The Zotac GeForce RTX 3080 Ti AMP HoloBlack is the company’s top graphics card based on the swanky new RTX 3080 Ti “Ampere” GPU by NVIDIA. Hot on the heels of its Computex 2021 announcement, we have with us NVIDIA’s new flagship gaming graphics card, a distinction it takes from the RTX 3080. The RTX 3090 is still around in the NVIDIA’s product stack, but it’s positioned as a TITAN-like halo product, with its 24 GB video memory benefiting certain quasi-professional applications, when paired with NVIDIA’s GeForce Studio drivers. The RTX 3080 Ti has the same mandate from NVIDIA as the RTX 3080—to offer leadership 4K UHD gaming performance with maxed out settings and raytracing.
Based on the same 8 nm “GA102” silicon as the RTX 3080, the new RTX 3080 Ti has 12 GB of memory, maxing out the 384-bit GDDR6X memory interface of the chip; while also packing more CUDA cores and other components—10,240 vs. 8,796, 320 TMUs, those many Tensor cores, 80 RT cores, and 112 ROPs. The announcement of the RTX 3080 Ti and its sibling, the RTX 3070 Ti—which we’ll review soon—may have been triggered by AMD’s unexpected return to the high-end gaming graphics segment, with its “Big Navi” Radeon RX 6000 series graphics cards, particularly the RX 6900 XT, and the RX 6800.
The GeForce Ampere graphics architecture debuts the 2nd generation of NVIDIA RTX, bringing real-time raytracing to gamers. It combines 3rd generation Tensor cores that accelerate AI deep-learning neural nets that DLSS leverages; 2nd generation RT cores that introduce more hardware-accelerated raytracing effects, and the new Ampere CUDA core, that significantly increases performance over the previous generation “Turing.”
The Zotac RTX 3080 Ti AMP HoloBlack features the highest factory-overclocked speeds from the company for the RTX 3080 Ti, with up to 1710 MHz boost, compared to 1665 MHz reference, a bold new cooling solution design that relies on a large triple-fan heatsink that, and aesthetic ARGB lighting elements that bring your gaming rig to life. Zotac hasn’t provided us with any pricing info yet, we’re assuming the card will end up $100 pricier than the base cards, like Founders Edition.
Palit GeForce RTX 3080 Ti GamingPro is the company’s premium custom-design RTX 3080 Ti offering, letting gamers who know what to expect from this GPU to simply install and get gaming. Within Palit’s product stack, the GamingPro is positioned a notch below its coveted GameRock brand for enthusiasts. By itself, the RTX 3080 Ti is NVIDIA’s new flagship gaming graphics product, replacing the RTX 3080 from this distinction. The RTX 3090 is marketed as a halo product, with its large video memory even targeting certain professional use-cases. The RTX 3080 Ti has the same mandate as the RTX 3080—to offer leadership gaming performance at 4K UHD, with maxed out settings and raytracing.
The GeForce RTX 3080 Ti story likely begins with AMD’s unexpected return to the high-end graphics segment with its Radeon RX 6800 series and RX 6900 XT “Big Navi” graphics cards. The RX 6900 XT in particular, has managed to outclass the RTX 3080 in several scenarios, and with its “XTXH” bin, even trades blows with the RTX 3090. It is to fill exactly this performance gap between the two top Amperes—the RTX 3080 and RTX 3090, that NVIDIA developed the RTX 3080 Ti.
The RTX 3080 Ti is based on the same 8 nm GA102 GPU as the other two top cards from NVIDIA’s lineup, but features many more CUDA cores than the RTX 3080, at 10,240 vs. 8,704; and more importantly, maxes out the 384-bit wide memory bus of this silicon. NVIDIA endowed this card with 12 GB of memory. Other key specs include 320 Tensor cores, 80 RT cores, 320 TMUs, and 112 ROPs. The memory ticks at the same 19 Gbps data-rate as the RTX 3080, but the wider memory bus means that the bandwidth is now up to 912 GB/s.
Palit adds value to the RTX 3080 Ti, by pairing it with its TurboFan 3.0 triple-slot, triple-fan cooling solution that has plenty of RGB bling to satiate gamers. The cooler is longer than the PCB itself, so airflow from the third fan goes through the card, and out holes punched into the metal backplate. The card runs at reference clock speeds of 1665 MHz, and is officially priced at the NVIDIA $1200 baselines price for the RTX 3080 Ti, more affordable than the other custom designs we’re testing today. In this review, we tell you if this card is all you need if you have your eyes on an RTX 3080 Ti.
Micron said on Tuesday that it had started volume production of DRAM chips using its latest 1α (1-alpha) process technology. Initially the company will use the new node for LPDDR4X and 8Gb DDR4 memory ICs, but the company will eventually use the new fabrication tech to produce all of its DRAMs.
Unlike its rivals, Micron does not plan to use extreme ultraviolet (EUV) lithography for at least a few years. However, as the company still needs to increase the density of its memory chips and reduce its per-bit costs, it has to rely on other innovations to shrink its DRAMs. Micron’s 1α node offers a 40% improvement in bit density and a 15% drop in power consumption compared to 1Z. It also offers higher performance potential. Micron had to improve array efficiency design and use new materials, including better conductors, better insulators, and new machinery to deposit, modify or selectively remove those materials.
Micron’s 8Gb DDR4 chips, which are already shipping in volume, are the first to use its 1α process technology. The company will begin shipping LPDDR4X chips made with the 1α node later this month. Micron will eventually transition virtually all of its DRAM to 1α, including DDR5, HBM2E, and GDDR6/GDDR6X.
Micron currently uses its 1α technology in its A3 fab in Taichung, Taiwan. In the future, the company will adopt the new technology at other fabs.
Micron has also completed validation of its 1α-based DDR4 on the latest server platforms, including AMD’s Third-gen EPYC ‘Milan’ processor. The same chips can also be used for notebooks.
Micron has introduced two new families of SSDs based on its latest 176-layer 3D TLC NAND. The new 2450- and 3400-series SSDs feature a PCIe 4.0 interface and are aimed at different classes of PCs and price points. Micron says the drives are in production and will be available at retail soon.
Micron’s 2450-series SSDs are designed to be affordable drives for mainstream PCs and are set to be available in M.2-2280, M.2-2242, and M.2-2230 form-factors. The family will offer 256GB, 512GB, and 1TB usable capacities.
In contrast, Micron’s 3400-series SSDs are aimed at high-performance applications and will be available in 512GB to 2TB configurations as well as an M.2-2280 form factor. Both SSD families are equipped with a thin (presumably graphene-based) heat-spreader, so they are compatible with both desktops and notebooks.
(Image credit: Micron)
Micron says that it uses in-house-developed NVMe 1.4-compliant SSD controllers for its 2450-series and 3400-series SSDs, but says that it wants to be flexible and could use third-party controllers if it needs to. Given that demand for storage devices is very high and ongoing shortages, Micron’s possible use of third-party SSD controllers isn’t surprising.
For unknown reasons, Micron isn’t disclosing the full specifications of its 2450-series and 3400-series SSDs at this time. Instead, the company says its 3400 drives “provides twice the read throughput and up to 85% higher write throughput” compared to its predecessors.
