[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [N8VEM-S100:1563] S100-68K-V3

Hi Andrew,

I understand.  I guess I would go directly to the 030 or 040 as the 020 does not give enough of a boost to make it worth while.  Until we can break the memory limit, there is not much of a reason to do this as CP/M is not going to exploit it - we have to get to Linux or BSD before it is going to be useful and that is going to require memory.  Might be able to do emuTOS and MINT but I don't think they really would exploit a 020 without more memory.  I think putting effort on processors without knowing what can exploit it is not efficient.  Don't build it because you can, but build it because you should is my motto.

I think in the end (as much as I don't like it), the S100 bus is just too limiting for 32 bit processors and doesn't make much sense.  I guess an SBC is a better approach but we have to have people that just don't build things but write software too.  My thoughts now are leaning towards a 68060 with some PSRAM or SDRAM as an SBC.  The reason I choose the 68060 is because it is 3.3 V and PSRAM or SDRAM are not 5 volt tolerant without some level translation so it would make the board simpler.  This also means there is probably going to be a cpld or fpga in the circuit as well - which is going to push towards SMT - which will not be as friendly to builders but I am thinking of some kind of mezzanine board maybe.  This is a ways out as I need to get CP/M 68K running and get the 68360 fully working (I have output to propeller working well but input is not reliable yet - need to throw the logic analyzer on it and see what is going on).

Dave

On Sunday, April 21, 2013 12:50:36 PM UTC-5, lynchaj wrote:

Hi Dave,

 

The 68020 on the S-100 bus would give you the full 32 bit ISA and up to 16MB of memory.  Still it would not have the MMU necessary for Linux or BSD type OS.

 

I am suggesting the PAK68/2 as an intermediate step to a PAK68/3 board.  The PAK68/2 is pretty simple and would lay the ground work for a PAK68/3 clone board.

 

A PAK68/3 (68030) would be fully 32 bit ISA capable and have the MMU on board for a Linux or BSD type OS.

 

However to fix the lack of memory issue, we would need an “over the top” memory board connector similar to what we have going for the S-100 80386 CPU + SRAM boards.  In fact, I suspect the S-100 80386 SRAM board could be readily adapted if not used outright for a 68030 CPU board.

 

These 32 bit CPU boards are really complicated.  The S-100 80386 CPU + SRAM boards are on their fifth revision or so.  I just did another revision with John this morning.  I suspect we are not done yet either.

 

Thanks and have a nice day!

Andrew Lynch

 

From: n8ve...@googlegroups.com [mailto:n8vem...@googlegroups.com] On Behalf Of yoda
Sent: Saturday, April 20, 2013 8:42 PM
To: n8ve...@googlegroups.com
Subject: Re: [N8VEM-S100:1563] S100-68K-V3

 

Not sure what we would do with it - without more memory.

On Saturday, April 20, 2013 6:32:23 PM UTC-5, lynchaj wrote:

Hi

 

I think once we get the S-100 68K CPU board working then next step should be a PAK68/2 clone.  Basically a 68020/68881 stuffed into a 68K socket.

 

http://www.bigbookofamigahardware.com/bboah/product.aspx?id=65

 

There is also a PAK68/3 (68030) design but it is a lot more complicated and requires PALs.  Probably doable but too big a leap for one fell swoop.

 

I have the schematics for the PAK68/2 around here somewhere.  It is not too complicated circuit and has a rectangular PCB layout.

 

Thanks and have a nice day!

Andrew Lynch

 

 

From: n8ve...@googlegroups.com [mailto:n8vem...@googlegroups.com] On Behalf Of yoda
Sent: Saturday, April 20, 2013 6:07 PM
To: n8ve...@googlegroups.com
Cc: mon...@vitasoft.org
Subject: Re: [N8VEM-S100:1561] S100-68K-V3

 

Good catch on the 5V regulator - glad my local Fry's had the insulators - picked up a couple.  I have most of the parts on except for the electrolytic caps and the regulator - going through all the jumpers now to see which ones are needed for bus master operation.  I will probably do some prelim testing tonight or tomorrow - quick check with ohm meter shows not shorts between 5V and gnd so that is a positive first step.



