Beginning of very long journey

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

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

Rustynutt
Atari Super Hero
Atari Super Hero
Posts: 593
Joined: Wed Mar 21, 2012 7:38 am
Location: Oregon
Contact:

Re: Beginning of very long journey

Postby Rustynutt » Sun Oct 14, 2018 5:57 pm

joska wrote:IDE is an integral part of the Falcon custom chipset, so replacing it does not make sense.


Once again I reported a post, not posted as a comment. Sorry about that.

For a couple bucks users can purchase a small iDE to SATA adapter.

Rustynutt
Atari Super Hero
Atari Super Hero
Posts: 593
Joined: Wed Mar 21, 2012 7:38 am
Location: Oregon
Contact:

Re: Beginning of very long journey

Postby Rustynutt » Sun Oct 14, 2018 7:23 pm

joska wrote:
mikro wrote:
mpattonm wrote:Switching DSP to 56301 will be a little setback, but hey - its a fun project and thats where the fun is. So back to the schematics.

Awesome. Just want to mention it - in this case some minor changes to the TOS are going to be needed.


If TOS needs to be changed to work with the 56301, how would compatibility with games and demos be?


When reading through the MOT docs, it states the 56301 is object code compatible

Is it the additional DSP memory and possible address difference your concern?

I have a 56301 PCI development card I was thinking of trying to use under Hatari, of course with change to access the PCI over the emulator, if possible. At a minimum, some of dml's bench test. Purchased it with the possibility of using it on the CTPCI. On a standard Falcon, if some how the DSP was replaced, think address changes in the GAL could redirect location.
Not sure if the CT60 uses the GAL's to interrupt.

mpattonm
Hardware Guru
Hardware Guru
Posts: 337
Joined: Mon Oct 21, 2002 8:52 am
Location: Czech republic
Contact:

Re: Beginning of very long journey

Postby mpattonm » Thu Oct 18, 2018 2:12 pm

mikro wrote:Your gut feeling was right, it can not work like that. I even guessed it, Videl is indeed using VID32 to sync with the Combel (for ST-RAM access). So *that* is the reason why they put 32.08... for compatibility on the external pin when master clock was changed. You are free to modify 25.175 or external clock (they work independently from the system bus) but not VID32, must be the same as master clock.

So you can rewire it a bit - instead of switching between 32.08.../custom clock on VID32 you keep the line from the master clock on VID32 and put the exact same switch on the external clock pin.

Not sure if I understand correctly, do you mean something like.... this?
You do not have the required permissions to view the files attached to this post.

mikro
Hardware Guru
Hardware Guru
Posts: 1784
Joined: Sat Sep 10, 2005 11:11 am
Location: Kosice, Slovakia
Contact:

Re: Beginning of very long journey

Postby mikro » Thu Oct 18, 2018 3:06 pm

mpattonm wrote:Not sure if I understand correctly, do you mean something like.... this?

Uhm, stop listening to me. Your v2.1 is totally fine, I don't know what I was thinking (it seems that I'd overlooked that the masterclock *is* shared with both the Videl and the Combel input).

If you like, you can add a similar switch (32.08/custom clock) to EXTCLK - similarly as on the CT2. So:

- if input to Combel == 32.08 MHz then EXTCLK would be wired from external source (video connector)
- if input to Combel == custom then EXTCLK = 32.08 MHz (for backward RGB compatibility)

mikro
Hardware Guru
Hardware Guru
Posts: 1784
Joined: Sat Sep 10, 2005 11:11 am
Location: Kosice, Slovakia
Contact:

Re: Beginning of very long journey

Postby mikro » Fri Oct 26, 2018 9:14 pm

I have remembered about one suggestion seen in another thread: viewtopic.php?f=97&t=33093#p340349

Maybe, when routing the PCB, you could somehow take this idea into account? So the next-gen expansion card wouldn't expand only CPU (and possibly TT RAM) but also ST RAM in one go. All that is needed is to somehow place the ST RAM headers close to the expansion slot headers so one PCB can be placed on all four headers at once.