In addition to 2450-series and 3400-series SSDs, Micron uses its 96-layer 3D NAND memory for its first UFS 3.1 automotive-grade storage devices. The unit offers 50% higher sustained write performance versus Micron’s UFS 2.1 drive and will hit production sometime in the third quarter.
High performance memory kits have evolved over the last few years, both in styling and technology. Styling has shifted to heavier heat sinks, LED light bars, and fancy RGB control software. The technology has done what it inevitably will by producing greater speeds and densities at generally lower cost as DDR4 has matured. The latest processors and graphics cards have been nearly impossible to get over the last six months, but memory pricing and availability has remained steady. With no end to the global shortages in sight, let’s dig into a product you can actually buy at MSRP!
Neo Forza started out as an I/C design house that then became an OEM for a variety of international brands over the past 15 years. Neo Forza possesses strong resources focusing on core technologies. From research and design to specific production to exceed the testing benchmark, Neo Forza keeps pushing over technical boundaries, providing top-notch performance and quality.
Neo Forza has leveraged that extensive experience to streamline their wafer screening process. Each Neo Forza design focuses on a single timing bin for each frequency step across the entire product stack. A Neo Forza 3200 MHz kit will offer the same timings and performance regardless of the naming convention or external appearance of the kit.
The Neo Forza Mars kit I have for testing today is one of their middle spec kits: 16 GB (2x 8 GB) at 3600 MHz, 18-19-19-39 timings, and 1.35 V. 3600 MHz has become the new gold standard for Ryzen builds, driving new focus into memory kits targeting a previously obscure specification. Let’s see how the Neo Forza Mars holds up in this ultra-competitive segment!
Asrock announced a new motherboard lineup called the Riptide series, and with it comes two brand new boards for AMD Ryzen CPUs, the X570S PG Riptide and the B550 PG Riptide. These motherboards are an off-shoot of the Phantom Gaming series, designed for gamers who use their systems for everyday use.
The most striking feature of both Riptide motherboards is the addition of a GPU anti-sag bracket built right into the motherboard itself. The bracket is installed right next to the chipset and SATA ports, and will prevent your graphics card from sagging in the front, where there’s the least amount of support.
(Image credit: ASRock)
The bracket is a nice feature to have, with how large graphics cards are getting these days. Many of the latest triple-slot graphics cards weigh around 1.5kg or more, including some of ASRock’s own models. Due to the bracket being situated behind the graphics card, this should give PC builders a very clean look, with the bracket tucked behind the graphics card and out of sight.
Speaking of aesthetics, both boards are very stealthy with a silver and black appearance. The only sight of color is the bright orange and purple Asrock logo on the chipset, which can easily be hidden with a large enough graphics card.
(Image credit: ASRock)
Other features include a 10 phase power delivery system, up to 4933MHz DDR4 memory support, and a special feature ASRock is calling ‘Lightning Gaming Ports.’ These ports are designed to give gamers the lowest latency possible for their keyboard and mouse.
We don’t know what kind of magic ASRock is doing to improve latency on these specific USB ports, but we believe it ensures you are plugging your mouse and keyboard directly into the USB ports wired to the CPU. This will allow for the lowest latency possible, as most of your USB ports are usually connected through the chipset.
Another interesting note is this is Asrock’s first-ever X570S motherboard, and it’s coming to the Riptide series. But we expect ASRock’s other lineups will get the X570S treatment soon. The biggest feature coming to X570S is the ability to run the chipset without a fan. This is great for reliability and acoustics, and something we’re excited to see coming back to AMD’s flagship chipset.
With Nvidia announcing the all-new RTX 3080 Ti and RTX 3070 Ti at Computex this year, AIB partners have wasted no time in announcing custom variants of the two GPUs. There are seven AIB partners so far that have listed custom variants of the RTX 3080 Ti and RTX 3070 Ti, with more to come.
The RTX 3080 Ti is Nvidia’s new gaming flagship for the Ampere generation, featuring 10,240 CUDA cores, 12GB of GDDR6X, a 1,395MHz base clock, and 1,695 Boost Clock. It’s just a hair slower than the RTX 3090, with the biggest tradeoff (between the two SKUs) being the VRAM capacity, which is shaved down from 24GB to 12GB.
The RTX 3070 Ti, is Nvidia’s new mid-range SKU that will slot in-between the RTX 3070 and RTX 3080. The 3070 Ti features 6,144 CUDA cores, 8GB of GDDR6X at 19Gbps, a base clock of 1,440 MHz, and a Boost frequency of 1,710MHz. Expect performance to lean more towards an RTX 3070 rather than the more powerful 3080, as the 3070 Ti uses the GA104 core, though the 35% boost in memory bandwidth should help.
Asus
Image 1 of 3
(Image credit: Asus)
Image 2 of 3
(Image credit: Asus)
Image 3 of 3
(Image credit: Asus)
Asus is bringing out three custom models for the RTX 3080 Ti as well as two custom SKUs for the lower end RTX 3070 Ti. At the top end will be the ROG Strix LC RTX 3080 Ti, featuring a 240mm AIO cooler to keep temperatures as cool as possible, the card is also decked out in a brushed metal finish, with the Strix language design, as well as a fully lit RGB shroud and fans.
For air cooling, Asus is dishing out the ROG Strix treatment to the RTX 3080 Ti and RTX 3070 Ti. For the RTX 3080 Ti ROG Strix, the cooler looks identical to the RTX 3090 variant, with a large triple-slot design, and triple 8-pin power connectors. Styling hasn’t changed either, with a fully lit RGB light bar on the side, and brushed aluminum finish all around the card.
Asus’ lowest-end offering, for now, will be the TUF series, which you will see on both the RTX 3080 Ti and RTX 3070 Ti. Similar to the ROG models, the RTX 3080 Ti TUF is identical in looks to the RTX 3090 TUF. So we wouldn’t be surprised if Asus simply installed the RTX 3090 cooler onto the RTX 3080 Ti cards since both the 3090 and 3080 Ti share the exact same GPU core.
Unfortunately, we don’t have pictures of the custom Asus RTX 3070 TI SKUs at this moment, however, we guess that the cards will be using a beefed-up cooler from the RTX 3070 class of cards, given the RTX 3070 Ti uses the GA104 core instead of GA102. We also don’t know what frequencies these cards will have but be sure these custom RTX 3080 Tis and RTX 3070 Tis will have higher frequencies than the reference specification.
Gigabyte
Image 1 of 3
(Image credit: Gigabyte)
Image 2 of 3
(Image credit: Gigabyte)
Image 3 of 3
(Image credit: Gigabyte)
Gigabyte’s offerings are very minimal for now, with the company currently offering the RTX 3080 Ti and RTX 3070 Ti in the Gaming OC SKU. The Gaming series in Gigabyte’s lineup represents the more budget-friendly level of SKUs rather than its top-end Aorus branded cards.
The RTX 3080 Ti Gaming OC design is identical to that of the RTX 3090 Gaming OC, with no changes to the shroud or cooler (what we can see of the cooler) at all. The card features a matte black finish with silver accents to add some extra styling to the shroud. The 3080 Ti Gaming OC features a factory overclock of 1710MHz.
Surprisingly the RTX 3070 Ti Gaming OC appears to have either a brand new cooler or an altered variant of the RTX 3070 Gaming OC. The heatsink has a different design with two heatsinks joined together by copper heat pipes, rather than three separate heatsinks found on the vanilla RTX 3070 variant, connected by two sets of copper heat pipes.