On Saturday, April 20, 2013 12:12:44 PM UTC-5, monahanz wrote:

I am currently assembling the board. S100 connectors not a problem. 

 

All users, do remember to put a mica insulator below the 5V regulator, there are traces in the area that could short to ground. Also the side of the heat sink needs to be clipped so the card will fit into bus slots. Unless you are in a rush give me a few days to build and test things before I write things up.

 

John

 

John Monahan Ph.D

e-mail: mon...@vitasoft.org

Text:    mon...@txt.att.net

 

 

From: n8ve...@googlegroups.com [mailto:n8ve...@googlegroups.com] On Behalf Of Andrew Lynch
Sent: Saturday, April 20, 2013 8:37 AM
To: n8ve...@googlegroups.com
Subject: RE: [N8VEM-S100:1557] S100-68K-V3

 

Hi Dave!

Thanks!  Yes, there are some issues with the S-100 68K CPU board.  I don’t think they’ll have an effect on functionality but do need some clean up.

 

The over pours near the top of the connector pads should not be an issue since the pads are significantly bigger than the connector wipers.  However I see a couple of traces and vias migrated down into the connector. 

 

Normally I hand route those away from the connector.  I’ve done that many times already with this board but during the last optimization they must have reappeared.  Still they shouldn’t affect functionality but it is not good practice.

 

Most likely there will be updates to the board and during the next respin we can clean up the minor trace anomalies.

 

Please post your progress on the board.  I am really looking forward to seeing how this board finally matures!

 

Thanks and have a nice day!

Andrew Lynch

 

From: n8ve...@googlegroups.com [mailto:n....@googlegroups.com] On Behalf Of yoda
Sent: Thursday, April 18, 2013 10:27 PM
To: n8ve...@googlegroups.com
Subject: [N8VEM-S100:1551] S100-68K-V3

 

I have received said boards and just a couple of things to be aware of.  It looks like there are some "over pours" of copper near the connector fingers which appear not to short but could possibly short when plugged into the S100 bus so please be careful to check this.  If there is a future production run of these boards the gerber files should be cleaned up - I did not notice in the original files but it is pretty obvious when you look at the real boards and this appears on both sides of the boards.

 

Another fix in the future would be to put the resistor number on the silk screen not the value, ie, R1 instead of 2200 - it makes it harder if there is a need to do debugging and you want to probe at a specific resistor it is hard to know which one is which when referring back to the schematics or vice versa.

 

I am starting to put together one of the boards and hope to be able to do bring up some time this weekend.   I plan to use initially in bus master mode.   I think I understand all the jumpers that need to be made and will post a bus master jumper configuration once I have a working board.  I have a V1 board that is working so that should aid in getting this board running,

 

From there I plan on finishing a BIOS for CP/M 68k and will publish that when working.   I plan to write the BIOS and monitor code using GCC cross compiler and 95+ per cent of the code should be in C with very little assembly code needed.  I have tested some of the code and it is working.  I have been waiting for this board to finish as I want to locate the stack disk buffer on the on board RAM away from main memory.  There were some issues with V1 that did not allow the on board memory to work without some surgery to the board.

 

Will post pictures, jumpers and code when I get a little further

--
You received this message because you are subscribed to the Google Groups "N8VEM-S100" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem-s100+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

--
You received this message because you are subscribed to the Google Groups "N8VEM-S100" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem-s100+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

--
You received this message because you are subscribed to the Google Groups "N8VEM-S100" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem-s100+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

--
You received this message because you are subscribed to the Google Groups "N8VEM-S100" group.
To unsubscribe from this group and stop receiving emails from it, send an email to n8vem-s100+...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.