Pcie x16 graphics card in x8 slot

*

*

This is a test that"s been put with the paces for simply around every generation of PCI Express, and it"s worth refreshing currently that the newest line of high-end GPUs has hit the market. The curiosity is this: Will a GPU be bottlenecked by PCI-e 3.0 x8, and also exactly how a lot affect does PCI-e 3.0 x16 have on performance?

We made a decision to test that question for internal research, yet finished up placing together a small report for publication.

You watching: Pcie x16 graphics card in x8 slot


PCI Express Theoretical Max Bandwidth

The theoretical maximum bandwidth of PCI-e 3.0 is 8GT/s, or practically 1GB/s per lane:

PCI-e 1.0PCI-e 2.xPCI-e 3.0PCI-e 4.x
x1250MB/s500MB/s985MB/s1969MB/s
x41000MB/s2000MB/s3940MB/s7876MB/s
x82000MB/s4000MB/s7880MB/s15752MB/s
x164000MB/s8000MB/s15760MB/s31504MB/s

For our test, we"re looking at PCI-e Gen3 x8 vs. PCI-e Gen3 x16 performance. That implies there"s a 66.7% difference in bandwidth obtainable in between the two, or a 100% rise from x8 to x16. But there"s a lot more to it than interface bandwidth: The tool itself need to exceed the saturation point of x8 (7880MB/s, before overhead is removed) in order to present any kind of meaningful benefit in x16 (15760MB/s, prior to overhead is removed).

Use Cases, Future Tests, & Test Setup

The use instances here are not that huge. Maybe you"ve acquired a thermal issue or a card that butts-up versus the CPU cooler, or some sort of liquid routing challenge. HSIO lanes are assigned to ancillary devices – like PCIe SSDs – and won"t eat into the CPU lanes accessible to the GPU. We"re additionally not testing multiple GPUs, which is where we"d favor to go next as soon as we"ve acquired two of the exact same GTX 1080 in the lab. Ideally, we test in x16/x16, x16/x8, and x8/x8 – yet that"s not possible right now. We"re additionally hoping to test dual-GPU, single-card configurations in between an x8 and also an x16 slot, as those may put more fill on the interconfront.

For the moment being, this test strictly looks at a single-GPU, single-card GTX 1080 Gaming X as it passes in between x8 and also x16 slots. If, for whatever reason, you"re debating the performance reduction from relocating to an x8 PCI-e slot through a single card, that"s what this test looks into.

We used our normal test bench (comprehensive below) for this research study. The EVGA X99 Classified motherboard is picky with its PCI-e slot utilization, and offers UEFI to clearly inform whether the linked gadget is receiving 1, 4, 8, or 16 lanes. We switched in between the initially x16 slot and also the first x8 slot for these numbers, then validated in BIOS and also software application.

PCI-e generations have the right to additionally be required in the EVGA UEFI, yet we did not explore the affect of PCI-e 2.x on the GTX 1080 at this time as it seemed even much less most likely of a use situation.

Game Test Methodology

We tested using our GPU test bench, in-depth in the table listed below. Our many thanks to sustaining hardware merchants for giving some of the test components.

NVidia"s 368.39 drivers were provided for game (FPS) testing. Video Game settings were manually regulated for the DUT. All games were run at presets characterized in their particular charts. We disable brand-sustained technologies in games, choose The Witcher 3"s HairWorks and HBAO. All other game settings are identified in corresponding game benchmarks, which we publish individually from GPU reviews. Our test courses, in the event hand-operated testing is executed, are likewise uploaded within that content. This allows others to replicate our outcomes by studying our bench courses.

Windows 10-64 develop 10586 was provided for trial and error.

Each game was tested for 30 seconds in an similar scenario, then repetitive multiple times for parity.

Mean FPS, 1% low, and also 0.1% low times are measured. We perform not measure maximum or minimum FPS outcomes as we think about these numbers to be pure outliers. Instead, we take an average of the lowest 1% of results (1% low) to show real-people, noticeable dips; we then take an average of the lowest 0.1% of results for significant spikes.

See more: How To Enable Spell Check On Discord, : Discordapp

GN Test Bench 2015NameCourtesy OfCost
Video CardThis is what we"re testing!--
CPUIntel i7-5930K CPUiBUYPOWER $580
MemoryCorsair Dominator 32GB 3200MHzCorsair$210
MotherboardEVGA X99 Classifiedvr-tab-quebec.com$365
Power SupplyNZXT 1200W HALE90 V2NZXT$300
SSDHyperX Savage SSDKingston Tech.$130
CaseTop Deck Tech Stationvr-tab-quebec.com$250
CPU CoolerNZXT Kraken X41 CLCNZXT$110

For Dx12 and Vulkan API trial and error, we use integrated benchnote devices and also count upon log generation for our metrics. That information is reported at the engine level.

Video Cards Tested

PCI-e 3.0 x8 vs. x16 FPS Performance

Let"s simply article all the charts initially, then talk numbers – they"re similar enough that this is the easiest method to read the information.

Metro: Last Light

*

*

Shadow of Mordor

*

*

Cevery one of Duty: Black Ops 3

*

*

GTA V

*

Ashes of Selfhood (Dx12)

*

Here"s what we"ve acquired for performance:

Between AVG FPS metrics in Metro: Last Light, we"re seeing a 1.05% gap (1440p) and also 0% gap (4K). Between 1% low metrics, that distinction is 0.95% (1440p) and also 0%.

For Shadow of Mordor, the numbers are similar – we"re seeing a 0.93% performance difference between AVG FPS metrics (or ~1% for 4K).

Babsence Ops 3, once tright here is a distinction, mirrors one likewise simply listed below 1%.

GTA V mirrors a difference of 0.52%. Ashes is similarly small.

Inconsequential Differences & Margins for Error

These numbers are close enough in some instances – prefer the GTA V 58.3 vs. 58 FPS output – that they"re effectively within margin of test error and also perform not definitively display a performance gap. When a reasonable performance gap is displayed – choose the ~1% difference in Metro: Last Light numbers – it is imperceptible to the user however measurable via our tools. And we carry out intend imperceptible – we"re talking 96FPS vs. 95FPS, for Metro.

Metro, by the method, is the a lot of reliable FPS benchmarking tool we have actually ever used. The game produces almost exactly the very same AVG, 1% low, and 0.1% lows through every single test pass, and so we trust these metrics as being external of test variance.

From a quick look, tright here is a little below a 1% performance distinction in PCI-e 3.0 x16 and also PCI-e 3.0 x8 slots. The difference is not also cshed to perceptible and must be ignored as inconsequential to customers fretting over potential slot or lane limitations. We are not certain just how this scales via SLI (specifically MDA "mode") or dual-GPU cards, however hope to research when we"ve obtained even more hardware in the lab.

See more: 4 4Gb Ram Or 2 8Gb - Two 8Gb Sticks Or Four 4Gb Sticks

We are additionally currently investigating the influence of PCI-e lanes on lower capacity VRAM cards, choose 4GB. Hits to system resources may anxiety the interconfront more.