Thinking of getting a new Atari 800XL board made....

[reifsnyderb]

Just now, scorpio_ny said:

Couldn’t one just short the leads per the keyboard matrix in a pinch?

You can.  Thinking more through this, maybe having some sort of pads on the board would be useful.....

[Overange]

3 minutes ago, reifsnyderb said:

You can.  Thinking more through this, maybe having some sort of pads on the board would be useful.....

or even just break out with headers 

 

[reifsnyderb]

Like this?

 

[Overange]

Just now, reifsnyderb said:

Like this?

 

 

  Yes, and I am thinking about maybe build a mini flying breakout board with tactile switches  

 

On my design, I went with going pin to pin to match the TKII stereo input, so one cable directly between the two with no criss crossing.

 

[ClausB]

1 hour ago, reifsnyderb said:

To follow-up:  I played around with the settings even more.  Still no change.  Other than this one issue, the picture is very nice.  I am also pleased with the sound.

Have you tried other values for R51? The 800 has 0 Ohms there, right? A quick test would be to short R51 with clip leads.

[reifsnyderb]

Just now, ClausB said:

Have you tried other values for R51? The 800 has 0 Ohms there, right? A quick test would be to short R51 with clip leads.

I have.  Originally, I didn't have a resistor there at all.  Right now, I have two resistors, in series, to come up with something pretty close to 750 ohms. 

[ClausB]

R46?

[reifsnyderb]

Just now, ClausB said:

R46?

It's at 1.6k.  I've tried different values there, as well.  1.5k makes no difference.  The 800 has 2.2k and that doesn't fix the problem, either.  (Actually, having 2.2k at R26 and 2k at R52 causes the video to randomly flicker on and off.)

[reifsnyderb]

Here's the current upper left quadrant:

 

[reifsnyderb]

Here's the top right quadrant:

 

 

I put a little chart under the ROM chip in case somebody wants to flash new OS's on the ROM.  Also visible, are the 4 jumpers that were mentioned in the Sweet 16 Spec.  I placed them outside of the shielding as it would make sense to do so.

 

[+kheller2]

Unrelated, are you using a custom MMU?  Why waste the other 64K in the SRAM?

[reifsnyderb]

40 minutes ago, kheller2 said:

Unrelated, are you using a custom MMU?  Why waste the other 64K in the SRAM?

I am using a ATF22V10 for the following reasons:

1.  PBI RAM can be added as per the Sweet 16 Spec.

2.  RW is ran into the chip as I have a plan to program it to only enable the ROM on a read.  This would allow for Axlon memory.

3.  Refresh, via the PBI, can be setup to act as an open-collector.  This will allow PBI devices to ground the Refresh without affecting the ANTIC chip.

 

The pinout is such that, with the exception of Refresh going to the PBI, a normal MMU could be installed if pin 10 is jumpered to ground.

 

The SRAM chip was used due to price.  There would require at least one more programmable logic chip to use the other 64k of SRAM.  Additionally, 3 pins from the PIA would have to be connected to the logic chip.  This would give you only 128k, total. 

[reifsnyderb]

4 hours ago, kheller2 said:

Unrelated, are you using a custom MMU?  Why waste the other 64K in the SRAM?

Ok, ok, ok.  I can see getting more questions on this.  64k can now be set by jumpering the 3 sets of pins marked by the arrows.  (Not completely visible.)  Otherwise, an ATF16V8B could be installed and programmed.  RAMBO banking could be done easily.  XE banking "might" even be possible by using the chip in registered mode.  (Phi2 and /halt have been ran to the ATF16V8B.)

 

Edited November 9, 2023 by reifsnyderb

[+scorpio_ny]

21 hours ago, reifsnyderb said:

Ok.  No matter what I do, I cannot get the two left-hand columns, on ACP, to show up with different hues.  So, I tried it on the 800 and can see a slight difference.  My 600XL, with a heavily modded S-Video circuit, is exactly the same (as far as colors are concerned) as the new board.  So, I got really suspicious and plugged in my original 800XL.  That is my very first computer.  The only mod it has is to add chroma to the monitor port and the removal of a capacitor in the video circuit to disable composite video for cleaner S-Video.  Otherwise, this computer is original.  (Quite frankly, I didn't even want to mod that but I also didn't want to setup RF output with an old TV, either.)  My original 800XL has the exact same color output in ACP.

