Motorola unveiled the Moto G10 and Moto G30 last month with big screens, big batteries, and quad rear cameras. Today the company announced it will bring the Moto G30 to India on March 9 alongside a new smartphone – Moto G10 Power.
The Moto G10 Power and Moto G30 will launch in India at 12PM local time (6:30AM UTC) and will be sold through Flipkart, which reveals both smartphones will run Android 11 out of the box and come with a feature called ThinkShield.
Since the Moto G30 was introduced last month, we know it will come with Snapdragon 662 SoC, 6.5″ 90Hz HD+ LCD, and a 5,000 mAh battery with 20W charging. The smartphone will feature a 13MP selfie camera, and its rear panel will be home to a quad camera setup consisting of a 64MP primary, 8MP ultrawide, 2MP macro, and 2MP depth units.
The Moto G30 also has a fingerprint reader on its back for biometric authentication. The G30 has two memory configurations – 4GB/128GB and 6GB/128GB. But it’s unclear what amount of RAM and storage it will have in India.
The Moto G10 Power, on the other hand, is a new smartphone, but considering the “Power” moniker, we believe it will be a Moto G10 with a bigger battery and faster charging.
The image of the Moto G10 Power shared by Motorola tells us the smartphone looks similar to the vanilla G10 – a notched display with a quad camera on the back accompanied by a fingerprint scanner.
The Moto G10 has a Snapdragon 460 SoC at the helm, and it’s built around a 6.5″ HD+ LCD. It packs a 5,000 mAh battery with 10W charging and has a total of four cameras on board – an 8MP selfie shooter with a 48MP primary camera on the back joined by an 8MP ultrawide, 2MP macro, and 2MP depth modules.
The iQOO Neo5 arriving on March 16 was rumored to support 88W wired charging, but 3C revealed that won’t be the case, and the smartphone will charge at up to 66W. Today, iQOO officially confirmed on Weibo that Neo5 will indeed feature 66W wired charging.
Additionally, iQOO confirmed that Neo5 will pack a dual-cell 4,440 mAh battery, which will go from flat to 50% in 10 minutes and get a full charge in 30 minutes with the bundled 66W adapter. The charger will also support 45W PD and allow you to charge other devices like laptops and cameras.
iQOO hasn’t detailed the Neo5’s specs yet, although the company previously confirmed that the smartphone will come with Snapdragon 870 SoC, 120Hz screen, and UFS 3.1 storage.
The Neo5 will have 12GB RAM onboard, but that will feel like 15GB thanks to the Memory Fusion technology.
With the official unveiling still over a week away, you can expect iQOO to confirm more features of the Neo5 in the days leading to its official announcement.
The Samsung Galaxy A52 5G has not been officially announced yet – that is expected to happen later this month – but a retailer in Saudi Arabia has put the phone on sale already, listing the full specs and everything.
The A52 5G has a 6.5” Super AMOLED display with 1,080 x 2,400 px resolution. We’re not seeing a refresh rate listed but it should be 120Hz (the rumors were right about everything else). The fingerprint reader is built into the display and the whole thing is guarded by Gorilla Glass (we’re not 100% clear on the version).
The Snapdragon 750G is the brains of the operation, as expected. It is hooked up to 8 GB of RAM and 128 GB storage and there’s a microSD slot for up to 1 TB more (the dual card slot has a hybrid design). Only one memory configuration is listed, but that’s something that varies by region.
Samsung Galaxy A52 5G (in Black)
Anyway, the Galaxy A52 5G has a 64MP main camera that can record 4K videos. Next up is a 12MP camera (presumably ultrawide) and two 5MP modules (macro and depth is our bet). The selfie camera has a 32MP sensor.
The phone’s 4,500 mAh battery is charged over USB-C at 25W. There’s no 3.5 mm headphone jack by the looks of it, but the phone does have NFC.
Note that the OS version is listed as Android 10, but we believe this to be a mistake – the phone ran Android 11 when it went through Geekbench.
The retailer is charging SAR 1,650 for the Galaxy A52 5G, which works out to $440/€365. This matches the rumored prices. There should be a 4G model as well, but that one isn’t showing up yet.
João Silva 8 hours ago Featured Announcement, Graphics
It has been a while since the RTX 3080Ti was last in the news. The latest rumours claim that the upcoming GPU will feature the crypto mining limiter that Nvidia added to the RTX 3060. Some details on memory configurations for the GPU were also shared.
This apparent leak comes from @kopite7kimi. According to their source, the RTX 3080Ti is still in development and Nvidia is planning to include 12GB of GDDR6X VRAM running at 19 Gbps across a 384-bit memory bus. This will bring the RTX 3080Ti’s memory performance somewhere in between the RTX 3080 and the RTX 3090.
It will have 12G 19Gbps VRAM and an ETH mining nerf too.
— kopite7kimi (@kopite7kimi) March 4, 2021
As anticipated, Nvidia is also said to be putting its new anti-mining tech into the RTX 3080Ti. This was first introduced with the RTX 3060 last month, which limits the hash rate when mining certain crypto currency, like Ethereum.
However, as pointed out by RedPandaMining, the limiter doesn’t work on all mining algorithms. Performance is reduced on the Ethhash and other memory-heavy algorithms but some other coins can still be mined at uncapped performance. Perhaps this is something that Nvidia will fix in future iterations of the limiter on future GeForce GPUs.