mpattonm
Hardware Guru
Hardware Guru
Posts: 337
Joined: Mon Oct 21, 2002 8:52 am
Location: Czech republic
Contact:

Re: Beginning of very long journey

Postby mpattonm » Sun Nov 18, 2018 12:30 pm

Just a quick update: schematics are complete (and no, I am not going to make any more changes this time, I swear), layout is fixed and I have completed first round of routing. Memory buses, clock signals, audio out, TV out and some other subsystems were routed manually, while autorouter took care of the rest. Still on a four layer board.
Next step is a manual cleanup and optimizations, it should keep me busy for couple of days.
You do not have the required permissions to view the files attached to this post.
Last edited by mpattonm on Sun Nov 18, 2018 5:34 pm, edited 1 time in total.

arf
Captain Atari
Captain Atari
Posts: 184
Joined: Thu May 17, 2012 9:56 pm
Location: Germany

Re: Beginning of very long journey

Postby arf » Sun Nov 18, 2018 1:58 pm

Keep on hackin’!

User avatar
TheNameOfTheGame
Atari God
Atari God
Posts: 1107
Joined: Mon Jul 23, 2012 8:57 pm
Location: Almost Heaven, West Virginia

Re: Beginning of very long journey

Postby TheNameOfTheGame » Sun Nov 18, 2018 4:21 pm

Thanks for the update!

Rustynutt
Atari Super Hero
Atari Super Hero
Posts: 593
Joined: Wed Mar 21, 2012 7:38 am
Location: Oregon
Contact:

Re: Beginning of very long journey

Postby Rustynutt » Sun Nov 18, 2018 7:52 pm

Visually check the SDMA CLKIN point. The common trace from the COMBEL to both the FPU clock and SDMA clock makes a small but severe dog leg turn around another IC's resistor. From normal view, the "jump" under the resistor is left of the SDMA clock pin, about 20mm.

Could you also consider a PGA or modified DSP pinout where the DSP can relatively easily swapped out?
Without a 56k sheet in front of me, recall the 56002 and upward compatible have a few pins swapped. They may also include increased memory management.

68030 PGA socket.

Kind of out from the scope of "the long journey", a nice additional feature would be to add separate oscillator sockets for IC's which in the past have had traces cut to insert a different clock. ie, DSP, FPU, MIDI and KB IC, VIDEL, VIDEL Genlock.
Pretty much any speeder, accelerator, CPU upgrade, basically all the reasons more Falcons didn't make it out from the System 7 wars :)

User avatar
calimero
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2250
Joined: Thu Sep 15, 2005 10:01 am
Location: STara Pazova, Serbia
Contact:

Re: Beginning of very long journey

Postby calimero » Sun Nov 18, 2018 8:20 pm

And what are final specs.? :)

Btw
Original Falcon board was also 4 layered?
Btw2
Will you made this opensource so anybody can make their Dream-Falcon? :D
using Atari since 1986.http://wet.atari.orghttp://milan.kovac.cc/atari/software/ ・ Atari Falcon030/CT63/SV ・ Atari STe ・ Atari Mega4/MegaFile30/SM124 ・ Amiga 1200/PPC ・ Amiga 500 ・ C64 ・ ZX Spectrum ・ RPi ・ MagiC! ・ MiNT 1.18 ・ OS X

mpattonm
Hardware Guru
Hardware Guru
Posts: 337
Joined: Mon Oct 21, 2002 8:52 am
Location: Czech republic
Contact:

Re: Beginning of very long journey

Postby mpattonm » Sun Nov 18, 2018 9:36 pm

Rustynutt wrote:Visually check the SDMA CLKIN point. The common trace from the COMBEL to both the FPU clock and SDMA clock makes a small but severe dog leg turn around another IC's resistor. From normal view, the "jump" under the resistor is left of the SDMA clock pin, about 20mm.

