Lake architecture for the Core i3 and below chips, so they probably won’t have much impact on our rankings.
The 200-series AMD chips are surprisingly capable at gaming even without a dedicated card. For more details about how the 200GE stacks up against Intel’s comparable budget chip, see our feature AMD Athlon 200GE vs. Intel Pentium Gold G5400: Cheap CPU Showdown.
For those looking for something with a bit more gaming prowess without having to resort to a dedicated graphics card, AMD’s Ryzen 5 3400G is tough to beat. While the Ryzen 4000 APUs are faster, they’re not readily available and cost a lot more. The 3400G is a solid option but doesn’t offer a big performance boost over previous-generation chips like the Ryzen 5 2400G.
If your budget is a bit more flexible and you’re looking to pair your processor with dedicated graphics for gaming, AMD’s Ryzen 3 3300X is great if you can find it at retail around the MSRP of $120, and Intel’s new Core i5-10600KF is impressive at about twice that amount. But like so many other PC components, both of these processors are hard to find in stock at reasonable prices. Hopefully availability will improve over time, but it probably won’t happen until later in 2021, at which time we’ll have new processors.
When choosing a CPU, consider the following:
You can’t lose with AMD or Intel: Both companies offer good budget chips, and overall CPU performance between comparative parts is closer than it’s been in years. You can see how the chips stack up in our CPU Benchmark Hierarchy. That said, if you’re primarily interested in gaming, Intel’s chips will generally deliver slightly better performance (and consume more power) when paired with a graphics card, while AMD’s Raven Ridge models (like the AMD Ryzen 3 2200G) do a better job of delivering gaming-capable performance at modest settings and resolutions without the need for a graphics card.
Clock speed is more important than core count: Higher clock speeds translate to snappier performance in simple, common tasks such as gaming, while extra cores will help you get through time-consuming workloads faster.
Budget for a full system: Don’t pair a strong CPU with weak storage, RAM and/or graphics.
Overclocking isn’t for everyone, but the ability to squeeze more performance out of a budget offering is enticing. Intel doesn’t have overclocking-capable processors for the sub-$125 market, but AMD’s processors allow for tuning, and in most cases the bundled AMD cooler is sufficient for the task. Automated overclocking features in most motherboards make the process easy, so even the least tech-savvy users can enjoy the benefits.
For even more information, check out our CPU Buyer’s Guide, where we discuss how much you should spend for what you’re looking to do, and when cores matter more than high clock speeds. If you can expand your budget and buy a mainstream or high-end processor, check out our lists of Best CPUs for Gaming and Best CPUs for Applications. Below, you’ll see our favorite budget picks.
The Ryzen 3 3300X unlocks a new level of performance for budget gamers with four cores and eight threads that can push low- to mid-range graphics cards to their fullest. This new processor wields the Zen 2 architecture paired with the 7nm process to push performance to new heights while enabling new features for low-end processors, like access to the speedy PCIe 4.0 interface. The 3300X’s four cores tick at a 3.8 GHz clock rate and boost to 4.3 GHz, providing snappy performance in lightly threaded applications, like games.
AMD includes a bundled Wraith Spire cooler with the processor. Still, you might consider budgeting in a better low-end cooler to unlock the full performance, particularly if you are overclocking. Speaking of which, the Ryzen 3 3300X can overclock to the highest all-core frequencies we’ve seen with a Ryzen 3000-series processor, making it a great chip for enthusiasts. Unlike AMD’s other current-gen Ryzen 3 processors, you’ll need to pair this processor with a discrete GPU, but the low price point leaves extra room in the budget for a more capable graphics card.
You can stick with the value theme and drop this capable chip into existing X470 of B450 motherboards, but you’ll lose access to the PCIe 4.0 interface in exchange for a lower price point. Better yet, AMD will have its new B550 motherboards on offer in June 2020. These new motherboards support the PCIe 4.0 interface but provide lower entry-level pricing that’s a better fit for this class of processor.
