Which IDE controller card?

Piyono

Piyono

0
Joined
Jan 25, 2002
Messages
599
Location
Toronto
I want to purchase a non-RAID PCI ATA133 card to increase the storage potential of my computer. All the cards I've looked at share very similar specs, and many use the same chipsets.
Is it reasonable to expect similar performance from any one of them or should I be particular as to which on I get?


Piyono
 
I have problems with my promise Ultra100 TX2 on a Tyan Tiger MPX system. But it seems to work fine with all other motherboards/chipsets i've had. I can't notice any performance difference between the Ultra100 and the onboard SiS 735 IDE controller. The Ultra100 is clearly faster than the VIA 686B based IDE controller.
 
zx said:
I have problems with my promise Ultra100 TX2 on a Tyan Tiger MPX system.
Click to expand...

What kind of problems?
The Promise Ultra133 TX2 is on my list.
Also up for consideration is this card from StarTech, which uses a SiliconImage chipset.


Piyono
 
I am currently using the Promise Ultra133 TX2 along with a Promise FastTrak TX2000 in my secondary computer. Both devices work flawlessly. I also bought 2x Ultra66 off eBay for a good price the other day; tested them briefly without any problems before they will be pressed into service on old PC's with 32 GB BIOS limits.
 

I use an Acard brand 2-channel ATA-133 host adaptor. It supports CD-R/W and DVD-R/W just fine as well as big ATA-133 hard drives. The Acard 6280 was the first ATA-133 card available. Acard makes their own chipsets for the ATA-133 card.

They are also the ones that created those handy little SCSI <---> ATA convertors (bridges) a few years back. They also produce PCMCIA chipsets. They have been a chipset designer for several years, but have slowly gotten into the PCI card business during the past 3 or 4 years.

 
I have two Maxtor D740X drives hooked on the Ultra100 TX2. The problems are high CPU utilisation and sound problems when there is disk activity. It seems to happen only with writes, like when I save a file or defrag. The computer becomes much less responsive (CPU wise).

There seem to be an issue with some IDE/SCSI/IDE RAID/SCSI RAID cards and some video capture cards with the MPX chipset. I have not found a solution, other than avoid those cards that have problems with the MPX like promise cards or to avoid the MPX.

Also, keep in mind that I have an SB Live sound card, which is not recommended for the any dual athlon system (or any system ? :lol: ).

This week end, I will test without the SB Live and I will tweak PCI latency to see if it solves the problem.

Keep in mind that I used this promise card in an Abit KT7 and a ECS K7S5A without any problems. So if you have a regular mobo, there should not be any problems.
 
I have a Promise Ultra 100. The last two versions of their drivers suck - slurring when playing audio. Each channel has a single drive on it, a WD 1200JB and a WD 800JB. Even playing a small MP3 will slur. In WMP and in the ATI file player.

Granted, W2k SP3 (which is what I run) and WXP SP1 have that "write-cache turned off" problem, so write throughput sucks (because the Promise is "seen" as a SCSI controller). But why do reads slur?

I got hold of the hot-fix for the write-cache issue, but guess what? It doesn't work because the hotfix says it can't read the current status of the drives. Something about driver implementation. BAH!

The only thing Promise has going for them is that the Ultra 100 works in Dos without drivers.

Hey .Nut, how about the Acard? Does it need drivers in DOS? Does it have the slurring problem? Have you tried applying MS's hotfix to the write problem?
 
Ok, so I bought the Promise Ultra133 TX2.
Anything I should know? Tweaks? Tips? Warnings?


Piyono
 
- Unless you have >1 drive on the same cable (bad idea), set the HD jumpers to Single mode if they have such a setting (WD drives) or the card will have problems detecting it

- Blue end of the cable must be attached to the card and black end to the drive or the card BIOS will not select the right ATA/UDMA transfer mode
 
e_dawg said:
- Blue end of the cable must be attached to the card and black end to the drive or the card BIOS will not select the right ATA/UDMA transfer mode
Click to expand...

Interesting. I've often wondered about that...
What's the difference, electrically, between the two ends of the cables? Are the conductors not wired straight through, pin for pin (ignoring the fact that the connector has 40 pins while the cable has 80 conductors)?

Anyway, it went in without a hitch and seems to be running fine so far. I may or may not run some benchmarks later on -- for what it's worth, the drive 'feels' just fine.


Piyono
 
Piyono said:
e_dawg said:
- Blue end of the cable must be attached to the card and black end to the drive or the card BIOS will not select the right ATA/UDMA transfer mode
Click to expand...

Interesting. I've often wondered about that...
What's the difference, electrically, between the two ends of the cables? Are the conductors not wired straight through, pin for pin (ignoring the fact that the connector has 40 pins while the cable has 80 conductors)?

