Cable matters? SATA vs. SATA II

[japanmanga2000]

Howdy, everyone! Just a quick question: I am currently using a SATA hard drive and I am planning on buying a new SATA II hard drive. Can I use my current cable and still have SATA II performance?

I tried looking around online for SATA cables. However, most of the products on the online stores are labelled as "SATA cable" (there were some that claimed to be SATA II cables as well but they cost more).

Does the cable really matter? If you use a regular SATA cable, does that mean that you won't get the fullest performance out of the SATA II hard drive? Any insights/feedbacks on these would be greatly appreciated! Thanks for your help!

 

[4745454b]

SATA, SATAII, and SAS all use the same cable. The SATAII cables might have a redisgned plug on the end to prevent it from accidently coming off. (if so, that might be worth it.) If you already have the SATA cable, and want to reuse it for the new drive, you should be good to go.
As a side note, remember that the difference between SATA and SATAII is the speed of the INTERFACE, not the speed of the drive. Harddrives of today are barely reaching the limits of ATA-100, and can't come close to filling the bandwith availible for SATA/SATAII. (What I'm trying to say is its a waste of money buying a SATAII drive as a similar model SATA drive should be just as fast.)

 

[wun911]

yep its all backwards compatible....

SATAII is a gimmick tho

 

[4745454b]

I don't view it as a gimmick, just something that isn't needed. Its like building a road where the posted speed limit is 700MPH. Who cares if you could drive that fast as there isn't a car out there that can do over 275. Same with SATAII. Who cares if the bandwith is 3Gbps, drives can't even fill ATA-100.

 

[linux_0]

yep its all backwards compatible....

SATAII is a gimmick tho

Yup SATAII is marketing and hype.

Your existing cables will work just fine :-D

 

[Codesmith]

Actually its more of a marketing lie.

There are no SATA II drives period. Anyone who labels their product SATA II is making up their own imanginary standards.

The SATA-IO board says thare are no SATA II devices and that the term SATA II should never be used to describe a product. They have the final word on such matters.

All SATA drives are from the same "genaration" and follow the same standards.

This issue has already been beaten into the ground.

 

[ZOldDude]

Howdy, everyone! Just a quick question: I am currently using a SATA hard drive and I am planning on buying a new SATA II hard drive. Can I use my current cable and still have SATA II performance?

I tried looking around online for SATA cables. However, most of the products on the online stores are labelled as "SATA cable" (there were some that claimed to be SATA II cables as well but they cost more).

Does the cable really matter? If you use a regular SATA cable, does that mean that you won't get the fullest performance out of the SATA II hard drive? Any insights/feedbacks on these would be greatly appreciated! Thanks for your help!

No such thing as SATA 2.

Z

 

[linux_0]

Tru64


Marketing at it's worst

 

[Codesmith]

The sad part is that when they do want to come out with a new generation of SATA drives compling to a higher set of standards they can't call it SATA II because people have been missusing the term.

Its going to have to be SATA Advanced or SATA 2nd Edition.

 

[wun911]

ok ok its not a gimmick its a "waste" "unnessary" "not fully utilized" my englash is poor

 

[Dade_0182]

Tru64


Marketing at it's worst

This suddenly makes AMD's marketing look good 8O
And we all know that thats total BS.

@japanmanga2000
Just get a couple of raptors and put them in raid 0 or something.

Even if there was a thing like SATAII and cables for it, would a silly little cable really hurt your pocket that badly???

 

[Dade_0182]

Tru64


Marketing at it's worst

This suddenly makes AMD's marketing look good 8O
And we all know that thats total BS.

@japanmanga2000
Just get a couple of raptors and put them in raid 0 or something.

Even if there was a thing like SATAII and cables for it, would a silly little cable really hurt your pocket that badly???

 

[Dade_0182]

Sorry for the double post, my internet connection at work is totally useless after 7:30AM.

 

[linux_0]

Tru64


Marketing at it's worst

This suddenly makes AMD's marketing look good 8O
And we all know that thats total BS.

@japanmanga2000
Just get a couple of raptors and put them in raid 0 or something.

Even if there was a thing like SATAII and cables for it, would a silly little cable really hurt your pocket that badly???



Cables sometimes can make a huge difference.

And sometimes not.

 

[Dade_0182]

I'm south african so i don't know what the prices are there for cables, I'm guessing maybe $4 or so. With all that real expensive hardware going into the build will ,lets be silly and say doubled for this situation, another $4 kill you? I don't think so but that don't matter coz SATAII drives don't exist.

 

[linux_0]

I'm south african so i don't know what the prices are there for cables, I'm guessing maybe $4 or so. With all that real expensive hardware going into the build will ,lets be silly and say doubled for this situation, another $4 kill you? I don't think so but that don't matter coz SATAII drives don't exist.

