Intel has only just launched its new Kaby Lake CPUs led by the Core i7-7700K, and yet we're hearing rumors already on the company's new HEDT, or high-end desktop processor.
The new Skylake-X will be the successor to the Core i7-6950X, while the lower-end 4-core/8-thread model arrives as the Kaby Lake-X family. The higher-end 10, 8, and 6-core models are based on the older Skylake architecutr, with a higher 140W TDP. The 4-core Kaby Lake-X processor feature a 112W TDP, while all of the new HEDT processors will be pushed as the Core i7-7000 series.
This means we won't see a Core i7-7950X, but rather a K series processor as the report mentions Intel will release 4 new K series products, and no X series product. So we should expect the the possibilities of Intel calling their new HEDT processor the Core i7-7950X.
Intel will also launch its new enthusiast-tuned X299 chipset that will house a LGA 2066 socket that will work with two generations of processors. Intel will reportedly use run with a native quad-channel DDR4-2667 kit of RAM. The new X299 chipset will also offer up 24 PCIe 3.0 lanes, with the new 'Basin Falls' PCB boasting 10 x USB 3.0, 8 x USB 2.0 ports, SATA 3.0, and Intel LAN. Nothing really revolutionary, at all.
We should expect Intel to launch its seemingly watered down competitor to AMD's revolutionary Ryzen processor - even compared to Intel's latest and greatest which is nearly a year away, with no mention of anything AI-related or next-generation over the current Z270 chipset, which is a much cheaper mainstream part. No games will use the will power of a 10-core processor, and there's only 24 PCIe 3.0 lanes, marking a shift away from multi-GPU setups (at least for now).
Intel looks to have an impressive value for money processor based on its new Kaby Lake architecture available, with the new Pentium G4560 costing just $64 - and it competes against processors worth 3x as much.
ComputerBase has released their full performance review on the Pentium G4560, seeing performance from the $64 processor that outperforms the 6-core AMD FX-6300, keeps up with Core i3-6100 processor, and even the $230 former gaming champion, the Core i5-2500K.
The site has a 'total rating' of application and games at 1080p, where you can see that the Pentium G4560 and Core i5-2500K are neck and neck.
When gaming at both 720p and 1080p, the new Kaby Lake-based Pentium G4560 beats the Pentium G4400, FX-6300 and A10-7890K - while just 5% slower than the more expensive Core i3-6100, and Core i5-2500K processors.
AMD is poised to unleash its new Ryzen CPUs sometime in Q1 2017, and from what we've heard it's going to be at the Game Developers Conference - which kicks off on February 27, and ends on March 3.
During GDC 2016, AMD held its huge Capsaicin event, where it better detailed Polaris at the time. Now we're seeing AMD teasing an 'Optimizing for AMD Ryzen CPU (Presented by AMD)' session at GDC 2017, which teases people to "join AMD Game Engineering team members for an introduction to the recently-launched AMD Ryzen CPU".
I've heard from my industry sources that AMD will be launching Ryzen during GDC 2017, and that the 8-core/16-thread Ryzen CPU will be the star of the show. I'm expecting a better, more detailed look at their upcoming Vega graphics card, as well.
AMD has reportedly begun sampling its new 4-core Ryzen CPUs, the SKU that has multi-threading disabled - these new Ryzen processors will take on Intel in the Core i3/i5 segment.
The new 4-core Ryzen CPUs will have unlocked multipliers, so you'll be able to overclock your heart away - and considering that Intel locks overclocking away on most of its lower/mid-range CPUs, AMD has a good chunk of the market that will be happy with an unlocked quad-core Ryzen processor.
We should expect the quad-core Ryzen CPU to have a base clock of 3.4GHz, with AMD's exciting new SenseMI technology to handle the overclocking on its own - depending on the cooling used. Stock cooler? It won't boost much. AIO cooler? Now we're talking. Custom water cooling? Yeah, baby. LN2? The sky is the limit. Obviously there's a 45W TDP in your way, but I'm sure we should expect a mainstream price on the 4-core Ryzen CPU when it launches in March.
AMD already has me excited over their Ryzen CPUs for consumers and gamers, but the Zen architecture is also going to do big things in the high-performance server market - starting with Naples.
Naples is AMD's high-end server part based on the Zen architecture that AMD is touting as an 'optimized GPU server platform' thanks to its 128 PCIe 3.0 lanes, allowing for 4-6 'direct attached GPUs'. This means it can take 32 x NVMe devices, and 4 x graphics cards - all on a 1U rack. This same 1U rack will sport 2 x InfiniBand EDR interconnects for super-fast data communication between the storage, and server systems.
As for specs, we're looking at up to 32 Zen CPU cores with 64 threads of performance, 64MB of L3 cache, a base clock of 1.4GHz and a turbo clock of 2.8GHz. A dual-CPU version of Naples would rock a massive 64/128 CPU cores/threads with some ultimate performance, especially when you throw Vega-based graphics cards and NVMe-based storage into the mix.
