Re: Z80 1st boot (wooot!)
On Sat, Jun 12, 2021 at 09:23 AM, joshbensadon wrote:
Heathkit used a similar model and it was successful.?
I think thats why I love all the stuff Lee (and you) have done here. When I was 13 I got some kits from Heathkit (and Eico) and I think that was so wonderful to do and explore. I knew crap about electronics back then and why something did what it did. Those kits gave me that knowledge years before I went to college. When heathkit went downhill so did a piece of me. I wish I could have bought more of them. You guys have the same approach here (and Lee mentions that a bit too in these topics) and I think not only for me but beginners will now know how that feels. The Z80MC and 1802MC are awesome. -Tony
|
Re: Z80 1st boot (wooot!)
Tony,
That's all Lee.? He's big on taking little steps.? He's always going on about grand builds that start at the bottom.? Building modules one at a time, testing them individually then combining them.? Heathkit used a similar model and it was successful.? Of course, step 1 was always "build the power supply".? ?:)
Josh
On Friday, June 11, 2021, 06:05:25 p.m. EDT, saturn5tony via groups.io <saturn5tony@...> wrote:
On Fri, Jun 11, 2021 at 05:51 PM, joshbensadon wrote:
Tony,? looking good!
Thanks Josh, Yea it was the Additional Operating details with ZMCv15 ROM you and Lee wrote that really helped alot. I saw how you guys set it up to test only the Z80 card by itself that got it working so fast. I knew I had a problem with that card at the 1st point looking at it with my scope. Yup Lee said an HC368 so yup it had to be one. No issue there at all. My HC stock is very low being I came from those days with LS's and such in my Lab, lol Great Docs from both of you. Man this is a lot of fun.
|
Re: Z80 1st boot (wooot!)
On Fri, Jun 11, 2021 at 05:51 PM, joshbensadon wrote:
Tony,? looking good!
Thanks Josh, Yea it was the Additional Operating details with ZMCv15 ROM you and Lee wrote that really helped alot. I saw how you guys set it up to test only the Z80 card by itself that got it working so fast. I knew I had a problem with that card at the 1st point looking at it with my scope. Yup Lee said an HC368 so yup it had to be one. No issue there at all. My HC stock is very low being I came from those days with LS's and such in my Lab, lol Great Docs from both of you. Man this is a lot of fun.
|
Re: Z80 1st boot (wooot!)
On Friday, June 11, 2021, 05:21:03 p.m. EDT, saturn5tony via groups.io <saturn5tony@...> wrote:
Here is the 1st boot of just the Z80MC CPU board!? yippee!
Finished it last week but it had no clock. Found out that my use of a 74LS368 was not a good Idea, so I put an order in to Mouser for an HC368 as Lee designed it for, got it today and in business with the zed80...sweet!!
@@@@ @ H?
Hard Reset 01
AF =1D68? BC =0840? DE =FF80? HL =FFFF
AF'=01FF? BC'=FFFF? DE'=FFFF? HL'=00C3
IX =FFFF? IY =FFFF? IR =001D? PC =FFFF? SP =FB00
Press M for Monitor..............................
????????????????????????????????????????????????? MICRO-BASIC COPYRIGHT 1983 BY DAVE DUNFIELD
READY
?NO PROGRAM ERROR
READY
10 PRINT "HELLO WORLD";
20 GOTO 10
LIST
0010 PRINT "HELLO WORLD";
0020 GOTO 10
READY
RUN
?SYNTAX ERROR IN LINE 0010
READY
10 PRINT "HELLO WORLD"
RUN
HELLO WORLD
HELLO WORLD
HELLO WORLD
HELLO WORLD
HELLO WORLD
HELLO WORLD
STOP IN LINE 0020
READY
Here is a pic of it (note to the top right with a discrete clock, now disconnected) also showed me that the chip enable's needed the HC368 as well.
Notice the (white) Mostek Z-80 from 1977! Wow, I was very careful bringing that up power wise too.
Well next I will need to replace the LS245 and LS273 with HC's as well but Mouser did not have any of these on hand.
Might work OK but well see.? Also I will be finishing the front panel card and the SIO card as well in the next few days so wish me luck.? 8-)
-Tony
|
Here is the 1st boot of just the Z80MC CPU board!? yippee!
Finished it last week but it had no clock. Found out that my use of a 74LS368 was not a good Idea, so I put an order in to Mouser for an HC368 as Lee designed it for, got it today and in business with the zed80...sweet!!
@@@@ @ H?
