System specs
13.3" 1920x1080 internal display
Intel I5-8350U 1.7 GHz 15W TDP with UHD 620 iGPU
16GB LPDDR3
Samsung EVO 970 NVME SSD
Windows 10 Pro 1803
eGPU hardware
Razer Core V2
Nvidia GTX 1080
Hardware pictures
(Pictures are coming when Wordpress allows, but its pretty much what you are imagining)
Installation steps
Using current firmware for Oct. 2018, plug in, automated update drivers.
Benchmarks
All benchmarks are on the internal display. I see a performance increase with an external monitor, but it seems more consistent to run them on internal hardware.
AIDA64 GPGPU reports 2258 MB/s memory write
Unigine 1920x1080 High quality: Heaven 3173/126.0 FPS, Superposition 11682/87.38 FPS, Valley 4030/96.3 FPS
(screenshots pending acct approval)
Comments
The Razer enclosure can power both the card (>180W) and the laptop (<65W), and it has a USB hub and wired NIC attached. For benchmarking I left USB and ethernet disconnected. I use a 2m Caldigit TB3 cable to connect the enclosure at my desk. Benchmarks are with Razer's 0.5m cable, but I haven't noticed any performance impact with the longer cable.
Pending: Add my system information and expected eGPU configuration to my signature to give context to my posts
I use a 2m Caldigit TB3 cable to connect the enclosure at my desk. Benchmarks are with Razer's 0.5m cable, but I haven't noticed any performance impact with the longer cable.
As you can see at my analysis https://egpu.io/forums/gpu-monitor-peripherals/tb3-cables-50cm-80cm-both-passive-and-150cm-active-tb2-cables-with-tb2-tb3-adapter-some-benchmarks there shouldn't be any big performance impact. The only notable impact with the active Tb3 cables it the fact, that they support only USB2.0 if used as USB-C cable.