The Nvidia RTX 3080Ti is currently expected to ship with a GA102 GPU, 10240 CUDA cores, 80 RT Accelerators and 320 Tensor Cores. A release date is currently unknown.
Discuss on our Facebook page, HERE.
KitGuru says: Do you agree with Nvidia’s decision to limit the hash rate of future GeForce cards? Are any of you waiting on the RTX 3080Ti before upgrading?
Become a Patron!
Check Also
KitGuru – Children’s Hospice Charity PC Auction
We can all agree that the last 12 months have been challenging and stressful for everyone, and I wanted to utilise Kitguru in a positive way for a great cause. Over the next two weeks, we are holding a silent auction via email for our readers and supporters, with 100% of the proceeds being donated to the Northern Ireland Children’s Hospice. Additionally KitGuru will be donating the same amount as the winning bid, effectively doubling the donation to the charity. This is a very worthy cause that we want to support this year and everyone who enters will know that they are potentially going to help children in real need.
Following on from AMD’s RX 6700 XT announcement stream last night, MSI revealed its line-up of custom-cooled solutions. Besides an AMD Reference design model, MSI will also be releasing Mech and Gaming series Radeon RX 6700 XT graphics cards.
Featuring the company’s Twin Frozr 8 thermal cooling solution, the dual-fan cooler of the MSI Radeon RX 6700 XT Gaming X uses the Torx Fan 4.0 design, connecting the centre of the fan to an outer ring through the fan blades, resulting in increased airflow. Additionally, the Gaming X cooling system also includes a metal backplate for additional passive cooling and protection, thermal pads for additional heat dissipation, and a Zero Frozr mode to stop fans when temperatures are low.
The Radeon RX 6700 XT Gaming X cards also feature MSI’s RGB Mystic Light lighting, which users can synchronise with other Mystic Light-compatible devices.
MSI Mech series cards also feature a dual-fan cooler, but instead of using Torx Fan 4.0 design, it employs the older Torx Fan 3.0. Similar to the Gaming X cards, the Mech series also features a metal backplate, thermal pads for extra heat dissipation, and Zero Frozr mode to prevent the fans from spinning at low loads.
Base and boost clocks, memory clock, and TGP of the Gaming X and Mech cards have not been disclosed yet. Both cards are powered by 2x 8-pin PCIe power connectors and MSI recommends the use of a 650W PSU. Available video outputs include 3x DisplayPorts 1.4 and an HDMI 2.1 port.
Custom MSI Radeon RX 6700 XT graphics cards will launch in mid-March 2021. Discuss on our Facebook page, HERE.
KitGuru says: Would you choose a custom MSI Radeon RX 6700 XT graphics card over the reference one? Which MSI Radeon RX 6700 XT graphics card do you prefer?
Become a Patron!
Check Also
Ghost of Tsushima’s developers will become ambassadors for the island
Ghost of Tsushima took many by surprise when it launched last Summer, offering a beautiful …
During the announcement of the Radeon RX 6700 XT graphics card, AMD also confirmed that it will bring Smart Access Memory support to the Ryzen 3000 series desktop processors, excluding the Ryzen 5 3400G and the Ryzen 3 3200G.
It seems AMD has taken the community’s feedback into consideration and decided to enable SAM (Smart Access Memory) on Zen2-based processors. The confirmation was made during the “Where Gaming Begins: Episode 3” stream yesterday. AMD also confirmed that SAM will bring about the same performance increase (up to 16%) on systems using Ryzen 3000 CPUs as it does on the ones using Ryzen 5000 CPUs.
AMD Smart Access Memory is a technology based on the PCIe Resizable BAR standard. By using SAM/PCIe Resizable BAR, the data channel between the CPU and the GPU is expanded, increasing bandwidth, which results in a theoretical performance increase.
This is a move that we did see coming eventually. We have previously shared screenshots of Asus and MSI motherboards with SAM enabled while using Zen and Zen2 processors, meaning that the ability to enable it was more of a restriction from AMD than an incompatibility.
Nvidia also has announced recently that it will bring PCIe Resizable BAR support to its RTX 30 series graphics cards. Nvidia mentioned that AMD 400-series motherboards would be compatible, but only when using Zen3 processors. Considering AMD’s announcement, Nvidia might extend it to the Zen2 processors as well.
Discuss on our Facebook page, HERE.
KitGuru says: Do you own an AMD Ryzen 3000 series processor? Will you enable SAM once AMD has released the appropriate BIOS for your motherboard?
Become a Patron!
Check Also
Ghost of Tsushima’s developers will become ambassadors for the island
Ghost of Tsushima took many by surprise when it launched last Summer, offering a beautiful …
DigiTimes today reported that DRAM prices are expected to rise throughout 2021 as Samsung, SK Hynix, and Micron maintain flash memory production levels despite increasing demand.
The report claimed that suppliers are cautious about building additional DRAM capacity which could result in limited global bit output growth. Demand for flash memory is expected to rise in the coming months.
DigiTimes reported that for the first quarter of 2021, spot prices for speciality DRAM chips have risen and supply has fallen short of demand. Things aren’t expected to improve in upcoming quarters.
