Open Source 68K Accelerator Project(s)

Troubles with your machine? Just want to speak about the latest improvements? This is the place!

Moderators: Mug UK, Zorro 2, Greenious, spiny, Moderator Team

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Open Source 68K Accelerator Project(s)

Postby terriblefire » Sat Nov 12, 2016 10:22 am

Hi Folks,

Apologies if this is not relevant but it was suggested to me that people on this forum may be interested in my 68K accelerator project and the video series I have about it. Although I am not focusing on the Atari ST i do have a few STs kicking around and I will be trying to bring the board up for the STFM soon. My channel is at

https://www.youtube.com/channel/UCu4uiU ... rribleFire

And all the code and sources are being published on GitHUB, Its all prototype at the moment but it is working on the A500 already. I have plans to make more complex boards in future as things progress. All the code is available at...

https://github.com/terriblefire/tf520

Currently I have a 68020 only board. I have plans to build a 68030 + FPU board in the near future. I will say though that I make no promises about anything except that I will keep everything open source and GPLd.

Regards
Stephen

User avatar
exxos
Hardware Guru
Hardware Guru
Posts: 4933
Joined: Fri Mar 28, 2003 8:36 pm
Location: England
Contact:

Re: Open Source 68K Accelerator Project(s)

Postby exxos » Sat Nov 12, 2016 10:33 am

Interesting stuff! Will check out your channel over the weekend.
4MB STFM 1.44 FD- VELOCE+ 020 STE - Falcon 030 CT60 - Atari 2600 - Atari 7800 - Gigafile - SD Floppy Emulator - PeST - various clutter

http://www.exxoshost.co.uk/atari/ All my hardware guides - mods - games - STOS
http://www.exxoshost.co.uk/atari/last/storenew/ - All my hardware mods for sale - Please help support by making a purchase.
http://ataristeven.exxoshost.co.uk/Steem.htm Latest Steem Emulator

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Re: Open Source 68K Accelerator Project(s)

Postby terriblefire » Sat Nov 12, 2016 11:09 am

I've heard that there are adaptor boards to adapt the STe PLCC CPU socket out to the big DIL sockets... Are these readily available?

User avatar
exxos
Hardware Guru
Hardware Guru
Posts: 4933
Joined: Fri Mar 28, 2003 8:36 pm
Location: England
Contact:

Re: Open Source 68K Accelerator Project(s)

Postby exxos » Sat Nov 12, 2016 11:55 am

terriblefire wrote:I've heard that there are adaptor boards to adapt the STe PLCC CPU socket out to the big DIL sockets... Are these readily available?


I'm not aware of anyone whos done that adaption no. Mostly dip to plcc are for the stfm like this one http://www.exxoshost.co.uk/atari/last/V22BOOSTER/index.htm
4MB STFM 1.44 FD- VELOCE+ 020 STE - Falcon 030 CT60 - Atari 2600 - Atari 7800 - Gigafile - SD Floppy Emulator - PeST - various clutter

http://www.exxoshost.co.uk/atari/ All my hardware guides - mods - games - STOS
http://www.exxoshost.co.uk/atari/last/storenew/ - All my hardware mods for sale - Please help support by making a purchase.
http://ataristeven.exxoshost.co.uk/Steem.htm Latest Steem Emulator

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Re: Open Source 68K Accelerator Project(s)

Postby terriblefire » Sat Nov 12, 2016 12:10 pm

exxos wrote:
terriblefire wrote:I've heard that there are adaptor boards to adapt the STe PLCC CPU socket out to the big DIL sockets... Are these readily available?


I'm not aware of anyone whos done that adaption no. Mostly dip to plcc are for the stfm like this one http://www.exxoshost.co.uk/atari/last/V22BOOSTER/index.htm


Thanks. No worries. I've got a reasonable knowledge of the ST machines from my FPGA clone days (I had an Atari ST FPGA clone build in the Digilent Atlys which was abandoned and the best bits incorporated into the MiST project) and a few contributions to the EmuTOS project which have been re-written now.

The idea with my board is to keep it open and cheap. I dont have plans to sell them myself simply ensure there is always the means to do it cheaply.

