Minimig Discussion Forum

Discussing the Open Source FPGA Amiga Project
It is currently Tue Dec 12, 2017 12:10 pm

All times are UTC




Post new topic Reply to topic  [ 39 posts ]  Go to page Previous  1, 2, 3, 4
Author Message
 Post subject: Re: minimig-mist AGA v1.2 release
PostPosted: Sun Nov 13, 2016 12:14 pm 
Offline

Joined: Sun Sep 25, 2016 4:05 pm
Posts: 29
MMrobinsonb5 wrote:
I'm using the Prodigy crack (probably a good idea to specify which crack when using ADFs so we can compare like-for-like) - and with Fast RAM enabled it won't boot beyond the cracktro. There's a quirk with the Minimig's Fast RAM in that it disappears immediately after a Guru, and comes back after a subsequent soft-reset - so with Aladdin, the sequence is:
Boot with Fast RAM -> Cracktro -> crash -> Fast RAM disappears -> boot -> Cracktro -> game
Then the game runs OK.

Yes, Prodigy crack with +3 trainer in my case too.
Configuration is 68020 CPU, 2MB chip, no slow and 2MB fast memory.
In my case it loads fine, no crash after cracktro, runs one time through the demo and crash upon return to the menu.
It might take a few more demo runs, but crashes shortly anyway.

Without fast memory demo could run much longer (and it runs noticeably slower), about an hour, but crashes all the same in the end.

Quote:
Again, the prodigy crack, and yes I can reproduce this. With Fast RAM enabled it boots, goes into attract mode, but seems to crash on returning to the menu.

Tried both prodigy +15 and melnok +3 variants, the same config (2 MB chip, no or 0.5MB slow and 2 MB fast), just the same as yours result.

Interesting that non-AGA design works just fine with the same options, so the memory problems is mostly related to latest AGA v1.2 release.

So my DE1-SoC port is works almost as the original Minimig-Mist AGA :)

Quote:
A tiny change in a completely unrelated part of the core can be the difference between Fast RAM working and not.

Yeah, changes in the working FPGA temperature is the cause also.
This is a result of badly written hardware, multiclocked design without proper constraints.
And I`am don`t know how to constraint it correctly either... :(


Top
 Profile  
 
 Post subject: Re: minimig-mist AGA v1.2 release
PostPosted: Sun Nov 13, 2016 12:55 pm 
Offline

Joined: Sun Oct 11, 2009 3:19 pm
Posts: 20
sonycman wrote:
A tiny change in a completely unrelated part of the core can be the difference between Fast RAM working and not.
Yeah, changes in the working FPGA temperature is the cause also.
This is a result of badly written hardware, multiclocked design without proper constraints.
And I`am don`t know how to constraint it correctly either... :(


Constraints are one way to do it right, better is to change the Instruction Decoder of the tg68K. To many logic levels and loops in there,
which drive the synthesizer nuts. So the synthesizer tries to make it with the given clock, and fails :(

Tried to look into it a year ago, but had to stop, day job got in the way and to much traveling.

Now got myself an DE1-SOC and some Xilinx Artix boards, (BTW, Xilinx tools have the same problem with the tg68k core),
and hopefully some time ;-)

Cheers!


Top
 Profile  
 
 Post subject: Re: minimig-mist AGA v1.2 release
PostPosted: Sun Nov 13, 2016 1:30 pm 
Offline

Joined: Tue Dec 13, 2011 7:48 pm
Posts: 341
e2kcpu wrote:
Constraints are one way to do it right, better is to change the Instruction Decoder of the tg68K. To many logic levels and loops in there, which drive the synthesizer nuts. So the synthesizer tries to make it with the given clock, and fails :(


Yeah, the problem is that with the clkena signal asserted only one cycle in 4 or 5, it's fine for some of those combinational chains to take 4 or 5 clocks to propagate, but it's extremely difficult to tell the tools which ones, and which ones have to be fast.

I did get the TG68 to run at a lower clockspeed in another project (without having to slow down the SDRAM controller) and managed to achieve timing closure, and also had the Minimig core using a similar arrangement - but the latter was still not even close to meeting timing.

sonycman wrote:
So my DE1-SoC port is works almost as the original Minimig-Mist AGA :)


Yup - great work!

