Jump to content
IGNORED

U1MB, Incognito, SIDE2, SIDE3, 1088XEL/XLD Firmware 4.0 Released


Recommended Posts

Hello Fjc,

I replaced the 74LS08 with the 74F08. It's an 800 XL.
That was a good tip. You are just a professional. ?
Now the flashing of the SIDE3 also works. Everything seems to be working. But I am not able to start an ATR.
As soon as the ATR file is integrated on D1 and the computer restarts, a BOOT ERROR appears. Have tried several files. the U1 MB also has the latest firmware.

  • Like 1
Link to comment
Share on other sites

5 minutes ago, Brandhotte said:

As soon as the ATR file is integrated on D1 and the computer restarts, a BOOT ERROR appears. Have tried several files. the U1 MB also has the latest firmware.

I suspect you have not enabled the HDD in the U1MB BIOS settings, or have not flashed the proper SIDE3 plugin and PBI BIOS.

 

Link to comment
Share on other sites

Just now, Level42 said:

Is there any news on the “FAT compatible” firmware release Jon ?

Funnily enough, I just finished emailing someone regarding the state of play here. :)

 

I decided to run with the feature creep I'm experiencing at the moment, since I want to release the beta of the SIDE3 loader with as many hard-hitting new features as possible to enable said features to be tested ASAP in the field. I have basically turned the file browser into a Windows File Explorer workalike, with a full complement of file management and cut and paste facilities.

 

Copy dialogue is half finished:

copy.thumb.png.940aa69e4aebfba677f7959a4a6b2b49.png

 

There are context menus (also accessible for joystick users, and enabling keyboard-less access to many functions) :

menu.thumb.png.cddf2c6507de7ea47f5259e75863c6f3.png

 

Now that entire folder trees can be tagged, I had to spend a good few hours writing a tree walker to produce the list of files to be copied:

1686067113_Fixedfilelist.png.b4b2e01655a288ccdd5ed8ba6771d6f4.png

So: I need to finish the file management functions and user-proof them, and:

  • Add '/R' switch to the DOS COPY command so you can copy folder trees there too
  • Add TIME/DATE commands to the DOS CLI
  • Figure out a user interface for the loader 'autorun' facility
  • Implement MBR partition renaming
  • Add a recursive delete function for non-empty folder deletion

...And so on.

 

There'll definitely be something released before Christmas, and hopefully before my birthday. :)

  • Like 9
Link to comment
Share on other sites

Hello Fjc,

As a matter of fact. Disabled the hard drive.
  Here are my data:
U1MB SIDE3 PBI-BIOS v.1.00
SpartaDos X 4.49e
SIDE3 Loader Vers.0.44
(I also noticed that I cannot open the integrated ATR file in the SIDE3 Loader with the CONTROL + F keys.
It then jumps to where I have mounted the file as D1:).
After the restart, the green LED and go briefly, then the red LED stays on and the cursor stops in the upper left corner. BOOT ERROR reappears after 2 minutes. I just do not know how to go on....

Link to comment
Share on other sites

On 11/27/2021 at 6:33 PM, Brandhotte said:

I replaced the 74LS08 with the 74F08. It's an 800 XL.

Is this still needed for SIDE3? I was hoping they mitigated for the "slurring" Phi2 during writes. Hiassoft designed a very simple 3-part circuit to shorten phi2 again during writes. If I buy a cartridge, I should not have to open my machine and replace an IC for it to work, IMHO.

 

Edited by ivop
Link to comment
Share on other sites

1 hour ago, Brandhotte said:

As a matter of fact. Disabled the hard drive.

Well, there you go: you're using the limited soft-OS ATR mounting. If you enable the HDD, you should enjoy better results.

1 hour ago, Brandhotte said:

(I also noticed that I cannot open the integrated ATR file in the SIDE3 Loader with the CONTROL + F keys.

The purpose of CTRL+F is to open the containing folder of a mounted disk image or search result.

32 minutes ago, ivop said:

Is this still needed for SIDE3? I was hoping they mitigated for the "slurring" Phi2 during writes. Hiassoft designed a very simple 3-part circuit to shorten phi2 again during writes. If I buy a cartridge, I should not have to open my machine and replace an IC for it to work, IMHO.

At least half of the XEs which have been through my hands during the past 14 months which arrived here with SIDE3 (so, about 6-8 out of a dozen or more machines) have required 74F08 in order to work with SIDE3. It's rare that any XL requires modification, which may be telling, but in any case: I happen to agree with you. Unfortunately, however, it appears that - for some mysterious reason - such stability issues never occur in Poland (which is odd, considering XE machines are absolutely prevalent over there), and the one machine exhibiting this issue which was actually sent (from outside of Poland) to Candle for inspection simply resulted in... Candle replacing 74LS08 to correct timing issues on the machine.