User avatar
exxos
Hardware Guru
Hardware Guru
Posts: 4933
Joined: Fri Mar 28, 2003 8:36 pm
Location: England
Contact:

Re: Open Source 68K Accelerator Project(s)

Postby exxos » Sun Nov 13, 2016 12:28 pm

Sounds cool :) so have you done a FPGA 68k then ? I think while the 68k maxes out at 32mhz, if a FPGA could bash out higher speeds then that would make a really interesting project.
4MB STFM 1.44 FD- VELOCE+ 020 STE - Falcon 030 CT60 - Atari 2600 - Atari 7800 - Gigafile - SD Floppy Emulator - PeST - various clutter

http://www.exxoshost.co.uk/atari/ All my hardware guides - mods - games - STOS
http://www.exxoshost.co.uk/atari/last/storenew/ - All my hardware mods for sale - Please help support by making a purchase.
http://ataristeven.exxoshost.co.uk/Steem.htm Latest Steem Emulator

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Re: Open Source 68K Accelerator Project(s)

Postby terriblefire » Sun Nov 13, 2016 9:12 pm

exxos wrote:Sounds cool :) so have you done a FPGA 68k then ? I think while the 68k maxes out at 32mhz, if a FPGA could bash out higher speeds then that would make a really interesting project.


I've patched up some to work correctly. I then moved on to do an Archimedes core and decided I wanted to get dirty building hardware again.

I've just hacksawed the 68000 off an old Atari STFM i found in my garage and plugged in my board running off the internal 8Mhz clock. Its sitting looping over some address range. Does anyone know if the original ST ROM (1.01 / 1.02) has compatibility issues with the 68020? If i jam the bus error pins on i see bombs so i know something can compute.

Cheers
Stephen

User avatar
exxos
Hardware Guru
Hardware Guru
Posts: 4933
Joined: Fri Mar 28, 2003 8:36 pm
Location: England
Contact:

Re: Open Source 68K Accelerator Project(s)

Postby exxos » Sun Nov 13, 2016 9:18 pm

terriblefire wrote:I've just hacksawed the 68000 off an old Atari STFM i found in my garage and plugged in my board running off the internal 8Mhz clock. Its sitting looping over some address range. Does anyone know if the original ST ROM (1.01 / 1.02) has compatibility issues with the 68020? If i jam the bus error pins on i see bombs so i know something can compute.

Cheers
Stephen


I'm not 100% sure, but I think you need TOS206 to use the 020 CPU.
4MB STFM 1.44 FD- VELOCE+ 020 STE - Falcon 030 CT60 - Atari 2600 - Atari 7800 - Gigafile - SD Floppy Emulator - PeST - various clutter

http://www.exxoshost.co.uk/atari/ All my hardware guides - mods - games - STOS
http://www.exxoshost.co.uk/atari/last/storenew/ - All my hardware mods for sale - Please help support by making a purchase.
http://ataristeven.exxoshost.co.uk/Steem.htm Latest Steem Emulator

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Re: Open Source 68K Accelerator Project(s)

Postby terriblefire » Sun Nov 13, 2016 9:28 pm

exxos wrote:
terriblefire wrote:I've just hacksawed the 68000 off an old Atari STFM i found in my garage and plugged in my board running off the internal 8Mhz clock. Its sitting looping over some address range. Does anyone know if the original ST ROM (1.01 / 1.02) has compatibility issues with the 68020? If i jam the bus error pins on i see bombs so i know something can compute.

Cheers
Stephen


I'm not 100% sure, but I think you need TOS206 to use the 020 CPU.


Thanks.. Thats going to be a pain on this ancient machine. Perhaps EmuTOS is my friend.

rpineau
Captain Atari
Captain Atari
Posts: 489
Joined: Wed Jun 29, 2011 6:39 am
Location: California / USA
Contact:

Re: Open Source 68K Accelerator Project(s)

Postby rpineau » Mon Nov 14, 2016 2:14 am

Yes you need a TOS 2.06 for the 68020 to work on a STE (see our 68020 card in a STE there : viewtopic.php?f=15&t=28277&p=303612#p300844).
We have both 68020 card working on a STE with TOS 2.06 and a MegaSTE with TOS 2.06.
There are "patched" version of previous TOS but yea, TOS 2.06 or EmuTOS at this point is the safest bet.
Regards, Rodolphe
Falcon + AB040 + Eclipse PCI + ATI Rage VGA card + NE2000 Ethernec + HxC Floppy Emulator
MegaSTE 4MB + CosmosEx / 1040 STF for hardware dev
http://www.rti-zone.org/atari.php

joska
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 3383
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: Open Source 68K Accelerator Project(s)

Postby joska » Mon Nov 14, 2016 7:11 am

terriblefire wrote:
exxos wrote:I'm not 100% sure, but I think you need TOS206 to use the 020 CPU.