When money is tight, being able to game without a graphics card can lead to serious savings. And with RAM prices continuing to soar, those working with small budgets need to tighten the strings anywhere they can.
That makes the four-core, four-thread Ryzen 3 2200G particularly appealing for budget gaming builders and upgraders. The $99 chip delivers solid 720p performance thanks to its Vega on-chip graphics, decent CPU muscle for mainstream tasks, and can be dropped into an existing inexpensive 300-series motherboard (after a requisite BIOS update), to form the basis of a surprisingly capable low-cost PC. It’s also unlocked, so with proper cooling you can tune the graphics or the CPU to best suit your needs.
Overclocking is possible, though officially unsupported
All models provide similar performance after overclocking
Reasons to Avoid
Graphics engine and memory can’t be overclocked
Weak single-threaded performance
AMD’s Athlon 240GE serves as the flagship of the company’s budget lineup, but it still packs a convincing punch for low-end gaming systems. The integrated Radeon Vega 3 graphics facilitate playable frame rates at lower resolutions and quality settings, but the 3.5 GHz base clock is the only differentiating feature between the Athlon 240GE and its counterparts. Due to the unofficial support for overclocking, that means you can tune the Athlon 200GE to the same top performance as the more expensive chips, but at a $20 price savings.
If overclocking isn’t in your plans, the Athlon 240GE is the best budget chip in its price band. Intel’s competing Pentium lineup lacks the graphical horsepower to be serious contenders for the extreme low-end of the budget gaming market, but they are attractive if gaming isn’t your primary goal. That is, of course, if you can find them.
Overclocking is possible, though officially unsupported
All models provide similar performance after overclocking
Reasons to Avoid
Graphics engine and memory can’t be overclocked
Weak single-threaded performance
AMD’s sub-$60 Zen-based Athlon is a good all-around value, thanks to its four computing threads and Vega 3 graphics that are capable of light gaming at lower resolutions and settings. Lightly threaded performance isn’t great, but when you’re spending this little on a CPU, you should expect compromises somewhere. And while it isn’t officially supported by AMD, if you have a compatible motherboard, this chip can be overclocked to eke out some extra CPU performance.
If your build budget can swing it, the $100 Ryzen 3 2200G is a much better chip with more cores and beefier graphics. But if you can only spend $60 or less on your CPU and you aren’t adding a dedicated graphics card, the Athlon 200GE is tough to beat. Intel’s competing Pentiums, the Gold G5400 and G4560, deliver better CPU performance. But they have higher MSRPs, and production shortages have made them hard to find unless you’re willing to spend close to $100 or more, making them incomparable in terms of budget CPUs.
Read: AMD Athlon 200GE Review
Integrated Graphics Gaming Performance
Image 1 of 11
Image 2 of 11
Image 3 of 11
Image 4 of 11
Image 5 of 11
Image 6 of 11
Image 7 of 11
Image 8 of 11
Image 9 of 11
Image 10 of 11
Image 11 of 11
You won’t find many game titles that will play well at the popular 1920X1080 resolution on the sub-$80 chips, but there are a few. As we can see, AMD’s $100 Ryzen 3 2200G is the undisputed king of the hill for 1080p gaming on integrated graphics, but the Athlon chips also push out playable frame rates in a few titles (if you’re willing to tolerate lower graphics quality settings).
Switching over to 1280×720 finds the Athlon processors providing up to 50 FPS at stock settings and experiencing a decent performance boost from overclocking. Remember, all of the Athlon chips will benefit equally from overclocking, meaning the Athlon 200GE and 220GE will achieve the same level of performance as the overclocked Athlon 240GE. That’s an amazing value for these low-cost chips. It should go without saying, but the Ryzen 3 2200G’s Radeon Vega 8 graphics engine blows through the 1280×720 tests with ease.
