It is, for me at least impossible to flash the aqua pumps via a PCI(-E) parallel port.

I have tried both a PCI and PCI-E card - both times being told that the 'pump can not be found'.

It is particularly worrying that a Aqua engineer informed me that the PCI port will work, even after alerting said engineer to a previous post highlighting exactly this problem (http://forum.aquacomputer.de/weitere-foren/english-forum/p20982-why-is-it-so-hard-to-update-the-fw-in-ac-xt-ultra/?highlight=flash ).

For me, and i suspect others, the PCI solution appears unworkable. Below i have pasted a section from the chipset manufacturer (Moschip) which seems to suggest that ECP/EPP modes cannot be guaranteed.
If true then maybe the aqua people should be honest, and not trick people into trying unworkable and possibly expensive solutions.17. Can the parallel port of MCS98XX fully support SPP/PS2/EPP/ECP mode?

We can support SPP and PS/2 Modes without any problems. EPP & ECP Mode support varies from machine to machine, and depends on the Operating System being used. Both Modes require two banks of registers to control the chip. The IEEE-1284 specification requires these be separated by an offset of +400h Bytes.

The PCI specification does not make any guarantees that two resource requests will have any specific relationship to each other. We request the two banks with the desired offset, but the system does not always honor those requests, and often returns the extended register bank at some entirely different address than what we requested. When this occurs, neither EPP nor ECP Mode will be usable.

An additional problem exists with EPP Mode. We have identified a timing problem in the chip that affects reading data from the remote device in EPP Mode. We are able to write data to the device (printer etc.) just fine under all conditions. The problem only affects reading data back from the device (scanner etc.) in EPP Mode. The symptoms result in the same Byte of data being returned for every read request. This problem does not affect ECP Mode. For this reason we recommend PS/2 or ECP Modes for applications that require bi-directional data transfers.
'

Zitat von »BetterRed«

It is, for me at least impossible to flash the aqua pumps via a PCI(-E) parallel port.

I have tried both a PCI and PCI-E card - both times being told that the 'pump can not be found'.

It is particularly worrying that a Aqua engineer informed me that the PCI port will work, even after alerting said engineer to a previous post highlighting exactly this problem (http://forum.aquacomputer.de/weitere-foren/english-forum/p20982-why-is-it-so-hard-to-update-the-fw-in-ac-xt-ultra/?highlight=flash ).

For me, and i suspect others, the PCI solution appears unworkable. Below i have pasted a section from the chipset manufacturer (Moschip) which seems to suggest that ECP/EPP modes cannot be guaranteed.
If true then maybe the aqua people should be honest, and not trick people into trying unworkable and possibly expensive solutions.17. Can the parallel port of MCS98XX fully support SPP/PS2/EPP/ECP mode?

We can support SPP and PS/2 Modes without any problems. EPP & ECP Mode support varies from machine to machine, and depends on the Operating System being used. Both Modes require two banks of registers to control the chip. The IEEE-1284 specification requires these be separated by an offset of +400h Bytes.

The PCI specification does not make any guarantees that two resource requests will have any specific relationship to each other. We request the two banks with the desired offset, but the system does not always honor those requests, and often returns the extended register bank at some entirely different address than what we requested. When this occurs, neither EPP nor ECP Mode will be usable.

An additional problem exists with EPP Mode. We have identified a timing problem in the chip that affects reading data from the remote device in EPP Mode. We are able to write data to the device (printer etc.) just fine under all conditions. The problem only affects reading data back from the device (scanner etc.) in EPP Mode. The symptoms result in the same Byte of data being returned for every read request. This problem does not affect ECP Mode. For this reason we recommend PS/2 or ECP Modes for applications that require bi-directional data transfers.
'

'...and often returns the extended register bank at some entirely different address than what we requested.'

Then why can they not detect this, and fool the parallel output into thinking that the addresses are correct?

On edit, they are saying that they cannot be sure of receiving the correct data, so it is not their fault.