Issues & Rosewill Rc-212 SATA/ATA Raid Cntrl

  • Thread starter Thread starter Flasherly
  • Start date Start date
F

Flasherly

Didn't take the 1T Samsung drive. Sort of.

Got it ID'd, but only after time-out errors with an earlier version of
a heavy-duty partition manager.

Then checked the transfer rates -- way out in whack-O Land -- so that
was it. I'm not taking chances with "that's how you destroy a HD"
scenarios.

The Samsung would, of course, work -just perfectly- on a relatively
AnCient 5+ yr-old Gigabyte/S-AM2 and its two SATA headers.

Guess-uhm, VIA/Rosewill, aren't quite up to speed with Samsung. I'm
certainly not -- up to messing with an already formated -and
populated- FAT32, if possible in otherly-case working scenarios.

Anyway - the Rosewill did, however, take a 750Gbyte Seagate just
perfectly. I can live with it, after all, I was putting in only one
other HD upon deciding two MB SATA wasn't going to cut it.

And that's just the way it happens that HDs get dicked and rotated
around in computers, willy-nilly (I've another 600Gbyte Seagate in
another system, if in need to fill/migrate into that extra Rosewill
slot).

So now you know if you've monster drives for what possibly Rosewill
wouldn't expect you to. I like that. I'll send them a letter to
paraphrase me on that on their box.
 
Flasherly said:
Didn't take the 1T Samsung drive. Sort of.

Got it ID'd, but only after time-out errors with an earlier version of
a heavy-duty partition manager.

Then checked the transfer rates -- way out in whack-O Land -- so that
was it. I'm not taking chances with "that's how you destroy a HD"
scenarios.

The Samsung would, of course, work -just perfectly- on a relatively
AnCient 5+ yr-old Gigabyte/S-AM2 and its two SATA headers.

Guess-uhm, VIA/Rosewill, aren't quite up to speed with Samsung. I'm
certainly not -- up to messing with an already formated -and
populated- FAT32, if possible in otherly-case working scenarios.

Anyway - the Rosewill did, however, take a 750Gbyte Seagate just
perfectly. I can live with it, after all, I was putting in only one
other HD upon deciding two MB SATA wasn't going to cut it.

And that's just the way it happens that HDs get dicked and rotated
around in computers, willy-nilly (I've another 600Gbyte Seagate in
another system, if in need to fill/migrate into that extra Rosewill
slot).

So now you know if you've monster drives for what possibly Rosewill
wouldn't expect you to. I like that. I'll send them a letter to
paraphrase me on that on their box.
Click to expand...

Yes, that is a VT6421A.

http://images10.newegg.com/NeweggImage/productimage/16-132-009-03.jpg

You could try one of these. $50

http://www.newegg.com/Product/Product.aspx?Item=N82E16816104006

I think that uses a SIL3124. That's the chip that matches the description.

http://www.siliconimage.com/products/product.aspx?pid=27

The Koutech comes in two versions. Koutech IO-PSA420 is for
regular desktop computers with 32 bit PCI. Works in 3.3V or 5V slots.
The IO-PSA421 is a 64 bit PCI card, and some server motherboards
have those slots. The IO-PSA421 won't give an advantage in a desktop
computer. Asus makes some older "workstation" motherboards with
a slot like that, but regular desktops are only 32 bit and run
at 32 bit speed limit (maybe 110MB/sec in terms of practical
transfer rate). The chip itself, runs the cables at SATA II rate,
so it should be compatible with a SATA II or SATA III drive. The
chip really works well, when the entire 64 bit interface is
available, as that gives 532MB/sec theoretical, and perhaps
300MB/sec practical. But when you need to connect a PCI computer
to SATA disks, you have to accept the shortcomings of the 32 bit
PCI bus (low speed).

So if VT6421A didn't work in Force150 mode, or if the drive
was SATA III, I might give that $49 card a try. Or look for
a company making a cheaper SIL3124 card.

In theory, you could also look for a SIL3132 bridged to a
PCI to PCI Express chip, but I doubt they'd be that much
cheaper if you could find one. I went looking on the Koutech
site, because years ago, they did some bridged versions of
cards. I didn't expect them to sell well, so for anyone in the
market, the best time to buy was when they first came out.

So these are the things I might look for, if I had a brand new
SATA III drive and an old PCI computer of some sort.

64 bit 3.3V PCI ----- SIL3124 ---- 4 SATA II (SATA II or SATA III drive)

32 bit 3.3/5V PCI --- SIL3124 ---- 4 SATA II (SATA II or SATA III drive)

32 bit PCI --- PCI_bridge_to_PCIE --- SIL3132 --- 2 SATA II (SATA II or SATA III drive)

Paul
 
Paul
Click to expand...

Gigabyte m61pme-s2 athlon x2 4200

I'm not being helpful especially [to you or anyone] if I'm being
relatively sarcastic about calling my MB 'relatively old.' My
apologies.

There shouldn't be anything 'that old' about its PCI bus standards:
What's old about it, in a very *odd* way I'm intending, is that its
BIOS and dual SATA/header bridge support take that terabyte Samsung HD
perfectly and without a hitch.

Now, in order to get the sarcasm down pat -- say, for instance, you
were find a 25% discounted Newegg Rosewill controller board (as I did)
-- which upon delivery yesterday, simply enough does not ...