The report meshes with comments made by Micron CFO David Zinsner at a Morgan Stanley technology conference on Wednesday. Zinsner said he expects supply to be taut all year, according to Seeking Alpha, which “bodes well” for DRAM pricing.
Micron previously said that first quarter memory bit shipments would be reduced by the power outage and earthquake that affected two fabs in Taiwan in December 2020. Those reductions were all but guaranteed to result in higher DRAM prices.
Sources believe speciality DRAM prices will continue to rally through 2022 because currently quotes for wafers are far lower than the $3,000 peak set in 2018.
DRAM pricing can be hard to predict, especially that far into the future. Does this seem like the year for anything to happen as expected in a market so volatile that a one-hour-long power outage at a single fab can raise prices?
Supply constraints have already limited the availability of CPUs, GPUs, and the devices they’re found inside. Now manufacturers will have to contend with limited flash memory supply, too, and it’s hard to see how that “bodes well” for anyone other than the companies enjoying these increased prices while they last.
Asus has introduced its family of custom graphics cards based on the AMD Radeon RX 6700 XT GPU with 2560 stream processors and a 192-bit memory interface. The new boards formally belong to higher-end/performance mainstream segment of the market, yet they have almost all the features that are found on enthusiast-grade graphics cards.
Asus’s family of custom Radeon RX 6700 XT video cards includes three models: the Dual Radeon RX 6700 XT, the TUF Gaming Radeon RX 6700 XT, and the ROG Strix Radeon RX 6700 XT. The ROG Strix-badged Radeon RX 6700 XT card naturally sits on top of the stack, the TUF-branded board combines durability and performance, whereas the Dual card is slightly more compact than the other two and can fit into Micro-ATX and Mini-ITX cases.
(Image credit: Asus)
All the products rely on their own custom PCB that is both tall and long, which most probably indicates an enhanced voltage regulating module (VRM). The cards also feature a hardware BIOS switch to automatically load different clocks and voltage settings, so it is safe to say that the three Radeon RX 6700 XT boards from Asus feature higher frequencies than those recommended by AMD.
As far as power delivery is concerned, the ROG and the TUF Radeon RX 6700 XT cards use two eight-pin auxiliary PCIe power connectors, whereas the Dual Radeon RX 6700 XT board comes with one eight-pin and one six-pin PCIe power connectors.
(Image credit: Asus)
All of the Asus Radeon RX 6700 XT cards are equipped with a large proprietary cooling system that is wider than two slots, feature two (in case of the Dual model) or three fans that stop at idle, multiple heat pipes, and a backplate to improve cooling and guarantee longevity. Since the new custom-built Radeon RX 6700 XT cards are designed mostly for DIY enthusiasts, their coolers are equipped with addressable RGB LEDs, making sure that the card blends in with your chosen aesthetic. The ROG has plenty of LEDs, whereas the TUF and Dual boards have more modest lighting.
(Image credit: Asus)
Asus yet has to announce final specifications of its Dual Radeon RX 6700 XT, TUF Gaming Radeon RX 6700 XT, and ROG Strix Radeon RX 6700 XT cards, so at this point it is impossible to tell how the new cards stack up against AMD’s reference design.
Asus did not disclose pricing or availability dates of its Radeon RX 6700 XT, though given the current deficit of GPUs and components, this is hardly surprising.
Nvidia might be preparing to launch a new RTX 3050 Ti GPU, according to a leak from laptop maker Asus. VideoCardz reports that Asus has published updated specifications for its latest TUF Dash F15 gaming laptop, and it mentions an unreleased Nvidia GeForce RTX 3050 Ti Laptop GPU.
The numbering scheme suggests Nvidia could be about to launch a fourth 30-Series laptop GPU to join the existing RTX 3060, RTX 3070, and RTX 3080 cards. Asus’ website mentions the card has 4GB of memory, but the company hasn’t included any further details on clock speeds etc. The RTX 3050 Ti is rumored to be an entry-level option, with a 128-bit memory bus and 60W TGP.
The Asus TUF Dash F15 laptop listing.
Nvidia’s launch of RTX 30-Series laptop GPUs has also been a little confusing. Nvidia is no longer adding “Max-Q” labels to the weaker, more power-efficient variants of its laptop-grade GPUs, and it’s more difficult to tell exactly what GPU you’re getting. Nvidia has started requiring laptop makers to be more transparent about RTX 30-series specs, but the previous Max-Q branding made it a lot easier to work out a laptop’s GPU performance without having to know its specific clock speeds.
We’ve reached out to both Nvidia and Asus to comment on the RTX 3050 Ti appearance, and we’ll update you accordingly.
With memory prices at one of the lowest points in years, now is a great time to be looking for memory upgrades. Summer hardware releases are in full swing, there is competition from both Intel and AMD, and the Red brand has thoroughly fixed the memory issues of generations past. No longer do users have to worry about memory compatibility or shopping for expensive AMD-branded kits. With 3200 MHz natively supported on the new Ryzen platform, options for enthusiasts have never been more open.
Corsair is a brand that needs no introductions. The California-based company has been a staple of the enthusiast community for decades, producing power supplies, peripherals, SSDs, coolers, and even entire pre-built systems. One of the brand’s most storied and successful family of products consists of their high-performance memory kits. Balancing high performance with style and impeccable quality, Corsair memory kits are a staple of the enthusiast hardware market.
