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

Started by ikonsgr, 09:17, 01 December 20

Previous topic - Next topic

entelle42 and 1 Guest are viewing this topic.

GUNHED

Shouldn't that be a personal thing? Depending what you have loaded into the FlashGordeon.
http://futureos.de --> Get the revolutionary FutureOS (Update: 2022.03.09)
http://futureos.cpc-live.com/files/LambdaSpeak_RSX_by_TFM.zip --> Get the RSX-ROM for LambdaSpeak :-) (Updated: 2021.12.26)

poulette73

I didn't go into details here, but I had done all long and wide tests (deleting each roms, one by one, with tries each time to find if a rom was in conflict, etc.).

In the end, even with an "empty" FlashGordon (only rom 0 for Basic and rom 7 for CPM) the problem was still present.

It was on advices of ikonsgr that I tried to find wich possible positions of USIfAC relocation.

This is only in slot 3, perfect for cohabiting with a FlashGordon (even virgin). 👍


Since then I've reloaded about ten roms in FlashGordon (low and high roms) and everything works perfectly.  :)

eto

Interesting: How can the ROM number have an effect on the compatibility, if there is no ROM number collision?

Is there a difference if the USIFAC physically before or after the Flash Gordon in the MX4?

Fran123

Quote from: eto on 12:26, 06 July 22Interesting: How can the ROM number have an effect on the compatibility, if there is no ROM number collision?
I guess when you write a rsx command the system searches the command in the ROMs in a specific order, and when it finds out, it stops

eto

Quote from: Fran123 on 12:50, 06 July 22
Quote from: eto on 12:26, 06 July 22Interesting: How can the ROM number have an effect on the compatibility, if there is no ROM number collision?
I guess when you write a rsx command the system searches the command in the ROMs in a specific order, and when it finds out, it stops
That is true, but I understood that there is no other ROM present, the FlashGordon is just physically present. 

poulette73

Quote from: etoIs there a difference if the USIFAC physically before or after the Flash Gordon in the MX4?
On the MX4, the USIfAC is connected in the 1st position physically.
I tried plugging the FlashGordon into positions 2, 3 and 4 and the problem persisted.

But since, everything is resolved with relocation of the USIfAC in rom slot 3.


Quote from: etoThat is true, but I understood that there is no other ROM present, the FlashGordon is just physically present.
That's absolutely correct! You understood well.

The problem was permanent, even with a FlashGordon empty of any rom.

GUNHED

Oh, that's a really nasty bug. Hope that someone will find out what's going wrong and fix it one day. Guess its was long and hard work to find out what was found out.
http://futureos.de --> Get the revolutionary FutureOS (Update: 2022.03.09)
http://futureos.cpc-live.com/files/LambdaSpeak_RSX_by_TFM.zip --> Get the RSX-ROM for LambdaSpeak :-) (Updated: 2021.12.26)

poulette73

For information, I found it amusing to do loading tests according to the type of support.  :D

AMSTRAD Comparative table


USIfAC II Direct Mode is amazing !!  :o

I will be curious to also test the Wifi access when I receive the module.


[EDIT] For testing, I started the timer when I hit the ENTER key, and stopped it when the game menu was displayed.
But for Sorcery in USIfAC II Direct access mode it's not representative, because the total time includes : loading time + software decompression time.

WacKEDmaN

looks like my replica board is incompatible with the USIFAC II
ive just finished testing it on an original 464...and it worked fine... (6e..without the diode! just the resistor..and with the same power supply)

i also tested a DDI5 on the replica board and it works fine...

there is something about the USIFAC II that the replica board really doesnt like :(

is anyone using the USIFAC on a Version 1 board? (PT No 270100, MC0001A with slashed zeros?!...not the Z70100!) which my replica was based off..


Devlin

Quote from: WacKEDmaN on 04:58, 07 July 22looks like my replica board is incompatible with the USIFAC II
ive just finished testing it on an original 464...and it worked fine... (6e..without the diode! just the resistor..and with the same power supply)

i also tested a DDI5 on the replica board and it works fine...

there is something about the USIFAC II that the replica board really doesnt like :(

is anyone using the USIFAC on a Version 1 board? (PT No 270100, MC0001A with slashed zeros?!...not the Z70100!) which my replica was based off..


Version 1 yes, but mine is an OG Z70100 not the MC0001A
Amstrad fan! | CPC464 + USIfAC II | Administrator of Amstrad Discord : https://discord.gg/ksWvApv

ikonsgr

Quote from: Devlin on 10:50, 07 July 22Version 1 yes, but mine is an OG Z70100 not the MC0001A
So, does USIfAC II works with your CPC464 replica?

