Author: batteryforpc

Building your own desktop is typically cheaper than buying a pre-built machine, but there’s definitely a reason why consoles are so popular. Some people would rather buy something that works out of the box, and sales can also turn pre-built PCs into a better value. One desktop from ABS with an RTX 2070 Super graphics card is currently on sale for $1,249.99, a $250 discount from the original MSRP. It’s also the cheapest PC on Newegg right now with that graphics card.

The desktop on sale has an Intel Core i7-9700 processor, which isn’t one of Intel’s newer 10th-gen CPUs, but it still has 8 cores and 8 threads. There’s also 16GB of RAM, a 512GB SSD for storage, and a B365 motherboard. It would have been nice to see a larger SSD, given that there are more than a few 1TB SSDs that only cost around $100, but at least that can be upgraded easily down the road.

The main selling point is the Nvidia GeForce RTX 2070 Super graphics card, which is an excellent GPU for 1440p or 4K gaming. You can read more about it in our original review of the 2070 Super, which also includes benchmarks from some popular games.

Finally, it wouldn’t really be a gaming PC without unnecessary lights, so ABS has added four 120M RGB fans to the front of the case. The desktop also uses 4mm tempered glass side panels, so you can see everything going on inside your PC.

Don’t fret if the pricing for B550 motherboards is preventing you from upgrading. VideoCardz received information that ASRock is reportedly preparing up to five different A520 motherboards for budget seekers. According to the publication, A520 motherboards will launch later this month, which differs from DigiTimes’ report of a September launch.

A520 is the entry-level chipset for AMD’s 500-series, and being A-series, CPU overclocking is pretty much off the table. It’s reasonable to expect A520 to arrive without PCIe 4.0 support, although AMD hasn’t officially said otherwise.

VideoCardz revealed two motherboards from ASRock’s upcoming A520 army. The A520M-ITX/ac comes in a mini-ITX form factor, while the A520M Pro4 has a larger microATX footprint.

ASRock A520M-ITX/ac

The A520M-ITX/ac’s resemblance to the existing B550M-ITX/ac is uncanny. If it wasn’t for the model name on the PCB, you could swear you’re looking at the B550M-ITX/ac.

The A520M-ITX/ac appears to feature an eight-phase power delivery subsystem, a configuration that’s commonly found on cheaper B550 motherboards. As expected of the mini-ITX form factor, the motherboard only has two DDR4 memory slots, but memory overclocking should be superb given the shorter traces to the processor.

The motherboard accommodates up to five storage devices. There are four SATA III connectors and a single M.2 slot up front. The A520M-ITX/ac only has one PCIe x16 slot so graphics card selection requires extra consideration.

Connectivity-wise, the A520M-ITX/ac is an identical copy of the B550M-ITX/ac. The rear panel reveals a DisplayPort and HDMI port for displays. Of course, you’ll have to pair the motherboard with an AMD APU to use either output since the normal Ryzen chips don’t come with an iGPU. An Ethernet port and wireless networking are also available for your usage.

The rear panel holds two USB 2.0 ports, one USB 3.2 Gen 2 Type-A port, two USB 3.2 Gen 1 Type-A ports, one USB 3.2 Gen 1 Type-C port, one PS/2 combo port and three 3.5mm audio jacks.

The B550M-ITX/ac retails for $129.99, so we expect the A520M-ITX/ac’s price tag to be a lot more friendly on the pockets.

ASRock A520M Pro4

The A520M Pro4 looks a little different to the B550M Pro4. For starters, the A520M Pro4 doesn’t have a fancy white heatsink over the power delivery subsystem. The images don’t uncover its design, however, the B550M Pro4 is equipped with an eight-phase configuration so we can expect the A520 variant to carry the same setup, if not less.

The A520M Pro4 puts four DDR4 memory slots at your disposal. We spot four SATA III ports in total, two less than the B550M Pro4. Nonetheless, the two M.2 slots remain untouched. Unlike the A520M-ITX/ac, the A520M Pro4 lacks wireless connectivity so you’ll be connecting to the Internet through the standard Ethernet port unless you purchase a wireless card.

In terms of expansion, the A520M Pro4 supplies two PCIe x16 slots. It’s only missing the PCIe x1 slot, which is present on the more expensive B550M Pro4.

Similar to the situation with the A520M-ITX/ac and B550M-ITX/ac, the A520M Pro4 has the exact same rear panel as the B550M Pro4. Display outputs consist of one D-sub port, one HDMI port and one DisplayPort output. The diverse mix of USB ports includes one USB 3.2 Gen 2 Type-A port, one USB 3.2 Gen 2 Type-C port, four USB 3.2 Gen 1 Type-A ports and two USB 2.0 ports. Lastly, the motherboard has three 3.5mm jacks for connecting audio devices.