Intel’s Pentium lineup, and even the Core i3-8100 for that matter, struggle tremendously under the weight of these titles. Gaming at 1920×1080 is a painful experience: You won’t find many games that are playable on Pentium processors at that resolution. Switching over to the 1280×720 resolution brings the Core i3-8100 and Pentium G5600 into acceptable territory, but those chips still can’t match the Athlon’s performance, not to mention the crazy good savings. Intel’s Pentium G5400 is particularly disappointing, though, due to its pared-down UHD Graphics 610 engine. We wouldn’t recommend this processor for gaming on integrated graphics.
But it’s hard to recommend Pentium processors at all right now. Intel is struggling with a shortage of 14nm production capacity, so these chips are extremely hard to find, and when you do find them, they are subject to severe price gouging.
Discrete GPU Gaming Performance
We focus primarily on integrated graphics gaming performance for ultra-budget chips, but these processors are also a great pairing with low-end discrete graphics cards. Below, we’ve tested the chips paired with an Nvidia GeForce GTX 2080 at the 1920×1080 resolution to remove any GPU limitations from our tests below. We tested with an Nvidia GeForce 1080 FE graphics card to remove graphics-imposed bottlenecks, but the difference between the processors will shrink with the cheaper graphics cards that are commonly found in budget builds. Provided the performance deltas are small, you can select less expensive models and enjoy nearly the same gaming experience with graphics cards on the lower-end of the GPU benchmarks hierarchy.
Image 1 of 11
Image 2 of 11
Image 3 of 11
Image 4 of 11
Image 5 of 11
Image 6 of 11
Image 7 of 11
Image 8 of 11
Image 9 of 11
Image 10 of 11
Image 11 of 11
Intel’s Coffee Lake Pentium models come with slight frequency improvements, a 3W increase in the TDP rating, and 4MB of L3 cache. These slight adjustments deliver a surprising boost to performance compared to the previous-gen Kaby Lake models. The Coffee Lake Pentium Gold G5600 even beats out the Kaby Lake Core i3-7100 in most of our gaming benchmarks, highlighting the impressive performance gains Intel made within a single generation.
The G5600 grapples with the Ryzen 3 2200G. The Ryzen 3 2200G is relatively simple to overclock with single-click options in the BIOS, and the bundled cooler provides enough headroom for all but the most extreme overclocking efforts. At stock settings, the 2200G trails the Intel Pentium Gold 5600, but the advantage of AMD’s unlocked multipliers is clear: At $99, the tuned Ryzen 3 2200G’s performance nearly matches the $117 Core i3-8100.
The Ryzen 3 2200G also comes with powerful integrated graphics that provide surprisingly strong gaming performance at lower resolutions and quality settings. That’s a feat the Core i3-8100 simply cannot match. If you’re seeking the absolute best gaming performance (when paired with a dedicated card) regardless of price, the Core i3-8100 fits the bill. If you want the most bang for your buck or plan on gaming on integrated graphics, the Ryzen 3 2200G is the clear value winner.
Productivity Performance
Image 1 of 22
Image 2 of 22
Image 3 of 22
Image 4 of 22
Image 5 of 22
Image 6 of 22
Image 7 of 22
Image 8 of 22
Image 9 of 22
Image 10 of 22
Image 11 of 22
Image 12 of 22
Image 13 of 22
Image 14 of 22
Image 15 of 22
Image 16 of 22
Image 17 of 22
Image 18 of 22
Image 19 of 22
Image 20 of 22
Image 21 of 22
Image 22 of 22
The Core i3-8100’s solid mixture of frequency and IPC throughput delivered to our expectations. The agile processor took the lead in several of our lightly-threaded applications, like the Adobe Cloud suite, but it is also surprisingly powerful in threaded workloads. The Intel Core i3-8100 also offers superior performance in applications that use AVX instructions, like HandBrake, which is a great addition to its impressively well-balanced repertoire. Much like we observed in our gaming tests, the Core i3-8100 offers the best overall performance.