The RTX 3070 Ti Gaming OC also features a large copper base plate that covers the GPU and all the GDDR6X memory modules. This is a big upgrade compared to the RTX 3070 Gaming OC which only has four copper heat pipes making direct contact with the GPU, paired with a metal base plate covering the memory modules.
Aesthetically, the card has also been noticeably altered. The Gigabyte logo that was at the rear of all Gaming OC cards is now near the front, and the “GEFORCE RTX” logo has its own silver badge on the top of the card. The silver accents on the shroud have also been switched, with silver accents to the top front and bottom rear of the card. With other Gaming OC cards, this was reversed. The RTX 3070 Ti also features a factory overclock of 1830MHz.
EVGA
Image 1 of 2
(Image credit: EVGA)
Image 2 of 2
(Image credit: EVGA)
So far, EVGA has the most custom SKUs announced for the RTX 3080 Ti and RTX 3070 Ti, with 8 custom models confirmed.
The RTX 3080 Ti alone will come in six flavors, the FTW3, FTW3 Hybrid and FTW3 Hydro Copper. The FTW models represent EVGA’s flagships, so expect robust power delivery and excellent performance from these models.
The remaining three consist of the XC3, XC3 Hybrid and XC3 Hydro Copper. These are EVGA’s budget and mid-range offerings, which should offer the best overall price to performance.
The RTX 3070 Ti will only come in two flavors for now, the FTW3 and XC3. Unfortunately, we don’t have specs or detailed pictures of any of EVGA’s SKUs at this time.
MSI
Image 1 of 3
(Image credit: MSI)
Image 2 of 3
(Image credit: MSI)
Image 3 of 3
(Image credit: MSI)
Similar to EVGA, MSI is announcing a ton of SKUs for both RTX 3080 Ti and RTX 3070 Ti. The models will consist of the Suprim, Gaming Trio, and Ventus variants. Each variant also gets a vanilla and factory overclocked model.
Overall the RTX 3080 Ti Suprim, Gaming Trio, and Ventus are identical to the RTX 3090 models with very very minor changes to the aesthetics of the card. The Suprim will be the top tier model, the Gaming Trio represents the mid-tier, and the Ventus is your ‘budget’ friendly RTX 3080 Ti.
The RTX 3070 Ti will also receive Suprim, Gaming Trio, and Ventus variants, but unfortunately, product pages for those cards are not available at this time.
The same goes for clock speed specifications on all of MSI’s RTX 3080 Ti and RTX 3070 Ti SKU, so we’ll have to wait until those become available.
Zotac
Image 1 of 4
(Image credit: Zotac)
Image 2 of 4
(Image credit: Zotac)
Image 3 of 4
(Image credit: Zotac)
Image 4 of 4
(Image credit: Zotac)
Zotac will feature five different SKUs for the RTX 3080 Ti and RTX 3070 Ti combined, consisting of the Trinity and Holo series. The RTX 3080 Tis are mostly identical in every way to the RTX 3090s, especially when it comes to the Trinity, where Zotac appears to have put the RTX 3090 cooler directly onto the RTX 3080 Ti.
For the RTX 3080 Ti Holo, there are a few things to note. The RTX 3080 Ti only has a single Holo SKU, while the RTX 3090 had two, the Core Holo and Extreme Holo. The RTX 3080 Ti holo seems to be its own SKU, with a slightly different aesthetic than any of the RTX 3090 Holos. The RTX 3080 Ti Holo features an elegant-looking RGB lightbar on the card’s side that goes from the top to almost the bottom of the card, with a grey color theme for the whole shroud.
The RTX 3080 Ti Trinity will receive a 1665MHz Boost clock (reference spec), the Trinity OC variant features a 1695MHz boost clock, and the Holo features the highest clock at 1710MHz.
The RTX 3070 Ti will also come in the Trinity and Holo flavors but will come with the same triple-fan cooling configuration as the RTX 3080 Zotac Trinity and Holo. This is very different from the vanilla RTX 3070 which maxes out at a twin-fan design.
We are not sure if the RTX 3070 Ti uses the RTX 3080 coolers from the Trinty and Holo series, but aesthetically they look nearly identical, making us believe this is probably true.
The RTX 3070 Ti Holo will come with a 1830MHz boost clock. and the Trinity will have an 1870MHz boost.
Colorful
Image 1 of 3
(Image credit: Colorful)
Image 2 of 3
(Image credit: Colorful)
Image 3 of 3
(Image credit: Colorful)
Colorful has the fewest amount of cards out of all the AIB partners so far, with only three SKUs announced, and only one of those being for the RTX 3080 Ti.
The only RTX 3080 Ti SKU Colorful has announced is the Vulkan OC-V, featuring a triple fan heatsink and a black and metal finish. Giving the card a very stealthy or ‘batman’ like appearance. The card will feature a base clock of 1365MHz along with a 1710MHz Boost Clock.
The first RTX 3070 Ti SKU announced is the 3070 Ti Advanced OC-V, a big chunky card measuring beyond two slots in thickness, and coming in with a rather unique color design consisting of a silver shroud, accented by purple and black, along with a red ringlit RGB fan in the middle. The card will come with a 1575MHz base clock and a 1830MHz boost clock.
Finally, the last SKU announced is the RTX 3070 Ti NB 8G-V, which appears to be the company’s budget-friendly 3070 Ti. The card features a dual-slot cooler, with a very boxy appearance. The shroud is covered in a matte black finish, accented by both glossy black and matte red finishes. The card will come with a 1575MHz base clock and a 1770Mhz boost clock.
PNY
Image 1 of 6
(Image credit: PNY)
Image 2 of 6
(Image credit: PNY)
Image 3 of 6
(Image credit: PNY)
Image 4 of 6
(Image credit: PNY)
Image 5 of 6
(Image credit: PNY)
Image 6 of 6
(Image credit: PNY)
Last but not least is PNY with four new SKUs planned for the RTX 3080 Ti and RTX 3070 Ti for now. The RTX 3070 Ti and RTX 3080 Ti will both come in Revel and Uprising editions. What we have pictured are the RTX 3080 Ti Revel Epic X, 3080 Ti Uprising Epic X and the RTX 3070 Ti Revel Epic X.
The RTX 3080 Ti Revel Epic X carries a two-toned design to the shroud, with a matte black covering the actual shroud, as well as a uniquely designed metal fan protector with a silver finish. Between the fans lies rings of RGB lighting. The same apparently goes for the RTX 3070 Ti as well, but the 3070 Ti is slightly smaller.
The RTX 3080 Ti Uprising Epic X features a grey finish with RGB accents near the middle of the fan. From what we can tell from the pictures, the card is absolutely gigantic, with a very wide heatsink, along with a length that is hard to describe. For perspective, the heatsink stretches out from the main PCB a good 4 inches, and the PCB isn’t compact at all. So this card is going to be a challenge for some PC cases.
Micron has introduced two new families of SSDs based on its latest 176-layer 3D TLC NAND. The new 2405- and 3400-series SSDs feature a PCIe 4.0 interface and are aimed at different classes of PCs and price points. Micron says the drives are in production and will be available at retail soon.
