Quote  

Is it possible to introduce some kind of check for the presence of the PSRAM at startup

There is but without the PSRAM CS pulled up the PSRAM will be corrupting flash access so the code simply doesn't run. Nothing the firmware can do about this.

Peter,

Thank you for the many improvements.
It works.

RP2040-Geek.pdf

Greetings,
Jan.

About PSRAM,

Although technically PSRAM can be attached to a RP2040, it is the RP2350 that is prepared for the use of PSRAM by means of a hardware interface. That hardware interface is shared with FLASH.

Early tests with Pimoroni Pico Plus 2 modules has shown that PSRAM works, and MMBasic implementation made the use of PSRAM transparent (not a RAM DISK or similar, but a large amount of data memory that can be used in conunction with built in RAM).

PSRAM comes with a disadvantage: speed.
1/ The PSRAM as such is nt as fast as the built in RAM. So prepare for slower access (execution) of your code. PSRAM is cached, and as long as you are using the cache, it is faster, but once you need a cache refresh, it is slower. The cache is 16kbyte in size.

2/ Because the PSRAM shares the bus with FLASH, the FLASH program cannot execute as fast.

3/ RP2350 has one big benefit over the RP2040, and that is the HDMI (DVI) interface. To use HDMI, it is required to "overclock" the RP2350 chip. At the moment the RP2350 equipped with PSRAM cannot be overclocked as much as a non-PSRAM module. I am not aware the exact cause, but fact is that by using PSRAM, HDMI may only be usable in 640x480, or not at all.

I absence of facts, I believe it is because of the extra chip (FLASH + extra PSRAM) on the QSPI bus. If this is the case, switching off the PSRAM by disconnecting the chip select pin (CSn) will still not allow you to overclock the RP2350 to HDMI speeds.

Summary: PSRAM makes the Picomite execute slower, PSRAM has impacton HDMI. There may be applications that require much RAM, and need PSRAM, be happy.. it is supported.

Volhout

EDIT: in the future it may be possible that a solution is found to solve the overclock issue. But the data access in PSRAM will per defintion always be slower.
Edited 2024-12-19 17:54 by Volhout

@Peter
I'm observing a strange behavior of the "List" command.
If the file has no file extension, then it is not found.
(Note: I think I've experienced the same issue with the "Open" command. But I can't reproduce it.)
isbox_raw.txt and isbox_raw1 have the same content.

If the file is zero bytes in size and I try a >List "filename"< the system freezes.
Maybe you want to take a look at it.



Screenshot:

  Quote  

> files
B:/test
00:01 01-01-2024        832  isbox_raw.txt
00:01 01-01-2024        832  isbox_raw1
00:02 01-01-2024          0  isboxraw.txt
0 directories, 3 files
> list"isbox_raw.txt"
00 52 09 6a d5 30 36 a5 38 bf 40 a3 9e 81 f3 d7 fb
10 7c e3 39 82 9b 2f ff 87 34 8e 43 44 c4 de e9 cb
20 54 7b 94 32 a6 c2 23 3d ee 4c 95 0b 42 fa c3 4e
30 08 2e a1 66 28 d9 24 b2 76 5b a2 49 6d 8b d1 25
40 72 f8 f6 64 86 68 98 16 d4 a4 5c cc 5d 65 b6 92
50 6c 70 48 50 fd ed b9 da 5e 15 46 57 a7 8d 9d 84
60 90 d8 ab 00 8c bc d3 0a f7 e4 58 05 b8 b3 45 06
70 d0 2c 1e 8f ca 3f 0f 02 c1 af bd 03 01 13 8a 6b
80 3a 91 11 41 4f 67 dc ea 97 f2 cf ce f0 b4 e6 73
90 96 ac 74 22 e7 ad 35 85 e2 f9 37 e8 1c 75 df 6e
a0 47 f1 1a 71 1d 29 c5 89 6f b7 62 0e aa 18 be 1b
b0 fc 56 3e 4b c6 d2 79 20 9a db c0 fe 78 cd 5a f4
c0 1f dd a8 33 88 07 c7 31 b1 12 10 59 27 80 ec 5f
d0 60 51 7f a9 19 b5 4a 0d 2d e5 7a 9f 93 c9 9c ef
e0 a0 e0 3b 4d ae 2a f5 b0 c8 eb bb 3c 83 53 99 61
f0 17 2b 04 7e ba 77 d6 26 e1 69 14 63 55 21 0c 7d
> list"isbox_raw1"
Error : Could not find the file
> list"isboxraw.txt"
(the system freezes)

Another thing: I recently started having automatic ".bak" files, like with CMM2. Is that new, can it be turned off?