The B550M Pro4 sells for $114.99, and the absence of a couple of the features on the A520M Pro4 should allow ASRock to price the motherboard more competitively.

Nuvia is not a household name for CPU designs, but this startup promises to revolutionize cloud server CPU performance in the coming years. 

According to the company, its Orion server system-on-chips (SoCs) based on its Arm’s Phoenix cores will deliver over two times higher single-thread performance when compared to existing x86 designs, such as AMD’s Zen 2 and Intel’s Sunny Cove using a third of power. 

Nuvia’s promises seem very bold at first, but the company’s founders — John Bruno, Manu Gulati, and Gerard Williams III — have a track record of developing successful processors as well as system architectures at Apple, AMD, Arm, and Google. The company was founded in 2019 with an aim to disrupt the market of cloud server processors by offering tangibly higher performance at a fraction of the power of x86 SoCs. 

Nuvia’s Orion SoCs use proprietary Phoenix cores that are presumably based on the Armv9 architecture, but feature a completely overhauled CPU pipeline as well as certain exclusive enhancements. According to the developer, it wants Orion/Phoenix to deliver the highest possible single-thread performance (at a given per-core wattage) at a consistently high utilization as well as frequency. 

Performance revolutionized?

Modern server processors feature tens or even hundreds of cores, but since thermal design power of CPUs is limited, multi-core monsters with tens of cores hardly ever draw more than 10W per core. According to Nuvia, the sweet spot for server processors lies in sub-5W per core range, which is where Arm can beat x86 in terms of single-thread performance. While x86 cores can scale well above 20W per core, performance of x86 solutions will be “only 40% – 50% faster”, according to Nuvia. 

“The optimal solution is one where a workload finishes in the shortest time possible while consuming the least possible power,” said John Bruno, founder and SVP of system engineering at Nuvia. “Nuvia is designing the Phoenix core to meet these targets. Built around Phoenix is the Orion SoC and hardware infrastructure specified to support peak performance on real cloud workloads without bottlenecks.” 

To prove its points, Nuvia demonstrated performance per watt of various modern CPUs in Geekbench 5. The company compared Apple’s A13, A12Z, AMD’s Ryzen 7 4700U (Zen 2), Intel’s Core i7-1068NG7 (Sunny Cove), Core i7-8750H (Skylake), and Qualcomm Snapdragon 865 SoCs. 

Based on the data demonstrated by Nuvia, its CPU cores can deliver about 50% higher performance in Geekbench 5 (2000 points vs. 1300 points) when compared to AMD’s Zen 2 and Intel’s Sunny Cove at 1/3 of power (4.40W vs. 4.80W). Furthermore, Nuvia says that it does not disclose the whole potential of its cores just now. 

At present, Nuvia demonstrates simulated performance of its Orion SoCs based on Phoenix cores, so it remains to be seen how a real product will work under typical workloads. Meanwhile, Nuvia is condifident that it will retain its single-thread performance per watt leadership and aims to bring the first Orion SoCs to the market over the next 18 months, sometime in late 2021 or early 2022, so the wait should not be too long.

Intel has finally given actual details on its upcoming Xe GPU architecture that will power its future CPUs with integrated graphics. While Tiger Lake, Intel’s upcoming 11th generation, 10nm mobile processor, will be the first product to leverage Xe, the company also said it was designing a more powerful mobile CPU with Xe integrated graphics (code-named DG1) for creators. During a press event, Raja Koduri, senior vice president, chief architect, and general manager of Architecture, Graphics, and Software at Intel, said it will ship this year, too. Intel is also working on a discrete GPU for enthusiast gamers, the Xe HPG, and yes, it will have hardware dedicated ray tracing. Koduri said it’s scheduled to ship sometime in 2021, but did not say whether or not it would be available in either desktop or mobile (or both) form factors at launch.

For some historical context, Intel’s first generation of GPUs was in its last discrete graphics card circa 1998, and Gen 5 through Gen 11 have been integrated into Intel CPUs for the past decade. (Gen 10 technically never existed though. It was scrapped alongside Intel’s Cannon Lake line of CPUs.) So, these Gen 12 graphics, totally redesigned and coming in both integrated and discrete packages, are kind of a big deal.

At an event earlier this week for press, Intel officially confirmed that its discrete Xe GPU will have ray tracing, but it also said it has improved its integrated graphics to the point where you can play graphically-intensive games at 1080p with a good frame rate. Intel will have more to say about its ray tracing GPU at a later date, as well as what laptop OEMs will have Tiger Lake chips, but it had plenty to share about its Xe graphics architecture and some of the neat, new features it brings with it.