Hard Reset 01
AF =1D68? BC =0840? DE =FF80? HL =FFFF
AF'=01FF? BC'=FFFF? DE'=FFFF? HL'=00C3
IX =FFFF? IY =FFFF? IR =001D? PC =FFFF? SP =FB00
Press M for Monitor..............................
????????????????????????????????????????????????? MICRO-BASIC COPYRIGHT 1983 BY DAVE DUNFIELD
READY
?NO PROGRAM ERROR
READY
10 PRINT "HELLO WORLD";
20 GOTO 10
LIST
0010 PRINT "HELLO WORLD";
0020 GOTO 10
READY
RUN
?SYNTAX ERROR IN LINE 0010
READY
10 PRINT "HELLO WORLD"
RUN
HELLO WORLD
HELLO WORLD
HELLO WORLD
HELLO WORLD
HELLO WORLD
HELLO WORLD
STOP IN LINE 0020
READY
Here is a pic of it (note to the top right with a discrete clock, now disconnected) also showed me that the chip enable's needed the HC368 as well.
Notice the (white) Mostek Z-80 from 1977! Wow, I was very careful bringing that up power wise too.
Well next I will need to replace the LS245 and LS273 with HC's as well but Mouser did not have any of these on hand.
Might work OK but well see.? Also I will be finishing the front panel card and the SIO card as well in the next few days so wish me luck.? 8-)
-Tony
|
Re: SD I/O 1.8MHz resonator
I’ve also seen ORANGE CONNEX on a few occasions. I’ve signed up for USPS notifications so that as soon as any package enters their system that targets my mailing address, regardless of origin, I know that a package has “arrived” on US shores … and I can see its subsequent progress. ?
toggle quoted message
Show quoted text
From: [email protected] [mailto:[email protected]] On Behalf Of SCOTT VITALE
Sent: Friday, May 28, 2021 3:16 PM
To: [email protected]
Subject: Re: [Z80MC] SD I/O 1.8MHz resonator ? I order parts off of eBay. In many cases, the supplier will ship, as Paul stated, via a "package bundle". Sometimes U.S. vendors do the same thing using DHL or FEDEX. Recently, I've seen the Chinese suppliers using ORANGE CONNEX that usually ends up in Bensonville, U.S. and is then relabeled with a USPS tracking number and sent via USPS. I've seen typically 2 weeks shipping time from China. Sometimes the vendor charges US$2 to US$4 for the shipping and sometimes its free. You can track the package from China to the U.S. via ORANGE CONNEX but afterwards, you don't get the USPS tracking number, so the package can't be tracked once it passes through U.S. customs, which can take 3 - 5 days (slower for caution due to China's attack on U.S. using bio-weapon CV-19). Once shipped via USPS, typically 3 days to arrive to your door.
Peace and blessings,
Scott
|
Re: SD I/O 1.8MHz resonator
I order parts off of eBay. In many cases, the supplier will ship, as Paul stated, via a "package bundle". Sometimes U.S. vendors do the same thing using DHL or FEDEX. Recently, I've seen the Chinese suppliers using ORANGE CONNEX that usually ends up in Bensonville, U.S. and is then relabeled with a USPS tracking number and sent via USPS. I've seen typically 2 weeks shipping time from China. Sometimes the vendor charges US$2 to US$4 for the shipping and sometimes its free. You can track the package from China to the U.S. via ORANGE CONNEX but afterwards, you don't get the USPS tracking number, so the package can't be tracked once it passes through U.S. customs, which can take 3 - 5 days (slower for caution due to China's attack on U.S. using bio-weapon CV-19). Once shipped via USPS, typically 3 days to arrive to your door.
Peace and blessings,
Scott
|
Re: SD I/O 1.8MHz resonator
Judging by the tracking on my two recent orders it appears that they bundle US (or maybe North American?) orders over some interval into a single package that is shipped to Colorado where it is disaggregated and then individual orders travel via USPS. Interestingly it appears that they "eat" the cost of the drop-ship to Colorado, and the Shipping Cost in the order roughly corresponds to the USPS rate from Colorado based on the package label (which in both cases was marked with a somewhat higher cost than was charged in the order).
I presume that the drop-ship is air mail as it was only a week between order-entry and arrival in Colorado. The second week was spent with USPS Ground.
A very efficient process; I'm impressed. I imagine that they have a similar set-up for other world regions.
I've seen what appears to be similar behavior with Hong Kong (and possibly Chinese) vendors on eBay -- after a pause the package appears to originate at a US location.