The Corsair Vengeance RGB Pro SL is a lower-profile take on the popular Corsair Vengeance RGB Pro. Corsair aims to give builders with potential cooler clearance concerns a no compromise option. As for performance, my test kit is a 2×8 GB kit of 3600 MHz at 18-22-22-42 and 1.35 V. Let’s see how this new edition of the Corsair Vengeance RGB Pro SL performs!
A number of European retailers have started to list Samsung’s yet-to-be-announced Samsung 980 (non-Pro) SSDs. Despite expectations, the new units are not inexpensive SSDs with a PCIe 4.0 x4 interface, but are rather unpretentious midrange PCIe 3.0 x4 drives that share some technologies with the flagship.
The Samsung 980 SSDs come in 250GB, 500GB, and 1TB versions. The drives are based on the company’s latest Pablo controller as well as 136-layer TLC V-NAND memory, just like the company’s top-of-the-range 980 Pro drives. The controller supports AES 256-bit encryption as well as TCG/Opal IEEE 1667 coding, so the new SSDs can be used for aimed at enterprises, government agencies and other entities that require advanced security. The new M.2-2280 SSDs do not use any DRAM cache and are therefore cheaper to produce than Samsung’s typical PCIe SSDs that usually come with a DRAM cache.
From sequential performance point of view, Samsung rates the new 980 drives for an up to 3,500 MB/s sequential read speed as well as an up to 3,000 MB/s sequential write speed. As for random performance, advanced versions of the Samsung 980 are claimed to be capable of up to 500,000 random reads (4KB, QD32) as well as up to 480,000 IOPS random writes (4KB, QD1).
(Image credit: Samsung)
Since the Samsung 980 SSDs feature a PCIe 3.0 x4 interface, unlike their 980 Pro counterparts, it is not surprising that the new drives are considerably slower when compared to the flagship products from Samsung (and other SSDs using a PCIe 4.0 x4 bus). What is surprising is that the new Samsung 980 SSDs are slower when compared to their predecessors, the Samsung 970 Evo Plus drives in such metrics as sequential write speed (3,000 MB/s vs. 3,300 MB/s), random read speed (500,000IOPS vs. 620,000 IOPS), and random write speed (480,000 IOPS vs. 560,000 IOPS).
Based on data from Geizhals.eu (via ComputerBase.de), Samsung’s 980 250GB drive costs €70, the 500GB version is priced at €90, and the 1TB model carries a €150 price tag. Meanwhile, Samsung’s 970 Evo Plus 250GB can be obtained for €54, the 500GB flavor is available for €74, whereas the 1TB version can be bought for €144.
Right now, Samsung’s 980 SSDs do not really look competitive with its direct predecessors at the given prices, but we’ll have to test them. But since the new drives are based on 136-layer TLC V-NAND and lack DRAM cache, it is possible that their manufacturing costs are lower than those of the 970 Evo, which will allow Samsung to lower prices of the 980 SSDs in the future.
European retailers intend to start sales of the Samsung 980 SSDs on March 30, 2021.
Light and snappy, a little too flashy and packing potential, the Marsback M1 is an interesting 75% Bluetooth mechanical keyboard that hit Kickstarter this week (early bird price of $159). It’s an attempt at the best gaming keyboard that experiments freely with a chunky, frosted base, over-the-top RGB and extra peppy switches while pulling in helpful features, like hot-swappable switches, that aren’t as common as we might like.
Marsback sent us a functioning prototype to test out. Ultimately, it’s a bold design that, at first glance, one Tom’s Hardware editor couldn’t decide if he loved or hated.
Marsback M1 Specs
Switches
Marsback MBS-I (tested), MBS-II or MBS-III&
Lighting
Per-key RGB
Onboard Storage
2 profiles
Connectivity
USB Type-C to USB Type-A, rubber
Additional Ports
None
Keycaps
PBT plastic
Construction
Polycarbonate plastic
Software
Marsback Pro
Dimensions (LxWxH)
12.6 x 5.2 x 1.7 inches (321 x 131 x 42.8mm)
Compatibility
Windows, Mac, iOS, Android, Dos, Linux, Unix
Who Is Marsback?
Let’s get this out of the way. Many are skeptical about buying something off crowdfunders like Kickstarter as there’s a chance it’ll never come to market. The good news is that this isn’t Marsback’s first product. The company, which started in 2018 with 3G consumer gadgets, also made the Zephyr gaming mouse with a built-in fan, which appears to be available outside of Kickstarter on Marsback’s website.
For better or worse, Marsback also acknowledged consumer skepticism about crowdfunding projects on the M1’s Kickstarter page, admitting that due to the keyboard’s large (6,000mAh) battery, some countries won’t allow it to be air shipped, which would result in slower delivery.
The company also pointed to budget concerns, stating that the keyboard’s hand-assembled, homemade mechanical switches are affected by increasing labor costs. It said that “keeping it affordable, yet high quality is never easy. But instead of aiming for making money, we aim for giving you an experience like no other”
A Marsback rep further addressed potential concern, telling us, “We will continue to sell the keyboard after the campaign [ends] on Kickstarter, both on our official website and Amazon store. And the Marsback M1 keyboard is not our first Kickstarter campaign. Before that, we’ve fulfilled the Zephyr gaming mouse.”
