Jump to content
IGNORED

XEL-CF Compact Flash Adapter for 1088XEL (formerly XEL-I3)


mytek

Recommended Posts

58 minutes ago, BigBen said:

To be able to use a cartridge at all I have to set "8k Cart Rom: Enabled" in the bios. Unfortunately the BIOS accesses $ D500 when booting and switches off my cartridge type "Megacart". In addition, a homemade cartridge does not run on it. Is it possible to add this entry for example: Cartridge Port Type: 8k Bios access, 16k without Bios access.

I don't own a MegaCart, but just to double-check, I emulated one in Altirra to see if writing to $D500-$D51F would affect the cartridge while the U1MB PBI device ROM/RAM is enabled via the device selection register. It does not (much to my relief). So if the emulation is accurate, I can only assume you have SpartaDOS X enabled (which writes to $D508, IIRC) and this is what's deactivating the MegaCart, because the BIOS certainly does not write to $D500-$D51F.

 

Hopefully '8K Cart ROM' is fairly self-explanatory; it's a feature I was asked by the beta testers to leave in place, and it will not change to disabled once you enable it. :)

Link to comment
Share on other sites

It took me a while to find all the bugs.

 

Fixed:
A Pokey had arbitrarily triggered an IRQ.
The Mouseatari Pic cyclically switched the 2nd joystick port on and off.
A character was sporadically displayed after switching on.
The Pokey was exchanged, all pics reprogrammed.

 

Everything runs until a game, Bounty Bob Strikes Back. Whether as a cartridge or file,

you cannot move Bob, only jumping is possible. All options in the BIOS are deactivated,

the memory is at 64k. Do you also have the problem?

 

The reset signal is triggered by the pic. When switched on for the first time, it is high,

then goes low and then it goes high again. For comparison, the same signal measured

on an NTSC 800XL and PAL 130XE. There is only one reset pulse. It was triggered on

the + 5V input voltage. Can the output switch-on state not be predefined without first going high?

 

 

I am very satisfied with the 1088XEL hardware, a big thank you to all developers.

 

 

1088XEL reset signal.

1077202515_1088XEL_PAL_POWER_ON_Yellow5V-Blue_Reset.png.thumb.png.11c8ab874c7f22b5e84c7e00afa01903.png

 

800XL NTSC reset signal

1529587745_800XL_NTSC_Yellow5V_Blue_Reset.thumb.png.a239da841e54eb3c0d0a2fc7266846a3.png

 

130XE reset signal

1383854107_130XE_PAL_Yellow5V_Blue_Reset.thumb.png.7f90777eb4a808a27548c63ee2fb2af3.png

 

Link to comment
Share on other sites

5 minutes ago, BigBen said:

Whether as a cartridge or file,

you cannot move Bob, only jumping is possible.

This is reproducible in the emulator. I'm not certain, but this may be caused by the fact the game is reading the joystick at $D3F0 instead of $D300. It's not just your machine, anyway.

 

Link to comment
Share on other sites

  • 2 weeks later...

Status update:

It's been over two months since I thought to myself: "I'll do it!. Hold my beer." ;-) So...

 

I finally had some time to test the prototypes. I built two.

The first one is based on Philips 74F** chips, the second is on Texas Instruments.

 

That's how it looks:

sally-cf-800-xl.thumb.jpg.ee092c78e1e6b49c37b0ddd74beb4d16.jpg

 

 

The results are promising, but there are still some issues.

 

- Prototype 1 (Philips)

  - it works pretty stable after a few minutes, when the chips get a little warmed up

  - sees a PQI 1GB 300x PQI card, and, as I thought, does not detect SanDisk 16GB

  - sometimes it is, sometimes it is not visible in SpartaDOS, but I managed to divide the card into partitions

  - always reads the card data with the BASIC program

  - I can load a couple of XEX files (e.g. River Raid, Zybex, a couple of demos) - sometimes the first time, sometimes not. (some of them never - like Gyruss, why?)

  - works much better when a fourth decoupling 100nF capacitor is added (so, one per IC)

 

sally-cf-xl-cap.thumb.jpg.50ba7ca2a95615a8531775e85f93c9e7.jpgxel-loader.thumb.jpg.8bfefc36fa89dbd34e38e759ca03e2a4.jpg  xel-loader-corrupted-dir.thumb.jpg.4ba67ac099a3c747a79e1be2ddd98863.jpg

 

 

 

 

  prot-1-bas-test.thumb.jpg.75409052881f5a9f369c1379e3a477bd.jpg  prot-1-sdx-test-1.thumb.jpg.ec34dc507c05eabf6dcbb4530445ac3e.jpg prot-1-sdx-test-2.thumb.jpg.d8a875e7e5f4e15356fff2fa61380880.jpg

 

- Prototype 2 (TI):

 - sees both cards, easily detects SanDisk 16GB

 - unfortunately, I'm not able to load any program

 

pqi-1mb-ti.thumb.jpg.4ba9da1b7d0451a02210664f7484c2b3.jpg sandisk-16gb-ti.thumb.jpg.ddb577e4ea1d608e9f72cebe0b666d9a.jpg

 

 

---

 

OK. I'm getting two signals from the motherboard:  
- D1xx - U2 (74L138) - pin 14
- B02 - Gated Phase 2 Clock - U18 (74LS08) - pin 11

 

800xl-wires.thumb.jpg.23e64263786841b120600cc0ca2f11d5.jpg        sally-cf-mounted.thumb.jpg.6d7fdd9bc478f4759f5c97850600d176.jpg

 

 

Do we have to use a buffered signal instead of Phi2 directly from the CPU?
 

