Geneve OS development discussion

[Jeff White]

1 hour ago, InsaneMultitasker said:

With a Genmod Geneve, the byte-transfer routines are in effect for both the upgraded and non-upgraded SCSI cards.  I have never "confirmed" that this mode is still needed, though I have had no reason to think that Michael and Harald would have coded it that way if word-transfer mode was possible.  Do you think word transfers should be possible?

I think word transfers with Genmod Geneve are possible with Rev E boards (upgraded or not), upgraded F and G boards, and upgraded ASCSI.

[+9640News]

An ANSI driver has been developed installing itself as a standalone XOP (>000D) into MDOS.  The MDOS Telnet command has been expanded to either use normal "ASCII/ADM3A" commands previously available in MDOS in the TTYOUT opcode, and now detects if the ANSI driver is installed when the user requests its use.  It will provide either ANSI 80 column text, or ANSI 80 column graphics for an individual's use.

 

Right now, there are 2 opcode calls to the XOP.  Opcode 0 sets text or graphics mode returning a successful setting with version number of the driver, and the other opcode passing a string to display with an absolute maximum length of 16K as long as it does not cross more than one memory page boundary.  There is a 32 byte memory reservation up around >FFC0 due to the two levels of XOP calls.

 

I spent one evening trying to interface the MDOS TYPE command to the ANSI driver, thus far without success.  The TYPE command has quite a bit going on with DV80 files, DF128 files, and the /M command complicating coding compounded by the issue it is using additional mapped memory pages I am trying to resolve.  I'm not sure the TYPE command will be compatible with ANSI usage as I think it is not working due to the combined use of additional mapped memory pages and using 2 levels of XOP's that is verboten in MDOS.  I will likely need to confirm with MAME and the debugger to see how everything is getting mapped to know if it is resolvable.

 

Credit for the ANSI code is from @InsaneMultitasker and Mike Maksimik.

 

If someone has an immediate use for the driver, let me know.

 

Beery

 

[+InsaneMultitasker]

@dheshared this photo with me;  since the 8-bit ramdisk detects properly (in so far as the screen depicts) I will look at removing the minimum size requirement in the next release of GenCFG.  The Geneve OS is already set up to support this scenario. 

[+dhe]

The photo discussed.

 

 

[+InsaneMultitasker]

On 7/1/2021 at 2:16 PM, 9640News said:

I spent one evening trying to interface the MDOS TYPE command to the ANSI driver, thus far without success.  The TYPE command has quite a bit going on with DV80 files, DF128 files, and the /M command complicating coding compounded by the issue it is using additional mapped memory pages I am trying to resolve

The TYPE command also invokes the scrollback buffer, further complicating that routine. 

[+InsaneMultitasker]

5 hours ago, dhe said:

The photo discussed

GenCFG now allows the user to format the ramdisk as a "floppy" partition if the total available sectors <=3200.  In some cases, GenCFG will also allow you to create a ramHD partition with as few as 800 sectors, though I don't see much value in this and may increase the threshold.   I intended to release the program for testing until I remembered that the floppy bitmap allocation routine takes some liberties based on the three fixed floppy partition sizes.  I'll poke at that tomorrow.  PM me if you are available for some pre-release testing as I do not want to reformat my ramdisks unless absolutely necessary. Which reminds me, I need to back up my files!   (I back them up to TIPI and to SCSI because the latter preserves the create/modify timestamps!)

 

For giggles I also added an (I)dentify option to allow the user to turn on a card from CRU 1000-1700.  This won't work for some cards like the rs232 because I'm only setting CRU bit 0 at the card base. 

 

 

[+9640News]

On 7/1/2021 at 3:16 PM, 9640News said:

An ANSI driver has been developed installing itself as a standalone XOP (>000D) into MDOS.  The MDOS Telnet command has been expanded to either use normal "ASCII/ADM3A" commands previously available in MDOS in the TTYOUT opcode, and now detects if the ANSI driver is installed when the user requests its use.  It will provide either ANSI 80 column text, or ANSI 80 column graphics for an individual's use.

 

Just a FYI for anyone following the thread.  I need to do some more testing as I encountered a versatility issue with the driver that it does not like certain workspace and buffer locations.  Thought I had it all worked out, but apparently not.

[+dhe]

I see to be having problem with Geneve OS.

I copied the new GDM2K to floppy a: - where they do run.

