SWUtils (SimCheck, SimTest, OSDump)

[Panther]

Time to revisit GTIA/CTIA detection again!  The timings seem to change easily, I'll need to figure out a more reliable method.

 

[+Stephen]

5 hours ago, Panther said:

Time to revisit GTIA/CTIA detection again!  The timings seem to change easily, I'll need to figure out a more reliable method.

I seem to recall there's a way using a specific PMG collision register - and it may involve checking results with the GTIA register on & off.  Similar to how in BASIC you can type POKE 623,64 and the screen should go black if you have a GITA, but it stays blue if you have a CTIA.

[Panther]

4 minutes ago, Stephen said:

I seem to recall there's a way using a specific PMG collision register - and it may involve checking results with the GTIA register on & off.  Similar to how in BASIC you can type POKE 623,64 and the screen should go black if you have a GITA, but it stays blue if you have a CTIA.

That's actually how I'm doing it.  Set the player 0 horizontal position, graphics mode 9, GPRIOR loaded with #$41, clear the collision registers, wait two jiffies, load each player and missile register with 0, and check for a player 0 vs. playfield collision.  It usually works.

 

[+Stephen]

4 minutes ago, Panther said:

That's actually how I'm doing it.  Set the player 0 horizontal position, graphics mode 9, GPRIOR loaded with #$41, clear the collision registers, wait two jiffies, load each player and missile register with 0, and check for a player 0 vs. playfield collision.  It usually works.

 

Ah - sorry, that was my best idea.

[Panther]

7 minutes ago, Stephen said:

Ah - sorry, that was my best idea.

I appreciate the help anyway.  The first method I tried when I first wrote SimCheck didn't work well at all!  I'm sure this will work, but just needs some tweaks.

 

[cbelcher]

I've tried SimCheck 1.5 burned to a cart and on several original systems - 600xl 800xl 65xe and 130xe it always calls the BASIC unknown.  They all have original ROMs.

Is this a known issue?

 

 

[Panther]

3 minutes ago, cbelcher said:

I've tried SimCheck 1.5 burned to a cart and on several original systems - 600xl 800xl 65xe and 130xe it always calls the BASIC unknown.  They all have original ROMs.

Is this a known issue?

 

 

When run from a cartridge, SimCheck cannot enable Basic to check the version since it would occupy the same memory space.

[cbelcher]

Oh that makes sense!

So the result is really "Unable" rather than "Unknown"

 

Great utility - thank you!

[Panther]

Good idea!  I've added having SimCheck show when it's unable to determine the Basic version.  This version also hopefully finally fixes the GTIA/CTIA detection.

 

[Beeblebrox]

@Panther Great software BTW. Use Simcheck all the time.

 

This may or may not be at all possible but can Simcheck be coded/configured to detect the presence of video boards like:

 

UAV

UGV

Spectre

?

I have UGV in a few of my XE's.

 

Edited May 30 by Beeblebrox

[Panther]

31 minutes ago, Beeblebrox said:

@Panther Great software BTW. Use Simcheck all the time.

 

This may or may not be at all possible but can Simcheck be coded/configured to detect the presence of video boards like:

 

UAV

UGV

Spectre

I'm afraid not, these do not have any hardware address presence on the Atari, they simply handle the output side.

Glad you find it useful.

[Beeblebrox]

5 hours ago, Panther said:

I'm afraid not, these do not have any hardware address presence on the Atari, they simply handle the output side.

Glad you find it useful.

No worries, thought that might be the case. Thanks.

[wildstar87]

Is it normal during Simtest, when checking base memory, for the display to get a bit corrupted during parts of the test?  The corruption flickers a few times, then the display is back to normal.

Edited June 2 by wildstar87

[reifsnyderb]

9 minutes ago, wildstar87 said:

Is it normal during Simtest, when checking base memory, for the display to get a bit corrupted during parts of the test?  The corruption flickers a few times, then the display is back to normal.

I believe it is as I use it extensively.  I think the display flicker is when the video memory is being tested.

[wildstar87]

I kinda figured something like that, but wanted to make sure.  The extended memory test runs through, but doesn't seem to have as extensive reporting as the base memory, just one line, with the address numbering counting.  I wasn't even sure if if passed, it just went to ED, but I'm assuming if there were any failures, it would call them out.

[Panther]

Indeed, the glitches in the display are perfectly normal as they occur first when SimCheck goes through the display RAM, at which point you'll see one screen character get changed at a time, moving down the list, though you won't see every character change due to the speed at which the tests occur, then you'll have the entire screen blanking out when the test gets performed in areas occupied by things such as the display list, which gets disabled intermittently.  SimCheck also uses two different test loops so that it can test memory space occupied by SimCheck, first using one routine, then using a different one in another part of memory.

 

[Ricky Spanish]

A great utility !! 

[Panther]

On 6/3/2023 at 3:02 PM, Ricky Spanish said:

A great utility !! 

 

Now you just need to update to version 1.6.

[Ricky Spanish]

20 minutes ago, Panther said:

Now you just need to update to version 1.6.

Works great ! I really appreciate your efforts with this !

[Panther]

I just updated SimTest to version 0.9.  No functionality changes to the memory tests themselves, but a few improvements to the interface:

• Allow additional tests after failed test (clear display and reset error count).
• Display message when memory test passes.
• Show extended memory map at startup.
• Removed redundant memory save location.
• Corrected display position error during memory tests.