At the event, Intel mostly focused on the capabilities of its Xe LP graphics microarchitecture, which covers will cover the spread between integrated and entry-level graphics, and mid-range graphics. The microarchitecture will “continue to drive the visual experience for PC, mobile, and ultra mobile form-factors,” said David Blythe, Senior Fellow and Director of Graphics Architecture at Intel. The challenge, according to Blythe, was to figure out a way to make a better GPU within the same-size die area of previous integrated GPUs, and it seems like Intel did it.

Intel’s Xe graphics will have up to 48 texels (similar to an image pixel), 96 EUs (execution units), 1536 flops (floating-point calculations), and up to 16MB of L3 cache, plus twice the amount of memory bandwidth. Compared to the current UHD Graphics 630 that now shares space on many Intel CPUs, it’s a decent jump in performance. The current Gen 11 has only 64 EUs, so the new architecture isn’t 50% larger as previous rumours indicated.

Yet despite the smaller pump in EUs that rumoured Intel re-designed the Xe LP microarchitecture from scratch, which means its 12th generation of graphics should not only be the most powerful it has ever built but the first time the company has substantially redesigned its graphics architecture for several generations.

Among many other things, Blythe said Intel’s new architecture will “enable much broader 1080p gameplay” thanks to enhanced power performance and efficiency in Xe, including “previously unachievable gameplay on some triple-A titles” in a mobile form factor with integrated graphics. In other words, gaming with integrated graphics is about to get a whole lot better.

At it stands now, playing something like Far Cry 5 at a 720p resolution and low graphics settings averages 17 frames per second with UHD Graphics 630, according to Tom’s Hardware. Intel’s Core i7-1065G7 Iris Plus Graphics variants fair better, but nowhere near the 1080p/60 fps gold-standard of modern gaming—averaging 27 fps for the 15W version of the CPU, and 37 fps for the 25 W version.

Intel showed off some 1080p gameplay footage, which included PlayerUnknown’s Battlegrounds, Grid, Mount & Blade II: Bannerlord, Doom Eternal, and Battlefield V, and while it was hard to make out the exact fps or graphics setting, the footage was consistently smooth and totally playable. Intel did show a comparison of Grid played on Gen 11 graphics at low quality and Xe graphics at high quality, and the frame rate seemed near-identical, but again, it’s hard to properly assess frame rate and graphics quality of streamed gameplay footage.

Intel also showed a side-by-side comparison of Battlefield V being played on a Gen 11 25W platform compared to Gen 12 Xe LP 15 W, and while the gameplay looked identical, a slowed-down version showed that the Xe LP microarchitecture was better at smoothing out the rolling action of a tank’s caterpillar track while it was in motion. The Gen 11 version, in contrast, was somewhat choppy.

On the software side of things, Xe graphics will also include a new adaptive GPU optimisation, which decides when and how to recompile the shading in any given scene. According to Intel, this is a general driver change, so not only is it a part of the Xe LP architecture, but Intel also plans to enable it for Gen 9 and Gen 11—although Intel did not say when exactly it would be available for Gen 9 and Gen 11 graphics.

Intel Xe also has a new feature called Instant Game Tuning, which is expected to keep games optimised without a full driver install. If you use discrete graphics you’re familiar with how often you need to update drivers to take advantage of all the features in a game. I feel like I’ve downloaded and installed a ton for my Nvidia graphics card in the last few weeks alone. The driver is needed to get the best performance out of your games, and often times is targeted at a specific game. Intel says it’s eliminating that by delivering driver updates through its cloud so users never have to worry about updating drivers themselves ever again.

To quickly add some changes on the media side of things: Intel increased the encode/decode throughput up to twice as much, added AV1 decoder support, HEVC screen content coding support, 4K/8K60 playback, HDR/Dolby Vision playback, and increased fidelity with full 12-bit end-to-end video pipelining. The Xe LP display engine also supports DisplayPort 1.4, HDMI 2.0, USB4 Type-C, and up to a 360 Hz refresh rate and Adaptive Sync.

All this means it seems something good will come out of Intel’s 10nm process being delayed for so long. CPU performance aside, Intel’s integrated graphics seems like it will get a major boost thanks to the Xe architecture. One major question left now is how will Xe LP compare to AMD’s integrated graphics, which have been traditionally superior to Intel’s. Apple is also getting ready to launch its own ARM-based processors with integrated graphics, so that will be another interesting point of comparison. Also, how will Intel’s ray tracing GPU compare to Nvidia’s, or even AMD’s?

So far we know that the custom GPUs powering the Xbox Series X and PlayStation 5 have ray tracing, but we haven’t heard much on the PC front. We know AMD’s next-gen GPUs with will have ray tracing, but no word on a release date yet. Nvidia’s 2nd generation of ray tracing GPUs might be around the corner as well. Still, this is a welcome shake-up to the GPU space that has long been dominated by Nvidia, and potentiality a massive change for the budget laptop space.