Other issues:
- The U1MB works without any problems when connected to the connector on the board, but it is much harder to read the contents of the CF card. I suspect that the Phi2 track on the PCB is causing interference.

- Probably the noise level is high, we have to think about effective filtering (need to check it with an oscilloscope)

 

Funny fact: connecting the SDrive to the SIO port makes it easier and more often to detect the card (CTRL+D) (the voltage drops from 5.02 to 4.99V)

 

What next: XE version - not sure how to do it without U1MB in my 130XE. I think we need an SDX rom image version with the XEL_CF.SYS file for use with i.e Ultimate Cart (or other like SIC!) ...and standalone Loader executable maybe?

 

Better to write in this thread (after all, it's still an XEL-CF mutation), or set up a new one for this project?

 

Greetings and all the best in the New Year!

  • Like 3
Link to comment
Share on other sites

4 hours ago, madness77 said:

  - works much better when a fourth decoupling 100nF capacitor is added (so, one per IC)

 

sally-cf-xl-cap.thumb.jpg.50ba7ca2a95615a8531775e85f93c9e7.jpg 

 

 

 

 

    

 

 

 

 

 

 

Yes at these speeds, each chip needs its own power reservoir as in a capacitor. And it should be as close and directly attached to the VDD of the chip lead as possible, with the other going to a ground plane. These squelch the power transits caused by the internal switching of the chip from getting onto the power bus.

 

4 hours ago, madness77 said:

Do we have to use a buffered signal instead of Phi2 directly from the CPU?

Highly recommended that it be buffered same as done on the XEL and XLD. It also wouldn't hurt to buffer the address lines going to the IDE port as well (reference 1088XLD schematics).

 

4 hours ago, madness77 said:

Other issues:
- The U1MB works without any problems when connected to the connector on the board, but it is much harder to read the contents of the CF card. I suspect that the Phi2 track on the PCB is causing interference.

- Probably the noise level is high, we have to think about effective filtering (need to check it with an oscilloscope)

You are likely fighting the poor power planes on the stock A8. Might even consider adding an LC power filtering circuit as was done on the XEL and XLD to take out noise on the +5V bus.

 

4 hours ago, madness77 said:

Better to write in this thread (after all, it's still an XEL-CF mutation), or set up a new one for this project?

You can keep sharing the project here as it develops, and maybe when it's in a more final form it would deserve a topic of its own. However its entirely up to you on what you wish to do.

 

  • Like 3
  • Thanks 1
Link to comment
Share on other sites

  • 1 year later...
On 1/6/2020 at 5:37 PM, madness77 said:

Status update:

It's been over two months since I thought to myself: "I'll do it!. Hold my beer." ;-) So...

 

I finally had some time to test the prototypes. I built two.

The first one is based on Philips 74F** chips, the second is on Texas Instruments.

 

That's how it looks:

sally-cf-800-xl.thumb.jpg.ee092c78e1e6b49c37b0ddd74beb4d16.jpg

 

 

The results are promising, but there are still some issues.

 

- Prototype 1 (Philips)

  - it works pretty stable after a few minutes, when the chips get a little warmed up

  - sees a PQI 1GB 300x PQI card, and, as I thought, does not detect SanDisk 16GB

  - sometimes it is, sometimes it is not visible in SpartaDOS, but I managed to divide the card into partitions

  - always reads the card data with the BASIC program

  - I can load a couple of XEX files (e.g. River Raid, Zybex, a couple of demos) - sometimes the first time, sometimes not. (some of them never - like Gyruss, why?)

  - works much better when a fourth decoupling 100nF capacitor is added (so, one per IC)

 

sally-cf-xl-cap.thumb.jpg.50ba7ca2a95615a8531775e85f93c9e7.jpgxel-loader.thumb.jpg.8bfefc36fa89dbd34e38e759ca03e2a4.jpg  xel-loader-corrupted-dir.thumb.jpg.4ba67ac099a3c747a79e1be2ddd98863.jpg

 

 

 

 

  prot-1-bas-test.thumb.jpg.75409052881f5a9f369c1379e3a477bd.jpg  prot-1-sdx-test-1.thumb.jpg.ec34dc507c05eabf6dcbb4530445ac3e.jpg prot-1-sdx-test-2.thumb.jpg.d8a875e7e5f4e15356fff2fa61380880.jpg

 

- Prototype 2 (TI):

 - sees both cards, easily detects SanDisk 16GB

 - unfortunately, I'm not able to load any program

 

pqi-1mb-ti.thumb.jpg.4ba9da1b7d0451a02210664f7484c2b3.jpg sandisk-16gb-ti.thumb.jpg.ddb577e4ea1d608e9f72cebe0b666d9a.jpg

 

 

---

 

OK. I'm getting two signals from the motherboard:  
- D1xx - U2 (74L138) - pin 14
- B02 - Gated Phase 2 Clock - U18 (74LS08) - pin 11

 

800xl-wires.thumb.jpg.23e64263786841b120600cc0ca2f11d5.jpg        sally-cf-mounted.thumb.jpg.6d7fdd9bc478f4759f5c97850600d176.jpg

 

 

Do we have to use a buffered signal instead of Phi2 directly from the CPU?
 