More good news comes from Marsback saying the M1 is “already in production,” with the first step being making and testing the switches. It’s expecting to ship the M1 in early July. The company first came up with the M1 in March 2019.
In a statement announcing the M1, Jack Walker, co-inventor of Marsback M1, said, “The Kickstarter campaign will help us manufacture our units and finalize our packaging components so M1 can arrive in homes across the world as soon as possible.”
Design
Image 1 of 4
(Image credit: Tom’s Hardware)
Image 2 of 4
(Image credit: Tom’s Hardware)
Image 3 of 4
(Image credit: Tom’s Hardware)
Image 4 of 4
(Image credit: Tom’s Hardware)
With its thick, heavy, white-frosted base, the M1’s keys look like they’re sitting in a block of dull ice. The body was CNC milled out of polycarbonate, because of the material’s translucent look and ”lack of tooling costs,” according to Marsback’s Kickstarter. This is actually the hardest and most time-consuming part of making the M1.
This makes for a block of a keyboard that looks much different than the black plastic bases and dark tops comprising the vast majority of mechanical clackers today, whether RGB is on or not. It also makes for a much heavier keyboard than you’d expect of a wireless keyboard with a 75% layout.
Lurking within the M1’s base are 21 surface-mounted LEDs. When brought to life, they produce a memorable blast of RGB. The frosty, diffused exterior often paints a cotton candy effect on the RBGs. But don’t mistake that for being dim. In fact, I’d advise you not to look directly into these lights for more than a couple seconds unless you’re after a quick headache. You can adjust brightness, but now matter how you slice it — be that with a rainbow effect, solid color breathing or a small patch of lights circling the perimeter — this is powerful, look-at-me RGB.
The downside is the visibility of the individual LEDs, which is so prominent that you can basically count each one. This isn’t an issue from a top-down view, but from the sides it makes the RGB glow look unnatural.
On top of that, in true gaming keyboard fashion, the keys themselves have per-key RGB for the ultimate dizzying array of RGB overload. These 84 LEDs for the keys are also very bright, casting a glow on the RGB base. Thankfully, RGB is controllable with hot keys (as are numerous other functions, such as volume control and a calculator launcher). But I wouldn’t remember them without the handy manual Marsback includes in the box. And there were times when some functions wouldn’t work on my prototype (hopefully this will be addressed with the final keyboards).
I tested the Sakura Pink version of this keyboard, which already looks busy sans RGB, with pink flowers and accents. It was hard to find an RGB effect that enhanced this design rather than overpowered it. And it was hard to tell what was the base’s RGB effect and what came from the keys. The black version and, especially, the white version with its white pudding keycaps seem to blend with the RBG effects better, based on the pictures Marsback has provided.
(Image credit: Marsback)
Sadly, I didn’t quite get the Cherry blossom feel I had hoped for with the M1’s Sakura Pink theme. The spacebar comes close, as do the pink petals sprinkled throughout, (although, I could use more of those). But many of the keyboard’s flowers come off more ‘60s and/or Barbie-themed.
Snappy Homebrewed Switches
(Image credit: Tom’s Hardware)
What first stood out to me about the M1 are the linear switches Marsback brewed for them. Despite the linear travel, Marsback promised a pleasantly tactile sensation and delivered.
As the M1’s Kickstarter page states, Marsback designed the switches to provide “better control [over] the keystrokes and bounce sound.” The vendor describes the switches as “light and short on touch” with lubrication applied to each switch’s mandrel and the spring connection for a smooth rebound. You should relube after 10 months though, a Marsback rep told us.
The M1 is available with three different types of Marsback M1 switches:
MBS-I (tested)
MBS-II
MBS-III
Total Travel
4mm ±, 0.6mm
4mm ±, 0.6mm
4mm ±, 0.6mm
Actuation Point
2mm ± 0.6mm
2mm ± 0.6mm
2mm ± 0.6mm
Actuation Force
45g ± 15g
50g ± 15g
55g ± 15g
I’ve only had a couple days with the M1, but typing with the MBS-I switches has felt perky and, as promised, satisfyingly bouncy. At first I did have to snap some switches more firmly into place, but since my test unit is a prototype this can be forgiven. Once the switches were properly installed, I was surprised at how responsive the keys felt. They certainly felt linear, but the quick bounce back did add a pleasant distinct feel to the typing experience while making it seem slightly more enjoyable for long-term typing than using Cherry MX Red linear switches.
The switches also felt very smooth. The lubrication seems effective, but I do wonder if the switches will lose a lot of their magic once this wears off. Combined with the keyboard’s PBT keycaps though, typing felt quick and slightly more stiff with a thinner, less hollow noise than what you get on other keyboards, including the Varmilo MA108M Moonlight keyboard, which has very smooth electrostatic capacitive mechanical switches.
Typing on the M1 felt very stable, with the keycaps showing very little wobble and only if I forced them. The MBS-I switches use 5-pins (compared to the 3 pins many switches, including Cherry MX Reds use), and M1 has a plate on top of its PCB, so this sturdiness isn’t surprising.
