Are My Writes Slow?

[idata]

In comparing my benchmarks with other I have seen on this forum, I find that my read speeds seem similar to others but my write speeds are lower. This has been the case since the first benchmark run, right after setting up the new PC, so I'm guessing it's nothing to do with TRIM. Win 7 64-bit and my driver is msahci.sys, no RAID. Any thoughts as to why the writes are slower than others have reported?

Thanks,

Sally

[idata]

@ KoiTsu

You need to reread the standard there is quite the difference between SATA I, II , and III. The changes are mostly in contact pressure, contact locking and noise.

@Mr. Gerlo

The problem is I have yet to run in to a cable that meets SATA II without meeting SATA III. That being said if the cable was at fault you would not achieve the 60MB if it was a problem. As long as your cable isn't a OKGEAR your probably will not have an issue. They are the only cable I've run into that does not meet any of the standards (I hate the problems I had with those cables). How may IO can your drive do? I noticed bad cables first show up as a massive drop in IO capabilities. Also if you install ISM 8.8 bad cables will show up as external drives instead of internal (undocumented?). I noticed performance issues when the drive reported external and changing the cable would fix the problem. Most cables that have speed issues will be physically damaged and are not made wrong.

And yes the Intel 80GB drives are quite the disappointment in throughput:(

I just notice your issue is with a 510. That should be around 150 - 200MB. Check the cables for kinks and sharp twists.

Try it on a different system. The G2 & G3 I have will not work at full speed on NVIDIA (NF 2200), AMD 770, and Intel ICH8M chipsets right around 70MB on all three.

[idata]

RBerles wrote:

@ KoiTsu

You need to reread the standard there is quite the difference between SATA I, II , and III. The changes are mostly in contact pressure, contact locking and noise.

I spend most of my time in the ATA8-ACS working draft spec.. Have you checked out the http://www.sata-io.org/documents/SATA-Revision-3.0-FAQ-FINAL.pdf official SATA-IO SATA600 (SATA rev. 3) FAQ? You know, from the people who participate in the T13 interface and ATA8 protocol specification? If so, you'd know one of the Q&A items was:

Q14: Were there any attenuation or jitter issues that had to be addressed with the jump to 6Gb/s?

A14: Ensuring signal integrity was the primary challenge in doubling the SATA transfer speed for a second time while using the same cables and connectors that were originally defined for the first generation 1.5Gb/s version.

Furthermore, in the http://www.sata-io.org/documents/SATA-6Gbs-Fast-Just-Got-Faster.pdf official SATA-IO SATA600 white paper, you can find the following:

SATA Revision 3.0 was designed to facilitate a much more elegant – and cost effective – migration to higher data rates. Rather than completely redefining the spec, SATA Revision 3.0 achieves its data rate increase by faster signal switching. While this has resulted in changes to the PHY, modifications to the protocol have been kept to a minimum. In addition, SATA Revision 3.0 is backward compatible so that the same connectors and cabling used for SATA 3 Gb/s under SATA Revision 2.6 can be used for SATA 6Gb/s.

Let's not forget http://www.sata-io.org/documents/SATA-6-Gbs-The-Path-from-3gbs-to-6gbs.pdf another SATA-IO whitepaper too:

The key factor driving the SATA Revision 3.0 specification is backward compatibility. While there are advanced technologies available for managing attenuation and jitter, such modifications to a specification can require non-trivial redesign across the development and supply chain – ASICs, PHYs, cabling, connectors, drivers, and applications. By allowing SATA 6Gb/s systems the ability to utilize the same connectors and cables as SATA 3Gb/s systems, most of the modifications required to achieve 6Gb/s are confined to the PHY layer. Except for doubling the data rate, the PHY is no more complex than that defined by the SATA Revision 2.6 specification. Changes to the protocol itself are minor, and the physical connectors and cables remain unchanged.

It's pretty cool that I read specifications, or else I might think there's actually physical differences. PHY != cabling. There is an entry in the 2nd whitepaper mentioning cable quality issues, but there is absolutely nothing in the interface connector section stating anything has changed. Because nothing has.

SATA600, however, does introduce two new connector types for small-form-factor systems or environments: a 7mm connector intended for use with optical drives, a low-insertion force connector intended for 1.8" devices (e.g. SSDs or hard disks). The connectors are physically slimmer and shorter in depth. They ARE NOT mandatory; they're 100% optional.

If you really don't believe me, you can read the https://www.sata-io.org/documents/SATA-6gbs-equipment-design-and-development-finisar.pdf official SATA600 Design and Development document from SATA-IO, Appendix # 1; read it in full, do not skim. If you want to talk about SATA300 vs. SATA150, we can talk about that too, and I'd more than happy to point you to the design guide and ECN documents so you can read them yourself. Again: no differences in cabling required. It's all the same, nothing has changed. Regarding EMI/ESD, possibly you're thinking of eSATA connectors, where these things actually matter most? If so, again: no cabling changes required, however http://www.serialata.org/documents/esd_control_for_esata_a02-rc-1.pdf SATA300 documenatation from SATA-IO does go over low-level interface specifics. Once more, this is for eSATA, not internal SATA.

Hope this helps.