Glad to hear that Tayda wasn't foolin' with their availability of the 1.8Mhz resonators :->.
toggle quoted message
Show quoted text
-----Original Message----- From: [email protected] [mailto: [email protected]] On Behalf Of Lee Hart Sent: Thursday, May 27, 2021 5:36 PM To: [email protected]Subject: Re: [Z80MC] SD I/O 1.8MHz resonator saturn5tony via groups.io wrote: Lee Hart wrote: I did find a source for more 1.8 MHz resonators. I ordered directly
from Tayda Electronics in Thailand, and they arrived. :-)
Oh great Lee. I just did my 1st order with them as well for these 1.8meg
cr's for my new z80 bd. How long did it take to get them? The shipping
was very reasonable. My order took just over 2 weeks to arrive. That's pretty good! I've shipped packages internationally that have taken 4-6 weeks to arrive. Of course, that's "good" compared to last year. Some packages sent in early 2020 took 3-4 months to arrive! Lee -- We cannot solve our problems with the same thinking we used when we created them. -- Albert Einstein -- Lee Hart, 814 8th Ave N, Sartell MN 56377, www.sunrise-ev.com -- This email has been checked for viruses by Avast antivirus software.
|
Re: SD I/O 1.8MHz resonator
saturn5tony via groups.io wrote: Lee Hart wrote: I did find a source for more 1.8 MHz resonators. I ordered directly
from Tayda Electronics in Thailand, and they arrived. :-)
Oh great Lee. I just did my 1st order with them as well for these 1.8meg cr's for my new z80 bd. How long did it take to get them? The shipping was very reasonable. My order took just over 2 weeks to arrive. That's pretty good! I've shipped packages internationally that have taken 4-6 weeks to arrive. Of course, that's "good" compared to last year. Some packages sent in early 2020 took 3-4 months to arrive! Lee -- We cannot solve our problems with the same thinking we used when we created them. -- Albert Einstein -- Lee Hart, 814 8th Ave N, Sartell MN 56377, www.sunrise-ev.com -- This email has been checked for viruses by Avast antivirus software.
|
Re: SD I/O 1.8MHz resonator
On Thu, May 27, 2021 at 11:14 AM, Lee Hart wrote:
I did find a source for more 1.8 MHz resonators. I ordered directly from Tayda Electronics in Thailand, and they arrived. :-)
Oh great Lee. I just did my 1st order with them as well for these 1.8meg cr's for my new z80 bd. How long did it take to get them? The shipping was very reasonable.
|
Re: SD I/O 1.8MHz resonator
SCOTT VITALE wrote: FYI: for non-standard baud rates;
I use Linux for all development work and?a USB-2-TTL serial converter for serial I/O. I tested with with CH340 to see how much baud-rate error would be tolerated at 57.6Kb and found up to +7% error was still functional but +10% was no-joy. I also found that anything more than -2% was no-joy.
I use PICOCOM for terminal emulation, which allows one to use ANY baud rate you tell it, so if you are using a 2 MHz oscillator, simply tell PICOCOM your ACTUAL baud rate is 62.5Kb (or whatever) and it'll work. Not sure about WINDOWS terminal emulators. Try it and see. Thanks Scott, The terminal programs I normally use (ProComm and TeraTerm) don't support odd baud rates; I'll have to try the one you suggested. It sounds like a good tool for testing baud rate accuracy. I did find a source for more 1.8 MHz resonators. I ordered directly from Tayda Electronics in Thailand, and they arrived. :-) Best wishes, Lee Hart -- We cannot solve our problems with the same thinking we used when we created them. -- Albert Einstein -- Lee Hart, 814 8th Ave N, Sartell MN 56377, www.sunrise-ev.com -- This email has been checked for viruses by Avast antivirus software.
|
Re: SD I/O 1.8MHz resonator
glad it's working for you.? Did you try other fixed baud rates?
toggle quoted message
Show quoted text
FYI: for non-standard baud rates;
I use Linux for all development work and?a USB-2-TTL serial converter for serial I/O. I tested with with CH340 to see how much baud-rate error would be tolerated at 57.6Kb and found up to +7% error was still functional but +10% was no-joy. I also found that
anything more than -2% was no-joy.
I use PICOCOM for terminal emulation, which allows one to use ANY baud rate you tell it, so if you are using a 2 MHz oscillator, simply tell PICOCOM your ACTUAL baud rate is 62.5Kb (or whatever) and it'll work. Not sure about WINDOWS terminal emulators. Try
it and see.?
Peace and blessings.
|
Re: SD I/O 1.8MHz resonator
FYI: for non-standard baud rates;
I use Linux for all development work and?a USB-2-TTL serial converter for serial I/O. I tested with with CH340 to see how much baud-rate error would be tolerated at 57.6Kb and found up to +7% error was still functional but +10% was no-joy. I also found that anything more than -2% was no-joy.