It took me some time to adjust to the M1 and avoid typos because the keys felt light and depressed easily. The M1’s Kickstarter goes as far as to say that average typing speed could increase by as much as 5% on the Marsback M1. I didn’t immediately see such results. My average speed was the same and my accuracy was slightly down, but I could see both, especially accuracy, improving over time.
(Image credit: Tom’s Hardware)
And, of course, should you dislike Marsback’s switches, they are hot swappable, making it fast and easy to change your keyboard’s mechanical switches without soldering.
Wireless Connectivity
The M1 has a leg up on productivity in its ability to connect to up to three Bluetooth devices and toggle between them using FN and F2, F3 or F4. I didn’t do a lot of back and forth between PCs in my short time with the M1, but when I did switch between PC to PC, I was able to do so quickly and without issues. The M1 uses Bluetooth 5.1 via a Nordic N52810 chip.
You could also use the M1 with its cable. Marsback even took the extra stepin making the M1 connect via USB-C, which is more common and faster than Micro-USB. However, my prototype annoyingly requires you use hot keys to switch to wired mode.
Marsback opts to pack the M1 with a 6000mAh battery that it says should last up to 2 months with the LEDs off and 12 hours with full RGB. Charging time is pegged at 6 hours. Although, my prototype lacks a power button, which complicates things and is another detail I hope is addressed with retail units.
Software
Mrsback’s M1 uses a 32-bit microcontroller and 128KB of SRAM and will be programmable via Marsback’s software, called Marsback Pro.
(Image credit: Tom’s Hardware)
Marsback Pro will let users make custom macros, set per-key RGB and store up to 2 onboard memory profiles, which you can activate without software by hitting FN and F9 or F10.
Final Thoughts
(Image credit: Tom’s Hardware)
If you’re a believer in crowdfunding and like the way the Marsback M1 looks, it may actually be a good contender in the increasingly flooded mechanical keyboard market.
For one, its feature set is hard to match. Reliable Bluetooth connectivity across three devices isn’t that common in a mechanical keyboard, let alone one with RGB. Add hot-swappable switches into the mix, and it becomes even rarer.
Compared to other Bluetooth gaming keyboards, the M1’s $159 early bird price isn’t horrible either. The Logitech G915 Lightspeed is currently $250, and the tenkeyless version, the Logitech G915 TKL, is on sale for about $200. The Hexgears Venture also started via Kickstarter and is now available for $160, and the Razer BlackWidow V3 Pro is $230.
At much above that, though, the M1 will probably have a hard time keeping up — especially if it doesn’t improve on build quality flaws with the final product. Some will also yearn for the option to connect via wireless dongle, especially for competitive gaming. Additionally, we can’t predict how long the M1 will be available at the early bird pricing or its final retail price.
But typing on this keyboard was impressive and distinct, largely thanks to Marsback’s bouncy linear switches. There are possibilities here. Particularly if you appreciate the phrase “moar RGB.”
Intel’s 11th Generation Rocket Lake processors aren’t due until March 30. However, some retailers are already shipping out orders. One user from the Chiphell forums has gotten his hands on a retail Core i7-11700K, and it would appear that Intel is using a similar memory overclocking concept as AMD’s Infinity Fabric Clock (FCLK), but with Rocket Lake chips.
If you’re not familiar with AMD’s Ryzen processors, many of which sit on our best CPUs list, the FCLK dictates the frequency of the Infinity Fabric, which serves as an interconnect across the chiplets. Adjusting this value allows you to hit higher memory frequency overclocks. By default, the FCLK is synchronized with the unified memory controller clock (UCLK) and memory clock (MEMCLK). Obviously, you can run the FCLK in asynchronous mode, but doing so will induce a latency penalty that negatively impacts performance.
Since Rocket Lake isn’t officially out yet, we’re not completely sure how Intel’s portrayal of the FCLK memory overclocking will work. The BIOS screenshot shows two operational modes for the CPU IMC (integrated memory controller) and the DRAM clock on MSI’s Z490I Unify. Apparently, Gear 1 runs both in a 1:1 ratio, while Gear 2 puts them in a 1/2:1 ratio. It’s similar to how the FCLK works on Ryzen processors.
According to the author of the forum post, his retail Core i7-11700K seems to hit a wall at DDR4-3733, suggesting that DDR4-3733 is the limit at which Rocket Lake’s IMC can run in a 1:1 ratio with the memory clock. In retrospect, the majority of AMD’s Zen 2 processors scale to a 1,800 FCLK (DDR4-3600) with some samples hitting a 1,900 MHz FCLK (DDR4-3800). If Rocket Lake has the same limits, it’s going to lose points since AMD’s latest Zen 3 processors have peaked as high as a 2,000 MHz FCLK (DDR4-4000) before breaking synchronous operation.
Image 1 of 2
Core i7-11700K with DDR4-4000 C18 (Image credit: Chiphell)
Image 2 of 2
Core i7-11700K with DDR4-3600 C14 (Image credit: Chiphell)
It’s too soon to pass judgement whether DDR4-3733 is a hard cap that’s built into the Rocket Lake silicon itself or it’s merely a product of an early and unoptimized microcode. We should point out that the user did his testing on a MEG Z490I Unify motherboard so a proper firmware is required to make Rocket Lake operate properly. The Chiphell forum user provided some RAM benchmarks that reportedly shows the performance impact.
