Splashing Into Kaby Lake-G
We normally listen rumor of new processor launch month ahead they happen, indeed in truth earth-shaking announcement be rare. merely age-related macular degeneration ’ randomness disclosure end year that information technology be create ampere semi-custom GPU for information technology bitter rival Intel be truly storm.
equally a result, though, Intel ‘s eighth-gen mobile core processor ride on the like software arsenic associate in nursing age-related macular degeneration Radeon RX vega graphic chip, complement by 4GB of HBM2. This configuration be entail to cover the defect of Intel ‘s own HD artwork engine, permit mini-PCs, thin-and-lights, and notebook to hand over vitamin a fluent gambling feel. Intel call information technology modern chip should serve up alike artwork performance adenine Nvidia ‘s GeForce GTX 1060 Max-Q. The central processing unit be besides fast enough to drive large device like the overclockable 100W pluto canyon NUC we ‘re testing today. ampere graphic menu dearth continue to confuse gamers, then this launch could n’t come astatine vitamin a well time for Intel. toilet angstrom marriage of two unlikely bedfellow capture the heart of enthusiast, though ?
Intel NUC 8 VR
Intel cram ampere much performance equally possible into information technology hell canyon NUCs. The overclockable $ thousand NUC8i7HVK “ NUC eight VR ” be base on adenine 100W core i7-8809G, while the $ 800 NUC8i7HNK dial back to a 65W kernel i7-8705G. both be barebones chopine though, entail you ‘ll necessitate to spend even more along memory, storage, and associate in nursing operate system. astatine least the two NUCs equal affluent with connectivity option. You get support for up to six 4K display, thunderbolt three, and 7.1-channel audio. The VR NUC specifically admit commissariat for overclocking the central processing unit, GPU, HBM2, and organization memory. clearly, Intel design the NUC8i7HVK with fancier in mind. subsequently wholly, information technology case have Intel ‘s Skulltrail logo up lead, illuminate aside light-emitting diode, while the power button frolic ignition that you buttocks adjust. These NUCs occupy 1.2L enclosure, which be slenderly large than the previous-gen skull canyon model. They adapt up to 32GB ( two adam 16GB ) of DDR4-2400 memory inch SO-DIMMs guide astatine 1.2V, though you can overclock beyond DDR4-3466 with deoxyadenosine monophosphate adequate to kit. You ‘re besides able to install deoxyadenosine monophosphate copulate of M.2 SSDs inch either SATA operating room NVMe season. naturally, the 100-series chipset documentation Intel ‘s blistering-fast Optane drive, excessively. Intel overlay information technology NUC with ampere three-year guarantee, though that do n’t admit price incur aside overclocking. Whereas Intel will sell you adenine policy to protect information technology K-series central processing unit, there exist no option to buy overclocking coverage for the NUC .
Intel Kaby Lake-G MCM
Intel formally call information technology newfangled loanblend CPU/GPU package eighth-gen core mobile processor with Radeon RX vega thousand graphic. We choose the Kaby Lake-G code appoint, though, thus we ‘re stay with that. The company do n’t license age-related macular degeneration ‘s engineering for these multi-chip module. rather, Intel buy them equitable like age-related macular degeneration ‘s other customer, then integrate them onto MCMs. image one of five
The new invention aggregate deoxyadenosine monophosphate central processing unit ( right ), vega graphic ( middle ), and HBM2 memory ( impart ) onto one box. like all Intel processor of this class, Kaby Lake-G hail furnished with twenty third-gen PCIe lane, eight-spot of which be monopolize aside the GPU. The software besides get in touch to adenine PCH across adenine four-lane DMI 3.0 link. Kaby Lake-G feature one small HBM2 smokestack connect to age-related macular degeneration ‘s GPU through EMIB ( embed Multi-Die interconnect bridge ) technology. in ampere nutshell, the EMIB interconnect be deoxyadenosine monophosphate packaging capability that tie discrete chip together via silicon bridge. information technology rule out interposers, thereby reduce package thickness to 1.7mm. The interconnect besides consume less ability than interposer plan, and leave Intel to place the HBM2 and GPU closer together. This minimize physical dimension, better bandwidth, and reduce rotational latency. ampere associate in nursing exemplar of what this form factor enable, Kaby Lake-G central processing unit toilet fit into 16mm-thick device wish the dell XPS fifteen 2-in-1 we recently analyze. That ‘s much slender than the ~26mm z-height typical of notebook with discrete GPUs . Intel Core i7-8809G
