|This is the talk page for discussing improvements to the Serial ATA article.
This is not a forum for general discussion of the article's subject.
|Serial ATA received a peer review by Wikipedia editors, which is now archived. It may contain ideas you can use to improve this article.|
|WikiProject Computing / Hardware||(Rated C-class, High-importance)|
|This talk page is automatically archived by Lowercase sigmabot III. Threads with no replies in 120 days may be automatically moved.|
User:Tom94022 - below are my responses to your three objections to my edit:
- "inappropriate for lede". First, it serves as an example that the Parallel ATA lead mentions Serial ATA. Second, MOS:LEAD says the lead "should identify the topic, establish context, explain why the topic is notable, and summarize the most important points, including any prominent controversies." Mentioning NVMe supplanting Serial ATA establishes context within the realm of drive interfaces.
- "Assertion 'As of 2018, computer manufacturers are increasingly switching to NVMe instead' is not supported by the reference ". On the contrary, it is supported by the quote in the reference "More and more makers of desktop PCs and laptops are designing slots, meant for much smaller SSDs, directly into the motherboards of their PCs." Perhaps I could use the quote field in the reference to make the reference more clear.
- "Assertion 'even for the data center' is not supported by the reference ". In the reference, the statement "The NVMe-based all-flash array market will grow very quickly" establishes that it is talking about enterprise data center applications, not laptops. Then the reference goes on to say "driving the strong growth of this market", which indicates adoption has already begun.
- Responding using the same numeration as above:
- We agree that WP:lede states, "The lead serves as an introduction to the article and a summary of its most important contents." This strongly suggests new material should first be introduced in the body and then only summarized in the lede if it is important. Most of the current lede is also inappropriate and I will probably fix it in the next day or two; it should be in a history section.
- Neither the reference nor the quote supports "switching" but possibly "adding." Most MOBs still have one or more SATA slots. Also many "smaller slots" are M.2 SATA slots, they are already in the article body (15 cites) so this proposed specific quote does not support the assertion at all. This could be added in the body but is not IMO significant for the lede. "Switching" approaches POV since we are a long way from NvME "supplanting" SATA in the way that SATA replaced PATA. Finally note that SAS appears multiple times in the body but not in the lede.
- The reference is about SAS, not SATA and therefore is not particularly relevant to a SATA article, much less in the lede.
- Tom94022 (talk) 08:29, 9 December 2018 (UTC)
- @Michaelmalak You should note that M.2 doesn't necessarily mean NVMe. M.2 supports the SATA and USB protocols as well as PCIe. A major part of M.2 SSDs is still using SATA, with USB used for low-end. --Zac67 (talk) 09:26, 9 December 2018 (UTC)
Which company/companies designed SATA?
Which company/companies designed SATA? We can add this info to the infobox and History section.
- It's a bit unclear which specific companies "designed" SATA. Seagate, APT and Vitesse were in 2000 the first to demonstrate, but the earliest version of the specification (2001) was copyrighted by APT, Dell, IBM, Intel, Maxtor and Seagate. At that time the Serial ATA Working Group had maybe 74 members. Tom94022 (talk) 19:43, 25 December 2018 (UTC)
- SATA 3: "It is backward compatible with SATA 3 Gbit/s."
OK, it is compatible with SATA 2. But, is it backward compatible with SATA 1 of 1,5Gbit/s?
The article doensn't specify.
- I've added 1.5 Gbit/s to that sentence as the now changed source states "backward compatibility with earlier SATA implementations will be maintained". But the fact itself was already mentioned in the article in the section Backward and forward compatibility. Digital Brains (talk) 13:43, 18 January 2019 (UTC)
SATA 3.2 speeds
I removed the claim from the article that with SATA 3.2, speeds of 16 Gbit/s or 1969 MB/s are possible. It is my opinion that this is not SATA but PCI-e. If we look at how the source puts it:
Initially introduced in January 2013, the SATA Express specification enables a client storage ecosystem that allows SATA and PCIe solutions to coexist. A host implemented to this specification will connect to and function with either a SATA or PCIe storage device. PCIe technology enables increased interface speeds of up to 2GB/s (2 lanes of PCIe 3.0), compared with today’s SATA technology at 0.6GB/s (6Gb/s)
It seems clear to me that they are not claiming that SATA reaches 2 GB/s, but that such hosts support both SATA at 600 MB/s and PCI-e at 2 GB/s. Granted, the cabling is identical (well, doubled), but it seems wrong to claim SATA reaches such speeds. For reference, the edits introducing it are this and this and this.
As an afterthought, I've reintroduced the 1969 MB/s figure back into the paragraph rather than the title. By the way, I love how originally the paragraph linked to the later section Serial ATA#SATA Express with the description "a more detailed summary"; progressively ever more detailed summaries until you end up at the complete specification, hehehe ;). Digital Brains (talk) 13:25, 18 January 2019 (UTC)