 

I then tried to copy from A: H:( SCSI2. )

   See g_copy.jpg

Copy said it went ok.

    See g_dir.jpg

 

But, when I try to run the files, I get an Error.

    See g_error.jpg

 

 

[+dhe]

I think I have it all squared away.

The trick seems to be, make sure you have your SCSMAP, REMAP and ASSIGNS all perfect before you start doing DIR {something}.

[+InsaneMultitasker]

I have posted source to GenCFG to the Horizon Ramdisk github repository.

 

Be advised that  @dhe and @9640News are helping me to track down what appears to be a problem with the detection routine that manifests only when a CorComp floppy controller is installed. 

[globeron]

GENCFG

1st time Corcomp controller as "undefinable"

When doing a S)can it sees Floppy controller

 

It also does not see my Rave99 card with Speech Synth. Installed.  

 

 

 

 

[+InsaneMultitasker]

59 minutes ago, globeron said:

1st time Corcomp controller as "undefinable"

When doing a S)can it sees Floppy controller

 

It also does not see my Rave99 card with Speech Synth. Installed. 

The re-scan result is an interesting development and may be a clue.  If you rescan again, does the card status return to 'undefinable'?  Without a Corcomp FDC at my disposal (and one that exhibits the issue at hand)  I have been unable to determine the root cause. 

 

The RAVE / speech does not have an associated CRU address (as would a floppy controller, ramdisk, etc) so it will not show up in the scan.  If there is a simple detection method, I am willing to try to incorporate the code into GenCFG. 

[globeron]

On 3/15/2021 at 10:01 AM, InsaneMultitasker said:

GENCFG lives just in time for PI Day --  8MB Horizon ramdisk successfully partitioned and formatted.  

A bit more testing is necessary before I replace the hard-coded sector IO/partition routines within the Geneve OS.  

 

      

 

 

 

How can I get GENCFG 1.3  (not sure how to do it via Github and the distribution is ver 1.2)

 

 

With Gencfg 1.2  I formatted the HRD 3000 on CRU >1400,  I can do the options for RAMDSK and RAMHD,

but it indicates as Horizon 16 bit with different memory allocation. 

 

(My HRD3000 is installed in my Geneve9640 system now, but in this video is the card)

I have this HRD3000 and think with RAMBO installed (the daughterboard at the bottom).

 

 

These are picts:

 

I am having some device errors (after format my A: is device error, this is a gotek usb drive). But after reboot from gotek mdos 7.3 then e: gives device error.    

 

With Form123 i do not have these issues and the system/sys is also copied as part of the process.

 

Not sure I am clear what the difference is between RamDSK (bootable and intention to put files on it) and RamHD (not too sure how to access it)

 

 

 

 

 

 

 

Edited August 22, 2021 by globeron
GENCFG 1.3

[globeron]

10 hours ago, InsaneMultitasker said:

The re-scan result is an interesting development and may be a clue.  If you rescan again, does the card status return to 'undefinable'?  Without a Corcomp FDC at my disposal (and one that exhibits the issue at hand)  I have been unable to determine the root cause. 

 

The RAVE / speech does not have an associated CRU address (as would a floppy controller, ramdisk, etc) so it will not show up in the scan.  If there is a simple detection method, I am willing to try to incorporate the code into GenCFG. 

 

The rescan stays to floppy controller status.

 

For the Rave not sure how to detect it. Maybe search if there is vocabulary of words? Not sure if there is a way to detect Speech.

 

 

 

[+InsaneMultitasker]

19 minutes ago, globeron said:

I am having some device errors (after format my A: is device error, this is a gotek usb drive). But after reboot from gotek mdos 7.3 then e: gives device error.    

 

With Form123 i do not have these issues and the system/sys is also copied as part of the process.

 

Not sure I am clear what the difference is between RamDSK (bootable and intention to put files on it) and RamHD (not too sure how to access it)

The A: device error is related to the CorComp card detection problem. A cold or warm restart is required. 

 

E: is not re-assigned to DSK6 unless you place the ASSIGN command in your autoexec. Be sure your autoexec contains all necessary ASSIGN, REMAP, and other commands to configure the OS for your unique system.

 

Note: The OS allows you to type "DIR DSK6."  (all caps) to catalog a device, even if you haven't assigned a drive letter.