I use PICOCOM for terminal emulation, which allows one to use ANY baud rate you tell it, so if you are using a 2 MHz oscillator, simply tell PICOCOM your ACTUAL baud rate is 62.5Kb (or whatever) and it'll work. Not sure about WINDOWS terminal emulators. Try it and see.?
Peace and blessings.
|
Re: SD I/O 1.8MHz resonator
Hi folks!
Just a quick update, it actually works if your terminal client supports custom baud rates. I got my kit fully assembled this morning, and had it working? in no time using GtkTerm under Linux, may work in Putty as well, but I haven't tried yet.
Thanks? for all the help.
|
Z80 is 4-22 clock cycles per instruction.
The longer ones are the are in the groups that increment or decrement
and repeat.
For Z80 the simple way is a RST (one byte call) to one of 8 locations.
The common one is RST-6 (call to location 0x0030)? I use the 8080
instruction form. Z80 form would be RST 30H.? That also works for
8080 and 8085.
Once at location 30H save the stack pointer, save the registers
and return to monitor calling point for single step.??
Cost in code is small and zero hardware.
Works best if the RST points to the same location as push button
interrupt (not reset). as then you can recover from a runaway program
or one caught in a loop.
Allison
|
Be careful Lee,? ?all that is going to go straight to my head!
The single step is definitely cool, but as I recall, it was your idea, I just coded it.
For those that don't know it.? Many single step operations write a jump to the next instruction(s) possible to be executed.? This jump returns control to the monitor program.? Ok, it might not exactly be a jump, some use Return, most use the Restart instructions, but it's all the same mechanics.
Lee suggested that the jump be externally generated though the timer interrupt.? In this fashion, it simplifies the need to write a jump into RAM and also permits single stepping through ROM (where you can't temporarily write a jump).? This single step code is in fact within the ISR and it simply wastes exactly the number of cycles so that when the ISR returns to main line code, only 1 instruction can execute before an interrupt brings the CPU back into the ISR.? That's the basic principle, but there's just a little more, a gory detail.??
Gory Details. The Z80 has variable length instructions, taking between 4 and 17 cycles more or less (check data sheet for exact numbers).? The interrupt works from a counter after 4096 counts I think (that's about 1mSec with a 4Mhz clock).? In order to perfectly align the interrupt for a single instruction, a halt instruction is used to wait for 1 timer interrupt interval before counting out the exact count of cycles.
Cheers, Josh
On Sunday, May 16, 2021, 02:21:54 a.m. EDT, Lee Hart <leeahart@...> wrote:
joshbensadon via groups.io wrote:
> Bob,
>
> Thanks!? That really means a lot to me!? I just kind of thought nobody
> ever looks at the source code.
>
> The part of the source code I love the most is the full duplex bit
> banged serial I/O.? There's been MANY bit banged I/O routines, but I've
> never seen a full duplex.
Josh's source code is a wonderful example of how to write and document
assembler. It's worth spending some time to read it with careful
concentration on the details.
There are many aspects that I like, but here are a few. One is his
full-duplex interrupt-driven serial I/O routine. Another is his
single-step routine. It can trace a program even through ROM. Yet
another is that it can bit-bang an SD-card, and even figure out the
byzantine PC FAT-16 director structure.
It's a masterpiece! :-)
Lee
--
A designer knows he has achieved perfection not when there is
nothing left to add, but when there is nothing left to take away.
? ? ? ? -- Antoine de Saint Exupery
--
Lee Hart, 814 8th Ave N, Sartell MN 56377, www.sunrise-ev.com
--
This email has been checked for viruses by Avast antivirus software.
|
joshbensadon via groups.io wrote: Bob,
Thanks!? That really means a lot to me!? I just kind of thought nobody ever looks at the source code.
The part of the source code I love the most is the full duplex bit banged serial I/O.? There's been MANY bit banged I/O routines, but I've never seen a full duplex. Josh's source code is a wonderful example of how to write and document assembler. It's worth spending some time to read it with careful concentration on the details. There are many aspects that I like, but here are a few. One is his full-duplex interrupt-driven serial I/O routine. Another is his single-step routine. It can trace a program even through ROM. Yet another is that it can bit-bang an SD-card, and even figure out the byzantine PC FAT-16 director structure. It's a masterpiece! :-) Lee -- A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away. -- Antoine de Saint Exupery -- Lee Hart, 814 8th Ave N, Sartell MN 56377, www.sunrise-ev.com -- This email has been checked for viruses by Avast antivirus software.