With a DDR4-4000 memory kit with 18-20-20-40 1T timings in asynchronous mode, the user got a latency of 61.3 nanoseconds in AIDA64. Switching over to a DDR4-3600 memory that has 14-14-14-34 2T timings allowed him to decrease the latency to 50.2 nanoseconds, which represents a 18.1% reduction. However, we have to take certain points into consideration. For one, the DDR4-4000 memory kit obviously has very sloppy timings that help contribute to higher latency. Furthermore, the user evidently overclocked the Core i7-11700K’s uncore frequency to 4,100 MHz on the DDR4-3600 run so that probably skewed the results in its favor as well.
We’ll have to wait until the Rocket Lake processors are available to investigate the matter thoroughly. So far, a DDR4-3733 limit certainly doesn’t bode well for Rocket Lake, especially when some of the really pricey Z590 motherboards are advertising memory support above DDR4-5000. In all fairness, Rocket Lake only natively supports memory up to DDR4-3200 so anything higher is technically overclocking in Intel’s book.
We now have official specs for the AMD Radeon RX 6700 XT, yet another poorly kept secret in the land of GPUs you can’t actually buy. We’ve been expecting Navi 22 to join the ranks of the best graphics cards and land somewhere near the RTX 3060 Ti in our GPU benchmarks hierarchy for several months now, and it will officially arrive on March 18, 2021, at 9am Eastern. It will be completely sold out by 9:00:05, and based on recent events like the RTX 3060 12GB, we doubt more than a handful of people will manage to acquire one at whatever MSRP AMD sets.
Speaking of which, AMD revealed that it plans to launch the RX 6700 XT with a starting price of $479. Considering AMD expects it to be faster than the RTX 3070, never mind the RTX 3060 Ti, that’s a reasonable target. The die size also appears to be relatively large, thanks to a still-sizeable Infinity Cache. Here’s the full list of known specs:
(Image credit: AMD)
The AMD Radeon RX 6700 XT comes in with the highest GPU clocks we’ve to date, 2424 MHz. The RX 6800 XT and RX 6900 XT both have 2250 MHz game clocks, though in actual benchmarks, we’ve seen speeds of more than 2500 MHz already — the Game Clock is more of a conservative boost clock. Even with a drop down to 40 CUs (from 60 CUs on the RX 6800), the higher clock speeds should prove relatively potent. Raw theoretical performance sits at 12.4 TFLOPS, and assuming AMD uses 16Gbps GDDR6 again (which is likely), it will have 384GBps of bandwidth. Except, it still has a honking 96MB L3 Infinity Cache.
We were very curious about how far AMD would cut down the Infinity Cache from Navi 21. The answer appears to be “not very much.” The Biggest Navi chip has up to 80 CUs and 128MB of Infinity Cache, so AMD cut the computational resources in half but only lopped off a quarter of the cache. That should keep cache hit rates high, which means effective bandwidth — even from a 192-bit memory interface — should be much higher than Nvidia’s similarly-equipped RTX 3060 12GB.
Let’s go back to that TFLOPS number for a moment, though. 12.4 TFLOPS may not sound like much, but it’s a big jump from the previous gen 40 CU part. The RX 5700 XT had a theoretical 9.8 TFLOPS, and we know the Infinity Cache allows the GPU to get closer to that maximum level of performance in games. That means a 40-50 percent jump in performance might be possible. On the other hand, the RX 6800 with 60 CUs, even at lower clocks, is rated for 16.2 TFLOPS, a 31% increase in compute potential. It also has 33% more memory bandwidth, which means on average it should be at least 20% faster than the 6700 XT, for about 20% more money (well, if MSRP was anything but a fantasy right now).
There are other indications this will still be a performant card, like the 230W board power (just 20W lower than RX 6800). And then there’s the die shot comparison.
(Image credit: AMD)
AMD didn’t reveal all of the specs, but based on that image, it looks like RX 6700 XT / Navi 22 will max out at 96 ROPs (Render Outputs), and the total die size looks to be in the neighborhood of 325mm square, with around 16-17 billion transistors (give or take 10%). That’s quite a bit smaller than Navi 21 (520mm square and 26.8 billion transistors), and perhaps the above images aren’t to scale, but clearly, there’s a lot of other circuitry besides the GPU cores that still needs to be present — the cores and cache only account for about half of the die area.
By way of comparison, Nvidia’s GA106 measures 276mm square with 12 billion transistors, while the GA104 has 17.4 billion transistors and a 393mm square die size. AMD’s Navi 22 should be competitive with GA104, but with a smaller size thanks to its TSMC N7 process technology. However, TSMC N7 costs more and is in greater demand, which leads back to the $479 price point.
Performance, as usual, will be the real deciding factor on how desirable the RX 6700 XT ends up being. AMD provided some initial benchmark results — using games and settings that generally favor its GPUs, naturally. Take these benchmarks with a grain of salt, in other words, but even reading between the lines, the 6700 XT looks pretty potent.
