USIfAC II:Convert a PC or USB stick to Amstrad HDD,access dsk's,and many more!

[ikonsgr]

Thanks to Toto,i now have a 40007, i'll replace 40010 on my CPC464 (and a cpc6128 with old rev.a board which has both 40010&40007 sockets installed) and verify the theory that USIFAC II has problems with 40007...

[NigeNigeNige]

sounds close to the same problem im having.....

@NigeNigeNige curious...which gate array is in your system? 40007?

@ikonsgr youve never tested with a 40007 correct? it was designed around the 40010? ...does your board have the 2nd slot for 40007? maybe i could send you one to test with? (i know there shouldnt be a difference, its looking like there is)

My board has a 40010, sorry this doesn't fit the theory. It's an MC0002C board with original Zilog Z80. 

[eto]

My board has a 40010, sorry this doesn't fit the theory. It's an MC0002C board with original Zilog Z80. 

And also we have reports from people with a 40007 where it works. 

I also tried several things, even got a new 464 in case it's just my particular one. But no change. Still happening with other 464s. It looks like under certain conditions, the USIFAC does not provide the right data from ROM and the CPU cannot continue with the boot process. 

[NigeNigeNige]

Today I tried the USB board + the ROM board + my DDI-1/FD-1. 
Now I consistently get a "USIFAC II Enhanced (6f) + 512Kb Ram" on the boot screen and no freezes!
Removing the DDI-1 goes back to the old state of hanging on the boot.

The RAM seems to work and Amstrad Diagnostics reports it correctly. 128K software works. I don't have anything that needs more than that right now.

But, the switch button for rom/ram only works about once in 10 tries. The other times it resets the CPC and again shows 512Kb RAM. So I can't really use the ROM functions.

[d_kef]

I opened up my 464 and the mainboard is a MC0001A PT NO 270100. I guess it's an old one.

But the strange thing is that when I replaced the on-board ROM with an EEPROM programmed with CPC664 OS+Basic the USIfAC boots up normally with or without the RAM/ROM board!!!
 
I then programmed the same EEPROM with CPC6128 OS+Basic and again the USIfAC boots up normally.

I thought maybe it's a problem with original the ROM (timing or whatever) and I programmed again my EEPROM with CPC464 OS+Basic.
To my surprise the USIfAC freezes the same way as with the original ROM.

When I have the CPC664 or CPC6128 ROM installed I can mount and run disk images normally but when I try to run/load/save directly from/to the USB then the CPC freezes and it comes back only with a power off/on.
Also |CAT works normally but CAT result in a freeze.

d_kef

[eto]

I then programmed the same EEPROM with CPC6128 OS+Basic and again the USIfAC boots up normally.

I tried the same and programmed a 6128 firmware and Basic 1.1. 

In my case it did not change.

[poulette73]

Thanks for everything John! 👍

I received my new USIfAC II very quickly with its new DC3 connector.
It works perfectly with the new firmware 6f.


And thank you also for this second chip in revision 6f, which I successfully installed on my other USIfAC II. 🙏😉

[NigeNigeNige]

great idea with that new connector... looks much more robust.. but the extra unconnected pins... do they really need to be there (probably hard to remove them)

i decided to remove the USB module... just to see if maybe that was causing issues....

well i didnt get nothing!.. total black screen on trying to boot....

not sure if newer firmware requires the module...

ill have to borrow the pic programmer again and try some older firmware without the module connected... id be happy if i just had serial and or esp8266 connection

It would be more structurally sound if every pin was soldered, like with a socket even if you're not using all the pins. On my board only 3 pins are attached at the connector that attaches to the USB module, which looks a little fragile to me

[GUNHED]

Maybe you got some hardware demon at your place. Better some it out! 

[GUNHED]

Sometimes it helps to have a break - well, honestly the weather isn't too helpful though. Good luck with all things CPC nevertheless. 

[ikonsgr]

I just tried the 40007 with an old rev.a CPC6128 which has dual sockets for GA chips (one for 40010 and one for 40007), but it seems that chip is damaged, as i don't get any responce (of course amstrad works perfect using the 40010 on the other socket).
In anycase, all these weird problems described lately, are very similar to the problems i had with first prototype RAM/ROM boards, which only resolved when i increased thickness and width  of signal traces, especially for gnd and 5v supply (although M4 boards still seem to have problems when connected to RAM/ROM board, so i have further increase trace widths in a new revision of the RAM/ROM board)
When i designed USIfAC II board (which is rather small in size,thus signal traces on board are rather short too) i used default widths for traces (0.33mm for signals and 0.4mm for gnd/5v supply), which prooved to be ok for the vast majority of Amstrad boards.
But it seems that some Amstrad boards, for some reason, are more "signal sensitive" regarding hardware connecting to expansion port. At some point in the future, i will order a new batch of pcbs for USIfAC II, where i will improve design (e.g. wider traces for all signals, especially for gnd) and hopefully resolve such problems.

[eto]

I'm not very familiar with oscillators but to me it seems that the NAND and the hex inverter, in combination with the crystal, create the 16MHz signal for the Gate Array. 

