The oddball Rev-E11 board (all models)

[dan_n]

I've noticed a couple of unusual things on the Rev-E11 boards (all models).

The first is a factory installed "bodge" wire on the back connecting a resistor and inductor behind the network chip. Rev-E12 boards have a trace there on the pcb so this was clearly a factory fix.

The second is the sram jumpers (0-ohm resistors) are set to 256kB for all models, even a Pro II. I've only seen a few E11 board pictures but they all seem to be set this way. I wonder why HAI did that because all model Rev-E12 boards have the sram jumpers set for 512kB.

Does anyone remember anything else odd about these boards?

 

[dwalt]

I've noticed a couple of unusual things on the Rev-E11 boards (all models).

The first is a factory installed "bodge" wire on the back connecting a resistor and inductor behind the network chip. Rev-E12 boards have a trace there on the pcb so this was clearly a factory fix.

The second is the sram jumpers (0-ohm resistors) are set to 256kB for all models, even a Pro II. I've only seen a few E11 board pictures but they all seem to be set this way. I wonder why HAI did that because all model Rev-E12 boards have the sram jumpers set for 512kB.

Does anyone remember anything else odd about these boards?

Does that imply the E11 boards will have any performance issues with less SRAM?

I have a 2E and an LTe both late production boards. They have the little adapter board for the eeprom. Looks like they quit bothering to fill out the type number, rev number and serial number. The number that goes below the sockets in the middle of the board is 00-00-01 rev. B

 

[dan_n]

The label on the E11 board shows Rev E11. Later board labels only show Rev E with the rev number hand written in marker on the pcb corner as you mentioned.

I'm also wondering if the E11 Pro II had any setup/programming limitations compared to the later boards. It might make firmware updates a few seconds slower since it could only buffer half as much data, but that's no big deal. What I learned from my reverse engineering (see engineering thread) is the firmware uses a task scheduler that probably maps a 4kB page of sram for each task (stack, etc.) and quickly switches tasks by changing the sram mapping. In theory an E11 board could only have half as many tasks as an E12, but only if HAI actually used the entire memory space.

What I really want to know is if the sram has to be set for 256kB on these boards. I've changed one to use all 512kB and didn't see any problem but I wasn't pushing it hard either.