Extended BASIC G.E.M.

[majestyx]

On 9/3/2020 at 4:44 PM, DavidC said:

Everything works here...Thank You!

Is yours working fine without having to enter the CALL LOAD lines? I'm wondering if it's an issue with either my system or my FinalGROM, based on Casey's post. I'm also using a nanoPEB so not sure if that has something to do with it.

 

EDIT: I may have stumbled upon the answer. I disconnected my nanoPEB & tried running XB 2.8 from The FinalGROM. Works fine - cyan screen with black standard text. Of course, my nanoPEB also has the 32K in it so I am unable to run the other versions like T40XB.

 

Casey (or anyone else) - could you verify your set up (PEB, nanoPEB, CF7+) and whether you are getting the same symptoms as I am? I'd  greatly appreciate it and I'm sure Harry would too, as I don't want him going on a wild goose chase for something that may only be happening for me.

Edited September 5, 2020 by majestyx

[senior_falcon]

I bet the nanoPEB is causing this.

@Majestyx: Try loading regular XB  or E/A and then CALL PEEK(-2,A,B)::PRINT A;B. 

I bet they are 86,223

Also, try putting in an actual XB, E/A or MiniMemory cartridge and CALL PEEK(-2,A,B)

 

If the nanoPEB is putting those values there, then it is important to know if changing them causes any problems. 86,223 would be >56DF which could be part of the disk system DSR.

 

 

[majestyx]

This just keeps getting stranger. On a whim, I tried a different CF card - the one that came with it which is 256 MB as opposed to the one I've been using which is 4 GB. I get different values now of 17 and 80. I have no idea what this means, but the numbers stay consistent at least. When I use the 256 MB card, the numbers are 17 & 80. When using the 4 GB card, the numbers are 86 and 223.

 

Since I have another 4 GB CF card available, I tried it as well. This time the numbers are 87 and 214. I reformatted this card, then used ti99dir 7.2a (I recalled that this was formatted with an earlier version of ti99dir) to make it CF7A+ usable. Now when I do the PEEKs, I get the same values as the other 4 GB card - 86 and 223. I repeated this once more and got the same numbers - 86 and 223.

 

I can confirm I get these values whether using the FG99 for XB 2.8 and TI XB, or a real TI XB cart. Not sure if this makes things clearer or even more confusing.

Edited September 5, 2020 by majestyx

[Casey]

I tried to attach my NanoPEB to see if I could answer a question above.  Either it no longer works or I don’t know where I’ve put the proper power supply for it, but I couldn’t get it to work.  When I reattached the TIPI, now, I get a purple screen with light blue letters, and CALL PEEK(-2,A,B) returns 93 and 1.  The only other change I made was I did an apt update to update the OS on the Pi.

[DavidC]

3 hours ago, majestyx said:

Is yours working fine without having to enter the CALL LOAD lines? I'm wondering if it's an issue with either my system or my FinalGROM, based on Casey's post. I'm also using a nanoPEB so not sure if that has something to do with it.

 

EDIT: I may have stumbled upon the answer. I disconnected my nanoPEB & tried running XB 2.8 from The FinalGROM. Works fine - cyan screen with black standard text. Of course, my nanoPEB also has the 32K in it so I am unable to run the other versions like T40XB.

 

Casey (or anyone else) - could you verify your set up (PEB, nanoPEB, CF7+) and whether you are getting the same symptoms as I am? I'd  greatly appreciate it and I'm sure Harry would too, as I don't want him going on a wild goose chase for something that may only be happening for me.

I still have to do the Call Load.  I also have a nano connected..Yes I have a nanoPEB with the 32k.  I dont know if this helps, but if I quit out and reload sometimes the screen colors are different.  Like, each time its a different color.  

 

Edited September 5, 2020 by DavidC

[senior_falcon]

There are a few addresses that might work instead of -2 and -1. But I need to know how these devices use memory because I don't want to change it to a different address that is also used by FinalGROM or TIPI.

 

(edit) Possibly >A000 could work.

Edited September 5, 2020 by senior_falcon

[+mizapf]

19 hours ago, senior_falcon said:

Majestyx reports that when choosing XB2.8 G.E.M. from the FinalGrom, memory at -2 is not zeroed out.

