Siig Ultra ATA 100?

[ohaya]

Hi,

I just got one of these cards, and I got it working in my system, but have a
couple of questions. Hopefully someone else here has used this...

1) When you boot with the card installed, an additional prompt (press F1 or
F11) appears. What does pressing the F1 or F11 do? It kind of looks like
F1 makes forces the card to check for drives, but I don't see what pressing
the F11 key does. I can't find any info on Siig's website or on web
searches about this.

2) The main reason that I got the card was that I was mainly interested what
difference it would make and also I am having a problem with a particular
standalone disk imaging program when using the onboard IDE controller.
After installing the card, I ran HDTach (the new beta version). Overall,
the numbers I got from HDTach were about the same, but I was puzzled because
burst read speed actually was lower with the Siig card vs. the onboard IDE
(same drive in both cases). With the onboard IDE, I'm getting ~70MB/s burst
read, and with the Siig, I was getting ~61MB/s. This was both with the
default IDE controller driver that Windows 2000 installed, and with the IDE
controller driver from the Siig diskette.

I haven't gotten around to seeing if the Siig solves my imaging program
problem because I don't know if I want to keep the card if it lowers my disk
performance .

Any thoughts or experience with this?

BTW, motherboard is a KT7E, which uses the KT133A chipset. I am using the
VIA IDE drivers with the PCI Latency patch installed (from
http://www.georgebreese.com).

Thanks,
Jim

 

[Eric Gisin]

Read burst is pretty meaningless, the OS never does reads from the drive's
cache.

If you put two drives on the two channels, and read both at once you would get
~100MB/s with current PCI IDE controllers.

 

[ohaya]

Eric,

Yes, I know that (mostly from posts here), but I'm just curious if you all
would've expected it to improve using the PCI IDE card vs. the onboard IDE
controller?

Jim

 

[Eric Gisin]

There are both on the same PCI bus in your case.

 

[Folkert Rienstra]

Eric,

Yes, I know that

Oh?

"... I don't know if I want to keep the card if it lowers my disk performance"

suggests otherwise.

(mostly from posts here), but I'm just curious if you all would've expected
it to improve using the PCI IDE card vs. the onboard IDE controller?

Oh, why is that, when both are UATA100?

Jim

(Now, let's see if you will wreck the restored quoting again on your next post).

 

[Folkert Rienstra]

There are both on the same PCI bus in your case.

PCI hasn't got anything to do with it.

 

[ohaya]

Oh?

"... I don't know if I want to keep the card if it lowers my disk performance"

suggests otherwise.


Oh, why is that, when both are UATA100?


(Now, let's see if you will wreck the restored quoting again on your next

post).


Folkert,

I'll try not to . Honestly though, no harm was intended on my part... it
wasn't even laziness .

It seems like posting preferences are NG-by-NG, and even person-by-person,
and to quote Abe Lincoln (I think), "you can't please all of the people all
of the time", and no matter how I post, seems like someone will feel
offended...


As to your question above about "why?", a couple of reasons:

1) I'd been tweaking the configuration of this new/old motherboard recently,
and I'd gotten some progress recently with improving IDE performance, after
installing George Breese's "PCI Latency" patch. Installing the Siig card
was kind of an experiment, to see if it got any better. I know that both
the onboard IDE and the Siig card are ATA 100, but I was thinking that the
Siig might work around any residual problems with the VIA chipset. More of
an experiment.

2) The Siig card info indicates that the card has a "128 byte FIFO per IDE
channel". I was thinking (probably incorrectly) that this might at least
increase read burst speed, since the disk transfers would be memory (FIFO on
the Siig card) to memory (cache on the drive). Like I said, I'm aware from
posts here that the read burst speed doesn't have much affect on overall
performance, but again, this was just curiousity.

Jim

 

[Folkert Rienstra]

performance"

Yup, you did it again.

expected

.... and again.

post).

.... and again.

Folkert,

I'll try not to . Honestly though, no harm was intended on my part...

it wasn't even laziness .

So it must be pure cluelessness then.

It seems like posting preferences are NG-by-NG,

There are some newsgroup that require you to wreck the quotings? Wow!

and even person-by-person,
and to quote Abe Lincoln (I think), "you can't please all of the people all
of the time", and no matter how I post, seems like someone will feel
offended...

.... and again, and it isn't even a quote this time.

As to your question above about "why?", a couple of reasons:

1) I'd been tweaking the configuration of this new/old motherboard recently,
and I'd gotten some progress recently with improving IDE performance, after
installing George Breese's "PCI Latency" patch. Installing the Siig card
was kind of an experiment, to see if it got any better. I know that both
the onboard IDE and the Siig card are ATA 100, but I was thinking that the
Siig might work around any residual problems with the VIA chipset. More of
an experiment.

2) The Siig card info indicates that the card has a "128 byte FIFO per IDE
channel".

With transfer sizes as big as 128 kB that obviously has little impact on the
greater scale.

I was thinking (probably incorrectly) that this might at least increase read
burst speed,

Oh, why?
And even so, which is doubtful, what is the FIFO size on the MoBo chipset controller?
And on the drive itself?

since the disk transfers would be memory (FIFO
on the Siig card) to memory (cache on the drive).

That will only make a difference if the CPU fails to keep
the FIFOs filled (or emptied, depending on direction).