Thanks.. Thats going to be a pain on this ancient machine. Perhaps EmuTOS is my friend.


exxos is right, TOS 1.x does not support the 68010+ stackframes. 192Kb EmuTOS is a direct replacement for TOS 1.x, although it's not much work to add an address decoder for the TOS 2 address range. I would say that is mandatory for any accelerator now :)
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

BlankVector
Captain Atari
Captain Atari
Posts: 385
Joined: Wed Oct 24, 2007 7:52 pm
Location: Paris, France
Contact:

Re: Open Source 68K Accelerator Project(s)

Postby BlankVector » Mon Nov 14, 2016 12:37 pm

terriblefire wrote:Perhaps EmuTOS is my friend.

Indeed,
- EmuTOS supports any CPU
- it automatically detects optional hardware
- it can be compiled to run at any address
- it is easily modifiable and compilable, traces can easily be added

So yes, EmuTOS is your friend to debug hardware issues, and more :)

User avatar
Cyprian
Atari God
Atari God
Posts: 1317
Joined: Fri Oct 04, 2002 11:23 am
Location: Warsaw, Poland

Re: Open Source 68K Accelerator Project(s)

Postby Cyprian » Mon Nov 14, 2016 1:50 pm

terriblefire great project. I keep finger crossed.


joska wrote:exxos is right, TOS 1.x does not support the 68010+ stackframes.

TOS 1.06 and 1.62 does support 68010
http://www.atari.org.pl/forum/viewtopic ... 40#p224340
Jaugar / TT030 / Mega STe / 800 XL / 1040 STe / Falcon030 / 65 XE / 520 STm / SM124 / SC1435
SDrive / PAK68/3 / CosmosEx / SatanDisk / UltraSatan / USB Floppy Drive Emulator / Eiffel / SIO2PC / Crazy Dots / PAM Net
Hatari / Aranym / Steem / Saint
http://260ste.appspot.com/

joska
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 3383
Joined: Tue Oct 30, 2007 2:55 pm
Location: Florø, Norway
Contact:

Re: Open Source 68K Accelerator Project(s)

Postby joska » Mon Nov 14, 2016 4:04 pm

Maybe, but not on an STFM :)
Jo Even

Firebee - Falcon060 - Milan060 - Falcon040 - MIST - Mega ST - STM - STE - Amiga 600 - Sharp MZ700 - MSX - Amstrad CPC - C64

matej
Atari freak
Atari freak
Posts: 56
Joined: Wed Jul 09, 2014 12:21 pm
Location: European Union

Re: Open Source 68K Accelerator Project(s)

Postby matej » Mon Nov 14, 2016 4:24 pm

Will those work on 1040STF too?

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Re: Open Source 68K Accelerator Project(s)

Postby terriblefire » Sat Nov 26, 2016 9:34 am

matej wrote:Will those work on 1040STF too?


I managed to get hold of a T_Board. Just need to set it up to run without the internal roms

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Re: Open Source 68K Accelerator Project(s)

Postby terriblefire » Sun Nov 27, 2016 3:14 pm

Latest update.. not working with TOS 2.06 :/

https://www.youtube.com/watch?v=xrdqksZ ... ture=share

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Re: Open Source 68K Accelerator Project(s)

Postby terriblefire » Sun Nov 27, 2016 11:19 pm

terriblefire wrote:Latest update.. not working with TOS 2.06 :/

https://www.youtube.com/watch?v=xrdqksZ ... ture=share


My feeling is that the write to disable interrupts from the MFP is failing.

PS. Atari-Forum is very slow for me lately

czietz
Captain Atari
Captain Atari
Posts: 334
Joined: Tue May 24, 2016 6:47 pm

Re: Open Source 68K Accelerator Project(s)

Postby czietz » Mon Nov 28, 2016 5:08 pm

Afair, the MFP under normal operation shouldn't cause a level 6 auto-vector interrupt but it should provide its own vector number according to the interrupt source. Maybe this interrupt acknowledge cycle (during which the MFP transmits the vector to the CPU) isn't working properly? (Also, unfortunately the decoded addresses in the video are very hard to see, so I can't tell to which part in the boot process you actually get before it hangs.)

czietz
Captain Atari
Captain Atari
Posts: 334
Joined: Tue May 24, 2016 6:47 pm

Re: Open Source 68K Accelerator Project(s)

Postby czietz » Mon Nov 28, 2016 5:42 pm

In such an interrupt acknowledge cycle the MFP puts the vector number on the data bus and asserts DTACK. On the 68020 your CPLD has to assert DSACK0 and/or DSACK1 instead of AVEC in this case. From looking at your Verilog source code I don't see this happening.