CALL PEEK(-2,A,B)::PRINT A;B gives the values 86 and 223.

Are those values the same for all FinalGroms?

Does anyone know the purpose of those values?

Is there any harm if they are changed?

 

The purpose of the addresses is to be part of the LOAD interrupt vector. However, I cannot say anything about the purpose of these particular values.

[senior_falcon]

Can you CALL PEEK(-8,A,B,C,D,E,F,G,H)::PRINT A;B;C;D;E;F;G;H

and report the results.

Edited September 5, 2020 by senior_falcon

[Casey]

One thing that happened that I didn’t expect. 

 

I typed CALL LOAD(-2,244,3) to change the screen colors to white on blue.  Leaving XB 2.8 with BYE and coming back in, it remembered that.  I turned off the power to 99/4A, turned it back on, and it still remembered that.  Then I remembered my Jedimatt 32K was plugged into power, so I unplugged it, plugged it back in after about 15 seconds, and XB 2.8 still remembers the color setting I made. 

 

I turned off the console, plugged in TI Extended BASIC instead of the FG99, entered in CALL PEEK(-2,A,B)::PRINT A;B and got back 244,3.

 

Where could these values be kept when the power is off?

 

If I CALL LOAD(-2,244,0) then inside TI Extended BASIC, turn the power off, put back the FG99, and go back into XB 2.8, CALL PEEK(-2,A,B)::PRINT A;B printed 244 and 0.

 

The result of CALL PEEK(-8,etc) you asked for, printed this for me:  145,0,0,16,7,240,244,0

 

[DavidC]

Fire up the TI, go into finalGROM menu and load xb2.8 GEM and I get the red screen.  I type CALL LOAD(-2,244,0) and do the CALL PEEK the numbers I get are 25, 90, 246, 244, 103, 210, 244, 0

 

I am using a nanoPEB.

 

Edit: If I power off the computer/nano the screen turns red again when I reboot, it does not remember the settings.

Edited September 5, 2020 by DavidC

[majestyx]

I feel like we are chasing ghosts here, as there doesn't seem to be much consistency in the results. I did no CALL LOADs, just the CALL PEEK & PRINT.

 

Two samples using my 256 MB card (did it 5 times):

102, 150, 152,0,117,4,17,16 (the first time)

102, 150, 152,4,117,68,17,80 (every other time)

 

Two samples using my 4 GB card (did it 5 times):

95, 198, 58, 119, 119, 0, 86, 223 (all but one time)

244, 0, 131,224,0,36,87,246 (this happened once & after getting the preceding values)

 

I'm guessing that the above values could change if I tried it enough times.

 

These results are all from cold starts using XB 2.8

Edited September 6, 2020 by majestyx

[senior_falcon]

Sure would be nice to know what is going on in those memory areas.

After changing the values in -2 and -1 have you noticed any problems with disk access? 

[senior_falcon]

Try this after starting XB:

10 FOR I=-24 TO -1::CALL LOAD(I,0)::NEXT I

This will zero out the memory from >FFE8 to >FFFF

Then see if FinalGrom and the nanoPEB work correctly.

 

Assuming all works as it should, then check to see whether memory from -24 to -1 is still set to zero.

Edited September 6, 2020 by senior_falcon

[DavidC]

Ok, this is weird. I disconnected my speech synth and plugged the nano directly in.  Boot up XB 2.8 GEM and it boots normally. Cyan screen, black letters, normal font.   I tried again with the speech hooked up....red screen, blue letters as in before.  Disconnect the speech and try again...Cyan screen as is normal....

[DavidC]

2 hours ago, senior_falcon said:

Try this after starting XB:

10 FOR I=-24 TO -1::CALL LOAD(I,0)::NEXT I

This will zero out the memory from >FFE8 to >FFFF

Then see if FinalGrom and the nanoPEB work correctly.

 

Assuming all works as it should, then check to see whether memory from -24 to -1 is still set to zero.

Yes, nano and FG seem to work fine, and everthing is still zero.   That was with speech disconnected.

 

EDIT: Seems to work fine with speech as well.

Edited September 6, 2020 by DavidC

[atrax27407]

So, does it work properly or not? 

[DavidC]

What part of "yes, nano and FG seem to work fine" confused you the most?

