The Intel E810 is a SFP28 25/100GbE adapter from Intel, on Not sourced this session - Intel's ark.intel.com product pages did not render for our sourcing pass (see the sources block below). Check the exact part number's own product page before buying if slot width matters to your build., using the ice driver that ships in the mainline Linux kernel. It is not marked end-of-life by its maker. 5 listings in stock, from $177.99.
Prices updated · live figures update every 4-5 hours
Both kinds of listing are in stock for this card right now. The cheapest Supermicro card is $496.99; the cheapest third-party board asserting the same silicon is $177.99, a 64% differencechipset attribution is the seller's. Which is right for you is a judgement about how much the maker's own board is worth to you, not a question we can answer from a listing.
| Card | Ports | Interface | Condition | Price | |
|---|---|---|---|---|---|
| Intel E810 H!Fiberseller claims this chipset | 1 | SFP28 | New | $177.99 | Buy |
| Intel E810 10Gtekseller claims this chipset | 1 | SFP28 | New | $194.99 | Buy |
| Intel E810 ipolexseller claims this chipset | 2 | QSFP28 | New | $405.99 | Buy |
| Intel E810 10Gtekseller claims this chipset | 2 | QSFP28 | New | $479.99 | Buy |
| Intel E810 Supermicro | 2 | SFP28 | New | $496.99 | Buy |
Passive direct-attach copper, by length. A DAC is a fixed-length cable with the transceivers moulded on - you buy the whole run at once rather than a cable plus two modules.
Every "compatible with" phrase below is the seller's own wording, recorded as a claim and badged. We have not tested any module against any switch.
| Module | Type | Brand | Seller compatibility claims | Price | |
|---|---|---|---|---|---|
| SFP28 25GBASE-SR module | 25GBASE-SR | 10Gtek | none stated | $19.99 | Buy |
| SFP28 25GBASE-SR module | 25GBASE-SR | 10Gtek | none stated | $19.99 | Buy |
| SFP28 25GBASE-SR module | 25GBASE-SR | 10Gtek | none stated | $19.99 | Buy |
| SFP28 AOC | AOC | 10Gtek | none stated | $19.99 | Buy |
| SFP28 25GBASE-LR module | 25GBASE-LR | 10Gtek | none stated | $22.99 | Buy |
| SFP28 AOC | AOC | ipolex | 25GBase Active Optical Cable, 25Gbps SFP28, SFP28-25G-AOC5Mseller claim | $31.99 | Buy |
| SFP28 AOC | AOC | 10Gtek | none stated | $35.99 | Buy |
| SFP28 AOC | AOC | 10Gtek | none stated | $37.99 | Buy |
Intel E810 is the card most third-party boards in the 25/40/100GbE tier are built around, and our own tracked listings show exactly that pattern: four tier-2 boards asserting the controller against one first-party Supermicro card. That is not a knock on the category - it is the same third-party-undercut economics the 10GbE cluster already documents for X520 and X710, one generation up.
We could not source Intel's own PCIe generation or formal end-of-life figures for this card in this sourcing pass - ark.intel.com did not render for us, returning a 404-style redirect on every product page we tried, including the bare ARK homepage. That is a tooling gap on our end this session, not evidence Intel has discontinued the part or hidden the spec; we are stating the gap rather than guessing a PCIe figure from the 700-series family pattern.
What we could confirm: the Linux kernel's ice driver, a separate driver module from the i40e driver that covers Intel's older 700-series (X710, XXV710). That distinction matters in practice - troubleshooting advice written for i40e-era cards does not transfer to E810, and a kernel old enough to lack ice support will not recognise this card even if it recognises an X710.
Our tracked listings split 25G SFP28 and 100G QSFP28 SKUs, the same shape as ConnectX-5. This page's cable tables default to the 25G SFP28 side; a 100G E810 build needs the cabling guide's QSFP28 section instead.
The Linux kernel's own ice driver documentation is titled for the Intel Ethernet Controller "800 Series" - the family E810 belongs to - though the rendered page does not spell out "E810" verbatim the way the i40e page names XXV710 directly. Stated as that limit rather than stretched into a direct quote naming this exact part. That matters more than any marketing claim, because a card whose driver is in the mainline kernel is a card your Linux-based NAS or hypervisor recognises without you doing anything.