Even after overclocking, the AMD Ryzen 3 1300X isn’t competitive enough with the Core i3-8100 to justify its higher price tag, and the lack of integrated graphics also restricts its appeal.
The Ryzen 3 2200G continues to impress with its lower price point and competitive performance, not to mention the integrated Vega graphics, making it the obvious choice for budget builders who are willing to spend a little extra time on tuning.
The Pentium lineup excels in most applications, but the Athlon processors also offer an impressive level of performance. It’s also noteworthy that Intel’s Pentium processors don’t accelerate AVX instructions, a staple in many types of rendering applications, while the Athlon processors fully support the densely-packed instructions. Intel’s chips lead in lightly-threaded applications, like web browsers, but the competing AMD chips also offer more than suitable performance in those workloads.
AMD’s new drivers bring official support for the Quake II RTX and Hitman 3, the latter of which promises to improve performance at its best 10%.
AMD has released new drivers for its graphics cards and APUs. Radeon Software Adrenalin 2020 Edition 21 1.1 drivers are available for Windows 7 and 10 operating systems and support graphics cards and APUs based on the company’s GCN and RDNA architectures.
The most significant updates to Radeon Software 21 1.1 drivers are official support for Hitman 3 and Quake II RTX games. According to the release bulletin, drivers will improve performance in Hitman 3 at its best 10% 4K resolution with Ultra settings when using a Radeon RX 6800 XT and the previous . 12 .1 drivers.
As usual, the drivers have also fixed issues with previous releases. This time, bugged bugs include possible crashes of Doom Eternal when Steam Overlay is enabled, possible problems with the Samsung Odyssey G9 display 5120 × 1440 resolution, and 240 at the Hertz refresh rate, and in GTA V, in some cases, lost reflections when the Reflection MSAA setting is enabled. Known issues include occasional Metro Exodus crashes when beam tracking is enabled, possible screen flicker when MSI’s Afterburner is enabled, and flickering in some applications and games when Radeon FreeSync is enabled if the program is set to borderless full screen mode. You can review all driver changes in their release notes.
AMD has launched a new graphics driver to support the new AAA title, Hitman 3, in all its glory. Adrenaline Driver 21.1.1, according to AMD, can add up to 10% more performance on Hitman 3 at 4K ultra settings with a Radeon RX 6800 XT, compared to previous driver 20.12.1. Driver 21.1.1 also adds support for Quake II RTX and features a barrage of bug fixes.
Bugs Squashed:
Several bug fixes include fixes for recording/streaming and performance metric OSDs not displaying info correctly. Plus, a few game fixes for Grand Theft Auto V, PlayerUnknown’s Battlegrounds, and Doom Eternal. Another several bug fixes are for the Odyssey G9 series of monitors, which had display issues or black screens when paired with AMD graphics cards. Finally, there are crash fixes for Oculus Link when running Polaris or Vega graphics products.
Known Issues:
Issues still at hand include brightness flickering with Freesync and playing a game in borderless windowed mode, as well as display flickering with Vega graphics and dual monitors, when one is set to a high refresh rate.
Another issue is with Metro Exodus. which may experience crashes with DXR Ray Tracing enabled. Radeon’s recording/streaming features might fail on HD 7800 series products, and, finally, Enhanced Sync might cause a black screen on some systems. Disable the feature to work around the problem.
Recently introduced AMD Ryzen mobile processors 5000, belonging to the fourth generation Cezanne APU. In our editorial office, we have already started testing one of the flagship Cezanne-H units and we are very curious how it will finally present a new system based on Zen 3 architecture. Meanwhile, AMD is preparing another premiere, which will be Cezanne desktop APU processors, belonging to the Ryzen series 5000 G. We still don’t know much about the processors themselves, as their debut is expected in the summer. The flagship model is to be 8-core and 16 – threaded AMD Ryzen 7 5800 G, which, according to the current information, is to be more efficient than the AMD Ryzen 7 5800 X unit. The GPU-Z database, in turn, provides some new information about the APU.