(Image credit: AMD)
That’s eight games, three with definite AMD ties (Assassin’s Creed Valhalla, Borderlands 3, and Dirt 5) and two with Nvidia ties (Cyberpunk 2077 and Watch Dogs Legion). AMD says “max settings,” but we suspect that means max settings but without ray tracing effects. Still, there are a lot of games that don’t use RT, and of those that have it, the difference in visual quality isn’t even that great for a lot of them, so rasterization performance still reigns as the most important factor. Based on AMD’s data, it looks like the RX 6700 XT will trade blows with the RTX 3070.
AMD had a few other announcements today. It’s bringing resizable BAR support, called AMD Smart Access Memory, to Ryzen 3000 processors. That excludes the Ryzen 3200G and 3400G APUs, which of course, are technically Zen+ architecture and have a limited x8 PCIe link to the graphics. AMD also didn’t mention any Ryzen 4000 mobile or desktop APUs (i.e., Renoir), so those may not be included either, but every Zen 2 and Zen 3 AMD CPU will have Smart Access Memory.
(Image credit: AMD)
AMD didn’t discuss future Navi 22-derived graphics cards, but there will inevitably be more products built around the GPU. From what we can tell, RX 6700 XT uses the fully enabled chip with 40 CUs. Just as we’ve seen with Navi 21 and previous GPUs like Navi 10, not all chips are fully functional, and harvesting those partial dies is a key component of improving yields. We expect to see an RX 6700 (non-XT) at the very least, and there are opportunities for OEM-only variants as well (i.e., similar to the RX 5500 non-XT cards of the previous generation). We’ll probably see the RX 6700 (or whatever the final name ends up being) within the next month.
Again, pricing and availability are critical factors for any GPU launch, and while we have no doubt AMD will sell every RX 6700 XT it produces, we just hope it can produce more than a trickle of cards. When asked about this, AMD issued the following statement:
“We hear, and understand, the frustration from gamers right now due to the unexpectedly strong global demand for graphics cards. With the AMD Radeon RX 6700 XT launch, we are on track to have significantly more GPUs available for sale at launch. We continue to take additional steps to address the demand we see from the community. We are also refreshing stock of both AMD Radeon RX 6000 Series graphics cards and AMD Ryzen 5000 Series processors on AMD.com on a weekly basis, giving gamers and enthusiasts a direct option to purchase the latest Ryzen CPUs and Radeon GPUs at the suggested etail and retail price.”
That’s nice to hear, but we remain skeptical. We’ve been tracking general trends in the marketplace, and it’s clear Nvidia continues to sell far more graphics cards than AMD, and it’s still not coming anywhere close to meeting demand. Will Navi 22 buck that trend? Our Magic 8-Ball was cautiously optimistic, as you can see:
(Image credit: redkid.net)
All joking aside, we’re looking forward to another likely frustrating GPU launch. There’s no indication that AMD will follow Nvidia’s example and try to limit mining performance on its future GPUs, but with or without high mining performance, the RX 6700 XT will inevitably sell out. There’s at least some good news in recent GPU mining profitability trends, however: Cards that were making $12–$15 per day last month are now mining in the $6–$8 range and dropping. That’s not going to stop mining completely, but hopefully it means fewer people trying to start up mining farms if the potential break-even point is more than a year away, rather than 3–4 months out.
The AMD Radeon RX 6700 XT officially launches on March 18. We’ll have a full review at that time. Given the pictures AMD sent along, we expect there will be dual-fan reference cards, but AMD will want to shift the bulk of cards over to its AIB partners. We should see various models from all the usual partners, and we’re eager to see how the GPU fares in independent testing. Check back on March 18 to find out.
Below is the full slide deck from AMD’s announcement today.
The bar has just been lowered (in a good way!) for Resizable BAR, the PCI-Express graphics feature that lets CPUs directly access a GPU’s onboard memory to improve gaming frame rates. That’s because AMD just announced it’s bringing the tech to its last-gen Ryzen 3000 series processors, not just the new Ryzen 5000 chips that initially launched with the feature.
AMD originally debuted the feature as “AMD Smart Access Memory,” and you specifically needed an AMD Ryzen 5000 CPU and an AMD Radeon RX 6000 graphics card to make it work. That wasn’t a particularly easy sell, considering both have been incredibly hard to find at retail since they first debuted.
But it was an easy sell for Nvidia and Intel, which announced in January that they’d be adopting Resizable BAR with initial support for Nvidia’s RTX 3000-series laptop GPUs, and later rolling out to RTX 3000-series desktops when paired with both AMD and a selection of both 11th Gen and 10th Gen Intel CPUs. Nvidia just launched support for the new RTX 3060 desktop graphics card last week, with its other new GPUs coming in late March (though you’ll need a motherboard update, too).
With Ryzen 3000, AMD’s actually promising up to 16 percent more performance, compared to the 10 percent both AMD and Nvidia previously offered, though it will really depend on the game. TechSpot discovered that some games could see a 20 percent boost on the AMD CPU+AMD GPU side of things, while other games actually had reduced performance.
Because of the possible downsides, Nvidia decided to only turn it on for certain games where there’s a benefit, with the first wave including Assassin’s Creed Valhalla, Battlefield V, Borderlands 3, Forza Horizon 4, Gears 5, Metro Exodus, Red Dead Redemption 2, and Watch Dogs: Legion.
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.