The 4MHz is then produced by the Gate Array.

[TotO]

Yes, the 74HCU04 is famous to be using into oscillator circuits.

[d_kef]

I see "(v3)" and "BASIC 1.1". Which ROM do you use?
Can you try with the original 464 ROM if you can find some time?

d_kef

[d_kef]

I see "(v3)" and "BASIC 1.1". Which ROM do you use?
Can you try with the original 464 ROM if you can find some time?

d_kef

thats the 6128 ROM....

just tried the 464 ROM... it locks up after 2nd boot (just b4 usifac ROM line)...but pressing the reset button makes it work like usual

EDIT: back on the 6128 ROM...soldered the usb module back on (with some wires)...
but i cant change DSK with the disc swap button... and hitting the reset button causes lockup just b4 the rom line... only way to get it back is to power down the system (and do |usb again)...
also with the 6F firmware, CAT doesnt work, just locks up...(requiring power cycle to get it back).. |CAT works tho
im going to try removing the diode (it was never needed before!)

Exactly the behavior I observe with my USIfAC!!!
The diode in or out made no difference in my case.
Firmware 6d is the only one that is adequately usable with my setup.

d_kef

[d_kef]

Sorry to hear about that. It must be a big disappointment.
Have you checked your PSU? Maybe you are lucky and it's just that.

d_kef

[d_kef]

Clock circuit maybe?
Have you checked the 16Mhz input and the 4Mhz output on the GA?

d_kef

[ikonsgr]

....this is the SECOND time USIfAC has killed something on this board....

I knew that USIfAC II is a gotek killer, but i never thought it as an Amstrad killer too...   
Seriously now, the fact that you have different behavior, when you twist the edge connector ribbon cable, might be a strong indication that this is were the problem lies!
 I remember once, i also had weird erratic problems (and again it was on testing board with CPC464, it seems that 464 is more "signal sensitive" than 6128 models afterall...), and when i changed the ribbon cable (which it was rather old and heavily used for testing all the boards i make), everything worked perfect again!
Most probable, either the edge connector or the ribbon cable connections with the edge/idc connectors, issues tiny parasitic resistances on signals, that cause tiny time delays thus create the erratic problems...
 Moreover, even the speed of rom/ram chips can cause such erratic problems. For example, RAM/ROM board (even earlier prototypes) seemed to work perfect on  Amstrads with EEPROM chips, but had problems with original rom chips. The differenece ofcourse is SPEED! eeprom is MUCH faster than original roms (~twice faster responce). Same goes for ram chips too, the very old CPC6128 from 1985 equipped with slow 250ns ram chips is the only one that gave me some video noise when used with RAM/ROM board.

[ikonsgr]

FINALLY:

You cannot view this attachment.

Yes, it is what it looks like, a CPC 464 with:
- Amstrad CPC 6128 FW ROM
- BASIC 1.1 ROM
- PARADOS ROM
- 2 ROM GAMES
- 256KB Expansion RAM...

1 lower rom+ 4 upper roms + 256KB RAM...ALLTOGETHER! 

And here is with FW3.15 variation:

You cannot view this attachment.


A simple RSX command will be incorporated into ULIfAC's firmware, that will transform a CPC464 to a CPC...6320 with PARADOS, in 2seconds! 

[TotO]

What is wrong with the BASIC version?

[ikonsgr]

What is wrong with the BASIC version?

I've change it, to know if basic 1.1 rom loaded into SRAM, is the one loaded, and not the original in the rom chip 

[pappous]

So Ulifac combines 2 boards, can you give us details and maybe when can make them for us

Also will it work for Amstrad 6128?

[ikonsgr]

So Ulifac combines 2 boards, can you give us details and maybe when can make them for us
Also will it work for Amstrad 6128?

Yes, the idea is to "merge" USIfAC II and RAM/ROM board into one single board with some extra functionality:
- Dual mode will support up to 16XRoms+256kb Ram, instead of 2X roms of the ram/rom board
- Many utilities/programs that loaded externally, will be incorporated into board's ROM as simple rsx commands. For example, file manager and rom configuration utility.Also, there will be extra RSX commands,for easy loading PARADOS Rom on a CPC6128, or a "full conversion packet" of 6128FW+BASIC1.1+PARADOS/AMSDOS for CPC464.
- Due to some better signal handling in dual mode, Symbos (and perhaps other programs) now loads in dual mode too.
- Single board will allow to use a 3d case.

Of course board works on both CPC464 and CPC6128, but the main benefits will be for the CPC464.
As for availability and price, if all goes well, i will probably have first boards by early next year, for ~35-40euros each.

[GUNHED]

Have you made any tests if it works with FutureOS too?

[ikonsgr]

Have you made any tests if it works with FutureOS too?

Future OS uses a lot of roms and if i remember correctly it also uses a custom program to place them, which unfortunately didn't seem to support USIfAC. If there is a way to place roms manually (e.g. what romfile goes to which rom slot),  give me a hint and i'll try it.