As long as the FIFOs are available the transfer should be seamless.
And even when that's not the case you have to ask yourself what
penalty is incurred by that (which is probably anybodies guess).

 

[ohaya]

Folkert,

At the risk of being forever flamed by others on this newsgroup, WHAT IS
YOUR PROBLEM?

Tell me how you want me to quote/reply, and I'll do my best to accommodate
you!! I really don't understand what it is that you want. Honestly!

When you mention about "wrecking quotes", are you talking about the way that
the ">"s are being pre-pended to my replies? That is the way that this
mailer (Outlook Express) works, I think.

Seriously, this haranguing from you is getting frustrating. I'm really
trying to understand what it is that you don't like.

Jim

 

[Timothy Daniels]

Folkert,

At the risk of being forever flamed by others on this newsgroup, WHAT IS
YOUR PROBLEM?

Folkert's problem is that he and Rod Speed and a handful of other
sock puppets here are all run by the the same guy with a medical
disability. Forget about it.

*TimDaniels*

 

[Folkert Rienstra]

Folkert's problem is that he and Rod Speed and a handful of other
sock puppets here are all run by the the same guy with a medical
disability. Forget about it.

Let's see if Will D. will join you two and then you three can form the club
of the 'living braindead'.

 

[Folkert Rienstra]

Quelle dilemma. Either you are a troll and just want me to shutup
in which case I would oblige you if I did, or, you are a complete
mental case and anything I would say would fall on dead brain matter.
I'm stuck, you got me.

 

[chrisv]

Tell me how you want me to quote/reply, and I'll do my best to accommodate
you!! I really don't understand what it is that you want. Honestly!

You really don't see anything wrong with the below, which is from an
earlier post of yours? Is it quite readable?


Overall,IDE

 

[ohaya]

You really don't see anything wrong with the below, which is from an
earlier post of yours? Is it quite readable?

chrisv,

Were you referring to the line breaks (again, this question is in
earnest)? Again, I wasn't doing anything intentional to mess up any
quotes, it was just the way that Outlook Express did it.

I'm posting this using Netscape, rather than OE. Is this better?

Jim

 

[ohaya]

Quelle dilemma. Either you are a troll and just want me to shutup
in which case I would oblige you if I did, or, you are a complete
mental case and anything I would say would fall on dead brain matter.
I'm stuck, you got me.

Folkert,

I wasn't trolling (I guess that it's up to you to believe that or not),
and I certainly didn't have any intention to imply that you should
's....up' (sorry, I can't even post that, as I consider it a bit
impolite). As to whether I'm what you suggest at the end of your post
above, I hope not .

I hope that this post (again, I'm using Netscape this time) doesn't
wreck the quote above, and again, if my earlier posts did that, it was
just the way that OE worked. Sorry.

If that's ok, and if we can get back to the subject of my original post,
I guess that I'm still curious about what the "press F1 or F11" does
with the Siig card. Also, based on earlier comments in this thread
(from both yourself and others), what are the advantages (if any) of the
non-RAID PCI IDE cards like the Siig then?

Thanks,
Jim

 

[Timothy Daniels]

.....I'm still curious about what the "press F1 or F11" does
with the Siig card. Also, based on earlier comments in this thread
(from both yourself and others), what are the advantages (if any) of the
non-RAID PCI IDE cards like the Siig then?

They are usually used to enable use of ATA/100 or ATA/133
hard drives at their maximum rated speeds, such as when one has
an elderly PC (like mine) that has an ATA/33 controller on the
motherboard. Other people might have a bunch of ATAPI
devices that use up most or all of the motherboard IDE controller
channels, and they need more channels for multiple hard drives.
As for the SIIG ATA controller card, it's a dumbed-down RAID
card, and the special function key is used to get into the card's
BIOS to set its RAID parameters when it's RAID enabled.
It has no function in the controller card.

*TimDaniels*

 

[chrisv]

chrisv,

Were you referring to the line breaks (again, this question is in
earnest)?
Yes.

Again, I wasn't doing anything intentional to mess up any
quotes, it was just the way that Outlook Express did it.

Google "OE quotefix", or, better yet...

I'm posting this using Netscape, rather than OE. Is this better?

Yes.

 

[ohaya]

..
snip
..
..

Google "OE quotefix", or, better yet...


Yes.

chrisv,

Thanks for the feedback. Then I'll try to use Netscape for posting, in
this NG.

Jim

 

[ohaya]

They are usually used to enable use of ATA/100 or ATA/133
hard drives at their maximum rated speeds, such as when one has
an elderly PC (like mine) that has an ATA/33 controller on the
motherboard. Other people might have a bunch of ATAPI
devices that use up most or all of the motherboard IDE controller
channels, and they need more channels for multiple hard drives.
As for the SIIG ATA controller card, it's a dumbed-down RAID
card, and the special function key is used to get into the card's
BIOS to set its RAID parameters when it's RAID enabled.
It has no function in the controller card.

*TimDaniels*

Tim,

Thanks for that info. I just couldn't find ANY information about what
the F1 did vs. the F11.

Jim

 

[Folkert Rienstra]

this question is in earnest)?

Yes.
Express did it.

Google "OE quotefix", or, better yet...

OE-QuoteFix doesn't fix your own booboos on send
if you haven't
noticed yet.
(Linewrap set at 50 for OE, 132 for OE-QuoteFix)

OE. Is this better?

Yes.

Let's see first. It's very likely to have a linebreak
setting too.