 

Form123 is limited to copying system/sys from a floppy disk device.   If you intend to boot from the ramdisk, copy the system file via the command line or use a program such as Directory Manager.  Form123 will exhibit the same issue with E: and ASSIGN.

 

RamHD is accessed using a "SCSx." device name/number.  The 'x' device number is mapped with SCSMAP. 

 

While sparse, the GenCFG documentation does refer to the remap and scsmap commands, and contains a simple overview and steps for formatting the ramdisks. 

 

 

 

[globeron]

On 3/15/2021 at 11:48 PM, Nick99 said:

I've seen someone else boot with a Myart picture in another thread, don't remember in what thread.

You need to convert from gif to Myart and the best program so far is Picture Transfer (MDOS program). 

PictureTransfer.dsk

90 kB · 22 downloads

 

I tried the picture transfer, but it cannot convert between 16 colours and 256 colours. 

 

I only got the Dragon to load (16 colours)

 

The picture loader as part of the SECTORONE disk has a loader, but it can only load 16 colour pictures (like HDRAGON,  but not the SWAN)

unless someone can change the tool to load 256 colours, the assembly source code is on the disk of the loader.

 

http://ftp.whtech.com/Geneve/9640 News disks v1-3/pc99 disks/9640V2%234.DSK

SECTORONE batch file has DISP as a loader for MYART file type "TITLE" (but only for low resolutions, so I still cannot load the SWAN)

but the concept is okay, load a picture, either wait a few seconds or press a button and continue in the .bat file)

 

 

 

 

[globeron]

19 hours ago, InsaneMultitasker said:

The A: device error is related to the CorComp card detection problem. A cold or warm restart is required. 

 

E: is not re-assigned to DSK6 unless you place the ASSIGN command in your autoexec. Be sure your autoexec contains all necessary ASSIGN, REMAP, and other commands to configure the OS for your unique system.

 

Note: The OS allows you to type "DIR DSK6."  (all caps) to catalog a device, even if you haven't assigned a drive letter.

 

Form123 is limited to copying system/sys from a floppy disk device.   If you intend to boot from the ramdisk, copy the system file via the command line or use a program such as Directory Manager.  Form123 will exhibit the same issue with E: and ASSIGN.

 

RamHD is accessed using a "SCSx." device name/number.  The 'x' device number is mapped with SCSMAP. 

 

While sparse, the GenCFG documentation does refer to the remap and scsmap commands, and contains a simple overview and steps for formatting the ramdisks. 

 

 

 

 

(I forgot) after the ASSIGN E=DSK6: it works now.

 

Gencfg (after 1st run) and quit, gives the device A: error, after the reboot it is okay to access A:

 

Question: How do I map and access the RAMHD (Harddisk) allocation on the HRD3000?

Found it add in autoexec. For HRD on CRU  >1400

 

SCSMAP 48

ASSIGN F=SCS4:

 

(and to use in GPL to load modules SCS4.filename (or SCS4.directory.filename)

 

 

 

Edited August 23, 2021 by globeron

[globeron]

(Not sure for this forum)

But i saw some differences in GPL.

 

MDOS 7.30   GPL and MYGPL

use lowercase characters. Parsec does not display the full character. 

 

Press tire....to begin.

 

Also GPL shows 2x meteroites on the right top corner when crashing, but MYGPL and GPL650 do not.

 

Furthermore Ships, LIFT, Top, arrow are displayed in blue colour.  Gpl650 shows White.

 

 

 

 

 

[+InsaneMultitasker]

5 hours ago, globeron said:

Question: How do I map and access the RAMHD (Harddisk) allocation on the HRD3000?

Found it add in autoexec. For HRD on CRU  >1400

 

SCSMAP 48

ASSIGN F=SCS4:

Good work, that's the correct approach

 

By the way, your questions will get more visibility and would be better served in the MDOS 7.30 release topic. 

 

Your Parsec pictures look like a likely hardware (video RAM) issue based on some of your other posts --Edit:   that's just a quick assessment and not based on any testing.  Post the files in the other thread and maybe someone else can try your files on their system. Stranger things have happened.  Not a hardware issue, see next post.

[+InsaneMultitasker]

2 hours ago, globeron said:

Also GPL shows 2x meteroites on the right top corner when crashing, but MYGPL and GPL650 do not.

 

Furthermore Ships, LIFT, Top, arrow are displayed in blue colour.  Gpl650 shows White.

