Page 2 of 2

Re: 32k Internal ram with m/c above 32k

Posted: Sat Sep 18, 2021 12:03 pm
by DereX81
Thanks for clearing that up Siggi :)

Does anyone know where I can find a test program that is bigger than 16k ?

Re: 32k Internal ram with m/c above 32k

Posted: Sat Sep 18, 2021 4:45 pm
by siggi
I don't think, that there is any usual BASIC program > 16K.

I needed to disable m/c above 32K, when I once compiled CHOPPERDROP3000, using MCODER II. The Basic source was < 16K, but during compiler run, the generated code was inserted into program line 2 (or 1?), which increased the size of the program to > 16K (display file above 32K). After deleting the compiled source code, the remaining (compiled) program then was again < 16K and could be run on any 16K Zeddy. But for compiling, more ram is necessary ...

Siggi

PS: Some of Wilf's programs need more than 16K, but use it for data, not BASIC source.

Re: 32k Internal ram with m/c above 32k

Posted: Sat Sep 18, 2021 6:38 pm
by GCHarder
"The Crystal Frog" requires 32K

viewtopic.php?t=334

Re: 32k Internal ram with m/c above 32k

Posted: Tue Sep 21, 2021 7:00 pm
by kmurta
DereX81 wrote: Sat Sep 18, 2021 12:03 pm Does anyone know where I can find a test program that is bigger than 16k ?
Super Star Trek is in BASIC and requires more than 16Kb:

https://sinclairzxworld.com/viewtopic.p ... 849#p36849

Re: 32k Internal ram with m/c above 32k

Posted: Mon Sep 27, 2021 9:46 pm
by DereX81
Thanks for all the feedback,

I managed to test the ram with the m1not circuit activated and de-acativated here are the results:

1. m1not de-activated
2. Poked 16389 with 255, then new, peeked 16389 and got 192, as expected ramtop now 49152
3. Made a REM line with 512 bytes, copied this all the way to 31000
4. Kept checking the end of the d_file (VARS) using peek of 16400 and 16401
5. Modifed the first REM line to 832 bytes (larger than d_file)
6. Added a couple of REM lines to get end of d_file to 32767
7. Copied the first REM line (832b), then enter
8. Whoopee ! we're past the 32k boundary
9. Carried on copying 512b REM lines all the way to 48000

10. m1not activated
11. Followed steps 2 to 6
12. After step 7 the ZX81 crashed just as siggi described :), sorry gozzo :(

I used the same process to test the vLA81 in mode 1 (16-48k), without the m1not circuit, I was able to use m/c above 32768, but not able to push basic programs past the 32k boundary