Micron’s 2450-series SSDs are designed to be affordable drives for mainstream PCs and are set to be available in M.2-2280, M.2-2242, and M.2-2230 form-factors. The family will offer 256GB, 512GB, and 1TB usable capacities.
In contrast, Micron’s 3400-series SSDs are aimed at high-performance applications and will be available in 512GB to 2TB configurations as well as an M.2-2280 form factor. Both SSD families are equipped with a thin (presumably graphene-based) heat-spreader, so they are compatible with both desktops and notebooks.
(Image credit: Micron)
Micron says that it uses in-house-developed NVMe 1.4-compliant SSD controllers for its 2450-series and 3400-series SSDs, but says that it wants to be flexible and could use third-party controllers if it needs to. Given that demand for storage devices is very high and ongoing shortages, Micron’s possible use of third-party SSD controllers isn’t surprising.
For unknown reasons, Micron isn’t disclosing the full specifications of its 2450-series and 3400-series SSDs at this time. Instead, the company says its 3400 drives “provides twice the read throughput and up to 85% higher write throughput” compared to its predecessors.
In addition to 2450-series and 3400-series SSDs, Micron uses its 176-layer 3D NAND memory for its first UFS 3.1 automotive-grade storage devices. The unit offers 50% higher sustained write performance versus Micron’s UFS 2.1 drive and will hit production sometime in the third quarter.
G.Skill has announced a special edition of the brand’s Trident Z memory that currently ranks as one of the best RAM kits on the market. The new Trident Z Maverik arrives as one of the key components for MSI’s latest MPG Gaming Maverik bundle.
G.Skill hasn’t drastically changed the Trident Z Maverik’s design. The memory maintains G.Skill’s distinctive tri-fin aluminum heat spreader with eye-catching RGB lighting, but the memory vendor did slightly redo part of the heat spreader to blend in with MSI’s industrial-looking theme.
Available only as part of the MPG Gaming Maverik bundle, MSI combines two Trident Z Maverik 16GB memory modules into a dual-channel setup. The memory modules are binned for DDR4-3600 with 18-22-22-42 timings. They’re rated for 1.35V and leave some overhead for memory clocking or optimization.
Image 1 of 3
Trident Z Maverik (Image credit: G.Skill)
Image 2 of 3
Trident Z Maverik (Image credit: G.Skill)
Image 3 of 3
Trident Z Maverik (Image credit: G.Skill)
MSI’s MPG Gaming Maverik bundle is practically an entire system, so it’s weird that the company refers to it as a bundle. At any rate, the bundle is comprised of Intel’s Core i7-11700K (Rocket Lake) processor, MSI’s own MPG Z590 Gaming Edge WiFi SP motherboard, MPG CoreLiquid K360 SP liquid cooler and MPG Velox 100 AirFlow SP case.
The MPG Gaming Maverik bundle will be available this month in limited quantities. MSI hasn’t revealed the pricing for the bundle.
After months of rumors and speculation, Nvidia has finally revealed the GeForce RTX 3080 Ti, the long-awaited heir to the GeForce RTX 2080 Ti. Forget the earlier 20GB VRAM rumors; the 3080 Ti is basically an RTX 3090 with half the VRAM. With specifications and performance so close to the GeForce RTX 3090, the new Ampere offering will have no problems competing for a spot with the best graphics cards on the market. Here’s the quick overview:
Nvidia GPU Specifications
Graphics Card
RTX 3090
RTX 3080 Ti
RTX 3080
RTX 2080 Ti
Architecture
GA102
GA102
GA102
TU102
Process Technology
Samsung 8N
Samsung 8N
Samsung 8N
TSMC 12FFN
Transistors (Billion)
28.3
28.3
28.3
18.6
Die size (mm^2)
628.4
628.4
628.4
754
SMs / CUs
82
80
68
68
GPU Cores
10,496
10,240
8,704
4,352
Tensor Cores
328
320
272
544
RT Cores
82
80
68
68
Base Clock (MHz)
1,395
1,440
1,440
1,515
Boost Clock (MHz)
1,695
1,665
1,710
1,545
VRAM Speed (Gbps)
19.5
19
19
14
VRAM (GB)
24
12
10
11
VRAM Bus Width
384
384
320
352
ROPs
112
112
96
88
TMUs
328
320
272
272
TFLOPS FP32 (Boost)
35.6
34.1
29.8
13.4
TFLOPS FP16 (Tensor)
142 (285)
136 (273)
119 (238)
108
RT TFLOPS
69.5
66.6
58.1
52.5
Bandwidth (GBps)
936
912
760
616
TDP (watts)
350
350
320
260
Launch Date
Sep 2020
Jun 2021
Sep 2020
Sep 2018
Launch Price
$1,499
$1,199
$699
$1,199
Armed with 80 out of a potential 84 Streaming Multiprocessors (SMs) from the GA102 GPU, the GeForce RTX 3080 Ti offers consumers 10,240 CUDA cores with a 1,665 MHz boost clock alongside 320 third-generation Tensor Cores for AI and 80 second-generation RT cores for ray tracing workloads. With the CUDA cores alone, it’s easy to see why Nvidia is boasting that the GeForce RTX 3080 Ti delivers up to 1.5 times higher performance than the GeForce RTX 2080 Ti. For those of you who love numbers, the GeForce RTX 3080 Ti pumps out 143%, 139% and 56% higher shader TFLOPs, Tensor TFLOPs and RT TFLOPs, respectively. It also has 48% more memory bandwidth.
We can also look at the more recent Ampere GPUs for comparison. The GeForce RTX 3080 Ti is only a hair behind the GeForce RTX 3090 in most areas, with the biggest difference being VRAM capacity. 3080 Ti wields 12GB of GDDR6X memory at 19 Gbps, on a 384-bit memory interface, meaning the GeForce RTX 3080 Ti lands with a maximum theoretical memory bandwidth up to 912 GBps. On paper, it’s only about 3–4% behind the 3090, except for having half the memory. Alternatively, the 3080 Ti has 14% more compute and 20% more bandwidth than the vanilla RTX 3080.
GeForce RTX 3080 Ti (Image credit: Nvidia)
For better or worse, the 3080 Ti takes after the
GeForce RTX 3080
in terms of design — if not for the “Ti” branding, we’d swear that we’re looking at a GeForce RTX 3080. The GeForce RTX 3080’s dual-slot cooling system is less robust than the GeForce RTX 3090’s triple-slot cooler, though really it’s the GDDR6X memory that gives us the most cause for concern. The 3090 and 3080 Founders Edition cards could run very hot on the memory, and the same goes for a lot of the third party designs. Adding two more GDDR6X modules isn’t going to improve the situation with the 3080 Ti.
The GeForce RTX 3080 Ti is also rated with the same 350W TDP as the GeForce RTX 3090, compared to the GeForce RTX 3080’s 320W TDP, so we’re looking at 30W overhead. Common sense would tell us that the cooling fans will have to work a bit harder, which may increase the operating noise level. There is evidence that replacing the stock thermal pads on the GeForce RTX 3080 and GeForce RTX 3090 helps better cool the memory chips. So, the big question is whether Nvidia substituted the thermal pads on the Ti. We suspect not, but let’s hope Nvidia proves otherwise.
