A8N-SLI Deluxe : using SATA_RAID connectors just for extra SATA drives (not RAID) ?

  • Thread starter Thread starter Toto
  • Start date Start date
T

Toto

I'd like to build a PC with seven or eight SATA Drives, without any RAID.
No doubt I can use 4 Drives on the SATA1 to SATA4 connectors (supported by
the NVidia chipset).

But, even after downloading & reading the manual, I'm not sure I can use
the SATA_RAID1 to SATA_RAID4 connectors (supported by the Sil3114
controller) just for adding 'ordinary' drives, without activating any kind
of Raid.

Can someone help me ?

TIA
 
Toto said:
I'd like to build a PC with seven or eight SATA Drives, without any RAID.
No doubt I can use 4 Drives on the SATA1 to SATA4 connectors (supported by
the NVidia chipset).

But, even after downloading & reading the manual, I'm not sure I can use
the SATA_RAID1 to SATA_RAID4 connectors (supported by the Sil3114
controller) just for adding 'ordinary' drives, without activating any kind
of Raid.

Can someone help me ?

TIA

If you look here,

ftp://ftp.asus.com.tw/pub/ASUS/misc/ide/silicon/3114/

you can see this driver. I believe this tiny download is the
key to running individual disks.

SiI3114_SATA_1007.zip

There is also a driver link on this page:
http://www.siliconimage.com/products/product.aspx?id=28&ptid=1

HTH,
Paul
 
(e-mail address removed) (Paul) wrote in
If you look here,

ftp://ftp.asus.com.tw/pub/ASUS/misc/ide/silicon/3114/

you can see this driver. I believe this tiny download is the
key to running individual disks.

SiI3114_SATA_1007.zip

There is also a driver link on this page:
http://www.siliconimage.com/products/product.aspx?id=28&ptid=1

HTH,
Paul

Thanks a lot, Paul, I've DL what you suggest. But as I just want to know
IF this configuration is possible (I haven't got the Mobo yet), I can't
test these drivers...
Can you (or someone else) just confirm me it's POSSIBLE to run individual
disks on these SATA_RAIDxx connectors ? Your Silicon Image Link suggests
it's possible, but...

Then I'll buy the mobo & use the drivers you suggest...
 
Toto said:
Thanks a lot, Paul, I've DL what you suggest. But as I just want to know
IF this configuration is possible (I haven't got the Mobo yet), I can't
test these drivers...
Can you (or someone else) just confirm me it's POSSIBLE to run individual
disks on these SATA_RAIDxx connectors ? Your Silicon Image Link suggests
it's possible, but...

Then I'll buy the mobo & use the drivers you suggest...

http://www.hardocp.com/article.html?art=NTU4LDM=

"The Sil3114 S-ATA Controller option controls the state of the
controller chip itself. The SATA option allows the tied SATA
ports to act as stand alone ports, while the SATA RAID option
enables the RAID hardware BIOS. Note that SATA drives will
operate in both stand alone and RAID mode with the Sil3114 S-ATA
Controller option set to the SATA RAID setting."

That is the best quote I can find on the topic.

HTH,
Paul
 
(e-mail address removed) (Paul) wrote in
http://www.hardocp.com/article.html?art=NTU4LDM=

"The Sil3114 S-ATA Controller option controls the state of the
controller chip itself. The SATA option allows the tied SATA
ports to act as stand alone ports, while the SATA RAID option
enables the RAID hardware BIOS. Note that SATA drives will
operate in both stand alone and RAID mode with the Sil3114 S-ATA
Controller option set to the SATA RAID setting."

That is the best quote I can find on the topic.

HTH,
Paul

That's it ! Thanks a lot, Paul...
 
Toto said:
(e-mail address removed) (Paul) wrote in


That's it ! Thanks a lot, Paul...

I have six SATA hard drives 4 on the Si3114 (red) as JBOD or individual
drives and 2 on the Nvidia SATA ports (black) No raid required

Works great, might buy 2 more hard drives
 
I have six SATA hard drives 4 on the Si3114 (red) as JBOD or individual
drives and 2 on the Nvidia SATA ports (black) No raid required

Works great, might buy 2 more hard drives
I presume that you use the RAID utility to set up a JBOD for each of
the single drives attached to eh SI3114 connectors (assuming you don't
want to span any of them).

