This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of
computers,
computing, and
information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join
the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles
The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Not every 10gbaseCX4 card has a modular PHY as claimed. The Myricom
card (
http://www.miricom.com) does not.
The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.
Cat 5e
Zac67 removed a mention of Cat 5e from the article. Reportedly this option is not mentioned in the IEEE standard. Here's
one ref discussing the possibility. I don't consider this to be a
reliable source. The reliable sources I've found don't mention Cat 5e. I haven't found a reliable source that specifically forbids it. ~
KvnG02:15, 26 December 2013 (UTC)reply
I've considered leaving it in for a while – but after all, stating that 10GbE over Cat5e is possible does require a very reasonable source. Personally, I have little doubt that it would work (for some 10-20m) just like FE over Cat3 works surprisingly well for short runs. Googling a bit turns out quite a few forums that quote WP on this, which is exactly why we require
WP:RS.
Zac67 (
talk)
12:19, 26 December 2013 (UTC)reply
I agree it should stay out for now. Better to have no information about 10 Gbit on Cat 5e than to have potentially wrong information. ~
KvnG16:47, 26 December 2013 (UTC)reply
Here's an interesting read relevant to the issue at hand: Hodge, Kenneth G. (2013).
"An investigation into 10GBASE-T"(PDF). Retrieved 2013-12-26. (My google query was "10GBase-T Cat5 OR Cat5e filetype:pdf") Slide 11: 45m over Cat5 UTP, but only 2 connectors (so no patching with connectors). Because that's given for a specific NIC, and for another NIC Cat5 is not mentioned, it looks to be taken from the NIC's
specs. Slide 12 has formulas that list Cat5 as having 6-15m less reach than Cat6. Slide 47 shows that Cat 5 passes the same set of tests as Cat6 and Cat6A UTP. The 10GBase-T specification does expressly not allow Cat5, as per slide 12
here, however that does not prevent it from working, like the FE/Cat3 example.
Conquerist (
talk)
21:15, 26 December 2013 (UTC)reply
And here's one more vendor
quote: "The Arista 7100T switches support 10GBASE-T over Category 6a cabling up to 100m, but also support Category 5e* and Category 6 cabling with distances up to 55m." Asterisk: "Performance of 10GBASE-T over Cat-5e cabling is not specified in the standard and thus cannot be guaranteed. Field testing is recommended before deployment to establish the feasibility of using existing Cat-5e cabling."
Conquerist (
talk)
04:00, 3 January 2014 (UTC)reply
The Arista one looks like a decent enough source to me – this could very well be included. The Bicsi paper is a very interesting read but somewhat unspecific.
Zac67 (
talk)
14:18, 4 January 2014 (UTC)reply
I have removed the mentioning of active optical cables (AOCs) in the 10GSFP+Cu part of the
Copper section, as they are optical cables. Also, the citation source was dead, and, to my knowledge, there is no such standardization of the cables. However, I do not doubt the cables do exist. That's why I also have left them in the respective table, adding some clarification about their medium (fiber). Still, it would be wishful to add sources, particularly for the range given in the table (100m).
The DAC part of the article better aligns with
Twinaxial cabling#Networking (Direct-Attach Copper) now. — Preceding
unsigned comment added by
Jooonsen (
talk •
contribs)
21:57, 12 February 2024 (UTC)reply