Image 1 of 3
GeForce RTX 3080 Ti Founders Edition (Image credit: Nvidia)
Image 2 of 3
GeForce RTX 3080 Ti Founders Edition (Image credit: Nvidia)
Image 3 of 3
GeForce RTX 3080 Ti Founders Edition (Image credit: Nvidia)
The GeForce RTX 3080 Ti’s other attributes align with those of the non-Ti. The Founders Edition is still happily rocking Nvidia’s 12-pin PCIe power connector. Display outputs still come down to one HDMI 2.1 port and three DisplayPort 1.4a outputs. Nvidia is set to launch the RTX 3080 Ti on June 3, with an official launch price of $1,199.
Considering the rest of the graphics card market right now, not to mention what we’ve seen of previous Ampere launches, that price is likely to be meaningless throughout 2021 and into 2022. The RTX 3080 continues to immediately sell out and generally goes for over $2,000 on eBay, and the RTX 3090 lands closer to the $3,000 range right now. Unless Nvidia can produce a lot more GPUs and cards, shortages and price gouging are likely to continue.
Nvidia GeForce RTX 3070 Ti
GeForce RTX 3070 Ti (Image credit: Nvidia)
Besides the RTX 3080 Ti, Nvidia also officially announced the GeForce RTX 3070 Ti. The graphics card utilizes the GA104 silicon, which is the same die inside the GeForce RTX 3070. However, the Ti variant comes with two additional SMs. This configuration allows the GeForce RTX 3070 Ti to sport up to 6,144 CUDA cores, 192 Tensor cores and 96 RT cores.
Nvidia clocks the GeForce RTX 3070 Ti higher than both the GeForce RTX 3070 and GeForce RTX 3080. It features a 1,575 MHz base clock and 1,770 MHz boost clock. With these clock speeds, the GeForce RTX 3070 Ti delivers a FP32 performance up to 21.75 TFLOPs.
The GeForce RTX 3070 Ti ends up being a moderately faster variant of the existing GeForce RTX 3070. In comparison to the GeForce RTX 2070 Super, however, Nvidia is claiming a performance uplift up to 1.5X.
Nvidia GPU Specifications
Graphics Card
RTX 3080
RTX 3070 Ti
RTX 3070
RTX 2070 Super
Architecture
GA102
GA104
GA104
TU104
Process Technology
Samsung 8N
Samsung 8N
Samsung 8N
TSMC 12FFN
Transistors (Billion)
28.3
17.4
17.4
13.6
Die size (mm^2)
628.4
392.5
392.5
545
SMs / CUs
68
48
46
40
GPU Cores
8,704
6,144
5,888
2,560
Tensor Cores
272
192
184
320
RT Cores
68
96
46
40
Base Clock (MHz)
1,440
1,575
1,500
1,605
Boost Clock (MHz)
1,710
1,770
1,725
1,770
VRAM Speed (Gbps)
19
19
16
14
VRAM (GB)
10
8
8
8
VRAM Bus Width
320
256
256
256
ROPs
96
96
96
64
TMUs
272
192
184
160
TFLOPS FP32 (Boost)
29.8
21.75
20.3
9.1
TFLOPS FP16 (Tensor)
119 (238)
87 (174)
81 (163)
72
RT TFLOPS
58.1
42
39.7
27.3
Bandwidth (GBps)
760
608
512
448
TDP (watts)
320
290
220
215
Launch Date
Sep 2020
June 2021
Oct 2020
Jul 2019
Launch Price
$699
$599
$499
$499
Nvidia only equipped the GeForce RTX 3070 Ti with 8GB of GDDR6X memory at 19 Gbps. This might seem like a disappointment for many consumers, who were expecting at least 10GB, like the GeForce RTX 3080. The GeForce RTX 3070 Ti also has a narrower memory bus, meaning it’s only good for 608 GBps of memory bandwidth.
It remains to be seen whether the GeForce RTX 3070 Ti can provide a sensible performance upgrade over the GeForce RTX 3070. The GeForce RTX 3070 Ti variant rocks a 290W TDP, a significant 31.8% increase over the non-Ti model. As usual, you’ll find the Founders Edition with the 12-pin PCIe power connector. Obviously, Nvidia’s partners will opt for a dual 8-pin PCIe power connector setup instead.
Image 1 of 4
GeForce RTX 3070 Ti Founders Edition (Image credit: Nvidia)
Image 2 of 4
GeForce RTX 3070 Ti Founders Edition (Image credit: Nvidia)
Image 3 of 4
GeForce RTX 3070 Ti Founders Edition (Image credit: Nvidia)
Image 4 of 4
GeForce RTX 3070 Ti Founders Edition (Image credit: Nvidia)
The GeForce RTX 3070 Ti gives Nvidia a chance to realign prices with the current market by bumping the price up to $599. It’s $100 cheaper than the GeForce RTX 3080.
We likely won’t see Nvidia’s new Ampere offerings at their respective MSRP at retail in sufficient quantities, but unlike graphics cards, dreams are free. We’re looking forward to testing and reviewing both the GeForce RTX 3080 Ti and GeForce RTX 3070 Ti in the coming weeks.
Nvidia is unveiling its latest flagship gaming GPU today, the GeForce RTX 3080 Ti. Based on Nvidia’s latest Ampere architecture, the RTX 3080 Ti will succeed the RTX 3080 and promises to deliver 1.5x more performance over the previous generation RTX 2080 Ti. Nvidia is making the RTX 3080 Ti available worldwide on June 3rd, priced from $1,199.
The RTX 3080 Ti looks very much like the RTX 3080, with an identical design and ports. The main difference is a jump in power and VRAM. The RTX 3080 Ti ships with more VRAM than the RTX 3080, with 12GB of GDDR6X in total. This new GPU is essentially as close as you can get to an RTX 3090 on paper, with half the VRAM. The $1,199 price matches the same pricing Nvidia used for the RTX 2080 Ti Founders Edition cards, and it’s $300 less than the giant RTX 3090.
RTX 3080 Ti specs
RTX 3090
RTX 3080 Ti
RTX 3080
RTX 3090
RTX 3080 Ti
RTX 3080
GPU clusters
82
80
68
CUDAs
10496
10240
8704
RTs
82
80
68
Tensor
328
320
272
ROPs
112
112
96
Boost clock
1695MHz
1665Mhz
1710Mhz
Memory
24GB G6X
12GB G6X
10GB G6X
Bus
384-bit
384-bit
320-bit
Bandwidth
936 GB/s
912 GB/s
760 GB/s
TDP
350W
350W
320W
Price
$1,499
$1,199
$699
You’re obviously losing out on an extra 12GB of VRAM if you opt for the RTX 3080 Ti over the 3090, and what will likely be a small improvement in performance for that $300 difference. But the RTX 3090 is giant because it has a far bigger cooler, and the RTX 3080 Ti has the same hardware design as the RTX 3080. That may prompt concerns around how hot the RTX 3080 Ti will run, but we’ll have to wait on reviews to find out if it’s really an issue.
Nvidia is also including its cryptocurrency nerf on the RTX 3080 Ti, much like new RTX 3080 and RTX 3070 cards. Nvidia offers a separate Cryptocurrency Mining Processor (CMP) for Ethereum miners instead. These cards include the best performance for mining and efficiency, but they’re not designed to handle games.