Anyway, it went in without a hitch and seems to be running fine so far. I may or may not run some benchmarks later on -- for what it's worth, the drive 'feels' just fine.


Piyono
Click to expand...

That rule only seems to apply to cables with 3 connectors on them.. the 2 connector cables don't seem to be affected. I don't know how the controller can tell the cable is backwards.

I assume the problem with using a cable backwards is that the distance between drive 1 and the controller is alot shorter than the distance between the controller and drive 2. Electronically this could result in problems in data integrity if the system wasn't specifically designed for that kind of operation. I would think that this issue wouldn't happen with SCSI.
 
IIRC the cable is 80-wire from the mobo connector to the 1st drive connector only. I could be VERY wrong... Please correct if that is wrong.

Cheers,

Jan
 
If someone could sacrifice a cable in the name of science, he could break open the three connectors (something I do inadvertently, sometimes when using too much extraction force) and inspect the little insulation-piercing teeth. Perhaps the blue connector terminates differently than the other two.

Just a guess


Piyono
 
I thought that the coding was because all of the 80 pin cables were set up for 'cable select' of master & slave so orientation becomes an issue as one of the pins is not connected on the slave position.

Jeff
 
JMP said:
I thought that the coding was because all of the 80 pin cables were set up for 'cable select' of master & slave so orientation becomes an issue as one of the pins is not connected on the slave position.

Jeff
Click to expand...

That's interesting, but the warnings in the Promise instruction book don't qualify their reasons. They just say "do it this way, period."

Piyono
 
blakerwry said:
Jan Kivar said:
IIRC the cable is 80-wire from the mobo connector to the 1st drive connector only. I could be VERY wrong... Please correct if that is wrong.

Cheers,

Jan
Click to expand...

The cables are 80 conductor all the way (or atleast the dozen or so I have are).
Click to expand...

That did come out wrong. I meant that the ground wires are connected only up to the first drive connector. The cable is 80-wire all the way. Still, I could be wrong...

Jan
 
Jan Kivar said:
blakerwry said:
Jan Kivar said:
IIRC the cable is 80-wire from the mobo connector to the 1st drive connector only. I could be VERY wrong... Please correct if that is wrong.

Cheers,

Jan
Click to expand...

The cables are 80 conductor all the way (or atleast the dozen or so I have are).
Click to expand...

That did come out wrong. I meant that the ground wires are connected only up to the first drive connector. The cable is 80-wire all the way. Still, I could be wrong...


That wouldn't make much sense, because if that were the case, then you could use 40 wire, 2 connector cables at speeds above UDMA 2 (which, of course, you can't)

There is an easy way for (i believe) both the drive and the controller to tell what kind of cable it's attached to, and to tell if the cable is inserted incorrectly. I believe it has to do with a simple wiring difference from the 40 wire cables, but I could be wrong.
Jan
Click to expand...
 
blakerwry said:
Jan Kivar said:
blakerwry said:
Jan Kivar said:
IIRC the cable is 80-wire from the mobo connector to the 1st drive connector only. I could be VERY wrong... Please correct if that is wrong.

Cheers,

Jan
Click to expand...

The cables are 80 conductor all the way (or atleast the dozen or so I have are).
Click to expand...

That did come out wrong. I meant that the ground wires are connected only up to the first drive connector. The cable is 80-wire all the way. Still, I could be wrong...


That wouldn't make much sense, because if that were the case, then you could use 40 wire, 2 connector cables at speeds above UDMA 2 (which, of course, you can't)

There is an easy way for (i believe) both the drive and the controller to tell what kind of cable it's attached to, and to tell if the cable is inserted incorrectly. I believe it has to do with a simple wiring difference from the 40 wire cables, but I could be wrong.
Jan
Click to expand...
Click to expand...
 
I'm having some problems with 2003 Server Enterprise. Several times now I enter my username/pwd and the desktop comes up, but it's not responsive. CTRL+ALT+DEL brings up the security window, and CTRL+SHFT+ESC brings up the task manager. But I can't kill the explorer.exe process. ANyway if I leave it alone for long enough (several minutes), it comes back with delayed write fail on pretty much all drives present.

The only thing I have recently done is changed a Promiseless ATA/100 PCI card for an ATA/133 card. 2003 seemed to recognise and install drivers for it, but I'm not sure if that isn't the problem. Any suggestions anyone?

Also the following errors in System Log:

The device, \Device\Scsi\ultra1, did not respond within the timeout period.

and

An error was detected on device \Device\Harddisk5 during a paging operation.

and

AMLI: ACPI BIOS is attempting to read from an illegal IO port address (0x71), which lies in the 0x70 - 0x71 protected address range. This could lead to system instability. Please contact your system vendor for technical assistance.
 
You must log in or register to reply here.
Back
Top