- parts of the disc routines have been replaced with the original ones from the ROM, which Zozosoft published a few pages ago; now I don't have any read/write problems anymore; changes are within preparing sector read/write; I still use INI for reading a sector; if this is still making problems, please tell me; the current solution is a little bit slow, it seems, that it requires some sector interleave; I will have to optimize it...)

It is more better, no retry, retry, retry... anymore
One note: ask the 73. EXOS variable (STEP RATE), and pass to the SEEK routines.

And also not works on turbo report error 26.
Probably some time out counter too short?

@Zozo: Your investigations helped a lot! There were stupid problems with some kernel routines.

Very interesting! Tell me more!

Don't know if it solves your problem with your memory expansion?

It is don't change anything
But I tested with third card, and it is working up to 7.12MHz! (Too slow chips for 10MHz)
Very-very strange...

What the loader do from finishing loading until own LPT displayed? The frozzen happen before the new LPT displayed. (Or immediately when the clock switching do while the SymbOS running.)

The working expansions are the modern SRAM types, the problematic are the old DRAM... I will try to test more types.

The SymbOS do any R register manipulation?

Little note, I see in the code:

EXOS already set the flags with OR A, then not needed do it again in the program code.
And XOR A instead LD A,00 one bytes shorter

I wasn't sure, what channel number to use.

When the machine code (header 5) program started all channels are closed, then free from 0 to 254. The 255 reserved for the Default Channel which number is stored in EXOS variable 4.

For a simple loading the 0 are the commonly used, because the XOR A trick
And another trick for more less bytes when more EXOS calls, for example:
Because the error checked, it is sure A=0 then not need set again for the channel number.

WOW ! Very nice the background is working on the CPC !!!

Oh Oh .. of course I meant a ENTERPRISE

Cool background

Thanks!! .. you can download it here BACKGROUND

Phew, that sounds good! I am still overwhelmed by all the different turbos/enhancements you have on the EP... Hope we get all the issues solved

Boxsoft mouse support is 90% implemented
Thanks again especially to LGB and gflorez. I have to add an autodetection (exactly the same like on the MSX) now and mouse button test. You will be able to use it in LGBs' JSEP Emulator soon!

Nice, but a real hw test is very important: JSep is a "stupid" tool for fast on-line presentation of an EP software without dealing/downloading/installing/ROM searching/configuring/etc of an emulator, this is more or less the main design concept of my JSep web based Enterprise-128 emulator. Thus, it lacks many things eg good timing, especially with things like WD/EXDOS access or the boxsoft mouse interface. If you forget some delays in you code, the mouse would work in JSep but maybe not on a real hardware.

I just write this here, so people with the "real thing" should be keen enough to test if it is available