 

Perhaps the issue simply cannot be fixed? If the worry is that by fixing one thing, another might become broken, then I can understand that, but such is the silence on this matter every time I bring it up, that I have come to accept it as a fact of life.

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

@candle seems aware of it:

 

 

And @HiassofT designed a simple circuit with an LS123, a capacitor and a resistor to shorten Phi2 during writes.

 

10 minutes ago, flashjazzcat said:

Unfortunately, however, it appears that - for some mysterious reason - such stability issues never occur in Poland (which is odd, considering XE machines are absolutely prevalent over there)

Yeah, that's pretty odd. Perhaps the earth magnetic field is different there ;)

Link to comment
Share on other sites

I'm sure there is nothing new under the sun as far as Candle is concerned, but in fairness, SIDE3 represents a move from Xilinx to Lattice FPGA hardware, so just because SIDE2 worked with a given machine, doesn't mean that SIDE3 will so the same, even if the same or similar well-established stabilisation methods are implemented. This would seem to me to represent an issue which would benefit from careful testing and collation of data (there being plenty of users out there now who could provide it), but VHDL and hardware isn't my department, so anything I say is pure helpless speculation. :)

9 minutes ago, ivop said:

Perhaps the earth magnetic field is different there ;)

Something is definitely different.

 

Link to comment
Share on other sites

20 minutes ago, flashjazzcat said:

I'm sure there is nothing new under the sun as far as Candle is concerned, but in fairness, SIDE3 represents a move from Xilinx to Lattice FPGA hardware, so just because SIDE2 worked with a given machine, doesn't mean that SIDE3 will so the same, even if the same or similar well-established stabilisation methods are implemented. This would seem to me to represent an issue which would benefit from careful testing and collation of data (there being plenty of users out there now who could provide it), but VHDL and hardware isn't my department, so anything I say is pure helpless speculation. :)

Even though I am a bit of a hardware guy, I never did much with verilog/vhdl and didn't like it when I did. But I assume candle has a high speed clock running and simulates the LS123 delay once Phi2 goes up. It can then count ticks and do the write after 180-200ns. Perhaps this shortening of Phi2 could be influenced through a register in the future. Like 0xff is do nothing, and then gradually shorten to 0x00 which means write immediately.

 

Quote

Something is definitely different.

I just checked if they still used 220V/50Hz (some countries like Chad and Russia still do), but they do not. It's 230V/50Hz, just like the UK.

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

Perhaps @HiassofT can post the schematic to shorten Phi2 again. I'm having a hard time finding both the small PCB I once made for breadboarding, and the schematic. It's probably not useful for candle, but it could be useful for future THT projects. @mytek's 1088XEL-CF didn't need it because of his four layer motherboards, but old skool XL/XE's need it. My fully socketed 800XL needed the LS08 to F08 fix, too.

Edited by ivop
Link to comment
Share on other sites

12 minutes ago, Brandhotte said:

"The purpose of CTRL + F is to open the contained folder of a mounted disk image or search result."

 

Well he doesn't. Always jumps back to the file selection ...

I don't really understand what you mean by this, since - by definition - you should end up back in the file selection. If you press CTRL+F on a disk image (in the list of drive mounts), it should take you to the folder which contains that image file, with the image file highlighted. If you press CTRL+F on a search result, it should take you to the folder which contains that file, with the file highlighted.

 

If you can explain your problem more clearly or perhaps with the help of screenshots, I'll be better able to help.

17 minutes ago, ivop said:

I just checked if they still used 220V/50Hz (some countries like Chad and Russia still do), but they do not. It's 230V/50Hz, just like the UK.

LOL. I didn't mean to imply the difference concerned mains voltage. :D

 

In any case: the 74F08 fix has worked in 100 per cent of cases (aside from a single odd-kid-on-the-block 600XL which had a problem which only a 5.5V PSU could fix), so regardless of whether we do this or implement Hias' fix, it's still alteration of the host machine, and I remember Candle not being a fan of this in the first place. And I can agree with that too, because when you 'fix' the machine to work with device A, who knows if you simultaneously 'un-fix' it with devices B, C and D.

  • Like 1
Link to comment
Share on other sites

4 minutes ago, flashjazzcat said:

LOL. I didn't mean to imply the difference concerned mains voltage. :D

LOL, I know, but I was just eliminating one thing that might be different :)

 

4 minutes ago, flashjazzcat said:

In any case: the 74F08 fix has worked in 100 per cent of cases (aside from a single odd-kid-on-the-block 600XL which had a problem which only a 5.5V PSU could fix), so regardless of whether we do this or implement Hias' fix, it's still alteration of the host machine, and I remember Candle not being a fan of this in the first place. And I can agree with that too, because when you 'fix' the machine to work with device A, who knows if you simultaneously 'un-fix' it with devices B, C and D.