Easeus Partition Manager Professional was hanging, whereupon an update
did not as much hang, although copy transfer rates through that card's
port at under 3000 Kbytes/per/sec are indicative of something
unacceptably, damnably wrong, and reason for getting it off that card,
as in pronto. The 750Gbyte Seagate, on the same card port, would hit
50,000 KB/ps bursts and run acceptably, normally, at sustained data
transfer rates over 20,000 to 30,000 KB/ps.

I'm not nor do I welcome risks. Either it's dead right or apparently
wrong.

The only thing I've risked was an initial price I shelled out for
reason to expectat a new Rosewill brand controller board from Newegg
would control a hard drive in not so imposing a manner as might a
similar 5-7 year-old Gigabyte m61pme-s2 function.

It's apparently an imposition within mistaken and false suppositions
for me to have made that assumption.

(The rest of the first post was sour-grapes rationalization. I do
have another 600G Seagate which conceivably could be migrated onto the
2nd port, a port in all honesty I presently have no use for, having
substituted one of the MB's SATA ports for the now restored, as
before, perfectly functional 1T Samsung, of course, alongside a newly
positioned 200G "beater" Seagate, by intent and purpose the broken-
back camel which prompted me to buy the controller board, in the first
place. I much prefer to smack around arbitrary boots over so many
arbitrarily sized partitions on that particularly hardy Seagate drive,
than even what upper limits may portend to a 750G Seagate happily
coexisting on the Rosewill. In other words, it was I who was simply
unhappy about using such large drives for anything less than long-term
storage.)

I've bought whole new MBs for less than I paid for that controller,
Paul. Granted, and again, relatively in a past tense, among great and
glorious smash-&-grab sales prices for *good* gear while Compgeeks.com
had it. So you can see why I'd as soon sit in my little pot here and
stew for awhile more before considering one red cent, or one moment
more, in addition to my purchase price to defray a slighted
substandard that never, actually, was one implicitly so stated. All I
believe Rosewill/Newegg has to say about this board is something along
that it does support LBA48(?) standards.

More than my last SI3112 Buslink controller, anyway, which would not
take the 750G Seagate, empathetically only the 200G Seagate, that is
before I killed the controller, to have rendered it clinically brain-
dead, with their support flash.
 
Didn't take the 1T Samsung drive. Sort of.

Got it ID'd, but only after time-out errors with an earlier version of
a heavy-duty partition manager.

Then checked the transfer rates -- way out in whack-O Land -- so that
was it. I'm not taking chances with "that's how you destroy a HD"
scenarios.

The Samsung would, of course, work -just perfectly- on a relatively
AnCient 5+ yr-old Gigabyte/S-AM2 and its two SATA headers.

Guess-uhm, VIA/Rosewill, aren't quite up to speed with Samsung. I'm
certainly not -- up to messing with an already formated -and
populated- FAT32, if possible in otherly-case working scenarios.

Anyway - the Rosewill did, however, take a 750Gbyte Seagate just
perfectly. I can live with it, after all, I was putting in only one
other HD upon deciding two MB SATA wasn't going to cut it.
Click to expand...

Some Samsung drives need a firmware update to make them compatible with
certain controllers:

http://knowledge.seagate.com/pkb_Home?language=en_US&c=Product_Family:Internal,Topics:Firmware

I have essentially the same VIA controller as you do, and it works fine
with a 2TB Samsung Spinpoint F4 model HD204UI. It's a generic controller
that came with ver. 4.95 firmware but was flashed with the Rosewill RC-212
card's ver. 5.20 BIOS:

http://www.rosewill.com/Mgnt/Upload...5- DOS flash utility and latest BIOS v5.2.zip

That's the only firmware upgrade I was able to find for the VIA VT6421A
controller, except for a version of 4.94 patched to fix bugs (Russian):

http://www.rom.by/forum/Vsja_podnogotnaja_VT6421A
 
Some Samsung drives need a firmware update to make them compatible with
certain controllers:

http://knowledge.seagate.com/pkb_Home?language=en_US&c=Product_Family...

I have essentially the same VIA controller as you do, and it works fine
with a 2TB Samsung Spinpoint F4 model HD204UI. It's a generic controller
that came with ver. 4.95 firmware but was flashed with the Rosewill RC-212
card's ver. 5.20 BIOS:

http://www.rosewill.com/Mgnt/Uploads2/AttachmentForProduct/RC_212, ...

That's the only firmware upgrade I was able to find for the VIA VT6421A
controller, except for a version of 4.94 patched to fix bugs (Russian):

http://www.rom.by/forum/Vsja_podnogotnaja_VT6421A
Click to expand...

Whoa...Russian patches? That's heavy. But, if it works, hey, that's
all it takes. It's been set up for one of the channels I missing (3
HDs in a 2 SATA header MB). So I'm basically cool with it (handles
the 700G Seagate just fine until things start getting bigger again).
Just struck me oddly flippant that a Gigabyte 5 yr. old MB BIOS takes
a 1T Samsung, whereas a fresh controller delivered a couple of weeks
ago turns its nose up on the same drive. Still feeling a little gun
shy after blowing the brains out of my last controller from a derailed
flash attempt on a Silicon Image Buslink board. But, thanks -- I'll
copy the links, get and put the updates with the controller drivers if
it's running an older firmware, and get the ser# on the Samsung and
see what Seagate is up to (think I saw mention they've recently bought
out Samsung's HD resources).
 
Back
Top