We state in-kernel driver names as documented facts because they are - they come from the kernel's own driver documentation, not from a forum thread reporting that something worked. Where community experience is the only evidence available for something, we say so rather than dressing it up.
This card is not marked end-of-life by its maker, which is the main practical argument for it over the older generation in the same interface family.
Windows support is a separate question from Linux support and varies by card age. Check the maker's current driver page for your specific part number before buying if Windows is the target, because an in-kernel Linux driver tells you nothing about it.
This is the most third-party-heavy card in our tracked catalogue: four of the five listings we hold are boards from 10Gtek, H!Fiber and Ipolex asserting an Intel E810 controller, and only one - a Supermicro AIOM card - is a named-maker board we could verify as first party. Every third-party claim is badged tier 2 in the tables above; we have not opened any of these boards to confirm the chip inside.
The general shape of the crossflash question is the same across every card in this category: an OEM board carries the server maker's firmware, that firmware sometimes behaves differently from the silicon vendor's, and there is a community practice of replacing it. We link the maintained community documentation rather than reproducing the procedure here. A stale copy of a flashing guide is actively dangerous, and the people who maintain the canonical versions do so because the details change.
If you are buying an OEM-branded card specifically because it is cheaper, factor in the possibility that you will need to do this. If that prospect does not appeal, the vendor-branded card at a higher price is buying you the absence of that job.
This is a SFP28 card, so the port is a cage rather than a socket and you choose what goes in it. For two machines near each other, a passive direct-attach copper cable is the cheapest, coolest and simplest answer, and it is what most builds using this card should use.
For longer runs you need optics - a module at each end plus a fibre between them - or an active optical cable, which packages the same thing as a single sealed object. Both cost more than direct-attach copper and both go much further.
The compatibility question belongs here rather than being buried. Third-party module sellers list the equipment they claim compatibility with, and those are claims we record and badge rather than verify. Vendor coding is a documented phenomenon: cards and switches can and do refuse modules that do not identify as approved parts. A direct-attach cable avoids the entire question, which is another reason it is the default recommendation for a two-machine build.
The cabling guide covers every SFP28 DAC, AOC and optic in one place · The 25/40/100GbE hub compares every card in this cluster and the switched-versus-point-to-point arithmetic · 10GbE vs 25GbE is the step-up decision, priced both ways · used servers almost always have a spare PCIe slot for one · NAS devices is usually what sits at the other end · rack disk shelves are the other common thing on the far end of a fast link.
We cannot verify the chipset claim ourselves - we have not opened any of these boards - so every 'Intel E810 controller' claim on a non-Intel-branded listing is recorded as a seller assertion and badged as one. It is the same trap the 10GbE cluster documents for third-party X520/X710 boards: often genuine, sometimes the better buy, but the claim is the seller's, not ours.
ice - a separate driver module from i40e, which covers Intel's older 700-series cards (X710, XXV710). Confirmed from the Linux kernel's own driver documentation, titled for the 800 Series family E810 belongs to.
We do not have a sourced figure for this session - Intel's own ARK product pages did not render for us. Check the exact part number's product page or box label before assuming your slot is wide enough, particularly for the 100G QSFP28 SKU.
Not sourced either way this session - we could not reach Intel's lifecycle/PCN pages. Do not assume either status from this page; check Intel's current documentation directly if that matters to your purchase timeline.
Both exist in our tracked catalogue and both appear in the live table above. The cabling sections on this page default to the 25G SFP28 side; a 100G QSFP28 build needs the cabling guide's QSFP28 section for matching cables.
That is simply what verified in the catalogue at seeding time - four of five listings are 10Gtek/H!Fiber/Ipolex boards asserting the E810 controller, one is a first-party Supermicro card. It may reflect how this generation reaches the secondary market more than anything about the chip itself; check the live table for the current mix.
Specification claims on this page were checked against the pages below on 4 August 2026. Where a widely-repeated figure did not survive that check, we report what the source says now and say so in the text rather than repeating the familiar number.