I agree with you $4 in a build is not a big deal.

However since the cables are the same it doesn't even matter that much.

:-D

 

[fainis]

SATA 2 is just a standard...the cable is the same except for a retainer clip.......it has the same pinout ....etc.....

sata 2 standard includes few benefits over the sata standard ..but not necesary faster .... i think ncq and tcq is better.......tcq is old ..i think it was introduces in 1997 in ata specifications....

anyway sata 2 is the way even if not an improvement in speed from sata ....it`s just a standard more

 

[fainis]

as for the speed it`s like that:

standard sata it`s 1,5gbps (bit) ... or 150mbytes per second.... 20 percent is used for error corection codes 150x1,2=180mega bytes per second
180 mega bytes per seconds x 8 = 1440 mega bits per seconds...

now you know why a raptor 150 only has a transfer rate of same 127....150 is maximum....


so don`t compare the ata 100 standard.....with this one.......

sorry for the spelling ..i was in a rush...

 

[JB123]

Currently I have been having many issues with my WD Caviar Black 640 gig hard drive performance (very slow transfer rates) and Western Digital Support claims the issue is because im using a SATA type 1 cable and I should be using Type II so they are sending me a free type II cable. I laughed at the support technician but he claims they have been solving many issues by switching the cable and the Caviar blacks require a type II cable. I really dont think the cable is my issue but Ill will update once I try it.

 

[raifdop]

The SATAII is working overtime with my Intel SSDSA2SH064G1GC and SSDSA2SH032G1GN, working well above the 1.5GB/s of the initial signaling rate of the physical layer of the SATA standard at least on the 32GB drive. SATAII cables are required for speeds above 1.5GB/s.

These SSD's are AMAZING for bootup performance. Complete boot from power up in under 30 seconds on Windows 7 x64. Just a little over 30 seconds running Vista 64.

 

[wuzy]

Firstly there's no such thing as "SATA II" as per defined by SATA-IO. Your Intel X25-M G2 has a SATA 3Gb/s interface working at a maximum theoretical 300MB/s (8b/10b encoding).
Any SATA 1.5Gb/s cables made within the spec. which is less than 1m are capable of 3.0Gb/s, period.

Lastly, I hope you realise this is an extremely old thread.

 

[letstrynl]

Anything on this thread may be true, but ...

I have a ASUS P6T Deluxe V2 board + Samsung Spinpoint F1 1 Tb drive connected to the on-board ICH10 interface.

After booting (Ubuntu GNU/Linux) and looking at SMART status I have consistantly Hardware_ECC_Recovered (+- 70000) errors.
On different controller SATA2 ports, on different cables.

Other SMART attributes are OK, so no disk fail.

This attribute tells the controller <-> disk communication has been corrected 70.000 times in about 30 seconds.

So, it might be a good idea to test some cables and see what the result is.

BTW: What are the best SATA cables around?

 

[letstrynl]

... and just to be sure ...

I added a 160 Gb SATA2 drive to my system.

Just did a "dd if=/dev/sda of=/dev/null bs=8M count=8" to read 64 MB of the 1 TB disk.

10.000 more ECC errors. Speed was 116 Mb/s (928 Mbit, one third of theoretical throughput).

After that I did the same on the 160 Gb Samsung disk.

No ECC errors. Speed was 64 Mb/s (512 Mbit, one sixth of theoretical throughput).

All this on the supplied (ASUS) SATA cables.

I'm puzzled.

I don't want to offend anyone who made strong remarks in this thread, but it seems speed does matter.

Another thought might be that the Intel ICH10 controller has issues?

 

[cjl]

Those speed numbers are perfectly fine. The interface has a theoretical maximum of 300MB/s (this is lower than 3Gb/s because of the 8b/10b encoding used). That doesn't mean that the drives are capable of that kind of transfer rate. On a modern hard drive, the only time you will ever see that kind of speed is when reading from the cache. The disk will have a transfer rate of ~100-130MB/s. Your 160GB is older, and 64MB/s is about right. Cables do not matter as long as they conform to the SATA standard, and the ICH10 is working fine.

 

[letstrynl]

Hi cjl, thanks for replying.

Yes, the faster Samsung F1 drive works like a charm, but that wasn't what the discussion was about ;-)
It was about better/worse cables.

I can imagine that the drive would work even faster if it hadn't to re-transmit 10.000 times in a few seconds.

And the cable is -as far as I know- the limiting factor here.

So, I still conclude there are cables that are said to be 300 MB/s but are in fact not.

Or would the drive be at fault here, maybe because it can't flush its cache on the 300 MB/s pipe because it's not fast enough?

That wouldn't be logical either, I should get ECC errors on the slower drive also then (also has a cache).

The (kinda interesting) discussion remains...

What's causing these ECC errors?