MiSTer PCXT

[wark91]

Sorry ... !

[akeley]

Alright, I've dragged my 31KHz monitor out. The core displays fine out of the box, just need to change Aspect Ratio to Full Screen as per usual.

IMG-20220731-213721.jpg (7.27 MiB) Viewed 8515 times

[MicroCoreLabs]

Archon is my favorite game, and one of the first programs I try when testing a new core! I enjoy watching the demo to see who wins: The computer or the computer!

[Mills]

Thanks for your input so far. I had no success with the CBM PC-I, but I got the CBM PC-10 BIOS to boot after padding the file with 32kb of zeros at the beginning:
pcxt-pc10.jpg

So that was half a success. I learned that this is not a good idea at all and will stop further approaches.

I tried that bios and I inserted the xtide at the first 12kb, it boots and tells me to press f1, then it should start xtide and load the img, But keyboard is not working .

[jordi]

Thanks for your input so far. I had no success with the CBM PC-I, but I got the CBM PC-10 BIOS to boot after padding the file with 32kb of zeros at the beginning:
pcxt-pc10.jpg

So that was half a success. I learned that this is not a good idea at all and will stop further approaches.

I tried that bios and I inserted the xtide at the first 12kb, it boots and tells me to press f1, then it should start xtide and load the img, But keyboard is not working .

I guess that computer doesn't have an special feature like Tandy has, so better use 86box for it as commented before.

[Mills]

@spark2k06
Timing constraint violations have been overlooked for a long time.
I have a suggestion to resolve the timing constraint violation.
I have considered two methods. You can choose one or the other.

1. Change the chipset frequency to 50 MHz for the PLL output.
https://github.com/kitune-san/PCXT_MiST ... hz-chipset

2. Insert F/Fs between combination circuits that cause delays. (E.g. before and after the RAM block)
https://github.com/kitune-san/PCXT_MiST ... adjustment

I compiled both versions.
1. The 50mhz one, worked as the previous versions, and landmark test showed a 4.99mhz 8088 . There are still some volume issues with the opl2 chip.

2.This second version has an issue because pc speaker and opl2 produce distorted noises.

[spark2k06]

I have updated the prebeta 2.0 branch with all the changes proposed by @newsdee regarding the video, plus those proposed by @kitune-san with the 50Mhz chipset, which is the one I have observed to be more stable:

https://github.com/spark2k06/PCXT_MiSTe ... rebeta-2.0

I also update the turbo version but here the SDRAM is still unstable...

https://github.com/spark2k06/PCXT_MiSTe ... -2.0-turbo

...so I suggest to keep using the prebeta 1.5 turbo version, release 04:

viewtopic.php?p=56624#p56624

Also remember that in order to use the OSD's integrated disk image selection option, you need to use the modified Main_MiSTer with integrated serdrive:

viewtopic.php?p=56797#p56797

It is not updated with the modifications of the latest official version, but you can see the changes here:

https://github.com/spark2k06/Main_MiSTe ... a2b5c41dc8

[spark2k06]

prebeta 2.0 branch update:

Overscan removal

• Missing lines are recovered on the right in CGA

https://github.com/spark2k06/PCXT_MiSTe ... a9e86f9ad5

The ghost line problem in Tandy 320x200x16 mode, a common problem in the original Graphics Gremlin project on which it is based, has not yet been solved.

Below is a video of the Area 5150 demo run in this PCXT project for MiSTer:

I will toss my hat in and say I'm working on adding true IDE support compatible with XT-IDE to this. VHDL/Verilog I understand, the MiSTer framework for accessing data I'm still reading through...

Have you made any progress on this, I haven't wanted to make any improvements to the serdrive integrated in Main yet because I was waiting to see if IDE was available, and it would be work that would be discarded.

[dontbugster]

Doesn't this provided some interface to it?
https://opencores.org/projects/sdcard_m ... controller

[spark2k06]

Doesn't this provided some interface to it?
https://opencores.org/projects/sdcard_m ... controller

I have seen several controllers, this one too. However, I have not found a way to make it "fit" into this project.

[spark2k06]

prebeta 2.0 branch update:

Setting timings for tandy_16_gfx mode

https://github.com/spark2k06/PCXT_MiSTe ... 2a53b69829

No more ghost lines in Tandy's 320x200x16 mode

no_ghost_lines_tandy.jpg (153.62 KiB) Viewed 7702 times

Soon, I will revise the latest stable version of the 1.5 turbo prebeta, to add this improvement.