I had a chance to load GPL 7.30 with a PARSEC cart dump from my Geneve folder; 'tire' and 'begin' are wrong and that big "T" stands out.   My copy of GPL 5.00 exhibits the same patterns and colors.  I found an old GPL version 1.04 that did the same thing and another 1.04 that shows the uppercase lower character set but the ships, lift, top and fuel, and arrow are blue.  When I crash, I get 1-2 ships or asteroids.  Speed 5 is WAY too fast. 

 

Please post your various versions in the 7.30 thread and I'll compare to my files as time permits. I don't know that I ever played Parsec on the Geneve - there may have been good/bad cart dumps.  Still conjecture but I wanted to at least let you know it's not your hardware.

 

[globeron]

@InsaneMultitasker 

 

With GENCFG is it possible to format the complete Horizon disk to a RAM-DSK format?  (or are there max. size limitations?)

because then I can do the mapping to a DSKx device. 

 

If I am correct RAM-HD can handle a larger disk size.  (but cannot do the DSKx device mapping).

 

[+9640News]

9 hours ago, globeron said:

@InsaneMultitasker 

 

With GENCFG is it possible to format the complete Horizon disk to a RAM-DSK format?  (or are there max. size limitations?)

because then I can do the mapping to a DSKx device. 

 

If I am correct RAM-HD can handle a larger disk size.  (but cannot do the DSKx device mapping).

 

You are limited to I believe 800K for a RAM-DSK with 3 subdirectories using REMAP.  Any extra memory beyond that is not useable.  If your HRD is larger than 800K, then to have a single device using all the memory, map it as a RAM-HD style with SCSMAP.

 

Beery

[globeron]

On 2/21/2021 at 5:40 AM, InsaneMultitasker said:

Playing around with some ramdisk-related code based on the /4a horizon ramdisk CFG routines. 

Short-term goal is to show the user the peripherals and current ramdisk format status.  From there I'll incorporate the "ram hard drive" routines so that I can retire the old FORM program.

 

 

Hi @InsaneMultitasker 

 

I was wondering how do you get the >1600 Horizon/RAMBO  displayed? 

(because mine only shows the Horizon 16-bit, but it has the 504K RAMBO installed or maybe it is not detected in Gencfg? 

(on a TI with ROS 8.34 I can detect it)

 

 

[+9640News]

1 hour ago, globeron said:

 

Hi @InsaneMultitasker 

 

I was wondering how do you get the >1600 Horizon/RAMBO  displayed? 

(because mine only shows the Horizon 16-bit, but it has the 504K RAMBO installed or maybe it is not detected in Gencfg? 

(on a TI with ROS 8.34 I can detect it)

 

 

@InsaneMultitasker answered elsewhere.  RAMBO memory is not available on the Geneve, only on the TI-99/4A.

[+InsaneMultitasker]

With many thanks to @9640News , I have been able to track down the GenCFG / CorComp controller card problem.

 

The CorComp floppy controller's DSR EPROM has two banks controlled by CRU bit 11 (base >1100). 

 

When the Geneve OS performs its initial powerup, it uses the CRU bit to distinguish between a TI card (one ROM bank) or CorComp card (two ROM banks).  Unfortunately for us, the OS does not restore the CRU bit to its default state rendering the DSR ROM header as no longer visible.  This error explains why GenCFG cannot identify the CorComp card during its first pass; it might also be a contributor to rompage issues experienced with the CorComp card.

 

The reason the card is detected the second time is because during the first pass, the ramdisk detection routine is triggered (when the DSR header can't be found) and CRU 11 is reset (to zero), making the DSR visible for the next pass.  Alas, the ramdisk detection routine also flips other CRU bits rendering the current drive unusable until a reset is performed OR until another drive is accessed.

 

As best I can tell, this segment of code is where the offense is committed and where we need to properly reset the ROM bank during the OS powerup.

 

File:  DSRPASS-S 

NOTMYR MOV  @H9FF0,@CTLBAS
       MOVB @H00+>2000,@CTLTYP
       SBO  11     ;set to page ROM bank 1
       CB   @>C000,@HAA+>2000
       JEQ  PWREX             it's a ti card (use 40 track, step setting=2)
**
; The fix:  add "SBZ 11" here!

-------

I don't want to add this 'fix' to GenCFG so for now, I'll create a 8-12 byte program to reset bit 11.  Once the OS is updated, the reset program will be unnecessary.