IDE mode Compact Flash problems...

Most likely a problem with the CF. Sometimes the I/O drivers are not strong enough to drive 5V IDE interface.

Have you tried an older and smaller CF? Have you tried other brands, i.e. Lexar?

IIRC, a CF will only configure as true-IDE mode from power-up, and not if it is hot-swapped (hence the extended pins on the CF socket) - perhaps there is a problem on power-up?

Regards,

--
|              Mark McDougall                | "Electrical Engineers do it
|     |   with less resistance!"

I doubt that's the problem. We have experienced same problem even with the ATA pin ground on our adapter. It is likely a power-up problem, but not for the mode selection. If the on-board CF controller can't communicate with the PC, the PC would not recognize the card.

That's the problem - this motherboard is not recognizing the card at all.

This is on an Intel D945GTP motherboard. Out of the box, the BIOS does not see the CF card. The BIOS finds no bootable devices and no drives are found when looking at the BIOS setup screens.

If I disable "Use Automatic Mode" in the Drive Setup in the BIOS it hangs at the Intel splash screen.

I have tried SanDisk, Lexar and PNY CF cards. All work in a different Intel motherboard model (a D865GVHZ) but not in this one.

Suggestions welcome.

Regards, Paul.

What size CF? One consideration is that the CF controller runs Power-on test. It might take longer to test the bigger chips. If it does not complete the test before the PC query the controller, it will not be recognized.

On the other hand, it could be wiring problems with your adapter. You can try ours with 32M CF as boot device and mount your 256M or bigger CF. We provide full refund if it does not work for you.

See:

FWIW, I recall an issue I had when the Sandisk started supporting DMA on their compact flash cards; the custom adapters I used then did not have DMA wired, but as the flashdisk reported DMA capability, the BIOS tried to use it and hung at boot.

(The workaround for me was a to use a utility from Sandisk to disable DMA on the compact flash device, until the adapters were fixed.)

/Rolf

Thanks Rolf. In this case none of the SanDisk, PNY and Lexar flash cards support DMA. They work on one Intel motherboard model (with a CF-IDE adapter that works in some other motherboards I've tried), but not this new 945 chipset one.

All the motherboards that work are similar (3GHz P4 or Athlon64 equivalent). Only this model motherboard (so far) is failing.

Could this be a BIOS issue, or a hardware fault/feature?

Regards, Paul.

One thing I have found is that CF cards are very sensitive to the IDE cable that is used. Quite often if the cable is the standard length there will be errors. The timing of the IDE interface can be changed on the motherboard settings, default is usually Mode 4 which is the fastest, if you use Mode 2 or 3, it will run a bit slower which can help.

regards, Johnny.

Either.

Try hooking up a standard disk to verify that the IDE interface on the motherboard works at all. (Cable/Power/motherboard problems possible)

Did you play with the 5V/3.3V settings on the adapter?

If a std IDE disk works, then it could be a BIOS issue; I see Intel has a recent BIOS update, posted 4th of Aug 05; though I couldn't read the release notes.

Another BIOS thingy is that some Phoenix BIOS I came across won't talk to a slave device on it's own; it will rely on the master to query the slave at bootup, and in some situations this can be a problem.

/Rolf

Thanks Rolf.

Standard IDE disk works fine.

3.3V and 5V settings on the adapter make no difference. The CF-IDE converter plugs straight into the IDE slot (no cable). Using the latest BIOS. The adapter runs in master mode.

Not looking good :-(

Paul.

Hmm...could you try using a cable? Bus timing may be different without.

Otherwise, I'd think it's time to call Intel support, possibly try an adapter they recommend.

/Rolf

Interesting.

I've had some bad Compactflash in the past that didn't work in IDE mode due to the fact that they didn't allow master/slave selection (I am always the master! I care not for your attempt to influence me by changing the state of a pin!), but your problem sounds very different.

Just an update - I tried two other 945-based motherboards: an ASUS and a Gigabyte. They both worked fine.

So I'll guess it's the BIOS.

A shame.