After reading the terribly scanty documentation at the SI3114 download
page, I'm deducing that by flashing the controller with the IDE BIOS
(as opposed to the RAID BIOS), that one might be able to simply
connect independent SATA drives to the Silicon Image controllers
without having to go through the hassle of the JBOD setup. Can anyone
tell me if this is the case? Is it, in fact, possible to flash the
SI3114 independently of the motherboard BIOS? I ask because the tiny
readme accompanying the SI Flash utility says it's for use with SI
"add-in cards." It warns against trying to use it with Macintosh
cards but says nothing at all about SI3114 controller chips built into
motherboards, so I'm reluctant to try it without confirmation that
it's meant for this purpose.

Thanks.

Ron
 
After reading the terribly scanty documentation at the SI3114 download
page, I'm deducing that by flashing the controller with the IDE BIOS
(as opposed to the RAID BIOS), that one might be able to simply
connect independent SATA drives to the Silicon Image controllers
without having to go through the hassle of the JBOD setup. Can anyone
tell me if this is the case? Is it, in fact, possible to flash the
SI3114 independently of the motherboard BIOS? I ask because the tiny
readme accompanying the SI Flash utility says it's for use with SI
"add-in cards." It warns against trying to use it with Macintosh
cards but says nothing at all about SI3114 controller chips built into
motherboards, so I'm reluctant to try it without confirmation that
it's meant for this purpose.

Thanks.

Ron

When the SIL3114 is used on an add-in PCI card, there are two
files that can be used for flashing. There is an IDE file and
a RAID file. I think the SIL3112 may have had the same kind of
option. (Hardware-wise, the 3114 is just a couple of 3112's
glued together. Software is what makes it all work.)

On a motherboard, the thing that is a file in the previous case,
is just a module buried in the BIOS flash image. That means
you cannot "flash" just that small portion of the BIOS. (A
BIOS file is a collection of compressed modules, in a sense
it is a tiny flat file system. The DMI, ESCD, boot block,
and microcode cache are exceptions and are not packaged quite
the same way.)

You would have to use a BIOS modification tool, to extract the
old module, and install whatever new module you happen
to have. The module has to have a format suitable for launch
by the motherboard BIOS. For example, I have a tool called
MMtool, that I use to extract and decompress modules inside
an AMI BIOS (to see, in fact, what version of RAID BIOS is
included with various versions of BIOS). I think that tool
can also insert a module, but I've never tried that.

Quite a while ago, I tried to modify an Award BIOS (with a
different tool), but there were a couple of details about the
way Asus sets up the BIOS, that caused the status display of
the tool I was using, to show garbage. I was not about to take
the output of that tool and try flashing a motherboard with it.

To give an example of the capabilities out there on the
Internet, there are people who have installed the 4.2.50
SIL3112 RAID BIOS into A7N8X BIOSes. That was because Asus
was not including that version in their release BIOS, choosing
to use 4.2.47 instead. So there are people who have succeeded
at doing this stuff. If I had a BIOS Savior for every board I
owned, I wouldn't mind being more adventurous.

Paul
 
milleron said:
I presume that you use the RAID utility to set up a JBOD for each of
the single drives attached to eh SI3114 connectors (assuming you don't
want to span any of them).

Yes I believe it was in the Si3114 BIOS setup screen
 
I'm deducing that by flashing the controller with the IDE BIOS
(as opposed to the RAID BIOS), that one might be able to simply
connect independent SATA drives to the Silicon Image controllers
without having to go through the hassle of the JBOD setup.

Is the 'JBOD' set up that difficult ? And can you confirm that in the JBOD
mode, you can really use the "single' disks separately (I had read
somewhere that in this 'bunch of disks'mode, you write the same thing on
all the 'single' disks -probably misunderstood...). Because flashing the
bios sounds difficult for me...
 
Toto said:
Is the 'JBOD' set up that difficult ?

No it was pretty straughforward with the Si3114 With the nvidia controller
I thought you needed the raid enabled. It turns out that you want raid
disabled if you want the nvidia controller to treaat each disk as a
separate drive. Spent three days figuring that one out.

And can you confirm that in the JBOD
mode, you can really use the "single' disks separately (I had read
somewhere that in this 'bunch of disks'mode, you write the same thing on
all the 'single' disks -probably misunderstood...)

With Si3114 they are single disks acting separately I believe its called
RAID ten. With Nviidia in raid mode it can go either way


.. Because flashing the
bios sounds difficult for me

Well it's bios within a bios

So far no one has come out with a hacked BIOS in any form for the A8N-SLI
AFAIK
....
 