Elsewhere, the RTX 3080 Ti has the same power requirements as the RTX 3090. You’ll need a 750-watt power supply, and the card can draw up to 350 watts of power. That’s the same as the RTX 3090, but the RTX 3080 draws less at up to 320 watts. Just like the RTX 3080 before it, the 3080 Ti also uses Nvidia’s new 12-pin connector. Nvidia will include an adapter that’s compatible with eight-pin cables.
Nvidia’s RTX 3070 Ti also arrives this month.
Nvidia is also launching a second GPU next week, the RTX 3070 Ti. The $599 RTX 3070 Ti will be available on June 10th, and is designed to offer 1.5x more performance over the previous RTX 3070 Super. It will include 8GB of GDDR6X memory.
Both new RTX cards will support all of Nvidia’s ray-tracing, DLSS, and Reflex technologies. More than 50 games now support Deep Learning Super Sampling (DLSS), offering AI-powered performance boosts to games.
While both of Nvidia’s new GPUs will be available this month, actual availability and pricing is obviously going to differ. Everyone has had a hard time getting hold of new RTX 30-series GPUs since their launch last year, and a flagship RTX 3080 Ti and more affordable RTX 3070 Ti isn’t going to help improve that.
A global chip shortage has pushed GPU prices up, and demand is still incredibly high during ongoing supply constraints. Nvidia has already warned these supply issues will continue throughout 2021, so don’t expect to easily be able to get hold of an RTX 3080 Ti or RTX 3070 Ti any time soon.
AMD CEO Lisa Su revealed two key new processors during the company’s Computex 2021 keynote. The $359 Ryzen 7 5700G and $259 Ryzen 5 5600G APU, both of which come to market August 5, 2021, will plug two glaring gaps in the company’s Ryzen 5000 product stack that currently leads our list of Best CPUs.
The new Cezanne chips mark the first new APUs for desktop PCs that you’ll be able to buy at retail since AMD launched the Zen+ “Picasso” models back in 2019. AMD did bring a refresh of those chips to market as the oft-maligned Ryzen Pro “Renoir” series, but in a disappointment to enthusiasts, those chips were destined for professional users and thus not available at retail.
In fact, AMD actually brought the very chips it’s announcing today to OEM systems a few months ago, meaning we already know most of the details about the silicon. The Cezanne APUs, which come with Zen 3 execution cores paired with the Radeon Vega graphics engine, feel like they’re a bit late to retail. The company’s first salvo of Ryzen 5000 processors delivered a stunning blow to Intel as it took the unequivocal lead in desktop PCs, but AMD’s pivot to premium pricing left it exposed with two massive gaps in its product stack. Unfortunately for AMD, Intel’s Rocket Lake blasted in a few months ago and plugged those gaps.
Now AMD’s retort comes as retail availability of a few of the Cezanne chips, though it’s noteworthy the company is still holding back several of its lower-end models from the retail market. Given the ongoing graphics card shortages, these newly revamped APUs are a welcome sight for the gaming market and serve as AMD’s “non-X” chips that traditionally offer more attractive price points at a given core count. That is if AMD can keep them in stock, of course. Let’s take a closer look.
AMD Ryzen 5000 ‘Cezanne’ G-Series Specifications
The Ryzen 5000G lineup spans from four to eight cores, but AMD is only bringing the eight-core 16-thread Ryzen 7 5700G and six-core 12-thread Ryzen 5 5600G to retail, while the Ryzen 3 5300G remains relegated to the OEM-only market (at least for now). AMD also isn’t bringing the 35W GE-Series models to retail, either, as it continues to focus on premium chips during the ongoing global semiconductor shortage.
AMD Ryzen 5000 G-Series 65W Cezanne APUs
CPU
Price
Cores/Threads
Base / Boost Freq.
Graphics Cores
Graphics Frequency
TDP
Cache
Ryzen 7 5800X
$449
8 / 16
3.8 / 4.7 GHz
N/a
N/a
105W
32MB (1×32)
Core i7-11700K (KF)
$374 – $349
8 / 16
3.6 / 5.0
UHD Graphics 750 Xe 32EU
125W
16MB
Ryzen 7 5700G
$359
8 / 16
3.8 / 4.6
RX Vega 8
2100 MHz
65W
20 MB
Ryzen 5 5600X
$299
6 / 12
3.7 / 4.6 GHz
N/a
N/a
65W
32MB (1×32)
Core i5-11600K (KF)
$262 (K) – $237(KF)
6 / 12
3.9 / 4.9
UHD Graphics 750 Xe 32EU
125W
12MB
Ryzen 5 5600G
$259
6 / 12
3.9 / 4.4
RX Vega 7
1900 MHz
65W
19 MB
Ryzen 5 3600
$200
6 / 12
Core i5-11400 (F)
$182 – $157
6 / 12
2.6 / 4.2
UHD Graphics 750 Xe 24EU
65W
12MB
Ryzen 3 5300G
N/a
4 / 8
4.0 / 4.2
RX Vega 6
1700 MHz
65W
10 MB
The 65W eight-core 16-thread Ryzen 7 5700G comes with a 3.8 GHz base, 4.6 GHz boost, and eight Radeon Vega CUs that operate at 2.0 GHz.
The Ryzen 7 5700G addresses the ~$350 price point to plug the sizeable gap between the $449 Ryzen 9 5800X and $299 Ryzen 5 5600X. That big gap left Intel’s Core i7-11700K with plenty of room to operate, but AMD says the new 5700G will plug that gap with CPU performance that slots in perfectly between the other Ryzen 5000 parts, not to mention the strengths borne of the integrated Vega graphics engine.
The 65W six-core 12-thread Ryzen 5 5600G comes with a 3.9 GHz base, 4.4 GHz boost, and seven Radeon Vega CUs that operate at 1.9 GHz.
The 5600G slots in at $259 to plug the gap between the $299 Ryzen 5 5600X and, well, the remainder of AMD’s sub-$299 product stack. AMD’s Ryzen 5 3600 is the only real relevant contender in this price range, and it launched two years ago with the Zen 2 architecture. The 3600 isn’t competitive with Intel’s Rocket Lake Core i5-11600K or -11400, leaving Intel plenty of room to roam uncontested in the budget market (as you can see in our Core i5-11400 review).
Based on suggested pricing, the 5600G contends with the Core i5-11600K and doesn’t do much to address the current value budget champ, the Intel Core i5-11400. That’s largely because AMD has decided not to include the 65W Ryzen 3 5300G, which it ships into the OEM market, in this round of chip releases. It also has yet to release the GE-series chips listed in the table below. AMD hasn’t indicated when the Ryzen 3 or GE-Series Cezanne chips will come to market.
AMD Ryzen 5000 GE-Series 35W Cezanne APUs
CPU
Cores/Threads
Frequency (Up to) Boost / Base
Graphics Cores
Graphics Frequency
TDP
Cache
Ryzen 7 5700GE
8 / 16
3.2 / 4.6
RX Vega 8
2000 MHz
35W
20 MB
Ryzen 5 5600GE
6 / 12
3.4 / 4.4
RX Vega 7
1900 MHz
35W
19 MB
Ryzen 3 5300GE
4 / 8
3.6 / 4.2
RX Vega 6
1700 MHz
35W
10 MB
Of course, integrated graphics are the big attraction for APUs. AMD continues to pair its APUs with the Vega graphics architecture, just as it did with the 4000-series APUs. AMD reworked the architecture for its last go-round — the revamped RX Vega graphics delivered up to ~60% percent more performance per compute unit (CU) than its predecessors, which equated to more graphics performance from fewer CU. We aren’t sure if AMD has made a similar adjustment this time around, but we’re sure to learn more as we get closer to launch.
