sil3114 versus sil3114a

[Gary D. Margiotta]

> is the man page for the driver out of sync (5.4, 6-current) ?:
>
> HARDWARE
>     The hptmv driver supports the following ATA RAID controllers:
>
>     ·	 HighPoint's RocketRAID 182x series
>
> so it does suport 18x0 and 18x0A (that is my understanding from the
> posts)? (it does according to highpoint but not according to the man
> page)
>
> i was not planning to buy hardware raid solution but the price is not
> that bad (considering that i can plug it to the existing machine now and
> having it running in a new motherboard later).

I won't go as far as to say that the 1820 is definitely supported, as I 
don't know exactly, but you could probably look through the driver source 
for our ATA system to see if the card ID is present, and then you'd know 
for sure.  I would personally think that the 1810 would be supported, as 
it's probably all int he same family of cards, so it's probably not too 
different to support all revisions.  However, I'm not a coder, so I don't 
want to give you bad advice.

The one benefit of the hardware raid card, as opposed to the 'software 
raid' ones, is that if my motherboard fails, I can in theory pull out the 
raid card and drives, move them into another system, and boot back up 
without missing a beat, since the raid metadat and config is contained on 
the actual card, and not controlled by a software driver or the operating 
system.  I've known others to have reported that this is easy, and does 
work fine, but I have no personal experience (yet) on that one.  Plus, 
with true hardware, you obviously get the performance gains as well, since 
you can (as is the case with my setup) use a lower standard system and 
CPU, since the hardware card does all the hard work for you, and you don't 
rely on the local system CPU to do it.

>
> well, the card is likely to move to and AMD64 bit machine i am planning
> to buy in about a year.
>

That will probably result in one quick system, I'd love to see performance 
numbers on that just to drool over.

>
> thank you all very much for responding, especially pointing to me that
> the sil chip-sets are not that good. it would be great if the (S)ATA
> maintainer summarised his opinions somewhere ;-).
>

You're welcome, I hate to have someone buy a solution that's not optimal, 
and end up havnig troubles with it down the road, and have wasted money 
and time on a solution when it could be avoided from the start.

And, actually, if you search the -current mailing lists, you'll see plenty 
of Soren's gripes about the crappiness of the SiI cards when he was 
writing support for them.  I've been tracking -current for a few years 
now, and that's how I know to avoid those cards.  Also, I haven't looked 
yet to confirm, but I'd bet that the ATA code is probably peppered with 
"choice comments" about what he needed to do to get the card to work.

Good luck... :)

>
> vlad
>

-Gary