Since HBM2E is no longer used in gaming GPUs due to its price, the highest performing graphic memory is now the PAM4-based GDDR6X, developed by Micron for Nvidia and reaching transfer rates of up to 21.0 GHz. However, it could now be outperformed by the classic GDDR6. Samsung has announced 20GHz and even 24GHz samples of GDDR6 (or GDDR6+?), that could bring about a 50% bandwidth increase when compared to current GDDR6. Read more “Samsung will make GDDR6 50% faster than today, already sampling”
We got a peek at the prices of cheaper B660 motherboards for Intel Alder Lake recently. So far, motherboards for the new desktop platform have generally been expensive. Why is that? According to TechPowerUp’s intel, it’s a combination of many factors and several features of Alder Lake processors. Some are quite surprising, for example the LGA 1700 socket itself is apparently much more expensive than you’d expect it to be. Read more “Causes of high Alder Lake mobo prices? PCB, VRM, even the socket”
Obscure hardware has been thriving lately, with AMD choosing to recycle APUs used in consoles into improvised PC CPUs. Following the Jaguar chip from the Xbox One, this year theAMD 4700S Desktop Kit, a CPU with 8 Zen 2 cores produced from defective PlayStation 5 APUs, has appeared. And AMD now seems to be preparing a similar CPU with a new 4800S designation. A mainboard fitted with such should be suited for gaming. Read more “AMD 4800S: Zen 2 console APU comes to PCs, now with PCIe 4.0”
GeForce RTX 2060 with 12GB of memory is officially here. Earlier leaks had it coming on December 7, which is the date of store availability. In the end Nvidia already “revealed” the card last week. But extremely quietly, the company just published the specs of the “new” on its website. The return of the RTX 2060, or arguably RTX 2060 Super is now a real thing. But unfortunately there are some questions around the price. Read more “GeForce RTX 2060 12GB quietly released. More memory costs more”
It has been over a year since Nvidia has announced a deal to buy ARM – source of the instruction set and CPU architectures that power embedded and mobile devices, phones and increasingly computers, laptops and servers. But now it looks like there is a major chance that this merger will not happen. US authorities have raised the loudest “NO” yet, aiming to stop the acquisition for fear it would stifle innovation and competition. Read more “Nvidia-ARM merger dead? US authorities seek to block the sale”
There was news that (some) boards for Alder Lake using B660 chipset won’t support PCI Express 5.0. But it looks that may not be an actual limitation of that chipset, but rather something that will depend on pricing of each particular model. It turns out that even some Z690 boards will be like this. Gigabyte has a first such board, and it seems that these models might be the key to building an Alder Lake PC relatively cheaply. Read more “Z690 Alder Lake boards can lack PCIe 5.0 too, for cost reasons”
DDR5 memory modules for PCs have arrived on the market only recently so their replacement should be many years away (the span between the introduction of DDR4 and DDR5 has been more than seven years). But Samsung, as one of the main memory modules producers, has now announced more distant plans for the successors of the DDR5 memory (DDR6), but also new graphic memory, to replace GDDR6 and GDDR6X at the Tech Day 2021 conference. Read more “Future of memory tech: Samsung on DDR6, GDDR7 and GDDR6+”
HDD capacities have been rising slowly lately and new larger HDDs often take very long to ship after being announced. But it looks like we are finally getting somewhere: the first 20 TB capacity HDDs (without SMR) are coming, from Western Digital. They’re actually coming to the market with general availability. What’s more, they are interesting because they are the first drives with the recently introduced OptiNAND technology. Read more “First OptiNAND-based WD HDDs go on sale. 20 TB without SMR”
It’s been a little over a year since Nvidia announced the 2.0 generation of their FPS-increasing AI upscaling DLSS, with significantly better quality thanks to temporal reconstruction and stabilization. Now it has been refined into DLSS 2.3, which brings a new AI for better quality, but at the same time Nvidia also introduces a purely spatial upscaler, which is actually a competitor or alternative for the similar AMD FSR technology. Read more “Nvidia launches improved DLSS and FSR competitor Image Scaling”
The new Z690 platform, with a LGA 1700 socket and Alder Lake CPU just launched by Intel is the first one to use DDR5 memory. Optionally the CPUs can still be used with DDR4 memory modules. You might feel like buying a DDR4 mainboard could be a mistake blocking the path to a later CPU upgrade. But we have good news for you – Alder Lake is not the last CPU with DDR4 support, such mainboards will also support upcoming CPUs. Read more “13th Generation Raptor Lake/Core CPUs to maintain DDR4 support”
Last week saw the reviews and launch of Alder Lake CPU and LGA 1700 mainboards. A big new feature of this platform is DDR5 memory allowing for a higher bandwidth, albeit potentially suffering from higher initial prices. Following the premiere of Alder Lake, we took a look at what’s the fastest available memory modules you can get. It looks like memory modules running on 7000MHz for PCs could be here sooner than later. Read more “Top DDR5 speeds? Currently at 6666, with 7000 MHz soon to follow”
Intel has unveiled the Alder Lake CPUs last week, and we covered the SKUs, features and specs here. There’s more further news, Intel has revealed a number of interesting details about the DDR5 memory support. There’s new generation of auto overclocking via XMP profiles coming, XMP 3.0, which has been significantly modified for DDR5 memory. And particularly the Dynamic memory Boost is a breakthrough feature for high-performance RAM. Read more “Alder Lake and DDR5: XMP 3.0, Dynamic Memory Boost and more”
GDDR7 is still far in the future (though Nvidia got ahead with their own Micron-made GDDR6X). But high-performance GPUs could get another new memory technology in the near future: HBM3 chips. Hynix has now announced that it has developed the first generation of this memory, which could lift graphics bandwidth by almost an order of magnitude. A single HBM3 chip (package) has a bandwidth higher than the entire RTX 3080. Read more “Hynix already has HBM3. Clocks up to 6.4GHz with huge bandwidth”
It seems that after the power consumption increase brought by Nvidia’s latest GPU (GeForce RTX 3000) was not the last time we’ll see this happening. We had news of 450W GeForce cards and even 550W future SKUs coming, but the crowning jewel will be the high-performance server GPU codenamed Hopper. Nvidia is supposedly planning a four-digit consumption exceeding 1000 W, requiring a completely new power supply solution. Read more “Extreme TDPs for GPUs? Nvidia Hopper is allegedly above 1000W”
Samsung’s semiconductor production has been a center of negative news lately. Apart form delays, Samsung processes are believed to be inferior to TSMC node-for-node. But despite this Samsung is still the number two in leading edge process nodes out of the Foundry producers. Now the company has announced 2nm node, which should go into production in four years, two to three years after the volume ramp of the 3nm technology. Read more “Samsung will start 2nm chip production 3 years after 3nm node”
Latest comments