Other issues:
- The U1MB works without any problems when connected to the connector on the board, but it is much harder to read the contents of the CF card. I suspect that the Phi2 track on the PCB is causing interference.

- Probably the noise level is high, we have to think about effective filtering (need to check it with an oscilloscope)

 

Funny fact: connecting the SDrive to the SIO port makes it easier and more often to detect the card (CTRL+D) (the voltage drops from 5.02 to 4.99V)

 

What next: XE version - not sure how to do it without U1MB in my 130XE. I think we need an SDX rom image version with the XEL_CF.SYS file for use with i.e Ultimate Cart (or other like SIC!) ...and standalone Loader executable maybe?

 

Better to write in this thread (after all, it's still an XEL-CF mutation), or set up a new one for this project?

 

Greetings and all the best in the New Year!

Just wanted to check in and see if you've found the time to work some more on this?

Link to comment
Share on other sites

  • 1 year later...
  • 8 months later...

I have a question about XEL-CF. Is it possible to use it completly without U1MB? Maybe not to boot from it, but just to store files? Like with a stand-alone SDX on a cart or AVG or IDE+2 with a driver of some kind? If yes, can you describe, how to do it or give me a link to a description? Thanks...

Link to comment
Share on other sites

12 minutes ago, Peri Noid said:

I have a question about XEL-CF. Is it possible to use it completly without U1MB? Maybe not to boot from it, but just to store files? Like with a stand-alone SDX on a cart or AVG or IDE+2 with a driver of some kind? If yes, can you describe, how to do it or give me a link to a description? Thanks...

Yes. There's a SpartaDOS X driver for it on the APT toolkit disk on my website. You can inject it into the CAR: volume of your SDX ROM of choice.

  • Thanks 1
Link to comment
Share on other sites

And it works. For a test I took my plain SDX cartridge, loaded the driver from an ATR image with SIO2SD and I could access the partitions on my CF card. Great! 

 

The thing is, 3 years ago, during several zoom meetings of our Polish atarionline.pl community, one of the members designed an IDE interface based on XEL-CF but to be connected to an ECI slot. A few of these were build. I decided to finally run mine. And it seems to work as expected.

  • Like 1
Link to comment
Share on other sites

5 hours ago, Peri Noid said:

The thing is, 3 years ago, during several zoom meetings of our Polish atarionline.pl community, one of the members designed an IDE interface based on XEL-CF but to be connected to an ECI slot. A few of these were build. I decided to finally run mine. And it seems to work as expected.

Good to see the design branched into an ECI aspect.  @Dropcheck made one for the PBI as well. And I built an 600/800XL prototype myself (Monster 600XL).

  • Like 1
Link to comment
Share on other sites

I think it was an issue with the pullup resistor, but recollections are foggy at best. I still have the ECI and PBI versions of the device here; I wasn't able to get IO to work well using one or the other, but once the button issue was rectified, the 'other' one (can't remember which) worked OK.

 

Note: you'll need to drive the device via the U1MB-hosted XEL-CF PBI driver in order to avail yourself of ATR mounting/swap functionality.

Edited by flashjazzcat
  • Like 1
Link to comment
Share on other sites

16 minutes ago, flashjazzcat said:

Note: you'll need to drive the device via the U1MB-hosted XEL-CF PBI driver in order to avail yourself of ATR mounting/swap functionality.

I'm aware of that :-) ATRs need a PBI bios, which is not available otherwise.

  • Like 1
Link to comment
Share on other sites

4 hours ago, flashjazzcat said:

I think it was an issue with the pullup resistor, but recollections are foggy at best. I still have the ECI and PBI versions of the device here; I wasn't able to get IO to work well using one or the other, but once the button issue was rectified, the 'other' one (can't remember which) worked OK.

The PBI version works perfectly, but the ECI version has the swap button problem.  The change in pull-up resistor worked for the PBI version (available here: CLICKY ), but we could never get the ECI version to work.  Well, I should say, it works, but once you press the swap button, if you have more than one ATR loaded (which is kind of the point of the swap button, right 🙂 ) they just cycle like a merry-go-round.  We tried for a bit and then it go put to the back burner.

 

-M

 

 

  • Like 1
Link to comment
Share on other sites

On 6/20/2023 at 12:54 PM, flashjazzcat said:

I recall that other than a few issues with ATR swap button actuation/debounce, Lenore's devices worked well with the driver (notwithstanding some IO stability issues dependent on the choice of host machine).

I checked today - the swap button does not work in this design. Well... no problem, I'll not use it :-)

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...