[somhi]

Great improvements @spark2k06!!!
I'm adapting those to Sockit (MiSTer) and NeptUNO (DeMiSTify).

By the way, congratulations to Kitune-San. Last changes in prebeta 2.0 reduces considerably the worst case slack timings.

[MicroCoreLabs]

Below is a video of the Area 5150 demo run in this PCXT project for MiSTer:

I'm happy to see that the MCL86 is able to run this new demo for the most part.

[wark91]

Thanks ! It is great !

[somhi]

I'm having good vibrations on the new core since prebeta 2.0 changes which led to more relaxed timings.

Now on a Cyclone IV FPGA I'm getting around -9ns worst case slack in comparison to -18ns I had before.
Before I couldn't even load the original IBM BIOS as the keyboard was not working. Now working Ok.
Compilation time has also reduced from more than 12 to 8 minutes.

This must be changed in a prior beta, but just noticed now the increase in the UART clock. Now I can also use serdrive at 460.8 in NeptUNO FPGA.

[spark2k06]

I have created a new branch based on the prebeta-2.0-turbo branch, in this case I have named it prebeta-2.0-turbo-stable:

https://github.com/spark2k06/PCXT_MiSTe ... rbo-stable

I've just removed the last 50Mhz chipset and SDRAM related improvements made by kitune-san, so that we get the stability back and at the same time take advantage of the latest Tandy mode display improvements.

I attach the binary so you can use it, I recommend to set the frequency to 14.318Mhz and save the configuration, as I remember that at 4.77Mhz in this version the UART does not work at 460.8Kbps, having to lower it to half:

tandy_14_318_mhz.jpg (164.45 KiB) Viewed 7432 times

[kitune-san]

Hi. Long time no see.

I have improved the READY signal.
This results in a clock cycle equivalent to PC/XT.

[AmintaMister]

Hi all!

I've tried to automatically mount the serial drive at MiSTer startup with:

Code: Select all

# Startup PC-XT serdrive
[[ -e /media/fat/Scripts/pcxt_uart_hdd.sh ]] && /media/fat/Scripts/pcxt_uart_hdd.sh

in "linux/user-startup.sh" but it doesn't work, why?

[bbond007]

Hi all!

I've tried to automatically mount the serial drive at MiSTer startup with:

Code: Select all

# Startup PC-XT serdrive
[[ -e /media/fat/Scripts/pcxt_uart_hdd.sh ]] && /media/fat/Scripts/pcxt_uart_hdd.sh

in "linux/user-startup.sh" but it doesn't work, why?

I have no idea why it does not work, but In my opinion that is not the ideal hack as it will conflict with all other cores UART functionality.

I'd suggest editing /sbin/uartmode and altering the script:

Code: Select all

#!/bin/bash

kill_all() {
        rm -f /tmp/uartmode*
        rm -f /tmp/ML_BAUD
        rm -f /tmp/PCXT_SERDRIVE
        killall midilink
        killall pppd
        killall agetty
        killall login
        killall mt32d
        killall fluidsynth
        killall mpg123
        killall serdrive
}

conn_speed=115200
[ -f /tmp/UART_SPEED ] && conn_speed=$(cat /tmp/UART_SPEED)
[ -f /tmp/CORENAME ] && core_name=$(cat /tmp/CORENAME)

if [ "$core_name" = "PCXT" ]; then
        kill_all
        sleep 1
        echo "1" >/tmp/PCXT_SERDRIVE
        taskset 1 /media/fat/games/PCXT/serdrive -c 1 -b 460.8Kb /media/fat/games/PCXT/hdd.img &
        exit 0
fi

echo "ttyS1: "$conn_speed

This fix will (hopefully) give you the results you want and (probably) should not screw up the UART functionality of other computer cores...

uartmode.zip

(923 Bytes) Downloaded 216 times

[spark2k06]

New prebeta 2.0 updates:

• Fix READY signals, by @kitune-san
• Memory map selection according to video mode

Thanks to this change, some parts of the area5150 demo that didn't look right, like the dancing elephant, are now displayed correctly:

area5150_elephant_before.jpg (101.19 KiB) Viewed 6928 times

area5150_elephant_after.jpg (80.54 KiB) Viewed 6928 times

Thanks @somhi for noticing this detail.

On the other hand, I have extended the serdrive feature integrated in the MiSTer Main:

Main_MiSTer_PREBETA_2_0_02.jpg (104.98 KiB) Viewed 6928 times

Hi all!

