After early performance numbers were spotted on Intel’s upcoming 6-Core Coffee Lake CPUs on the SiSoft Sandra benchmark database, new Coffee Lake benchmarks are now appearing in the GeekBench database.
While GeekBench results page shows the CPU as Kaby Lake, the specs show a 3.19GHz, 6-core/12-thread processor with 1.5MB of L2 and 12MB of L3. This is consistent with the specs from the previously leaked SiSoft Sandra benchmarks.
According to the benchmark page, the CPU scored 4619 single core score and 20828 multi-core score.
As expected, this puts single core performance higher than AMD Ryzen and multi-core score somewhere in line [...]
Own an Skylake or Kaby Lake processor with Hyper-Threading? It may be a good time to update your system UEFI/BIOS and CPU microcode. A critical flaw in Intel’s implementation of Hyper-Threading was recently revealed by the Debian Linux mailing list.
The critical flaw, which exists on all Intel 6th Generation “Skylake” and 7th Generation “Kaby Lake” processors with Hyper-Threading may cause “unpredictable system behavior” under a specific set of circumstances.
According to the Intel’s official documentation:
Errata: SKZ7/SKW144/SKL150/SKX150/SKZ7/KBL095/KBW095 Short Loops Which Use AH/BH/CH/DH Registers May [...]
A mysterious “Genuine Intel CPU 0000” has appeared in SiSoft Sandra benchmark database, said to be likely an engineering sample of Intel’s upcoming Coffee Lake CPU.
The chip in question has six cores featuring a base clock of 3.1GHz and a turbo clock of 4.2GHz. It carries 256KB of L2 cache per core and 9MB of shared L3 cache. The chip reportedly has a die size of 149mm².
What tier this chip sits at is unknown, but considering the competitive pressure from AMD, it could very well slot into the consumer Intel Core i7-8700K. A quad-core variant will also be available but will be manufactured under a smaller, 126mm² die.
Coffee Lake will be [...]
AMD’s Zen based server platform, EPYC (codename Naples), has been in the works for quite some time and it finally appears ready for prime time. At AMD’s recent EPYC launch event in Austin, Texas, AMD fully detailed their new server platform designed to challenge Intel’s dominance in the server market.
AMD EPYC 7000 Series Processor
Launching today is no less than twelve CPU SKUs in the AMD EPYC 7000-Series processor lineup. The new CPUs will feature up to 32-cores/64-threads, 8-channel DDR4 memory support (up to 2TB of memory per CPU), 128 PCIe lanes, a dedicated security subsystem, and an integrated chipset.
Canard PC revealed in a tweet that AMD is preparing B2 stepping of its Summit Ridge CPUs.
The revision is expected to improve uncore features such as PCIe and memory controllers. Canard PC hasn’t revealed exactly which areas of the design will be affected, but it did mention that it hasn’t seen any improvement proposals made towards the CPU core design themselves.
A stepping upgrade is significant as stepping refers to when the design of the chip is revised in hardware which extends beyond what microcode adjustments could fix.
Ah tiens : stepping B2 en approche pour Zen. Il fixe de nombreux bugs hardware.
As AMD had already made all their big announcements at Computex, their presence at E3 was fairly limited to supporting their OEM partner, Alienware, which had recently announced their Area 51 Gaming PC supporting AMD’s new HEDT Ryzen Threadripper platform.
At the show, AMD brought along some of their new Threadripper CPUs to show off, giving us an opportunity to take a closer look at an early engineering sample of the massive Threadripper chip in all its glory.
Aside from the sheer size, which is due to Threadripper being essentially two Ryzen dies on a single chip, the most interesting aspect of the Threadripper chip is that AMD [...]
IBM Research Alliance recently announced that they’ve developed a process to produce the world’s first 5nm transistors. The new transistors will allow IBM to squeeze as much as 30 billion transistors into a chip just 50mm² resulting in as much as 40% improvement in performance at the same power or a 75% reduction in power at the same level of performance when compared to current generation cutting edge 10nm technology.
In order to shrink the transistors down to the 5nm feature size, IBM researchers spent 10 years developing a new technology which they call Stacked Nanosheet Gate-All-Around Transistor (GAAFET) technology.This simpler [...]
Intel recently released their Skylake-X and Kaby Lake-X processors to great fanfare, with the new flagship Skylake-X based Intel Core i9-7980XE featuring a massive 18-cores and 32-threads. However, overclocking enthusiasts are already up in arms as the new processors which Intel claims are designed for “overclocking” may not actually be as overclocking friendly as Intel claims.
In a recent de-lidding video released by world renowned overclocker der8auer, it was revealed that Intel is now skimping on the thermal interface material even on their high end X-Series processors. In the video, we can see that Intel is using their cheap TIM (Thermal Interface [...]
With Intel cancelling their IDF (Intel Developer Forum) event this year, Intel is making their big reveal at Computex, introducing their highly anticipated, high end Intel Core i7 and Core i9 X Series CPUs. The new Intel Core i7 and Core i9 X-Series CPUs are expected to be Intel’s answer to AMD’s upcoming Threadripper CPUs which feature up to 16-cores and 36-threads.
According to Intel, the new Intel X-Series CPUs will be designed for content creation, 4K gaming, streaming, and overclocking. The new processors will feature up to a 15% performance increase in single threaded workloads and a 10% performance increase in multi-threaded workloads [...]
As with any new hardware, AMD’s Ryzen platform launched earlier this year with plenty of room for improvement.
The latest update AMD has released is an AGESA (AMD Generic Encapsulated System) update which updates the code to v188.8.131.52. For those unfamiliar, AGESA is the base code from AMD which is what motherboard BIOS makers build their BIOSes off of. The new update largely improves memory support and overclocking, but also adds some new virtualization features as well.
The biggest update with AGESA v184.108.40.206 is improved memory support and overclocking. AMD has added 26 new parameters that can be adjusted in the BIOS for memory that doesn’t adhere [...]