The Intel X540 is a RJ45 10GbE adapter from Intel, on PCIe 2.1 x8, using the ixgbe driver that ships in the mainline Linux kernel. It is not marked end-of-life by its maker. 4 listings in stock, from $38.39.
Prices updated · live figures update every 4-5 hours
Both kinds of listing are in stock for this card right now. The cheapest Intel card is $90.00; the cheapest third-party board asserting the same silicon is $38.39, a 57% 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.
The X540 is the copper card: 10GBASE-T over ordinary RJ45, no modules, no cages, no compatibility question. If your 10GbE has to travel through walls on cabling that already exists, this is the category of card you need, and the X540 is its cheap end.
The trade for that convenience is heat, and it is the defining characteristic of this card rather than a footnote. 10GBASE-T draws more power per port and runs hotter than SFP+ direct attach - that direction is documented at product level - and the X540 in particular has a reputation for running warm enough that airflow over the heatsink is not optional. In a server chassis with front-to-back airflow this is a non-issue. In a quiet desktop case with a passive card sitting under a graphics card, it is something to plan for.
We are deliberately not quoting a wattage here. Per-port figures for 10GBASE-T are widely repeated, and we have not fetched and logged a source we can point you at, so a number from us would be one you could not check.
The other consideration is cabling. 10GBASE-T over Cat6 has a shorter supported reach than over Cat6a, and if you are running new cable for a 10GbE link there is no reason not to run the better specification. If you are reusing existing Cat5e in the walls, that is where the arithmetic gets uncertain and testing the actual run is the only way to know.
The X540 runs on ixgbe in the mainline Linux kernel, the same driver as the X520 and X550. 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.
Dell and HPE branded X540 cards are widespread and inexpensive, and RJ45 cards do not carry the transceiver-compatibility question at all - there are no modules involved.
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 an RJ45 card, so it takes an ordinary Ethernet patch lead and the entire transceiver compatibility question does not arise. No modules, no vendor coding, no seller compatibility claims to evaluate.
What you get instead is the cabling specification question. 10GBASE-T's supported reach depends on the cable: Cat6a carries the full standard distance, Cat6 carries less, and Cat5e may work over short runs with the only reliable test being the actual run in your actual walls. If you are pulling new cable, pull Cat6a.
And you get the heat. 10GBASE-T draws more power per port and runs warmer than SFP+ direct attach - a documented product-level characteristic of the standard. In a server chassis with directed airflow this never comes up. In a quiet desktop case it is worth planning for. We publish no wattage figure because we have not fetched a source we can cite for one, and a number you cannot check is not worth printing.
The 10GbE hub compares every card we track and covers the SFP+ versus 10GBASE-T decision · the two-machine build prices two of these plus a cable, with no switch · 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.
Because 10GBASE-T does considerably more signal processing than a passive copper SFP+ link, and it draws more power per port as a result - a documented product-level characteristic of the standard rather than a fault in the card. The X540 in particular wants airflow over its heatsink. In a server chassis this is automatic; in a quiet desktop case it is something to arrange deliberately. We do not publish a wattage because we have not sourced one we can cite.
Sometimes, over short runs, and the only way to know is to test the specific run. 10GBASE-T's supported distance depends on the cable specification, with Cat6a supporting the full standard distance and Cat6 supporting less. If you are pulling new cable, pull Cat6a - the cost difference is small next to the labour of doing it twice.
The X550 is the newer generation of the same idea and generally the better buy if the price gap is small, but both run on the same ixgbe driver and both do 10GBASE-T. Check the live prices above - in this category the older card is often not cheaper by enough to matter.
No, and that is the whole point of it. RJ45 cards take an ordinary Ethernet patch lead. The entire transceiver compatibility question - vendor coding, seller compatibility claims, module heat - simply does not arise.
ixgbe, in the mainline Linux kernel, the same as the X520 and X550. Support across TrueNAS, Proxmox, Unraid and standard Linux distributions is broad and long-standing.
Only if you will use both ports. For connecting one machine to one other machine or to one switch, a single port is sufficient and usually cheaper. The T2 is common on the used market because that is what servers were specified with, so sometimes the dual-port card is the cheaper listing anyway.
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.