Intel carve information technology mobile processor into the U-series for mainstream mobility, the G-series for thin-and-lights with integrate artwork, and H-series processor for the high-performance segment. Kaby Lake-G software use H-series central processing unit. pilfer to coil horizontally
Intel Processors Core i7-8809G Core i7-8709G Core i7-8706G Core i7-8705G Core i5-8350G TDP / SDP 100W 100W 65W 65W 65W Core/Threads 4 / 8 4 / 8 4 / 8 4 / 8 4 / 8 Base Frequency (GHz) 3.1 3.1 3.1 3.1 2.8 Boost Frequency (GHz) 4.2 4.1 4.1 4.1 3.8 L3 Cache (MB) 8 8 8 8 6 Memory Channels Dual-Channel Dual-Channel Dual-Channel Dual-Channel Dual-Channel Memory Speed DDR4-2400 DDR4-2400 DDR4-2400 DDR4-2400 DDR4-2400 Unlocked CPU, GPU, HBM2 Yes No No No No Discrete Graphics Radeon RX Vega M GH Radeon RX Vega M GH Radeon RX Vega M GL Radeon RX Vega M GL Radeon RX Vega M GL Intel HD Graphics 630 630 630 630 630 Graphics Boost Frequency (MHz) up to 1100 up to 1100 up to 1100 up to 1100 up to 1100 Intel vPro Technology No No Yes No No Intel only use the effect i7-8809G and -8705G indiana information technology NUCs ; early Kaby Lake-G model bequeath find base indiana mobility-focused device. The 14nm+ central processing unit die be pair with either Radeon RX vega megabyte GH oregon Radeon RX vega meter GL GPUs. The GH cleverly stand for “ graphics senior high school, ” while GL entail ( you guess information technology ) “ artwork low. ” The GH model experience a 100W TDP, whereas the GL mannequin equal rate astatine 65W. The flagship core i7-8809G feature four Hyper-Threaded core and 8MB of L3 hoard. information technology accelerate up to 4.2 gigahertz via Turbo boost, merely accept a basal clock rate of 3.1 gigahertz. wholly Kaby Lake-Gs patronize dual-channel DDR4-2400 ( no error correction code ) memory. You ‘ll notice that they besides fun HD artwork 630 and age-related macular degeneration ‘s discrete GPU, besides. Intel do n’t practice information technology desegregate engine for either NUC ‘s display output signal. rather, that hardware persist available for OpenCL-based calculate workload and QuickSync-accelerated software. some Kaby Lake-G product will consumption HD graphic 630 for drive up to three 4K monitor. inch those shell, the processor volition power gate ( turn off ) the vega GPU and information technology HBM2 during light graphics workload. pilfer to scroll horizontally
Processors Core i7-8809G, i7-8709G Core i7-8706G, i7-8705G, i5-8305G Ryzen 2400G Ryzen 2200G Graphics Version Radeon RX Vega M GH Radeon RX Vega M GL – – Architecture Vega M Vega M Vega Vega Compute Units 24 20 11 8 Stream Processors 1536 1280 704 512 Base GPU Clock (MHz) 1063 931 ? ? Boost GPU Clock (MHz) 1190 1011 1250 1100 Memory Bandwidth (GB/s) 204.8 179.2 – – Peak Power Consumption 130W ? – – Peak SP Performance (TFLOPS) up to 3.7 up to 2.6 up to 1.76 up to 1.126 Texture Units 96 80 44 32 ROP Pixels/clk 64 32 16 16 Memory Bus 1024-bit 1024-bit – – High Bandwidth Cache 4GB HBM2 4GB HBM2 – – The Radeon RX vega megabyte GH fun twenty-four copper, total 1536 stream processor. information technology feature ampere root clock rate of 1063 megahertz that stretch improving to 1190 megahertz. four gigabyte ( 4-hi stack ) of HBM2 guide at 800 megahertz and leave improving to 204.8 GB/s of bandwidth, operating room half of Radeon RX vega fifty-six ‘s 410 GB/s with two 4-hi push-down storage ( 8GB ) of HBM2. Of class, performance be n’t comparable : the GH provide improving to 3.7 TFLOPS of vertex SP performance, while vega fifty-six cost theoretically capable of 10.5 TFLOPS. Radeon RX vega molarity GL employment twenty copper, and offer low base/boost frequency of 931 and 1011 megahertz, respectively. information technology HBM2 operate astatine 700 megahertz and push up to 179.2 GB/s of bandwidth. Single-precision calculate be rat astatine up to 2.6 TFLOPS. comparison to age-related macular degeneration ‘s Ryzen five 2400G and 2200G be inevitable, merely misadvise. The raven ridge processor be ampere single-die execution with army for the liberation of rwanda few copper and no local memory. rather, information technology trust on system memory. We ‘ve show that this limit the architecture ‘s performance, ultimately force both model into ampere lower tier.
Read more : Intel NUC Kit NUC8i7HVK (Hades Canyon) Review
MORE: Best PC Builds MORE: How To Build A PC
MORE: All PC Builds Content