As with all Ryzen 5000 processors, Cezanne fully supports overclocking, which includes memory, graphics and CPU cores. AMD also says that the auto-overclocking Precision Boost Overdrive (PBO) and adaptive offset features are also supported. The Cezanne chips drop into the same motherboards as the current-gen Ryzen 5000 processors, so X570, B550, X470 and B450 are all supported. As with the other Ryzen models, memory support weighs in at DDR4-3200, though that does vary by DIMM population rules.
(Image credit: AMD)
The new APUs hail from the Ryzen 5000 Mobile family (deep dive here), so they have physically identical silicon that has been transitioned from the FP6 BGA-mounted arrangement found in laptops to the AM4 socket on desktop PC motherboards. AMD then simply tunes the silicon for the more forgiving power limits and thermal conditions of the desktop, meaning that it can uncork the power settings and be more aggressive with boosting activity while being less aggressive with power-sharing/shifting between the CPU and GPU units.
The Zen 3 architecture grants higher L3 cache capacities than we’ve seen with AMD’s past APUs. For instance, the eight-core 16-thread Ryzen 7 5700G now has 20MB of L3 cache compared to its eight-core predecessor that came with 12MB. These are the natural byproducts of the Zen 3 architecture and should benefit general iGPU performance, too.
However, in contrast to the existing Ryzen 5000 chips for the PC, the APUs come as a single monolithic die. That results in a less cache than we see with the chips without integrated graphics, like the eight-core Ryzen 5 5600X. The 5600X comes with 32MB of L3 cache, which is significantly more than the 16MB of L3 cache found on the eight-core Ryzen 7 5700G. We’ll be sure to poke and prod at the cache when the silicon lands in our labs.
Additionally, the 5000G chips have the same I/O controller on the SoC as the mobile parts, so the chips are limited to 24 lanes of PCIe 3.0, as opposed to the 24 lanes of PCIe 4.0 found on the other Ryzen 5000 parts. This comes as the tradeoff of bringing the mobile architecture to the desktop PC, with AMD’s initial decision to stick with PCIe 3.0 for its mobile parts largely being driven by battery life concerns.
AMD Ryzen 5 5700G Gaming and Productivity Benchmarks
(Image credit: AMD)
AMD shared a surprisingly slim selection of its own benchmarks to compare the Ryzen 5 5700G with Intel’s Core i7-11700. AMD’s test notes are also lacking. As with all vendor-provided benchmarks, you should view these with the requisite amount of skepticism.
As expected, AMD’s benchmarks show notable performance advantages across the board, especially when gaming on the 5700G’s Radeon Vega 8 graphics compared to the -11700’s UHD Graphics 650 with the Xe architecture. AMD’s last batch of 5000G comparative benchmarks were much more expansive when it compared Cezanne to the Comet Lake chips, but the Rocket Lake comparisons are far more limited. We’ll suss all that out in the review.
Ryzen 5000G Pro Series Desktop Processors
Image 1 of 4
(Image credit: AMD)
Image 2 of 4
(Image credit: AMD)
Image 3 of 4
(Image credit: AMD)
Image 4 of 4
(Image credit: AMD)
AMD also released its Ryzen 5000G Pro series today. As you can see in the slides above, aside from a few extra professional features, they’re identical to the client chips.
Thoughts
Overall the Cezanne desktop APUs look promising, and AMD’s pricing goes a long way to addressing the notable price gaps that come from its lack of value “non-X” chips with the Ryzen 5000 generation, an exclusion that has received plenty of criticism from the enthusiast community.
AMD’s timing for desktop APUs could be a bit better — Intel’s value Rocket Lake chips have been on the market for several months, and the continuing chip shortage coupled with cryptomining has destroyed any chance of scoring a reasonably priced GPU, at least for now. That means a chip with competitive 1080p gaming performance will be a hit with enthusiasts looking to wait out the current GPU crisis.
That said, we’re still seeing a complete lack of AMD’s cheap chips on the market, so the company’s decision to keep the Ryzen 3 and 35W GE-Series models off the retail market is disappointing. It makes good business sense given the state of the market (AMD sells every single high-end chip it punches out), but we’d like to see some improvement on the lower end of the market.
The Ryzen 5000G chips come to market on August 5, 2021. As you can imagine, we’ll have the full story when reviews arrive near that same time.
AMD introduced its new Radeon RX 6000M-series laptop graphics at Computex, during a keynote by AMD’s CEO, Dr. Lisa Su. The new mobile graphics lineup is made up of the top-end AMD Radeon RX 6800M, a mid-range RX 6700M and the entry level RX 6600M. For now at least, the GPUs are being paired in systems from laptop vendors with AMD’s Ryzen processors for what the company calls “AMD Advantage.”
These are the first laptop GPUs from AMD that use its RDNA 2 architecture, with Infinity Cache for higher memory bandwidth, low power consumption (AMD claims near 0 watts at idle) and high frequencies even when the system is running at low power. The company is claiming up to 1.5 times performance over last-gen RDNA graphics and up to 43% lower power consumption.
AMD Radeon RX 6800M
AMD Radeon RX 6700M
AMD Radeon RX 6600M
Compute Units
40
36
28
Game Clock
2,300 MHz
2,300 MHz
2,177
Memory
12GB GDDR6
10GB GDDR6
8GB GDDR6
Infinity Cache
96MB
80MB
32MB
AMD Smart Access Memory
Yes
Yes
Yes
AMD Smart Shift
Yes
Yes
Yes
Power Targets
145W and above
Up to 135W
Up to 100W
Resolution Targets
1440p
1440p/1080p
1080p
The most powerful of the new bunch is the AMD Radeon RX 6800M, which will be available starting June 1 in the Asus ROG Strix G15 Advantage Edition. It has 40 compute units and ray accelerators, along with a 2,300 MHz game clock, 12GB of GDDR6 memory and a 96MB cache. It will also be compatible with AMD SmartShift and Smart Access Memory.
Image 1 of 4
(Image credit: AMD)
Image 2 of 4
(Image credit: AMD)
Image 3 of 4
(Image credit: AMD)
Image 4 of 4
(Image credit: AMD)
AMD compared the ROG Strix G15 with the RX 6800M and a Ryzen 9 5900HX to a 2019 MSI Raider GE63 with a 9th Gen Intel Core i7 processor and an RTX 2070, claiming up to 1.4 times more frames per second at 1440p max settings in Assassin’s Creed Valhalla and Cyberpunk 2077, 1.5 times the performance in Dirt 5 and 1.7x more frames while playing Resident Evil: Village.
In closer comparisons, to an RTX 3070 (8GB) and RTX 3080 (8GB), AMD claimed its flagship GPU was typically the top performer – within a frame or so – in several of those games, as well as Borderlands 3 and Call of Duty: Black Ops Cold War, though it’s unclear which settings and resolutions were used for these tests.
Unlike Nvidia, AMD isn’t aiming for 4K gaming. The most powerful of the cards, the RX 6800M, aims for a power target of 145W and above and is designed for 1440p.