AMD launched an offensive late last year with the massive reveal of Ryzen, and had a more detailed tease at CES 2017 last week - but now, Intel has seemingly reacted by adding Hyper-Threading technology to some its new Kaby Lake-based range of Pentium processors.
Not all of the new Pentium CPUs will receive HT, but TechPowerUp reports that the list includes the "Pentium G4620 (3.7GHz base frequency and integrated graphics HD 630, $93) and G4600 (3.6GHz base frequency, $82), both at 51 W TDP; G4560 (54W TDP at 3.5GHz and integrated HD 610 graphics, $64); and the Pentium G4600T (3GHz, HD 630, $75) and G4560T (2.9GHz, HD 610, $64) slot in as the low power Pentium offerings with a 35W TDP".
But now that some of the new Pentium processors feature Hyper-Threading support, how will Intel make consumers see a difference between the Core i3 models? The company Advanced Vector Instructions 2 (AVX2) extension set, which is mainly used for database processing and video editing - so Intel is hoping there is still a market for a slice of consumers who will buy a HT-enabled Pentium processor over the slightly cheaper Core i3.
The pricing is also a little cheaper, with the new Kaby Lake-based Core i3 processors having a $50+ premium over the new Pentium chips, but will the performance be much better? Well, the new Core i3-7100 and Pentium G4620 are only 200MHz apart and with AVX2 being the only difference, why would you pay more?
It looks like Intel could be in for a world of hurt, and a massive explanation for the reason behind some of its new processors being stupidly easy to hack into a PC - through a simple USB port.
Positive Technologies, a security vendor, has discovered that some of Intel's new CPUs have a debugging interface that can be accessed through USB 3.0 ports. This provides the hacker - and at the point of using as USB flash drive, any person in the world that can hold a USB stick in their hands - full control over the PC. Worse yet, it is completely undetectable by current security tools.
The possibilities of this are virtually unlimited, as someone could plug a USB 3.0 stick in and upload malicious code, keyloggers, and virtually anything else they want. The same flaw allows someone to make the PC completely inoperable, where it is capable of rewriting the entire BIOS. Yeah, that's a pretty big deal.
CES 2017 - Although Qualcomm has just created the world's first 10nm chip with its Snapdragon 835 SoC, and TSMC is working on 7nm chips, Intel affirms that its own 10nm processor will still release in 2017 despite original delays.
Intel's eight-generation 10nm Cannon Lake CPUs were originally supposed to release in 2016, but the chips were delayed when Intel had some troubles shrinking down the chip's transistors. Now we're right at the beginning of 2017 and Intel has only just launched its seventh-generation Kaby Lake CPUs on the new 200-series Z270 chipset, leading many to believe Cannon Lake would be pushed into 2018--but Intel has set the record straight.
At CES 2017 Intel CEO Brian Krzanich demoed a prototype 2-in-1 device running an eight-gen 10nm Cannon Lake CPU that displayed a promo video. "For those who are wondering if Moore's Law is alive, if 10 nanometers going to be here, the answer is absolutely yes," Krzanich said on stage.
CES 2017 - When we first reported about Intel's new Kaby Lake-based Core i7-7700K hitting 7GHz, it was only with 2 cores enabled - so impressive to see 7GHz, but not so impressive without all 8 cores enabled.
During CES 2017 - overclockers were able to hit 7GHz on the 7700K with all of the CPU cores enabled - using LN2 of course. They hit 7026MHz specifically, with a CPU multiplier of 70, and an FSB of 100.
The new Core i7-7700K has a base clock of 4.2GHz and boost of 4.5GHz, with most people able to hit 5GHz without a problem. The CPU retails for $305 in the US, and obviously changes throughout the world - so for me in Australia, it'll cost $1500 or something ridiculous (jokes, jokes - probably $1200).
CES 2017 - Qualcomm has officially introduced their Snapdragon 835 processor with X16 LTE. Back in November, we reported that the Snapdragon 835 is the first 10nm SoC and we also put an end to the speculations that Qualcomm's next chip will be named Snapdragon 830.
As of today, the Snapdragon 835 is officially out and we, as the phone manufacturers, have high expectations from the Qualcomm's latest processor. The Snapdragon 835 features a new Kryo 820 CPU with four high-performance cores running at 2.45GHz and four efficiency low-power cores running at 1.9GHz. These are backed by dual channel LPDDR4x memory at 1866MHz.
The Snapdragon 835 also has a new Adreno 540 GPU with support for DX12, Vulkan, OpenCL 2.0 and OpenGL ES 3.2. Qualcomm also upgraded the DSP inside of the Snapdragon 835 by including the Hexagon 682, a machine learning-focused upgrade to the Hexagon 680 in the Snapdragon 820.