But the Phi2 shortening during writes can be done in the device/cart itself. That's the beauty of it :) No need to alter the host machine.

Link to comment
Share on other sites

32 minutes ago, ivop said:

But the Phi2 shortening during writes can be done in the device/cart itself. That's the beauty of it :) No need to alter the host machine.

Yes, agreed, but we come full circle then, since I'm sure efforts were made to do this with SIDE3, but we still have a problem.

29 minutes ago, Brandhotte said:

Isn't there some test that I can use to check why I keep getting BOOT ERROR?

Test a stock DOS 2.5 ATR and turn off any extended RAM on the machine. If that boots (which it should), then your problems are caused by the choice of ATR file.

Link to comment
Share on other sites

1 hour ago, ivop said:

Perhaps @HiassofT can post the schematic to shorten Phi2 again. I'm having a hard time finding both the small PCB I once made for breadboarding, and the schematic. It's probably not useful for candle, but it could be useful for future THT projects. @mytek's 1088XEL-CF didn't need it because of his four layer motherboards, but old skool XL/XE's need it. My fully socketed 800XL needed the LS08 to F08 fix, too.

You can find that eg in the Turbo Freezer 2005 schematic (it's in the lower right corner) https://www.horus.com/~hias/freezer/turbo-freezer-2005/hardware/schematics-xl.pdf

 

74HCT123 with 5k6 and 33pF will work, too, but keep in mind that these '123 aren't exactly precision devices (esp in the configuration we use them here, it's also mentioned in the datasheet) so if in doubt better check PHI2 vs shortened PHI2 with a scope (and if it's not short enough use a lower value resistor or pot).

 

Also keep in mind that this only affects writes, but if rogue reads can alter state in a device/chip (eg checking for toggle-bit programming done in a flash chip or reading data from an IDE device) this won't help - in this case you need to latch address signals or shorten reads (the latter isn't recommended, but usually works well enough and the simple "stabilizing PHI2" mod with the wire on LS08 in the Atari will do it just fine).

 

so long,

 

Hias

  • Thanks 1
Link to comment
Share on other sites

2 hours ago, flashjazzcat said:

Yes, agreed, but we come full circle then, since I'm sure efforts were made to do this with SIDE3, but we still have a problem.

I remember having a discussion about this with Candle and yourself and did some tests to compare the CF Card access on a 1088XEL to what Candle had given me.

 

IDE_scope_compare.thumb.jpg.8e6c715a78261469ab5b28d552cc494d.jpg

 

B02 = Buffered Phi2 via 74F08.

 

You can see where the IDE's CE1 appears earlier on the XEL vs Candle's example. However the IORD and IOWR matches more closely, although it does appear that the Phi2 has been shortened - going high a bit later. Its been so long ago that I forget what Candle's baseline machine was that he was using, but it does appear that he was adjusting the PHi2 pulse length.

 

Now I tried something similar in an internal CF drive I was experimenting with, and it not only didn't seem to be doing much good for the CF drive reliability, it was actually worse. However I did observe something most interesting after I took out the pulse shaping circuit... If I added 10K pull-ups to all the data lines on the 600XL I was using for my tests, I completely lost the ability to even see many of my CF cards in the Loader. It would still work with some Chinese SanDisk fakes I got, and actually did quite well with those. So this has me thinking that perhaps the reliability problem might be associated with the un-buffered data lines, because adding 10K pull-ups to the data lines shouldn't be an issue. One of these days I intend to find out if buffering the data lines right out of the CPU has any effect.

 

  • Like 1
Link to comment
Share on other sites

5 hours ago, flashjazzcat said:

In any case: the 74F08 fix has worked in 100 per cent of cases (aside from a single odd-kid-on-the-block 600XL which had a problem which only a 5.5V PSU could fix)…..

ah, so now I’m an odd-kid-on-the-block…

 

Really ?

Link to comment
Share on other sites

8 minutes ago, Level42 said:

ah, so now I’m an odd-kid-on-the-block…

 

Really ?

100 per cent of machines I worked on, aside from the 600XL I mentioned. If you want to add yourself to the list of exceptions without being incredulous about the fact the fix works almost all of the time, consider it noted.

Link to comment
Share on other sites

12 minutes ago, mytek said:

The 74F08 also made the 1088XEL a success. Because prior to that discovery, it was having some serious PMG issues during the beta test phase when using a 74HCT08 as well as a 74LS08.

 

And only on PAL machines further narrowing down the issue.  Crazy just what a "house of cards" these machines are.

  • Like 1
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...