[atrax27407]

The word "seem" is not an unequivocal determination.

[DavidC]

Fair enough. I did not partake in a 8 hour debugging session.  I did what senior_falcon requested.   I loaded XB 2.8 GEM. I did this..

 

"Try this after starting XB:

10 FOR I=-24 TO -1::CALL LOAD(I,0)::NEXT I

This will zero out the memory from >FFE8 to >FFFF

Then see if FinalGrom and the nanoPEB work correctly.

Assuming all works as it should, then check to see whether memory from -24 to -1 is still set to zero."

 

I tried CALL CAT("DSK1.") and the disk catalog worked.   I loaded a few programs and ran them.  They worked.    I "BYEd" out of xb2.8 loaded TML it loaded normally.  I quit out of that and reloaded XB 2.8 GEM it loaded normally.   I called PEEK(-8,a,b,c,d,e,f,g,h):: PRINT a;b;c;d;e;f;g;h and got   0,0,0,0,0,0,0,0 as a result.   Therefore, everything seems to work fine.

[atrax27407]

OK

[DavidC]

When I boot up xb 2.8 without the speech it boots normally, cyan screen, black letters, normal font.  The call peek returns these: 29,214,247,20,87,213,85,1   

 

[senior_falcon]

  85 should be:

 -64=try to load Force Command on BYE

  21   2(black)*16+6(light blue)-17 = 21

So you are actually Black on Light Blue and using font #1

 

I'll probably zero out >FFE8 to >FFFF at startup. Then you have a known starting point. To use custom fonts and colors you'd have to type them in or use a LOAD program to load them. They should persist as long as you don't quit or bye. Either of those would reset it to the default font. 

There's another way that perhaps could be made to work by using an address in low memory, but this would require a bunch of rewriting.

[DavidC]

1 hour ago, senior_falcon said:

  85 should be:

 -64=try to load Force Command on BYE

  21   2(black)*16+6(light blue)-17 = 21

So you are actually Black on Light Blue and using font #1

 

One time the screen was indeed black, with light blue font.  The other 5 or six times I tried it, it was cyan and black.  

 

EDIT:  I tried again without the Speech...this time it booted up with the red screen and blue font. 

Edited September 6, 2020 by DavidC

[senior_falcon]

The number at -2 tells XB 2.8 what screen colors to use. I did make it so transparent cannot be used and so that you can't accidentally use the same foreground and background colors.

 

Can you do one other test for me. Select 2 for XB in the XB 2.8 G.E.M. cartridge. Type CALL POKEV(16111,99,99,99,99,99,99)

Use the computer to load stuff, run programs, etc. After you have wrung it out, Quit or BYE to get back to the master title screen. Press 2 again, and in XB type:

CALL PEEKV(16111,A,B,C,D,E,F)::PRINT A;B;C;D;E;F

 

(edit) actually, I only need 2 bytes, so just do 16111,A,B

The information I have for that memory location (>3EEF) is (?? not used any more was for the 99/4 ??)

If these are not used I should be able to move the color and font bytes there.

These values are retained through a Quit, at least in Classic99.

Edited September 7, 2020 by senior_falcon

[DavidC]

11 hours ago, senior_falcon said:

The number at -2 tells XB 2.8 what screen colors to use. I did make it so transparent cannot be used and so that you can't accidentally use the same foreground and background colors.

 

Can you do one other test for me. Select 2 for XB in the XB 2.8 G.E.M. cartridge. Type CALL POKEV(16111,99,99,99,99,99,99)

Use the computer to load stuff, run programs, etc. After you have wrung it out, Quit or BYE to get back to the master title screen. Press 2 again, and in XB type:

CALL PEEKV(16111,A,B,C,D,E,F)::PRINT A;B;C;D;E;F

 

(edit) actually, I only need 2 bytes, so just do 16111,A,B

The information I have for that memory location (>3EEF) is (?? not used any more was for the 99/4 ??)

If these are not used I should be able to move the color and font bytes there.

These values are retained through a Quit, at least in Classic99.

Ok, after playing a couple disk loaded games, mounting different disks into vol. 1 of the nanoPEB, saving and loading small programs, newing and reloading. I quit to the master title screen, returned to XB2.8 and the values from the PEEK are:  76 and 49.