According to the information provided by the GPU-Z software, the AMD Ryzen 7 processor 5800 G will have 16 PCIe 3.0 line for the graphics card and integrated Radeon Graphics (Vega 8).
AMD Ryzen 9 5900 HX – New APU from Reds, the most efficient mobile chip in the PassMark test
AMD Ryzen 7 5800 The G is the most efficient representative of the new generation APU Cezanne. Contrary to last year’s models, this time the processors are based on the Zen 3 architecture. Nevertheless, we will still be “stuck” to the Vega graphics chips. This is confirmed by the entry from the GPU-Z software, which states the use of iGPU Radeon Vega 8 in the Ryzen 7 system 5800 G. This is the same Vega 8 that was already available in APU Renoir. Although the program reports the clock speed at 1800 MHz, we are sure that it will eventually be a higher frequency. Vega 8 in the top APU Cezanne-H reaches 2100 MHz, so for the desktop version with a higher TDP (65 W), we expect an even higher clock.
AMD Ryzen 5000 – premiere of Cezanne processors for laptops. Zen 3 architecture enters top notebooks
The processor still has a DDR4 memory controller clocked at 3200 MHz. The test platform on which the processor was installed had DDR4 memory 3600 MHz, which at 128 – the iGPU bit memory bus gives a bandwidth of 57, 6 GB / s. The new Cezanne APU, as in the mobile version, uses the PCIe 3.0 interface to connect to the graphics card. Unlike the Cezanne-H, however, the desktop APU has 16 PCIe 3.0 lanes for the GPU. For comparison, Ryzen series mobile processors 5000 only have 8 PCIe 3.0 lanes for dGPU.
There are many ultra-compact form-factor PCs Intel-based PCs around, but when it comes to AMD-based UCFF systems, the choice is somewhat limited, so pricing is usually inflated. Minisforum, which specializes in inexpensive ultra-compact PCs, recently started to sell its UM250/UM300-series UCFF computers, which pack an AMD Ryzen processor, starting at $299.
Minisforum’s UM250/UM300-series PCs come in an aluminum chassis that measures 128x127x46 mm (according to Notebookcheck.net), which is just slightly larger than Intel’s NUCs but still fits in the palm of your hand.
The tiny system can be powered by either AMD’s Ryzen 5 Pro 2500U (4C/8T, 2.0GHz – 3.60GHz, 6MB cache, Radeon Vega 8 GPU, 15W, 14LPP) or AMD’s Ryzen 3 3300U (4C/4T, 2.10GHz – 3.50GHz, 6MB cache, Radeon Vega 6 GPU, 15W, 12LP) APU. These aren’t AMD’s latest processors, but they are more than good enough for productivity and basic entertainment tasks. However, you should keep in mind that AMD’s Radeon Vega GPUs have rather outdated media processing capabilities.
(Image credit: Minisforum)
The UM250/UM300 can also pack up to 32GB of DDR4 memory using two SO-DIMM slots, an M.2 SATA SSD, and a 2.5-inch SATA SSD or HDD.
As for connectivity, the systems feature an Intel AX200 Wi-Fi 6 and Bluetooth 5.1 adapter, two GbE ports (which will be particularly handy for corporate users), four USB 3.1 Gen 1 connectors, one USB Type-C port that can be used to connect a monitor or any other device, two additional display outputs (DisplayPort 1.2, HDMI 2.0), and a 3.5-mm audio jack.
(Image credit: Minisforum)
Minisforum’s UM250 barebones with the Ryzen 5 Pro 2500U can be purchased directly from the company for $299. Meanwhile, a UM250 machine with 16GB of memory and a 512GB SSD is priced at $499 at Amazon.
