4 ms·

PCIe compliant device must support using less lanes than it can Is that true in practice or is it like USB where manufacturers can wildly violate specs and jus

PCIe compliant device must support using less lanes than it can Is that true in practice or is it like USB where manufacturers can wildly violate specs and just expect consumers to deal with it?

f001 5y ago
It is true in practice. The older style of mining rigs for certain types of cryptocurrency rely on a ton of gpus (6-18) connected via risers to every available pci-e slot on a motherboard, most of them being 1x slots. There are special motherboards for this that don’t even come with 16x slots, instead breaking that out into 16 1x slots to allow for more gpus to be connected.
giobox 5y ago
It generally always works in practice. Common real-world example is “External GPUs” connecting via thunderbolt: these are actually just generic PCI express 4x slots on a cable (4x is max throughout thunderbolt 3 achieves, can’t remember what cap is on new thunderbolt 4…). Although graphics cards always generally should go in a 16x slot for best performance, they all work in the external 4x, even the latest most high end power/bandwidth hungry GPU parts. performance may suffer in some workloads though due to reduced bandwidth of the 4x slot though. You can stick any PCI express device in an external GPU enclosure, will pretty much “just work” despite being a 4x slot - these things are really external PCIe enclosures despite the GPU focus.
- Dylan16807 5y ago
  In terms of PCIe 3 lanes, the max on thunderbolt 3 is 2.75x, and the max on thunderbolt 4 is 4x. Out of a roughly-5x total bandwidth. Why was 3 capped? Nobody seems to know.
- novok 5y ago
  AFAIK thunderbolt has an overhead that it adds to plain old PCIe, making it slower in practice. I kind of wish it was just plain PCIe, you could directly connect cheap-ish pcie adapters to it then.
magicalhippo 5y ago
When the PCIe device is initialized, it starts up in x1 mode. The controller then negotiates the number of lanes with the device. So yeah it pretty much has to work in x1.

PCIe compliant device must support using less lanes than it can Is that true in practice or is it like USB where manufacturers can wildly violate specs and just expect consumers to deal with it?

f001 5y ago

It is true in practice. The older style of mining rigs for certain types of cryptocurrency rely on a ton of gpus (6-18) connected via risers to every available pci-e slot on a motherboard, most of them being 1x slots. There are special motherboards for this that don’t even come with 16x slots, instead breaking that out into 16 1x slots to allow for more gpus to be connected.

giobox 5y ago

It generally always works in practice. Common real-world example is “External GPUs” connecting via thunderbolt: these are actually just generic PCI express 4x slots on a cable (4x is max throughout thunderbolt 3 achieves, can’t remember what cap is on new thunderbolt 4…). Although graphics cards always generally should go in a 16x slot for best performance, they all work in the external 4x, even the latest most high end power/bandwidth hungry GPU parts. performance may suffer in some workloads though due to reduced bandwidth of the 4x slot though. You can stick any PCI express device in an external GPU enclosure, will pretty much “just work” despite being a 4x slot - these things are really external PCIe enclosures despite the GPU focus.

Dylan16807 5y ago
In terms of PCIe 3 lanes, the max on thunderbolt 3 is 2.75x, and the max on thunderbolt 4 is 4x. Out of a roughly-5x total bandwidth. Why was 3 capped? Nobody seems to know.
novok 5y ago
AFAIK thunderbolt has an overhead that it adds to plain old PCIe, making it slower in practice. I kind of wish it was just plain PCIe, you could directly connect cheap-ish pcie adapters to it then.

magicalhippo 5y ago

When the PCIe device is initialized, it starts up in x1 mode. The controller then negotiates the number of lanes with the device. So yeah it pretty much has to work in x1.