SSD in SCSI mode instead of AHCI using X79 chipset

[lucbujold]

HI to you,

I'm running with an Asus P9X79 Pro, BIOS updated to the latest version and a Corsair Neutron 120GB...
The SATA mode in BIOS is set to AHCI.

But in Windows 7, all drives (1 SSD, 1 WD HDD and 1 Seagate HDD, and Asus DVD Drive) are seen as SCSI devices...

I tried to use only windows 7 64 bits native AHCI drivers and I got the same things as I used the IRST drivers provided by Asus... As I've done many search on the web about AHCI and X79 chipset I've found so many issues and no real answer, I'm deseperatly need help.

As I want to get all the performance and stability of my system I just want to make sure that everthing work as it is design to...

thanks in advance for your help...

 

[pbcopter]

That is normal for those controllers/drivers to report SCSI. The drives are functioning normally as AHCI.

If you are connected to the Intel controller, what IRST driver version are you using?

 

[lucbujold]

The RST 12.8.0.1096...

I know that those drivers can be reported as SCSI but why?
is it because the X79 use the SCSI and ACHI the same way or is it because there is no good or specific drivers?

And as I saw many issues with the RST, could it be better to use some native windows 7 AHCI drivers?

Other little question, as you said that it is operating as an AHCI, how can I make sure that all benefits of AHCI are effectives?

If the BIOS sees 125GB instead of 120GB, could it be a clue of a SSD problem?

 

[essenbe]

IRST (Intel Rapid Storage Technology) is Intel's raid driver, but it also contains IaSror.SYS which is Intel's AHCI driver. It is the only way to get it that I know of. MSAHCI is Microsoft's AHCI driver. I have found that I seem to get better performance slightly from using the Intel driver. If you will go to device manager and expand the IDE ATA/ATAPI Controllers, you should see one or the other there. If it says MSAHCI or has an Intel chipset AHCI driver listed, you are in AHCI mode. Those are your Sata port controllers, if it says you have the AHCI drivers, then you are in AHCI mode.

 

[pbcopter]

Go to this site and download and install the drivers under Intel RAID/AHCI Drivers C. 1. called

Universal 64bit Intel RST AHCI & RAID driver v11.2.0.1006 mod by Fernando

(The only mod done is to let the installer recognize the X79)

To me, these are the best drivers for the X79 and the drives are not ID'd as SCSI.

Other AHCI and RAID Drivers » Recommended AHCI and RAID Drivers

 

[lucbujold]

Thanks Pbcopter...!

It works...!

But I'm not able to do firmware upgrade of the SSD... I was'nt able to do before... I've read the it was not possible to do it when the drive is seen as SCSI... So my goal was to get it seen as an ATA: goal achive.

I'm thinking that Corsair may have remove the new m208 firmware (I have m206), or I have some issue with the drive...

and I have another question: why a WD black sata3 (pluged in a Sata3 port) is slower than a Seagate Barracuda sata3 (pluged in a Sata 2 port).

WD: Write: 90MB/S Read: 96 MB/S
Seagate: Write: 140MB/S Read: 182 MB/S.
My SSD Corsair: Write: 306MB/S Read: 500 MB/S

tested with AS SSD

Theses results are the same after the installation of the new drivers (by Fernando)

 

[pbcopter]

Is one of your drives connected to the Marvell controller? If so, that might be the difference. That controller will not yield the same speeds as the Intel.

What drivers are you using for the Marvell?

 

[lucbujold]

The Western Digital was on the SATA1 (6G) Port...
And the Seagate was on the SATA2 (3G) port...

They are using the Intel controller...
I have a DVD writer (sata III) on the SATA3 (3G) port...

I did'nt tried to use the Marvel controller as I was thinking that it was only for raid configuration...

 

[pbcopter]

I really can't say why you are getting those benchmarks.

What do you get if you reverse the connections on the WD and Seagate?

 

[essenbe]

I really don't think it matters whether they are connected to a sata 2 or a sata 3 as long as it is controlled by the Intel controller. I do not believe and have never seen a mechanical hard drive that can come close to saturating the bandwidth of sata 2. You must remember, it was only the past few years that SSDs were capable of that.

 

[lucbujold]

I know, Essenbe that either HDD can saturate SATA2 but my concern is more the fact that a WD Black which is supposed to be a good performer with compare to a Seagate Barracuda...

And I forgot to mention that the Seagate is splitted in 2 1TB partitions... And what I've just discoverd is that one of the partition have sames readings that the WD... And the other got much faster readings... Even though both partitions have same amount of free space...

Of Pbcopter, I've tried the WD on the Marvell Controller (which is a SATA3) and I got the same results...

Tried to swap the WD and Seagate ports and the results are the same...
So I don't think that the motherboard as an issue, but I will try to find somes benchmarks to compare my drives...

By the way, the SSD will be replaced by Corsair, they suspect something wrong about the firmware...
And for the SSD replacement, what would be the best solution? A new clean install or I can just do a system backup with Macrium?

 

[pbcopter]

It is up to you. I typically use a image I take periodically with Acronis but if you don't have much installed a clean install is always good.

If you do a clean install, I recommend that you disconnect the other drives first, to eliminate the possibility of problems.

 

[lucbujold]

OK' I will take note of that...

And is there is a kind of tutorial for SSD optimisation?

 

[lucbujold]

I've found this: Test Results - UserBenchmark.

iit is my tests and what I've discovered is that my Seagate is in the target...
But my WD is not...
And for the SSD I'm on target...

Do you think that the amount of free space can affect the results?

So I will try to contact WD to check out if there is a issue...

 

[essenbe]

If it helps here is a HD Tune of a 1TB WD Black I did a while ago connected to a Sata 3 Port and my current 2TB Seagate Barracuda connected to a Sata 2 Port.

 

[essenbe]

I think another thing you will find is it matters which test you run on the drive. Different tests measure differently and use different types of data for their tests. Some drives respond to different tests differently.

 

[lucbujold]

Hi, Essenbe it is a really good luck that you had the same drives than me!!! AND used on the same ports!!!

With HD Tune Pro 5.50,

On the WD I got 96 MB/S

On Seagate I've got just a 5mb/s faster than you.

So there is only two things I see:
First, you and I got a "defective" WD that is not performing as it should.
Second, the WD is performing as designed and it is just not the performer it claims to be...

I've contacted WD to get some light on it...
Stay tuned!

And again, all I have to tell about you guys (and all sevenforum) is that are just the best we can find on the web!

 

[essenbe]

lucbujold, thanks and it could be possible that both of us have a defective drive, but I seriously doubt it. I no longer have the Western Digital. It was in a PC I built and sold. But it's still running and is about 5 or 6 years old now. Again, the numbers are really about 'ball park' numbers. They will be affected by what processes you have running in the background, many of which you are not even aware of. You will most likely get different numbers if you run it again, but similar. The numbers also vary with which test you use. I believe you would grt quite different numbers if you ran ATTO benchmark test. When I had the drive I ran several tests with different drives I owned at the time. The Samsung Spinpoint 1 TB HD103SJ consistently beat every drive I owned including the WD Black. Sadly, Seagate bought Samsung's Hard drive division and you can no longer bet the Spinpoints, other than the Laptop drives. But again, it depends a lot on what test you run as to the results.

 

[essenbe]

lucbujold, This is what I mean about which test you run. This is my Samsung 840 Pro 256GB. If you didn't know that and looked at each test, you'd think it was a different drive. It shows how much difference there is in which test you run.

​