WacKEDmaN

Quote from: ikonsgr on 12:17, 07 July 22
Quote from: Devlin on 10:50, 07 July 22Version 1 yes, but mine is an OG Z70100 not the MC0001A
So, does USIfAC II works with your CPC464 replica?
seems Devlin is using an original 464...not a replica and the Z70100 not the 270100 (mine has a 2 not a Z!) as can been seen on the mainboard versions page..

eg
https://www.cpcwiki.eu/imgs/c/ca/CPC464_270100_Grimware_crop.jpg <the board my replica is based off
https://www.cpcwiki.eu/imgs/6/60/CPC464_Z70100_Tomdalby.jpg < Devlins board

Devlin

Quote from: WacKEDmaN on 13:06, 07 July 22
Quote from: ikonsgr on 12:17, 07 July 22
Quote from: Devlin on 10:50, 07 July 22Version 1 yes, but mine is an OG Z70100 not the MC0001A
So, does USIfAC II works with your CPC464 replica?
seems Devlin is using an original 464...not a replica and the Z70100 not the 270100 (mine has a 2 not a Z!) as can been seen on the mainboard versions page..

eg
https://www.cpcwiki.eu/imgs/c/ca/CPC464_270100_Grimware_crop.jpg <the board my replica is based off
https://www.cpcwiki.eu/imgs/6/60/CPC464_Z70100_Tomdalby.jpg < Devlins board
I am indeed using an original CPC464 - not a replica.
Yep, that's the one - although I believe mine is even slightly different to that, as mine has no components with that white tubing on it, and the heat-sink (now replaced) is a full square of metal with the wee tabs and no cut-outs on it.
It might be worth prodding Bryce or someone in the know as to why it might be occurring.
Amstrad fan! | CPC464 + USIfAC II | Administrator of Amstrad Discord : https://discord.gg/ksWvApv

WacKEDmaN

Quote from: Devlin on 13:20, 07 July 22Yep, that's the one - although I believe mine is even slightly different to that, as mine has no components with that white tubing on it, and the heat-sink (now replaced) is a full square of metal with the wee tabs and no cut-outs on it.
It might be worth prodding Bryce or someone in the know as to why it might be occurring.
thats interesting it doesnt have the 2x 12ohm resistors in the sleeves (altho my replica doesnt have the sleeves either, just the 2 1/2W 12ohm resistors on the board like all the other 1/4W resistors)...they are used to power VDD2 on the gate array..so R138 and R139 has to be there somewhere!

back on topic... if anyone has any ideas what might be left to try im all ears!
(considering ive tried cleaning the edge connector multiple times, tried different voltages, tried another edge connector to M4(?) ribbon cable, tested it working on another machine, and other expansion cards work fine)

eto

I guess it's save to say, that the USIFAC has issues with some 464s. It's totally unclear though, if there is a common factor behind it or if there are several issued that can cause this. I doubt it's just bad luck as I had 2 different 464s with exactly the same behaviour.

If there is any way to help with debugging this, I would be happy to do so. 





WacKEDmaN

#915
Quote from: eto on 14:30, 07 July 22I guess it's save to say, that the USIFAC has issues with some 464s. It's totally unclear though, if there is a common factor behind it or if there are several issued that can cause this. I doubt it's just bad luck as I had 2 different 464s with exactly the same behaviour.

If there is any way to help with debugging this, I would be happy to do so.





im not sure where to go from here to debug it..
this usifac is tested working on 3 machines now..it was tested by the supplier on both 464 and 6128 and ive also tested it on a friends 464...worked fine on all 3...

i guess i should add another observation...
when i use the 1.1 ROM (patched to show 64k on boot) with the usifac it will actually boot...BUT..the screen gets all screwed up (like the CRTC timings are being messed with).. it gets to Ready and cursor, but no ROM string, and trying to use the RSX commands, it will either just lock up or throw a few random pixels on the screen and then lockup!

basic 1.0.. (always a few red pixels under the o in electronics)
You cannot view this attachment.

Basic 1.1
You cannot view this attachment.

eto

Quote from: WacKEDmaN on 15:56, 07 July 22im not sure where to go from here to debug it..
that was more towards iconsgr ;-) he might have an idea what to test next. And since I have a non-working CPC/Usifac combination, I would be happy to help with some tests.

Quote from: WacKEDmaN on 15:56, 07 July 22when i use the 1.1 ROM (patched to show 64k on boot) with the usifac it will actually boot...BUT..the screen gets all screwed up (like the CRTC timings are being messed with).. it gets to Ready and cursor, but no ROM string, and trying to use the RSX commands, it will either just lock up or throw a few random pixels on the screen and then lockup!
If I am not mistaken, the USIFAC uses the firmware contents to check if it's a 464. Not sure if it's initialized correctly with the 1.1 firmware. 

ikonsgr

Well,i believe that the root of the problem might has to do with the handling of ready signal from USIfAC.
When USIfAC is activated, a PIC's pin (the output of the master Configurable Logic Cell) goes to gnd, and through a diode to READY signal. This suppress the READY pulse from Gate Array chip, and practically freeze Amstrad, exactly like when you press the pause switch. This method seemed to work on all Amstrad CPC's i've tried (~10 different Amstrad's both CPC 6128 and CPC464), but for some reason it might not work on specific Amstrads.
One thing i don't know for sure, is if USIfAC works ok with older 40007 GA chip, for example my CPC 464 has the newer 40010 chip installed, and a couple of other CPC464 i've tried i don't know what GA chip had.
So, first, anyone with this kind of wierd erratic problems on CPC464, can check what GA chip is installed.
Then,you might experiment with 10k pull-up or pull-down resistors on either READY signal (pins in green circle), or master CLC output (pins in Red circle).
For a "deeper" analysis, if someone has an oscilloscope can check Amstrad's READY signal with USIfac on board.