|
joshbensadon via groups.io wrote: You've had a lot of experience on assembler, I wonder what you think about how flexible assemblers should be? To be exact, let's talk about
the Z80 here.? Forgive me if I can't remember this straight off the top
of my head, but there are some instructions that look funny. One problem is human nature; everyone has their own idea of what an assembler "should" do. Another is that manufacturers copyright their assembler mnemonics. That forces subsequent designers to invent their own. I think that's what required Zilog to come up with their own set for the Z80 instead of using/extending the Intel 8080 set. So, we wind up with every micro having a unique set of mnemonics. And half a dozens different assemblers for each CPU! Which is "better"? Whichever one you have, or like, or happen to use the most! Aren't standards great? We have so many to choose from! If the CPU runs an OS (8080/8085/Z80, 6502, 1802, PDP-8, PDP11, VAX)
I usually have a ASM native to that on the platform as well as cross assemblers.
I did the same years ago with teco compatible visual editor (Vteco, Vedit,
KED, EDT, TPU, LSE and friends) so a file could go from platform to platform and still look and edit in a similar way on a VT100 or VT220/320
terminal or PC. I prefer to spend time of the code not dealing with new
editors. Editors are another can-o-worms! Early ones that came with the OS were often terrible. Like you, I discovered VEDIT early on, and standardized on it. One nice feature is its macro language was powerful enough to automatically translate one set of mnemonics into another. I.e. it could convert 8080 assembler to Z80. It was a way to deal with the endlessly changing syntax of various assemblers. Lee -- A designer knows he has achieved perfection not when there is nothing left to add, but when there is nothing left to take away. -- Antoine de Saint Exupery -- Lee Hart, 814 8th Ave N, Sartell MN 56377, www.sunrise-ev.com -- This email has been checked for viruses by Avast antivirus software.
|
I just wanted to give a recommendation for the ASXXXX family of cross-assemblers, as I use them regularly. They are full-featured, open-source, actively updated, and work well for a very large number of chips.
-Frank
On Fri, May 14, 2021 at 12:41 PM saturn5tony via <saturn5tony= [email protected]> wrote: On Fri, May 14, 2021 at 11:23 AM, Bob Kay wrote:
Of course there are many others out there, but those are the ones I use most often.
Oh thanks Bob!
Yea I will try all the ones you mentioned. I dont have any serious time limits and I love to try out all this new stuff. I will look into that the more I build my new Z80MC. Thanks a bunch.
-Tony
|
Hi Allison,
You've had a lot of experience on assembler, I wonder what you think about how flexible assemblers should be? To be exact, let's talk about the Z80 here.? Forgive me if I can't remember this straight off the top of my head, but there are some instructions that look funny.
eg: AND? ?A, r vs AND r
I think the data sheet shows one form, but not the other.? Oh, I might be confusing this with the 8080.? I'm losing it lately.
Anyway, my question is, should an assembler treat the two instructions above as the same?? or should it complain that the code is not exactly as per data sheet?
I think it should consider both the same and not stop on a simple syntax difference.? But that might promote people not memorizing each code to the letter.
Thoughts?
Regards, Josh
On Saturday, May 15, 2021, 02:09:52 p.m. EDT, ajparent1/kb1gmx <kb1gmx@...> wrote:
I use a small list of assemblers due to the abundance of assemblers in the
world and very few using a standard or even similar syntax.
So on Z80 CPM:
DRC ASM, MAC?
ASMZ (old version from 40years ago)
Avocet ASM80 (also for 6800, 6502, and a dozen more).
?And just about all of the ones on the Walnut creek CP/M CD.
For dos/dosemu a very similar set most picked to be compatible with the CP/M kit.
(dosemu is because I mostly run linux on intel or ARM).
Same for PDP-11 under rt11 and RSX11 and VAX (OpenVMS).? I used all of those
systems and more and have a lineup of CP/M systems, PDP-8f, several PDP11
and VAX based systems to run on.? Add to that 6502, and 1802 and many others
as well.
Most of the CPUs if it runs an OS (808/8085/z80, 6502, 1802, PDP-8, PDP11, VAX)
I usually have a ASM native to that on the platform as well as cross assemblers.
I did the same years ago with teco compatible visual editor (Vteco, Vedit,
KED, EDT, TPU, LSE and friends) so a file could go from platform to platform
and still look and edit in a similar way on a VT100 or VT220/320 terminal
or PC. I prefer to spend time of the code not dealing with new editors.
In the end is more about what you like or the code if someone elses
assembles on.? Likely the native platform is PC (predominent) so thats
a fairly easy one to fine assemblers of multiple different origins for.
Allison
|
Lee Hart
Zanchius