 

As the new board's Luma circuit is almost identical to the original XL circuit, I am thinking that maybe my TV is the problem.  I am not sure at this point.  The new board has a 33pF blur capacitor as per the 800 spec and is configured with a 10uF tantalum capacitor as per a 130XE mod.  Neither of these Luma changes should create a problem.  So, I am thinking this is probably going to be ok.

 

Here's the Luma circuit:

 

 

Note:  C55 is the capacitor that sits on the only power rail going into the entire video circuit.  So, it also is before the 4050 and color circuit.

 

Hi @reifsnyderb,

 

Doing some research on the video you are trying to adjust. Your color circuitry, even with some of the mods, is still basically based on the Atari 800 (my assumption).I came across an article that stated the Atari 800 video's Luma is over driven. It got me thinking that you may have adjusted your video to fix highlights the instead of the shadows. That is why you are seeing the dark areas crunched up as you posted. If you adjust to fix the shadows, then, the highlights are blown out.

 

See this post especially towards the end. It leads to the fix as explain in this second post in section I6.

[reifsnyderb]

34 minutes ago, scorpio_ny said:

Hi @reifsnyderb,

 

Doing some research on the video you are trying to adjust. Your color circuitry, even with some of the mods, is still basically based on the Atari 800 (my assumption).I came across an article that stated the Atari 800 video's Luma is over driven. It got me thinking that you may have adjusted your video to fix highlights the instead of the shadows. That is why you are seeing the dark areas crunched up as you posted. If you adjust to fix the shadows, then, the highlights are blown out.

 

See this post especially towards the end. It leads to the fix as explain in this second post in section I6.

Hello!

 

I just tried dropping in a 220 ohm resistor.  After doing so, the first 3 columns are now identical.  So, I played with the TV settings and no change.  After that, I tried a 1k ohm resistor and it's a little brighter and back to 2 identical columns on the left.  I am now putting the 750 ohm resistor equivalent (2 resistors in series) back in place.  Thanks for looking!

 

Brian

[+scorpio_ny]

1 minute ago, reifsnyderb said:

Hello!

 

I just tried dropping in a 220 ohm resistor.  After doing so, the first 3 columns are now identical.  So, I played with the TV settings and no change.  After that, I tried a 1k ohm resistor and it's a little brighter and back to 2 identical columns on the left.  I am now putting the 750 ohm resistor equivalent (2 resistors in series) back in place.  Thanks for looking!

 

Brian

Did you adjust the color for the showdown/darks first? In other word, on your tv, brightening to show the proper steps for the dark area and then change the resistors to fix the highlights?
 

[reifsnyderb]

2 minutes ago, scorpio_ny said:

Did you adjust the color for the showdown/darks first? In other word, on your tv, brightening to show the proper steps for the dark area and then change the resistors to fix the highlights?
 

I tried changing the brightness with no difference.  Changing out that resistor affects how far the sync signal pulse is pulled down....which I think changes the blanking level.

[+scorpio_ny]

Hmmm... well have your tried swapping out the 4050? In the post they also mentioned that also would affect the color output.

[reifsnyderb]

5 minutes ago, scorpio_ny said:

Hmmm... well have your tried swapping out the 4050? In the post they also mentioned that also would affect the color output.

The 4050 is brand new.  Even so, I swapped it out to test.  No change. 

[+mytek]

The 4050 as far as I recall only buffers Luma and Csync, not the color signal. I don't see how that would affect the color you are seeing. However the resistor DAC that follows the 4050 does influence the color you see because of how much intensity it creates for different luminance levels. And it's not just a factor of the resistor DAC values in play, but also the +5V and GND bias resistors that follow. And finally, the transistor(s) used to feed the video output also play a role, requiring the proper biasing related to their gain factor. Basically everything needs to be just right to have an even scaling across the board.