"Leadfoot" said:
No it was pretty straughforward with the Si3114 With the nvidia controller
I thought you needed the raid enabled. It turns out that you want raid
disabled if you want the nvidia controller to treaat each disk as a
separate drive. Spent three days figuring that one out.

And can you confirm that in the JBOD

With Si3114 they are single disks acting separately I believe its called
RAID ten. With Nviidia in raid mode it can go either way


. Because flashing the

Well it's bios within a bios

So far no one has come out with a hacked BIOS in any form for the A8N-SLI
AFAIK
...

Leadfoot, have you been able to set up disks individually ?

There are some threads where people claim to have trouble with this.

http://forums.pcper.com/showthread.php?t=390717
http://forums.pcper.com/showthread.php?t=392193

The claim is, that at one point the A8N-SLI had an option to set the
SIL3114 to IDE mode, versus RAID mode. I don't see that option in
the manual, so perhaps it was only in early versions of the
actual BIOS ?

You cannot download a datasheet for the SIL3114, but for the
SIL3112 I did get a datasheet. On that chip, there is a pin
called "IDE_CFG", used for IDE-RAID config.

"When PCI_RST_N is deasserted, this pin is an output and represents
flash memory address bit 0. During reset, it is sampled to configure
Mass Storage class or RAID mode in the PCI Class Code register. A
high on this pin sets Mass Storage class, a low sets RAID mode. The
configuration state is latched internally when PCI_RST_N is
deasserted. This pad is internally pulled high to enable Mass
Storage class if left unconnected."

Now, _if_ that pin existed on the SIL3114, it could be driven by a
GPIO on the motherboard, under BIOS control. Or, it could be set
permanently to RAID mode. In any case, if people are having trouble
installing the "vanilla" IDE driver for the SIL3114, this might
be the reason. If the IDE driver is installed, and that pin is
sampled in RAID mode, then the enumeration of the 3114 would
indicate "RAID mode", and the installer could not work.

In RAID mode, the options I could see are: setting a single disk
to a RAID 0 stripe of one disk, or using JBOD. I thought one
of the side effects of using JBOD, was an OS like WinXP might
deal with the disk in "dynamic disk" mode, and for some third
party tools, this can cause problems.

So, there appear to be a few unanswered questions about SIL3114
on A8N-SLI, such as whether it ever really supported the
vanilla IDE driver, and why Asus would remove it, if the
ability to do that was present. Since there is only one
motherboard BIOS module available for SIL3114, that module
would have to deal with such a pin in both states, so that
should not be an issue. (And, BTW, _all_ the BIOS files I
checked for A8N-SLI Deluxe have the same 5139.bin module to
control the SIL3114 chip in the BIOS. That has never been
updated apparently, but neither do I see changing that
making any difference.)

Without a datasheet for the SIL3114, I cannot be 100% certain that
the same kind of pin exists, but architecturally, the SIL3114 is
just two SIL3112 chips slapped together. So the same pin
should exist somewhere on the chip. The SIL3114 has more pins than
the SIl3112, so it is not possible to guess at which pin it is.

Paul
 
When the SIL3114 is used on an add-in PCI card, there are two
files that can be used for flashing. There is an IDE file and
a RAID file. I think the SIL3112 may have had the same kind of
option. (Hardware-wise, the 3114 is just a couple of 3112's
glued together. Software is what makes it all work.)

On a motherboard, the thing that is a file in the previous case,
is just a module buried in the BIOS flash image. That means
you cannot "flash" just that small portion of the BIOS. (A
BIOS file is a collection of compressed modules, in a sense
it is a tiny flat file system. The DMI, ESCD, boot block,
and microcode cache are exceptions and are not packaged quite
the same way.)

You would have to use a BIOS modification tool, to extract the
old module, and install whatever new module you happen
to have. The module has to have a format suitable for launch
by the motherboard BIOS. For example, I have a tool called
MMtool, that I use to extract and decompress modules inside
an AMI BIOS (to see, in fact, what version of RAID BIOS is
included with various versions of BIOS). I think that tool
can also insert a module, but I've never tried that.

In fact, the download of the flashing tool at SI does include a
separate program to modify BIOS files. Their scanty documentation
describes a BIN file that is intended to be inserted into a
motherboard BIOS file, but the corresponding firmware download seems
to contain only an rXXXX.bin (RAID) file and a bXXXX.bin (IDE) file
for flashing PCI cards. Unfortunately, it does not include the
XXXX.bin file one would need to actually utilize their UpdBIOS utility
to modify a motherboard BIOS. This stuff is way over my head and
something I'd not try even WITH a BIOS Savior. It irks me to realize
that Asus could easily provide an A8N-SLI BIOS that would allow users
to connect single SATA drives, apparently including SATA optical
drives, to the SI3114. What a silly, customer-hostile decision to
leave it as a RAID-only controller!
I'd be assuaged if I knew for sure that it's easy to connect them as
single-disk JBOD "arrays," but I don't have a couple of SATA drives
lying around with which to perform the experiment.