AMD launched the Radeon RX 6800 XT and RX 6800 on November 23, 2020. The first of the new RDNA2 architecture graphics cards had plenty to offer, ranking near the top of our GPU benchmarks hierarchy and earning a place on our list of the best graphics cards. AMD does particularly well with games that don’t support ray tracing. In such cases, there are quite a few games where the 6800 XT leads the (theoretically) more expensive RTX 3080, though enabling ray tracing or DLSS quickly turns the tables. The biggest problem, as we’ve seen with all of the recent GPU launches, is actually finding one in stock. Now we’re looking at three third-party custom cards, from ASRock, Asus, and Sapphire, to see what they bring to the table.
The core features and RDNA2 architecture are all unchanged, so the main differences between the cards will be in clock speeds, cooler designs, and aesthetics. There are also a few third-party add-ons, in the way of software, that might sway your purchasing decision. But let’s be real: Finding any of these cards in stock can be an exercise in futility, and with the recent surge in cryptocurrency mining, it could be months before supply is anywhere close to matching demand. In other words, if you want an RX 6800 XT as soon as possible, the brand and model of card will be far less of a consideration than whatever you can actually lay your grubby little mitts on.
The good news is that performance across all of the tested RX 6800 XT cards is very close. At factory stock settings, the speediest of the cards we’ve tested is only 2-3 percent faster than the reference RX 6800 XT. Between the three custom cards, the performance deltas are even smaller, to the point of being effectively non-existent. But that doesn’t mean the cards are all equal, as the cooling designs and other elements come into play. Here’s a quick overview of the specs before we get into the individual card analysis and benchmark results.
AMD Radeon RX 6800 XT Specifications
ASRock Taichi RX 6800 XT
Asus ROG Strix LC RX 6800 XT
Sapphire Nitro+ RX 6800 XT
Reference RX 6800 XT
Architecture
Navi 21
Navi 21
Navi 21
Navi 21
Process Technology
TSMC N7
TSMC N7
TSMC N7
TSMC N7
Transistors (Billion)
10.3
10.3
10.3
10.3
Die size (mm^2)
251
251
251
251
SMs / CUs
72
72
72
72
GPU Cores
4608
4608
4608
4608
Ray Accelerators
72
72
72
72
Boost Clock (MHz)
2360
2360
2360
2250
VRAM Speed (Gbps)
16
16
16
16
VRAM (GB)
16
16
16
16
VRAM Bus Width
256
256
256
256
ROPs
128
128
128
128
TMUs
288
288
288
288
TFLOPS FP32 (Boost)
21.7
21.7
21.7
20.7
Bandwidth (GBps)
512
512
512
512
TDP (watts)
350?
350?
350
300
Pricing
$829
$899 ($1,080)
$769 ($999)
$649
Let’s first address the elephant in the room: The pricing is either fantasy land or, in the case of actual ‘street’ pricing, egregious. Theoretically, the Sapphire Nitro+ initially launched at $769, but of course, it was sold out — just like every other GPU. Newegg currently lists it at $999, and it’s still out of stock. The Asus Strix LC is a similar story, with a launch price of $899 but a current Asus store price of $1,080, and it’s also sold out. ASRock gave a launch price of $829, but retail prices are much higher than that, and naturally, the reference AMD RX 6800 XT can’t be had for anything close to $649.
Beyond price, the only difference in specs is the TDP. Sapphire lists 350W, while Asus and ASRock don’t give any value. We put in 350W with a question mark based on our testing. All three AIB cards have the same 2360 MHz Boost Clock, which they can exceed in some cases. That’s where the cooling solutions come into play.
That’s it for the introduction. Let’s get to the individual cards, and we’ll dig into the finer points of each one, including any extra features that can help it stand out. We’re going to dispense with actual scores on these cards, mostly because they all feel like ghost launches. Yes, they technically went on sale, but both pricing and availability are so limited that we don’t know where they’ll really land. They’re all more or less equal, depending on your wants and needs.