This is completelly new board with different layout and certainly, different routing. All clock signals were routed manually (as the only signals actually) on a GND layer. Thus the above does not apply on the new board.

Rustynutt wrote:Could you also consider a PGA or modified DSP pinout where the DSP can relatively easily swapped out?
Without a 56k sheet in front of me, recall the 56002 and upward compatible have a few pins swapped. They may also include increased memory management.

F030NG is designed with 56303 DSP, which, while code compatible, is significantly faster then 56001 and is coupled with 128kx24b memory. Compared to three 32kx8k modules on original Falcon030, it has four times bigger memory capacity.
At one point I was also considering a dedicated slot for DSP add-on card, but it was lef out from the design afterall. It might appear in a future tho.
Rustynutt wrote:68030 PGA socket.

Kind of out from the scope of "the long journey", a nice additional feature would be to add separate oscillator sockets for IC's which in the past have had traces cut to insert a different clock. ie, DSP, FPU, MIDI and KB IC, VIDEL, VIDEL Genlock.
Pretty much any speeder, accelerator, CPU upgrade, basically all the reasons more Falcons didn't make it out from the System 7 wars :)

Again, clock generation is way different form original Falcon. For example, instead of plain 32.089MHz crystal, F030NG features PLL programmable from 32 to 50MHz in 10 steps. There is also a switch to bypass Combell clock to CPU and FPU, allowing these two to be fed from PLL directly. Meaning you will have number of options and combinations to overclock your system, there will be no use for the basic speeders. With the exception of CT family, of course.

mpattonm
Hardware Guru
Hardware Guru
Posts: 337
Joined: Mon Oct 21, 2002 8:52 am
Location: Czech republic
Contact:

Re: Beginning of very long journey

Postby mpattonm » Sun Nov 18, 2018 9:42 pm

calimero wrote:And what are final specs.? :)

Yes, I will make a wiki page eventually.
calimero wrote:Btw
Original Falcon board was also 4 layered?

Really? I did not know that. All I have ever seen was 6 layer PCB versions. That would perhaps explain some of the weird routing artefacts that make no sence at all.
calimero wrote:Btw2
Will you made this opensource so anybody can make their Dream-Falcon? :D

Eventually yes, but not from the day one.

Galvez
Captain Atari
Captain Atari
Posts: 240
Joined: Fri Oct 19, 2007 7:49 am

Re: Beginning of very long journey

Postby Galvez » Mon Nov 19, 2018 8:23 am

mpattonm wrote:
calimero wrote:Btw
Original Falcon board was also 4 layered?

Really? I did not know that. All I have ever seen was 6 layer PCB versions. That would perhaps explain some of the weird routing artefacts that make no sence at all.

I think he was asking, see the question mark.

User avatar
Greenious
Hardware Guru
Hardware Guru
Posts: 1277
Joined: Sat Apr 24, 2004 5:39 pm
Location: Sweden

Re: Beginning of very long journey

Postby Greenious » Mon Nov 19, 2018 11:23 pm

I'm impressed by the work so far and really like where this is going!

A thought... given that most ppl interested in this new mobo probably want something faster than 030 & FPU, how about lifting them off the mobo to a daughterboard for the expansionbus instead? You can save some more space and people dont have to buy a CPU they will not use...
Updated my guides as of june 28th, 2016. Check'em out and feedback!
viewtopic.php?t=5040

User avatar
calimero
Fuji Shaped Bastard
Fuji Shaped Bastard
Posts: 2250
Joined: Thu Sep 15, 2005 10:01 am
Location: STara Pazova, Serbia
Contact:

Re: Beginning of very long journey

Postby calimero » Tue Nov 20, 2018 12:01 am

Yes, I was asking :)
but as I remeber from 90s that I read somewhere that Falcon MB is 4 layered... I also remeber that comparing to Amiga1200 I was thinking that Falcon motherboard and components were made in much more "modern" manner :D