(Image credit: AMD)
The middle-tier AMD Radeon RX 6700M is designed for 1440p or 1080p gaming, depending on the title. It has 36 compute units with a 2,300 MHz game clock, 10GB of GDDR6 RAM and an 80MB infinity cache, as well as the same support for SmartShift and SAM. AMD says these will ship in laptops “soon.’ It also said that the GPU will allow for 100 fps gaming at 1440p and high settings in “popular games,” though didn’t specify which games it was referring to.
Image 1 of 3
(Image credit: AMD)
Image 2 of 3
(Image credit: AMD)
Image 3 of 3
(Image credit: AMD)
The RX 6600M sits at the bottom of the stack for gaming at 1080p. AMD compared it to an RTX 3060 (6GB) on 1080p max settings, and found that it led in Assassin’s Creed Valhalla, Borderlands 3 and Dirt 5. It was five frames behind in Call of Duty: Black Ops Cold War in AMD’s tests, and there was a one-frame difference playing Cyberpunk 2077. Like the RX 6800M, the 6600M will start shipping on June 1.
(Image credit: AMD)
AMD Advantage Laptops
AMD is now referring to laptops with both AMD processors and graphics as offering the “AMD Advantage.” The company says these designs should offer great performance because of power sharing between the CPU and GPU.
Image 1 of 2
(Image credit: AMD)
Image 2 of 2
(Image credit: AMD)
AMD says its technologies can achieve up to 11% better performance in Borderlands 3, 10% in Wolfenstein Young Blood, 7% in Cyberpunk 2077 and 6% in Godfall.
(Image credit: AMD)
Additionally, the company says AMD Advantage laptops will only have “premium” displays — either IPS or OLED, but no VA or TN panels. They should hit or surpass 300 nits of brightness, hit 144 Hz or higher and use AMD FreeSync.
Image 1 of 3
(Image credit: AMD)
Image 2 of 3
(Image credit: AMD)
Image 3 of 3
(Image credit: AMD)
Each laptop should come with a PCIe NVMe Gen 3 SSD, keep the WASD keys below 40 degrees Celsius while gaming and allow for ten hours of video on battery. (AMD tested this with local video, not streaming.)
(Image credit: AMD)
The first of these laptops is the Asus ROG Strix G15, with up to a Ryzen 9 5900HX and Radeon RX 6800M, a 15-inch display (either FHD at 300 Hz or WQHD at 165 Hz) with FreeSync Premium, liquid metal for cooling both the CPU and GPU along with a vapor chamber. It will launch in mid-June.
(Image credit: AMD)
The HP Omen 16 will also come with a 165 Hz display with up ao a Ryzen 9 5900Hx and AMD Radeon RX 6600M for 1080p gaming. It will launch sune on JD.com, then become available worldwide.
(Image credit: AMD)
In June, we should see more releases from HP, Asus, MSI and Lenovo.
AMD has announced its long-awaited Radeon RX 6000M series of mobile GPUs, featuring its RDNA 2 architecture.
Today’s release consists of three chips: the RX 6800M (configurable at 145W and above), the RX 6700M (up to 135W), and the RX 6600M (up to 100W). AMD says the flagship 6800M delivers the fastest AMD graphics for laptops yet; it claims the 6800M will run modern AAA games at frame rates that are comparable to or better than those of Nvidia’s mobile RTX 3080. It’s also purported to outperform Nvidia’s chip while gaming on battery.
AMD says the RX 6700M will deliver up to 100fps “in popular games” at 1440p resolution. The 6600M is better for “epic 1080p gaming.” Keep an eye out for independent reviews of these chips in the coming weeks for better idea of the performance you can expect from each one.
The 6000M series will be available starting on June 1st.
Radeon RX 6000M series
GPU
Power target
Compute units / ray accelerators
Game clock (MHz)
Memory (GDDR6)
Infinity cache
GPU
Power target
Compute units / ray accelerators
Game clock (MHz)
Memory (GDDR6)
Infinity cache
Radeon RX 6800M
145W and above
40
2300
12GB
96MB
Radeon RX 6700M
Up to 135W
36
2300
10GB
80MB
Radeon RX 6600M
Up to 100W
28
2177
8GB
32MB
AMD also announced AMD Advantage, a new “design framework initiative” meant to encourage OEMs to include certain features on their AMD-powered systems, and to indicate to consumers which Ryzen- and Radeon-powered laptops AMD thinks are the best. It appears to be a similar idea to Intel’s Evo program, but it’s just for gaming laptops, and the standards look much more stringent. It AMD Advantage laptops are expected to include the following:
AMD Ryzen 5000 mobile processors, Radeon 6000 graphics and Radeon software
Support for AMD’s Smart Acess Memory and Smart Shift technology
A display that reaches at least 300 nits of brightness, covers either 100 percent of the sRGB gamut or 72 percent of the NTSC gamut, has at least a 144Hz refresh rate and low latency, and supports AMD Freesync
At least one NVME PCIE Express Gen 3 SSD
The ability to maintain a surface temperature under 40 degrees Celsius on the WASD keys
Over 10 hours of video playback on battery
It’s unclear how many laptops will actually meet all of these standards. Forty degrees Celsius is close to as hot as keyboards commonly get in the center. But there aren’t too many gaming rigs that reliably break 10 hours of video playback on battery, and plenty of the best gaming laptops out there max out below 300 nits of brightness. That said, all kinds of Intel Evo-certified laptops also don’t meet all the Evo requirements in my testing — units and methodologies can vary.
The first AMD Advantage laptop to be announced is Asus’ new ROG Strix G15. This can be configured with up to a Ryzen 9 5900Hx, a Radeon RX 6800M, and a 15-inch WQHD 165Hz display with 3ms response time. The G15 will be available at Best Buy in June.
TeamGroup’s Xtreem ARGB memory, which holds a spot on our list of best RAM, now arrives with a kit capacity up to 256GB. Conscious that not everyone chases speed, the memory vendor has cooked up a big memory kit for Intel and AMD HEDT owners that want to maximize the memory on their systems.
The Xtreem ARGB 256GB memory kit is comprised of eight memory modules that are 32GB each. Therefore, you’ll need a motherboard that has eight DDR4 memory slots to leverage this new kit. TeamGroup is playing it safe and binned the memory kit for DDR4-3600, which is the sweet spot for AMD’s Ryzen processors. The memory modules have their timings configured to 18-22-22-42.
Admittedly, DDR4-3600 at C18 isn’t the best of what TeamGroup has to offer. The brand sells the Xtreem ARGB DDR4-3600 in both 16GB (2x8GB) and 32GB (2x16GB) flavors with 14-15-15-35 timings. A 256GB memory kit is already expensive as it is. TeamGroup probably chose a lower bin to help reduce the final cost of the memory kit.
Although the manufacturer didn’t specify the operating voltage, we suspect that the Xtreem ARGB DDR4-3600 C18 memory modules only pull 1.35V so there is headroom for optimizing the timings. However, your mileage will depend on what kind of integrated circuits TeamGroup is utilizing for the DDR4-3600 C18 variants. If it’s not Samsung B-die, there’s less chance of getting these down to C14.
TeamGroup didn’t reveal the availability or pricing for the Xtreem ARGB DDR4-3600 C18 256GB (8x32GB) memory kit. Similarly-specced offerings start at $1,499.99 so that’s basically the baseline for a memory kit of this caliber.
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.