A few weeks ago, our furnace (technically a heat pump) stopped working at the worst possible time of the year. But at least one room stays nice and toasty all winter long — thanks to my gaming PC.
Now, The Wall Street Journal is reporting this might actually be a trend: some bitcoin miners and gamers are warming their rooms, growing tomatoes, and even heating a chicken coop during the pandemic.
It made me wonder: how many Verge readers are doing the same?
Poll
Are you (intentionally) heating your home with gaming hardware?
0%
No
(0 votes)
0%
Yes, with my gaming PC
(0 votes)
0%
Yes, with my mining rig
(0 votes)
0%
Yes, with something else (share in the comments!)
(0 votes)
0 votes total
Vote Now
Me, I keep my PC running all night long, folding proteins to help study COVID-19, which I highly recommend. (I just crossed 175 million points this week.) But I have to admit the room hasn’t been as warm since I swapped my own GeForce GTX 1080 back in. The impossible-to-find AMD Radeon RX 6800 definitely put out more heat, and we actually had to crack a window when I had an Nvidia RTX 3080 folding those proteins. I can’t speak for every model, but the Founder’s Edition runs hot.
We have already reported on the current rapidly advancing development in China’s semiconductor industry. Even though it will take a few more years in production to catch up with the competition, the development of the designs seems to be going much faster.
As the colleagues from Golem (via Toms Hardware) now report that the Chinese company Tianshu Zhixin alias Iluvatar has developed and already had a GPGPU (General Purpose GPU) manufactured that can almost compete with chips from AMD and NVIDIA. The GPU on any graphics card can be referred to as a GPGPU. The micro-architecture is able to make certain calculations. For games and 3D applications, these are FP 32 – and INT 32 – calculations. These have an accuracy of 32 Bit, but can also be expressed in 16 Bit, 8 bit and lower accuracy break up or on 64 Inflate bit. Depending on the design of the pipeline and the front and back end, such non-native computing steps are no longer particularly effective.
Accordingly, a certain trend has developed in the development of GPU accelerators, namely separate designs for gaming and HPC / AI products. NVIDIA uses an adaptation of the Ampere architecture for the GA 100-Accelerator. With the RDNA and CDNA architecture (Radeon Instinct MI 100) also two separate paths with a special design. Ultimately, Intel also provides for different expansion levels and specializations within the Xe families – Xe-LP, Xe-HP, Xe-HPG, Xe-HPC.
But now back to the GPU baptized the Big Island from China: The development should only have started 2018 and in May 2020 the tape-out was done. 7 nm are mentioned as the production size and since there is also talk of 2.5D CoWoS, only TSMC is actually considered as a contract manufacturer. It remains to be seen whether the manufacturing of the GPU will be affected by the threatened sanctions and ban imposed by the US on companies that do business with China and companies in the US. Last reported TSMC to have manufactured far less volume for customers from China.
Big Island should be native with the data formats FP 32, FP 16, Bfloat 01, INT 32, INT 16 and INT8. This covers the typical area of a GPGPU, but for example dispenses with the FP that is important in the HPC segment 64 – calculations. With Bfloat 01 and INT8, the AI formats important for inferencing and training are supported. For FP 16 – calculations Tianshu Zhixin speaks of a computing power of 147 TFLOPS. The Radeon Instinct MI 100 from AMD comes for FP 16 on 185 TFLOPS, NVIDIAs A 096 even manages 312 TFLOPS. So one is not quite at eye level in China yet, but the development steps are considerable.
Tianshu Zhixin does not give any technical details for the Big Island GPU. Four HBM memory chips can be seen next to the central GPU on the chip’s renderings. At least a 4th 64 bit wide memory interface is therefore considered to be secured. Beyond that, we currently know little about the GPU.
The Chinese company is currently supposed to validate its chip so that it can soon be used in various areas of application. China is currently trying to make itself independent from abroad in numerous areas of the semiconductor market. In addition to industry giants such as Huawei, this is also happening through countless other projects. With regard to CPU development, our test of the Zhaoxin KX-U 6780 A showed that the development steps are quite large, but still a lot to do there.