Btw
Regarding 32.089MHz, someone, I think Mikro, mention that PAL and NTSC Falcons have different clock? In FalconNG you will put which one?
using Atari since 1986.http://wet.atari.orghttp://milan.kovac.cc/atari/software/ ・ Atari Falcon030/CT63/SV ・ Atari STe ・ Atari Mega4/MegaFile30/SM124 ・ Amiga 1200/PPC ・ Amiga 500 ・ C64 ・ ZX Spectrum ・ RPi ・ MagiC! ・ MiNT 1.18 ・ OS X

User avatar
shoggoth
Nature
Nature
Posts: 951
Joined: Tue Aug 01, 2006 9:21 am
Location: Halmstad, Sweden
Contact:

Re: Beginning of very long journey

Postby shoggoth » Wed Nov 21, 2018 12:12 pm

Greenious wrote:I'm impressed by the work so far and really like where this is going!

A thought... given that most ppl interested in this new mobo probably want something faster than 030 & FPU, how about lifting them off the mobo to a daughterboard for the expansionbus instead? You can save some more space and people dont have to buy a CPU they will not use...


This probably rules out the CT6x, since it needs the original 030 to be present during boot (god knows why).
Ain't no space like PeP-space.

mikro
Hardware Guru
Hardware Guru
Posts: 1784
Joined: Sat Sep 10, 2005 11:11 am
Location: Kosice, Slovakia
Contact:

Re: Beginning of very long journey

Postby mikro » Wed Nov 21, 2018 6:02 pm

shoggoth wrote:This probably rules out the CT6x, since it needs the original 030 to be present during boot (god knows why).

I'll speculate here and say it is because R. Czuba didn't want people to manipulate the 030 in any way. AFAIK you have to ground pin 17 (BGACK-, Bus Grant Acknowledge) to signalise that another device (CT60) has become the bus master.

User avatar
MacFalcon
Captain Atari
Captain Atari
Posts: 279
Joined: Sat Oct 11, 2003 1:00 pm
Location: Muenchberg, Germany
Contact:

Re: Beginning of very long journey

Postby MacFalcon » Wed Nov 21, 2018 7:54 pm

Are you sure that the 030 is needed? Rodolphe wrote me one day that my Falcon could run even with dead 030 in 060 Mode (when I killed my board during some testing for flash060.prg...)

But that's offtopic ;)

Great Work! A new Falcon Board would be awesome!

Chris
Falcon060, 100/25 MHz, 256 MB SD-RAM, PhantomPBS, CTPCI with Radeon, USB and Network Card. atari rulez!

Rustynutt
Atari Super Hero
Atari Super Hero
Posts: 593
Joined: Wed Mar 21, 2012 7:38 am
Location: Oregon
Contact:

Re: Beginning of very long journey

Postby Rustynutt » Sun Nov 25, 2018 7:00 am

Rustynutt wrote:Could you also consider a PGA or modified DSP pinout where the DSP can relatively easily swapped out?
Without a 56k sheet in front of me, recall the 56002 and upward compatible have a few pins swapped. They may also include increased memory management.

F030NG is designed with 56303 DSP, which, while code compatible, is significantly faster then 56001 and is coupled with 128kx24b memory. Compared to three 32kx8k modules on original Falcon030, it has four times bigger memory capacity.
At one point I was also considering a dedicated slot for DSP add-on card, but it was lef out from the design afterall. It might appear in a future tho.
[quote="Rustynutt"]

Have yet the chance to work with a couple of Motorola DSP developer cards picked up last year.

They may or may not be of interest.
After upgrading my Milan, will install the PCI DSP card in this machine for some simple experiments.

Installing the CTPCI on a Falcon is down the road a bit more.

The audio development board looks as if it could be interfaced directly to the CT card, honestly have not looked into that possibly, yet still has other interface designs.

The photos are located here open to public if you have a Facebook account.

https://m.facebook.com/michael.grove.39 ... 952377519/


Social Media

     

Return to “Hardware”

Who is online

Users browsing this forum: No registered users and 9 guests