rpineau
Captain Atari
Captain Atari
Posts: 489
Joined: Wed Jun 29, 2011 6:39 am
Location: California / USA
Contact:

Re: Open Source 68K Accelerator Project(s)

Postby rpineau » Mon Nov 28, 2016 6:07 pm

On a ST, only the ACIA should trigger AVEC.
Here is our VHDL code from our working 68020 card :
AVEC <= not FC2 or not FC1 or not FC0 or not A19 or not A18 or not A17 or not A16 or VPA;

In the case of the MFP, it triggers a level 6 interrupt and wait for IACK from the GLUE (which is asserted when the GLUE detect an interrupt ack cycle from the CPU), it then put the interrupt vector on D0-D7 and assert DTACK. This ends up on the 68020 D24-D31 so you should assert DSACK1 in that case (16 bit transfer and the CPU know to get the 8 high bits it needs for the vector).

Regards, Rodolphe
Falcon + AB040 + Eclipse PCI + ATI Rage VGA card + NE2000 Ethernec + HxC Floppy Emulator
MegaSTE 4MB + CosmosEx / 1040 STF for hardware dev
http://www.rti-zone.org/atari.php

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Re: Open Source 68K Accelerator Project(s)

Postby terriblefire » Mon Nov 28, 2016 9:32 pm

czietz wrote:In such an interrupt acknowledge cycle the MFP puts the vector number on the data bus and asserts DTACK. On the 68020 your CPLD has to assert DSACK0 and/or DSACK1 instead of AVEC in this case. From looking at your Verilog source code I don't see this happening.


Can you point me to where this is mentioned in the 68020 manual. The bits i read said AVEC needed to get asserted but i may have missed this bit. Sorry about the video quality. I need to work more on the open broadcaster settings.

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Re: Open Source 68K Accelerator Project(s)

Postby terriblefire » Mon Nov 28, 2016 9:41 pm

rpineau wrote:On a ST, only the ACIA should trigger AVEC.
Here is our VHDL code from our working 68020 card :
AVEC <= not FC2 or not FC1 or not FC0 or not A19 or not A18 or not A17 or not A16 or VPA;


OK i'm pretty sure i do this.

wire CPUSPACE = &FC;

(which is the logical reduction of FC)

wire IACK = CPUSPACE & ({A[19:16]} == {4'b1111});

Then

assign AVEC = IACK & ~VPA;

Note that my CPLD doesnt have crazy inverting logic like PALs.

rpineau wrote:In the case of the MFP, it triggers a level 6 interrupt and wait for IACK from the GLUE (which is asserted when the GLUE detect an interrupt ack cycle from the CPU), it then put the interrupt vector on D0-D7 and assert DTACK. This ends up on the 68020 D24-D31 so you should assert DSACK1 in that case (16 bit transfer and the CPU know to get the 8 high bits it needs for the vector).

Regards, Rodolphe


I may be missing this bit then. How does this cycle get distinguished from other bus cycles if it happens after the IACK part of the cycle?

Cheers
Stephen

terriblefire
Atari freak
Atari freak
Posts: 68
Joined: Sat Nov 12, 2016 10:15 am

Re: Open Source 68K Accelerator Project(s)

Postby terriblefire » Mon Nov 28, 2016 9:52 pm

Actually. I'm pretty sure the above is all correct because I see the CPU take the interrupt vector and execute the service routine. My strong feeling is that writes to over the synchronous bus are not being latched and therefore the MFP etc are not resetting their interrupts.

I'll double check the Interrupt vector behaviour in the simulator.

czietz
Captain Atari
Captain Atari
Posts: 334
Joined: Tue May 24, 2016 6:47 pm

Re: Open Source 68K Accelerator Project(s)

Postby czietz » Mon Nov 28, 2016 9:56 pm

For the 68000 CPU, the interrupt acknowledge cycle is finished when a device asserts either VPA (for an auto-vector int.), DTACK (for a non-auto-vector int.) or BERR (in which case a spurious interrupt is generated). So you distinguish the cycle by the line asserted by either the device or GLUE. You just have to translate this into the lines the 68020 expects, i.e. AVEC, DSACK.

As for your second posting: As already written, the CPU is not supposed to take the level 6 auto-vector interrupt as shown in your video but the vector indicated by the MFP. Which MFP interrupt do you see taken then?


Social Media

     

Return to “Hardware”

Who is online

Users browsing this forum: No registered users and 1 guest