At this stage, the theoretical FP 16 performance is known from the performance of a computing circuit called Big Island, which falls between the computational circuits of AMD and NVIDIA.
Over the years, China’s efforts to become self-sufficient in technology have certainly become familiar to everyone who follows the industry. Now China’s Tianshu Zhixin Semiconductor has announced that it has completed its first GPGPU computing circuit for data centers.
Named Big Island by Tianshu Zhixin, the GPGPU is built from as many as 24 billion transistors and is manufactured using a 7-nanometer manufacturing process. According to Tom’s Hardware, the circuit manufacturer is not public information, but the information provided about the process is consistent with TSMC’s process. The so-called district. tape-out, ie the completion of the finished masks used in its manufacture, took place in May last year. According to a published rendering of the circuit, it is supported by four stacks of HBM memory.
So far, Big Island’s performance readings are still partially obscured. According to Tianshu Zhixin, the circuit would achieve TFLOPS performance with FP 16 accuracy 147 when its most direct competitors AMD’s Radeon Instinct MI 100 achieves 184, 6 TFLOPS and NVIDIA A 100 77, 97 TFLOPS with the same accuracy. In addition to FP 16 accuracy, Big Island also supports all other general calculation accuracies, including the relatively newcomer BF 16 or BFloat 16.
The manufacturer Shuttle has now announced the XPC DA 320. Said barebone has the current AMD Ryzen CPUs in socket AM4 and can be equipped with up to 32 GB of RAM (2 x 16 GB SO-DIMM). DDR4 – 2933 – RAM (Athlon: DDR4 – 2666) is officially supported. The housing has a volume of 1.3 liters. According to Shuttle, the AMD Radeon graphics unit integrated in the processor enables three UHD displays to be operated. These can be connected via HDMI 2.0 or using a DisplayPort.
In addition, two gigabit network cards and COM ports are available to users. The mini-PC can also be started remotely with the help of the remote power-on connection. The DA 320 can also be used with the supplied 75 / 100 – mmVESA bracket can be mounted behind a monitor or on the wall, among other things.
In addition to a 2.5-inch drive, an M. 2 module can be installed in the housing. Please note here that the M.2 2280 M slot supports PCIe x4 NVMe and SATA. The M.2 – 2230 E-Slot, on the other hand, is suitable for the installation of an optional WLAN module (WLN-M). USB and audio connections are available on the outside of the device. There are a total of six USB 3.2 Gen1 ports and two USB 2.0 interfaces on the front and rear. An SD card reader can also be used. A double fan cooling system with heat pipes ensures sufficient fresh air. The built-in power supply has 120 W.
Optional accessories include feet for vertical operation (PS 02) as well as a connection cable for the remote power-on connection (CXP 01) available. Users can also purchase a WLAN / Bluetooth module (WLN-M) directly from Shuttle. The recommended retail price for the XPC Barebone DA 320 is around 236 euros. The barbone is now available in stores.
We use cookies on our website to give you the most relevant experience. By clicking “Accept”, you consent to the use of ALL the cookies.
This website uses cookies to improve your experience while you navigate through the website. Out of these, the cookies that are categorized as necessary are stored on your browser as they are essential for the working of basic functionalities of the website. We also use third-party cookies that help us analyze and understand how you use this website. These cookies will be stored in your browser only with your consent. You also have the option to opt-out of these cookies. But opting out of some of these cookies may affect your browsing experience.
Necessary cookies are absolutely essential for the website to function properly. This category only includes cookies that ensures basic functionalities and security features of the website. These cookies do not store any personal information.
Any cookies that may not be particularly necessary for the website to function and is used specifically to collect user personal data via analytics, ads, other embedded contents are termed as non-necessary cookies. It is mandatory to procure user consent prior to running these cookies on your website.
