Hardware news this past week has only partially slowed, with an uptick in security notices responsible for most of the coverage we've found interesting. Researchers at Eclypsium have identified vulnerabilities in more than 40 drivers from 20 different vendors, something we'll talk about in today's coverage. We also talk about Ryzen 3000 binning statistics posted by Silicon Lottery, the CPU binning company.

Show notes continue after the embedded video.

Hardware news this week is largely focused on new product launches, or rumors thereof, with additional coverage of Intel's plans to launch 10nm Ice Lake CPUs in some capacity (for real, this time) by end of year. The XFX RX 5700 XT "THICC" was leaked -- yes, that's a real name -- and it's accompanied by other partner model cards coming out in the next week.

Show notes continue after the embedded video.

This week's hardware news was filmed prior to our trip to Vancouver for LTX, which we're covering in a lot of content pieces coming up. HW News discusses CCX overclocking, 3nm and 5nm process progress, DRAM revenue dropping hard, and industry topics like Origin's sale to Corsair. We also talk about 5.2GHz 3900X overclocking results, but that'll be in the video only for this one. The rest is in the written section below, as always.

Hardware news for this week is a bit sluggish, with Amazon’s Prime Day -- and the ensuing unrepentant consumerism -- seeming to occupy more than its share of headlines this week. Still, we’ve curated some of the more interesting stories including the latest report from Digitimes and an elucidating interview where Intel CEO Robert Swan cites being “too aggressive” as a key factor in Intel’s CPU shortage. Other topics include information on AMD’s Arcturus GPUs and what form they could take, a Toshiba Memory rebrand, and NZXT adding to its pre-built machines catalog. 

In recent GN news, we’ve delved ever further into Ryzen 3000 and the Zen 2 architecture, including a deep dive into AMD’s Precision Boost Overdrive algorithm, looking at how Ryzen 3000 frequencies scale with temperature, and our R9 3900X overclocking stream.

GN just notched one of its busiest weeks ever, thanks to relentless product launches from AMD and Nvidia. We’ve recently reviewed Nvidia's RTX 2070 Super and RTX 2060 Super, in addition to AMD’s Ryzen 5 3600, Ryzen 9 3900X, and Radeon RX 5700 XT. We also have multiple videos further analyzing Ryzen 3000 boost clocks and the RX 5700 XT cooling solution.

If you’ve enjoyed this coverage, please consider supporting our focused efforts through a GN store purchase.

For mostly non-AMD related news this week, Intel has announced multiple new technologies focused on chip packaging, in addition to hiring a new CCO in Claire Dixon. MSI is updating its AM4 400-series of motherboard to include a larger BIOS chip, there’s a new PCIe 4.0 SSD coming, with a presumably cheaper 500GB capacity, and we’re expecting custom Navi cards in August. The news stories follow the video embed, per the usual.

Alongside the 3900X and 3700X that we’re also reviewing, AMD launched its R5 3600 today to the public. We got a production sample of one of the R5 3600 CPUs through a third-party and, after seeing its performance, we wanted to focus first on this one for our initial Ryzen 3000 review. We’ve been recommending AMD’s R5 CPUs since the first generation, as Intel’s i5 CPUs have seen struggles lately in frametime consistency and are often close enough to AMD that the versatility, frametime consistency, and close-enough gaming performance have warranted R5 purchases. Today, we’re revisiting with the R5 3600 6-core, 12-thread CPU to look at gaming, production workloads with Premiere, Blender, V-Ray, and more, power consumption, and overclocking.

This week has been the busiest in our careers at GN. The editorial/testing team was two people, working in shifts literally around the clock for 24/7 bench coverage, and the video production team was three people (all credited at article's end, as always). We invested all we could into getting multiple reviews ready for launch day and will stagger publication throughout the day due to YouTube's distribution of content. We don't focus on ad revenue on the site these days and instead focus on our GN Store products and Patreon for revenue, plus ad revenue on YouTube. If you would like to support these colossal efforts, please consider buying one of our new GN Toolkits (custom-made for video card disassembly and system building, using high-quality CRV metals and our own molds) or one of our system building modmats. We also sell t-shirts, mousepads, video card anatomy posters, and more.

Notable changes to our testing methods, other than overhauling literally everything (workstation overhaul, gaming overhaul) a few months ago, would include the following:

  • Windows has all updates applied on all platforms, up to version 1903
  • All BIOS updates and mitigations have been applied
  • For new AMD Ryzen CPU testing, we are using a Gigabyte X570 Master motherboard with BIOS version FC5 installed, per manufacturer recommendations
  • We have changed to GSkill Trident Z RGB memory at 4x8GB and 3200MHz. The 32GB capacity is needed for our Photoshop and Premiere benchmarks, which are memory-intensive and would throttle without the capacity. 

The memory kit is an important change for us. Starting with these new reviews, we are now manually controlling every timing surfaced. That includes secondary and tertiary timings. Previously, we worked to control critical timings, like primary and RFC, but we are now controlling all timings manually. This has tightened our margin of error considerably and has reduced concern of “unfair” timings being auto-applied by the various motherboards we have to use for CPU reviews. “Unfair” in this instance typically means “uncharacteristically bad” as a result of poor tuning by the motherboard maker. By controlling this ourselves, we eliminate this variable. Some of our error margins have been reduced to 0.1FPS AVG as a result, which is fantastic.

Leading into the busiest hardware launch week of our careers, we talk about Intel's internal competitive analysis document leaking, DisplayPort 2.0 specifications being detailed, and Ubuntu dropping and re-adding 32-bit support. We also follow-up on Huawei news (and how Microsoft and Intel are still supporting it) and trade tensions.

Show notes continue after the embedded video.

Hardware news this week has a few more AMD rumors -- one of which we're debunking (X590) and another we're re-highlighting (B550) -- with additional news coverage of the US tariffs and impact to consumer pricing. On the topic of pricing, aside from an overall increase as a result of tariffs, Intel has expressed interest to reduce its desktop CPU prices by 10-15% with the launch of Ryzen.

The show notes continue below the embedded video.

Computex 2019 is next week -- a few days from now, technically -- and hardware news has been alight with PCIe 5.0 and DDR5 discussion for Intel platforms, Huawei's ban from the US, DDR4 overclocking close to 6GHz, and more. Intel's biggest news is certainly the PCIe 5 and DDR 5 discussion, which will be our leading story for today's news.

Written show notes are below the video embed.

Computex is coming up. We'll be at the show next week, which means most of our time will likely be spent covering new X570 motherboards for AMD and AMD's announcements, with NVIDIA's presence sparse at the show and Intel's focused on mobile. Beyond the motherboards and CPUs, we already have numerous cooler and case manufacturers lined-up for meetings, alongside some GPU board partners. In the meantime, this past week has set the stage for Computex with a focus on security vulnerabilities, tariffs, and X570 chipset sightings.

The show notes are below the embedded video.

Page 1 of 24

We moderate comments on a ~24~48 hour cycle. There will be some delay after submitting a comment.

Advertisement:

  VigLink badge