I've tried to automatically mount the serial drive at MiSTer startup with:

Code: Select all

# Startup PC-XT serdrive
[[ -e /media/fat/Scripts/pcxt_uart_hdd.sh ]] && /media/fat/Scripts/pcxt_uart_hdd.sh

in "linux/user-startup.sh" but it doesn't work, why?

Now, it is not only possible to load an HDD image, but it can also be saved in the settings, so you don't have to worry about the script, because the feature is integrated in Main

Lines must be added to the MiSTer.ini configuration file:

Code: Select all

[PCXT]

uart_baudrate=460800   ; 115200, 230400 or 460800
uart_hdd_cyl=62
uart_hdd_sect=63
uart_hdd_head=16

As long as we don't have a real IDE on XTIDE, this is the most friendly solution... however, before I make a pull request to sorgelig to bring it to the official Main, there are still some things to do.

We still need to be able to add FDD image for serdrive, and we would have to rework these changes to make them cleaner, I don't have a deep knowledge of the Main project either, maybe you can give me a hand here, @bbond007:

https://github.com/spark2k06/Main_MiSTe ... t_serdrive

For example, right now when selecting the image, although it is applied, it hides the menu, and it would be nice to see how it has been applied keeping the menu active.

In addition, the code has several warnings that need to be checked. And I would also like to replace F12 as Win+F12 to access the menu, like ao486 does... I have to check how that works.

Once all these problems are solved and the pull request from Main is accepted by sorgelig, I hope to make this prebeta 2.0 official as beta 2.0 as well, in order to reach more users.

[AmintaMister]

Thanks a lot @bbond007 and @spark2k06!

[Malor]

If I understood sorgelig correctly, he won't accept this core until the IDE controller works. He seemed pretty horrified by the serdrive idea.

[spark2k06]

If I understood sorgelig correctly, he won't accept this core until the IDE controller works. He seemed pretty horrified by the serdrive idea.

Well integrated in the Main, except that it will take up a bit more than usual, I don't see what the drawback could be, the serdrive would only be active when the core running... for other cores, it wouldn't have any repercussions.

It is clear that the ideal is real IDE, but as long as we don't have it (it seems easy to implement, but it is not), it is the best we have at the moment.

[dmckean]

If I understood sorgelig correctly, he won't accept this core until the IDE controller works. He seemed pretty horrified by the serdrive idea.

Well integrated in the Main, except that it will take up a bit more than usual, I don't see what the drawback could be, the serdrive would only be active when the core running... for other cores, it wouldn't have any repercussions.

It is clear that the ideal is real IDE, but as long as we don't have it (it seems easy to implement, but it is not), it is the best we have at the moment.

Have you looked into his suggestion of trying to integrate the floppy drive from the Atari ST into this core?

[spark2k06]

Have you looked into his suggestion of trying to integrate the floppy drive from the Atari ST into this core?

I have already looked at several projects, yes, that one too. They're all proving to be a real challenge for me, unfortunately... one day we'll have some integrated, for sure, but at least for the moment they're proving to be unattainable for me.

[dmckean]

Have you looked into his suggestion of trying to integrate the floppy drive from the Atari ST into this core?

I have already looked at several projects, yes, that one too. They're all proving to be a real challenge for me, unfortunately... one day we'll have some integrated, for sure, but at least for the moment they're proving to be unattainable for me.

Floppy support is probably the quickest way to get the core accepted into Main as IDE will most likely take longer. Unfortunately, it's proven difficult to recruit anyone to help with it.

[Malor]

I'd help if I could, but I barely know a resistor from a capacitor.

The Amiga/Minimig core has two-controller, four-channel IDE emulation you might be able to poach. I think it may be pretty complete, because the virtual drives on Minimig work very well.

That's probably all 16-bit, though, and adapting it to run on an 8-bit bus might be painful.

[spark2k06]

New Main with floppy disk loading support via integrated serdrive:

boot_b.jpg (84.08 KiB) Viewed 6569 times

To boot by floppy disk, you must press the 'B' key... loads booters without problems, here is an interesting website with many booters games.

http://retrograde.inf.ua/index2.htm

[akeley]

Great news about the FDD support. Look for "gruby ibm" on archive.org, it's probably the most complete MSDOS floppy collection out there (in various formats).

[suww37]

Thank you . Now I think the most important thing left is mouse support.