Kind regards
Michael
Edited 2024-12-19 21:02 by twofingers

  twofingers said  

can it be turned off?
Michael

Why ? It is a feature, not a bug...
After you have finished editting, type KILL "*.bak"

Volhout
Edited 2024-12-19 21:11 by Volhout

  Volhout said  

... Why ? It is a feature, not a bug...

Hi Harm,
I see it differently. It was introduced on the CMM2 because there was a temporary lack of reliability when saving files. It was retained. It seems to be new on the Pico. I'll take a look at the manual...
I found it: Page 106

  Quote  

If editing an existing file a backup with .bak appended to the
filename is also created on exit.

()
Apart from that, I find the behavior rather annoying.
Kind Regards
Michael
Edited 2024-12-19 21:23 by twofingers

I'll fix list so you can list files without an extension - currently appends .bas. Can't replicate any lockup with empty files. I'm not changing .bak. One day you will need it after saving a file and then realising you made a big mistake and then you would ask for it.

Hi Twofingers,

It is new. It did not exist in the previous 5.09.00 series. And given the fact that the A:/ drive is quite small in size (600-800kbyte) it could present an issue when there is not room to put the .bak file. In that case it could cause more harm. In case of the Game*Mite where the A:/ drive is nearly full, editting is not possible anymore.

I think it is worth testing exactly that problem case...

Volhout
Edited 2024-12-19 21:36 by Volhout

It used to happen in DOS too. :) In fact, most commercial programs create BAK files of some sort. SL6 does, as does TinyCAD not to mention the proper stuff like AutoCAD. There are very few reasons not to produce them, apart from them looking a bit untidy.

As Harm says, it's easy enough to delete them after. They can be very useful if you mess something up though! Rename your working file, rename the BAK file to BAS and load it.

  Quote  

In case of the Game*Mite where the A:/ drive is nearly full, editting is not possible anymore.

let's be clear you ONLY get a .bak file when using EDIT fname$. Editing a loaded file does not save to disk.

  matherp said  

... Can't replicate any lockup with empty files.

Tested several times ...
Tested again and I can now go back to the prompt with CTRL-C.

  matherp said  

...  I'm not changing .bak. One day you will need it after saving a file and then realising you made a big mistake and then you would ask for it.

Well, in my workflow I save extremely frequently, so backup files are unnecessary.
But I can live with it.
Regards
Michael

@Peter
You write in your comment on the new AES function:

  Quote  

The maximum number of elements in 'in' and 'out' is 256 when defined as arrays or 128 if one or other is defined as a string

What are the reasons for the restriction to 128 characters when 'in' and 'out' are defined as strings? In my tests, more than 128 seem to be feasible.
Kind regards
Michael

Its not 128 - brain fade - it is 240. The point is that it is not 256

  Quote  

If editing an existing file a backup with .bak appended to the
filename is also created on exit.

()

is this behaviour for all variants as I cannot reproduce it onWebmite V6.00.01 RC9. Editing a .bas file does not create a .bak file.

If you are editing a file that's already in flash then no BAK file is created. It's only if you use EDIT <filename>.Is that what you are doing?

Anyone running the I2C @ >100KHz? Manual still recommends 100KHz.

By default OPTION SYSTEM I2C runs the I2C at 400KHz. You can override this by using
OPTION SYSTEM I2C clkpin,sdapin [FAST/SLOW]
This is in the manual.
Edited 2024-12-20 04:37 by matherp

  Mixtel90 said  

If you are editing a file that's already in flash then no BAK file is created. It's only if you use EDIT <filename>.Is that what you are doing?

yes, correct

Oh, I don't know then. :)  It's something new so I don't know if it's in that particular version.

According to Michaels post it should be built in beginning with V6.00.01RC9

-------snip -----------

-------snip -----------

Another thing: I recently started having automatic ".bak" files, like with CMM2. Is that new, can it be turned off?

-------snip -----------

so i thought it should apply to all variants. Can EDIT "filename.bas" without problems, but no .bak file.