Devlin

Quote from: ikonsgr on 12:11, 08 July 22Well,i believe that the root of the problem might has to do with the handling of ready signal from USIfAC.
When USIfAC is activated, a PIC's pin (the output of the master Configurable Logic Cell) goes to gnd, and through a diode to READY signal. This suppress the READY pulse from Gate Array chip, and practically freeze Amstrad, exactly like when you press the pause switch. This method seemed to work on all Amstrad CPC's i've tried (~10 different Amstrad's both CPC 6128 and CPC464), but for some reason it might not work on specific Amstrads.
One thing i don't know for sure, is if USIfAC works ok with older 40007 GA chip, for example my CPC 464 has the newer 40010 chip installed, and a couple of other CPC464 i've tried i don't know what GA chip had.
So, first, anyone with this kind of wierd erratic problems on CPC464, can check what GA chip is installed.
Then,you might experiment with 10k pull-up or pull-down resistors on either READY signal (pins in green circle), or master CLC output (pins in Red circle).
For a "deeper" analysis, if someone has an oscilloscope can check Amstrad's READY signal with USIfac on board.
I have a 40007 in my 464 and have no issues with 6d fw
Amstrad fan! | CPC464 + USIfAC II | Administrator of Amstrad Discord : https://discord.gg/ksWvApv

eto

Quote from: ikonsgr on 12:11, 08 July 22So, first, anyone with this kind of wierd erratic problems on CPC464, can check what GA chip is installed.
Then,you might experiment with 10k pull-up or pull-down resistors on either READY signal (pins in green circle), or master CLC output (pins in Red circle).
will do on Monday. 

does the big red circle with the diode have a meaning for this task?

ikonsgr

Quote from: eto on 21:17, 08 July 22does the big red circle with the diode have a meaning for this task?

 Nope,i just used a previous photo of board's back :-)

WacKEDmaN

i have a 40007 :picard:

ill have a look on the sillyscope and play with the resistor on READY pin later today...

Edit: havent had a chance to scope it yet.... ill get to it asap

WacKEDmaN

well ...this is strange AF!  :picard:

..ive been running a soak test on Noels diag with a DDI5 (512Kb) for the last 3 days straight with zero errors.. so the expansion works fine, 128kb games run great...

so i setup the USB oscilloscope, hooked up the USIFAC.. and it did somethin it hasnt done before, it made noise speakers (the stereo connector).. so i switched power off and restarted..it did it a few times.. but then.. the ROM string started to show!..
after a few more power cycles with the reset (some faster than others!).. it eventually booted! and the RSX commands worked fine!

ok then!.. so i plugged in usb... and it crashed...
i tryed switching of and powering back up..multiple times.. and the same old bootloops or corruption...

so played around with power supply... slowly increasing the voltage until i seen some change in the pattern...
well i eventually got up to 6.2v!.. and the USIFAC booted first time... i enabled USB..but it crashed with corruption again...

..i increased the voltage to 6.5v (yes SIX POINT FIVE VOLTS!).. and it accessed the files on the USB (with |CAT)

soooo... it looks like im good to go....

but... whyyy is it requiring 6.5v on my board when it accepted 5.2v (with the same psu) on the other CPC464 i tested it working perfectly on?! .... im really stumped!

..im hoping 6.5v isnt going to hurt any other the other chips too much...they chips are quite very warm..(anyone know upper voltage limits?!)

btw.. im seeing 6.4v at legs of barrel jack on the board.. and 6.3v at IC115.. the serial header on the USFAC is measuring 6.2v

TL;DR: needed more voltage on my board....

EDIT: spoke to soon... was accessing the usb... did a |464, it rebooted and showed parados (that i had selected)..
but when i tried to access the usb it crash...
..powered off and restarted...and back to boot loops...

....I REALLY DONT GET THIS....

EDIT2: WTF bumped voltage to 6.8 volt and now its working again... i REALLY dont wanna go any higher...(we'll see how long this lasts...)

eto

I have just realised that there is now a diode but no longer the pull-up resistor. Should I remove the resistor and add the diode? Or does it depend on the firmware, which modification is required? The comments on the previous page are a bit confusing, and I would prefer to have the right modification applied before doing further tests. 

Maybe a suggestion: hardware modifications could be documented in the first post, like the firmware updates. 

WacKEDmaN

Quote from: eto on 13:09, 13 July 22I have just realised that there is now a diode but no longer the pull-up resistor. Should I remove the resistor and add the diode? Or does it depend on the firmware, which modification is required? The comments on the previous page are a bit confusing, and I would prefer to have the right modification applied before doing further tests.

Maybe a suggestion: hardware modifications could be documented in the first post, like the firmware updates.
ive just gone back to the resistor..no diode... seems to work ok...
..been running the USIFAC since i posted last (about an hour)...and havent had an issue..except for the excess heat! (at 6.8v!)

Powered by SMFPacks Menu Editor Mod