Quote:
Yeah, changes in the working FPGA temperature is the cause also.
This is a result of badly written hardware, multiclocked design without proper constraints.
And I`am don`t know how to constraint it correctly either... :(


Do you at least have the SDRAM constrained?

_________________
~ Amiga 4000/030 ~ Amiga 1200 030/50MHz ~ Turbo Chameleon 64 ~ Altera DE1 with Minimig core ~
Details of my projects: http://retroramblings.net


Top
 Profile  
 
 Post subject: Re: minimig-mist AGA v1.2 release
PostPosted: Sun Nov 13, 2016 2:25 pm 
Offline

Joined: Sun Sep 25, 2016 4:05 pm
Posts: 29
MMrobinsonb5 wrote:
Do you at least have the SDRAM constrained?

Of course I do, without that even old non-AGA minimig did`nt work correctly on DE1-SoC.
The latest AGA core also have very good constraints for sdram and some other clocks.
But it seems that is not enough.

This is bad I don`t know how to debug amiga side software - if only we have some kind of debugger to see memory contents and disassembly from 68k side.
That HRTmon - what is it, exactly?


Top
 Profile  
 
 Post subject: Re: minimig-mist AGA v1.2 release
PostPosted: Sat Nov 19, 2016 11:45 am 
Offline

Joined: Sun Feb 22, 2009 5:02 am
Posts: 175
Not really related, but I have noticed vast differences in behaviour of my MIST depending on what power supply I use.

_________________
-- kolla


Top
 Profile  
 
 Post subject: Re: minimig-mist AGA v1.2 release
PostPosted: Sat Nov 19, 2016 6:41 pm 
Offline

Joined: Sun Oct 11, 2009 3:19 pm
Posts: 20
kolla wrote:
Not really related, but I have noticed vast differences in behaviour of my MIST depending on what power supply I use.


The you have really bad power supplies lying around ;-)


Top
 Profile  
 
 Post subject: Re: minimig-mist AGA v1.2 release
PostPosted: Sun Nov 20, 2016 12:57 am 
Offline

Joined: Sun Feb 22, 2009 5:02 am
Posts: 175
e2kcpu wrote:
kolla wrote:
Not really related, but I have noticed vast differences in behaviour of my MIST depending on what power supply I use.


The you have really bad power supplies lying around ;-)


Yes, it seems that most 5V USB chargers tend to have a quality drop after a while, still good enough for charging batteries (albeit slower), but not good enough to keep the MIST (or a Raspberry pi for that matter) running well.

_________________
-- kolla


Top
 Profile  
 
 Post subject: Re: minimig-mist AGA v1.2 release
PostPosted: Sun Nov 20, 2016 1:48 pm 
Offline

Joined: Sun Oct 11, 2009 3:19 pm
Posts: 20
kolla wrote:
Yes, it seems that most 5V USB chargers tend to have a quality drop after a while, still good enough for charging batteries (albeit slower), but not good enough to keep the MIST (or a Raspberry pi for that matter) running well.


I'm ****** for years now, that a 5V external power supply is not a wise thing to to, in powering a 5V board.
And I'm glad, the newer eval boards are using 9v or 12V. It is few bucks more for the power supply,
but it is really worth it. Less problems, less thing to explain.

And somehow the newer "5V" power supplies are getting towards 5.2V or 5.3V to charge faster, more reliable on USB.
So they are getting dangerous for 5V boards ...


Top
 Profile  
 
 Post subject: Re: minimig-mist AGA v1.2 release
PostPosted: Tue Nov 22, 2016 11:48 am 
Offline

Joined: Sun Feb 22, 2009 5:02 am
Posts: 175
And all boards are really 3.3 V boards today anyways :)

For my own 5V systems at my desk, I use an old dual PSU made for servers, with one module hooked up via UPS. It delivers steadily and plenty :)

_________________
-- kolla


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 39 posts ]  Go to page Previous  1, 2, 3, 4

All times are UTC


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group
Translated by Xaphos © 2007, 2008, 2009 phpBB.fr