Quite a while ago, I tried to modify an Award BIOS (with a
different tool), but there were a couple of details about the
way Asus sets up the BIOS, that caused the status display of
the tool I was using, to show garbage. I was not about to take
the output of that tool and try flashing a motherboard with it.

To give an example of the capabilities out there on the
Internet, there are people who have installed the 4.2.50
SIL3112 RAID BIOS into A7N8X BIOSes. That was because Asus
was not including that version in their release BIOS, choosing
to use 4.2.47 instead. So there are people who have succeeded
at doing this stuff. If I had a BIOS Savior for every board I
owned, I wouldn't mind being more adventurous.

Paul

Ron
 
milleron said:
In fact, the download of the flashing tool at SI does include a
separate program to modify BIOS files. Their scanty documentation
describes a BIN file that is intended to be inserted into a
motherboard BIOS file, but the corresponding firmware download seems
to contain only an rXXXX.bin (RAID) file and a bXXXX.bin (IDE) file
for flashing PCI cards. Unfortunately, it does not include the
XXXX.bin file one would need to actually utilize their UpdBIOS utility
to modify a motherboard BIOS. This stuff is way over my head and
something I'd not try even WITH a BIOS Savior. It irks me to realize
that Asus could easily provide an A8N-SLI BIOS that would allow users
to connect single SATA drives, apparently including SATA optical
drives, to the SI3114. What a silly, customer-hostile decision to
leave it as a RAID-only controller!
I'd be assuaged if I knew for sure that it's easy to connect them as
single-disk JBOD "arrays," but I don't have a couple of SATA drives
lying around with which to perform the experiment.



I have 6 sata drives and have each drive on the Si3114 as it' s own array
i.e 4 JBOD arrays with one drive in each array.

Works great
 
I have 6 sata drives and have each drive on the Si3114 as it' s own array
i.e 4 JBOD arrays with one drive in each array.

Works great

That's very good to know. You don't perchance have an optical SATA
drive in that group, do you? I don't really see how you could
configure one as a JBOD, but I thought I'd ask in case there's a way.

Thanks


Ron
 
That's very good to know. You don't perchance have an optical SATA
drive in that group, do you? I don't really see how you could
configure one as a JBOD, but I thought I'd ask in case there's a way.

Thanks

You must have a plextor, only SATA one I've heard of and have never tried
one
 
In fact, the download of the flashing tool at SI does include a
separate program to modify BIOS files. Their scanty documentation
describes a BIN file that is intended to be inserted into a
motherboard BIOS file, but the corresponding firmware download seems
to contain only an rXXXX.bin (RAID) file and a bXXXX.bin (IDE) file
for flashing PCI cards. Unfortunately, it does not include the
XXXX.bin file one would need to actually utilize their UpdBIOS utility
to modify a motherboard BIOS. This stuff is way over my head and
something I'd not try even WITH a BIOS Savior. It irks me to realize
that Asus could easily provide an A8N-SLI BIOS that would allow users
to connect single SATA drives, apparently including SATA optical
drives, to the SI3114. What a silly, customer-hostile decision to
leave it as a RAID-only controller!
I'd be assuaged if I knew for sure that it's easy to connect them as
single-disk JBOD "arrays," but I don't have a couple of SATA drives
lying around with which to perform the experiment.

Ron

There is actually a download with all three files in it.
Look for the 5.0.39 BIOS version on the SiliconImage website.
The 52KB file is the motherboard one, while the two 128KB
files are the add-in card versions.

And, if you are tired of fighting problems with Plextor
SATA, there appear to be adapters that will convert a
SATA drive to an IDE host.

http://www.satagear.com/SATA-IDE-34_SATA_Adapter.html
http://www.satagear.com/SATA-MM-A1_SATA_Adapter.html

One of those devices is based on this chip:

http://www.jmicron.com.tw/serialata/satabridge.htm
http://www.jmicron.com.tw/serialata/pdf/JM20331P1_PB.pdf

Apparently, that chip can convert a SATA disk to an IDE host,
and by changing a mode pin, it can also do a PATA disk to
a SATA host.

Paul
 
Back
Top