[reifsnyderb]

7 minutes ago, mytek said:

The 4050 as far as I recall only buffers Luma and Csync, not the color signal. I don't see how that would affect the color you are seeing. However the resistor DAC that follows the 4050 does influence the color you see because of how much intensity it creates for different luminance levels. And it's not just a factor of the resistor DAC values in play, but also the +5V and GND bias resistors that follow. And finally, the transistor(s) used to feed the video output also play a role, requiring the proper biasing related to their gain factor. Basically everything needs to be just right to have an even scaling across the board.

I now have the Luma side almost exactly as per the 800XL schematic.  The only addition is the 10uF Tantalum capacitor and the blur capacitor is 33pF as per the 800.  What's odd is that even my all but stock 800XL has the same issue.  The 130XE has the same issue.  Only the 800 has a slight difference.  The only common denominator is the TV.  Unfortunately, it's a real pain to test on another TV at the moment.

 

(I just noticed an error.  R53 is 33 ohms.  I've tried this with and without R53, without any change.)

 

 

 

 

 

 

[reifsnyderb]

Since I just posted this elsewhere, here's the schematic for the banking logic:

 

 

I'll need to use one of the pins (i.e. 13) for a /halt latch and still have 2 other macro cells available in the event I need extra logic.

 

In the schematic above, PB2-PB5 are from the PIA chip.  SRA14-SRA16 go to pins A14-A16 on the SRAM chip.

 

For normal operation, if PB4 and PB5 are high, SRA16 will be set low, SRA14 will be equal to AR14, and SRA15 will be equal to A15.

 

With CPU banking, PB4 will be set low, /halt will have to be high, SRA16 will be set high, SRA14 will be equal to PB2, and SRA15 will be equal to PB3.

 

With ANTIC banking, the /halt latch will be checked to see if /halt is called.  If PB5 is set low and /halt is called, SRA16 will be set high, SR15 will be equal to PB2, and SRA15 will be equal to PB3.

 

The /halt latch, on Pin 13, can be configured such that .d is from pin 9 and it is triggered by sending a pulse out pin 16 when Phi2 drops.  The capacitor, C75, will ensure a slight delay on the clock pulse.  The value of this capacitor will have to be determined through experimentation.  However, it shouldn't take much.  I've used a 4.7nF capacitor for a Read/Write Late delay on an 800 board.  Something similar should work fine.  I'll have to fine tune it with the scope.

 

I also positioned the inputs and output so that if the ATF16V8B is omitted, A14 is across from SRA14, A15 is across from SRA15, and ground is across from SRA16.  This will allow for easy jumpering without the chip.   🙂

 

 

 

 

 

[reifsnyderb]

PBI RAM will be able to be enabled or disabled by a jumper.  If PBI RAM is disabled, a normal Atari MMU could be used with the only caveat being that the PBI Refresh line will be disabled.  The Atari MMU will have to be installed in the MMU socket such that 4 empty pins are on the right and the Atari MMU uses the left 20 pins.

 

Edited November 9, 2023 by reifsnyderb

[+scorpio_ny]

2 minutes ago, reifsnyderb said:

PBI RAM will be able to be enabled or disabled by a jumper.  If PBI RAM is disabled, a normal Atari MMU could be used with the only caveat being that the PBI Refresh line will be disabled.  The Atari MMU will have to be installed in the MMU socket such that 4 empty pins are on the right and the MMU uses the left 20 pins.

 

I am little confused. Does one still need to use the Atari MMU with your design or can use a GAL MMU instead?

[reifsnyderb]

10 minutes ago, scorpio_ny said:

I am little confused. Does one still need to use the Atari MMU with your design or can use a GAL MMU instead?

All of the above.   🙂

 

You can use an Atari MMU if you install it so that it's all the way to the left in the socket and set the PBI RAM to disabled.  (An Atari MMU won't handle PBI RAM anyhow.)  I kept those pins identical so that this is possible.  Setting the PBI RAM to disable will set pin 10, of the MMU socket, to ground.  So, an Atari MMU can be dropped in.  An ATF22V10 can be programmed as the MMU, and used so that all the features are available.  Optionally, an ATF16V8B (or C) could be programmed in combinetorial mode so that it works as an Atari MMU and installed exactly like an Atari MMU.  (Though, if you are going to program your own MMU, you may as well use an ATF22V10.)

 

I'll release the MMU WinCupl code in the future.

 

 

Edited November 9, 2023 by reifsnyderb