this post was submitted on 04 Jan 2024
43 points (90.6% liked)

Selfhosted

40347 readers
330 users here now

A place to share alternatives to popular online services that can be self-hosted without giving up privacy or locking you into a service you don't control.

Rules:

  1. Be civil: we're here to support and learn from one another. Insults won't be tolerated. Flame wars are frowned upon.

  2. No spam posting.

  3. Posts have to be centered around self-hosting. There are other communities for discussing hardware or home computing. If it's not obvious why your post topic revolves around selfhosting, please include details to make it clear.

  4. Don't duplicate the full text of your blog or github here. Just post the link for folks to click.

  5. Submission headline should match the article title (don’t cherry-pick information from the title to fit your agenda).

  6. No trolling.

Resources:

Any issues on the community? Report it using the report flag.

Questions? DM the mods!

founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
[–] TCB13@lemmy.world 12 points 10 months ago (1 children)

Everything was fine until...

these are PCIe Gen3 x2 only

Fucks sake. I've seen ARM board with PCI better than that.

[–] aBundleOfFerrets@sh.itjust.works 2 points 10 months ago (1 children)

What arm board :p
Honest question. All the ones I have seen are really awful and I would love to tinker with something that has real pcie (Ampere workstations do not count)

[–] TCB13@lemmy.world 4 points 10 months ago* (last edited 10 months ago) (1 children)

Both the ROCKPro64 and the NanoPi M4 from 2018 has a x4 PCIe 2.1 interface. Same goes for almost all RK3399 boards that care to expose the PCIe interface.

Update: there's also the more recent NanoPC-T6 with the RK3588 that has PCIe 3.0 x4.

This boards seems extremely poorly designed, have a look at the CPU specs: https://www.intel.com/content/www/us/en/products/sku/97926/intel-atom-processor-c3758-16m-cache-up-to-2-20-ghz/specifications.html

They could've exposed more SATA ports and / or PCI lanes and decided not to do it.

And... let's not even talk about the SFF 8087 connector that isn't rated to be used as an external plug, you'll likely ruin it quickly with insertions and/or some light accident.

[–] stalfoss@lemm.ee 1 points 10 months ago (1 children)

PCIe 2 x4 is the same speed as PCIe 3 x2, no?

[–] TCB13@lemmy.world 1 points 10 months ago* (last edited 10 months ago)

Generally, there's a small difference in speeds:

PCIe 2.0 x 4 > 2.000 GB/s PCIe 3.0 x 2 > 1.969 GB/s

But we also have to consider the suggested ARM CPU does PCIe 2.1 and we've to add the this detail:

PCIe 2.1 provides higher performance than the PCIe 2.0 by facilitating a transparent upgrade from a 32-bit data path to a 64-bit data path at 33MHZ and 66MHz.

I shouldn't also have a large impact but maybe we should think about it a bit more.

Anyways I do believe this really depends on your use case, if you plan to bifurcate it or not and what devices you're going to have on the other end. For instance for a NAS I would prefer the PCIe 2.1 x 4 as you could have more SATA controllers with their own lanes instead of sharing lanes in PCIe 3.0 using a MUX.

Conclusion: your mileage may vary depending on use case. But I was expecting to have more PCI lanes exposed be it via more m.2 slots or other solution. I guess that when a CPU comes with everything baked in and the board maker "only has" to run wires around better do it properly and expose everything. Why not all SATAs for instance?