New (alt) BIOS for Ultimate 1MB/Incognito

[MARIO130XE]

After updating to upbi180 I cant run the sio2sd disk selection/cfg tool by holding down the shift key on the sio2sd device, when the pbi is enabled.

if pbi is disabled it's still working. also in fast mode, when I have selected the XL/XE HISPEED SIO patched OS ROM it's working.

 

is there a solution to fix my problem, without downgrading the U1MBROMs? I would prefer to use unpatched XL/XE OS ROM and still have fast loading from my sio2sd again.

 

greetings Mario

[flashjazzcat]

What are your settings in the high speed SIO menu?

 

EDIT: just tested with my SIO2SD and I can't get the menu to boot even on a machine with no U1MB in it at all. I never used the boot menu facility before, though. I'm using the new (3.x) firmware, but is there something I missed regarding setup? Even when I set SIO2SD up so it boots to the loader every power-up, nothing happens.

Edited September 17, 2016 by flashjazzcat

[MARIO130XE]

You need the SIO2SD.xex File in the root of the SD Card and in the Setup of the SIO2SD you need to set the option 'cfgtool mode: startup+shift'

Then you must press the Shift key of the SIO2SD-Device and power on the Atari. (yes, Power on. only coldstart does not work)

 

In older Bios / Pbi versions it was working fine. But after the last update it's only working, if the pbi is disabled, so you can have fast loading only by using a patched OS.

 

Greetings, Mario

Sio2SD.xex

Edited September 18, 2016 by MARIO130XE

[flashjazzcat]

You need the SIO2SD.xex File in the root of the SD Card and in the Setup of the SIO2SD you need to set the option 'cfgtool mode: startup+shift'

Then you must press the Shift key of the SIO2SD-Device and power on the Atari. (yes, Power on. only coldstart does not work)

OK: this is pretty much how I expected it to work, but for some reason it doesn't work here, which is a little frustrating (the Shift key certainly works). Will see what I can do regarding reproduction and diagnosis.

[flashjazzcat]

In older Bios / Pbi versions it was working fine. But after the last update it's only working, if the pbi is disabled, so you can have fast loading only by using a patched OS.

Found a bug (I knew there'd be one somewhere). HDD boot drive number was still used even when the HDD was disabled, so if you had the PBI on solely for the purpose of running HSIO, but had left the HDD boot drive at (say) D3:, the machine would fail to boot from disk. Hopefully this is the problem you were encountering.

 

Please try the following update and let me know if it fixes the issue. I'm able to boot the SIO2SD XEX loader (only by configuring it to always boot to the loader, since I still can't get it to load by holding the device's shift key on power-up, regardless of whether U1MB's PBI ROM is on or off), and boot from ATRs mounted in RespeQt, which I could not do with 1.80.

 

upbi181.zip

 

Sorry about that.

[+Stephen]

Found a bug (I knew there'd be one somewhere). HDD boot drive number was still used even when the HDD was disabled, so if you had the PBI on solely for the purpose of running HSIO, but had left the HDD boot drive at (say) D3:, the machine would fail to boot from disk. Hopefully this is the problem you were encountering.

 

Please try the following update and let me know if it fixes the issue. I'm able to boot the SIO2SD XEX loader (only by configuring it to always boot to the loader, since I still can't get it to load by holding the device's shift key on power-up, regardless of whether U1MB's PBI ROM is on or off), and boot from ATRs mounted in RespeQt, which I could not do with 1.80.

 

upbi181.zip

 

Sorry about that.

That explains the trouble I was having yesterday.

[Kyle22]

J

 

Found a bug (I knew there'd be one somewhere). HDD boot drive number was still used even when the HDD was disabled, so if you had the PBI on solely for the purpose of running HSIO, but had left the HDD boot drive at (say) D3:, the machine would fail to boot from disk. Hopefully this is the problem you were encountering.

Please try the following update and let me know if it fixes the issue. I'm able to boot the SIO2SD XEX loader (only by configuring it to always boot to the loader, since I still can't get it to load by holding the device's shift key on power-up, regardless of whether U1MB's PBI ROM is on or off), and boot from ATRs mounted in RespeQt, which I could not do with 1.80.

upbi181.zip

Sorry about that.

Jon, Does this affect Incognito, or is it a U1M only thing?

[MARIO130XE]

Found a bug (I knew there'd be one somewhere). HDD boot drive number was still used even when the HDD was disabled, so if you had the PBI on solely for the purpose of running HSIO, but had left the HDD boot drive at (say) D3:, the machine would fail to boot from disk. Hopefully this is the problem you were encountering.

 

Please try the following update and let me know if it fixes the issue. I'm able to boot the SIO2SD XEX loader (only by configuring it to always boot to the loader, since I still can't get it to load by holding the device's shift key on power-up, regardless of whether U1MB's PBI ROM is on or off), and boot from ATRs mounted in RespeQt, which I could not do with 1.80.

 

upbi181.zip

 

Sorry about that.

Thanks for all your hard work, mate. NOOOOOO need to say sorry!!!

I'll test it later today.

 

Greetings Mario

[Dmitry]

Has anyone else had an issue with uFlash where the U1MB is not recognized?

 

I recognize that I might have a lemon U1MB, but thought I'd check.

 

Like for example if there is an alternate path, first flash to a more recent bios before attempting uFlash - or does it work with every U1MB no matter how old? This U1MB might be a few years old.

[flashjazzcat]

Jon, Does this affect Incognito, or is it a U1M only thing?

Yes: Incognito too. Here's the fixed version:

 

ipbi181.zip

 

If you have the HDD boot drive set to something other than D1: but can still boot from an ATR on D1: with the HDD disabled (HSIO on or off), the issue is fixed.

 

Thanks for all your hard work, mate. NOOOOOO need to say sorry!!!

I'll test it later today.

Great - thanks. As soon as the issue is confirmed fixed, I'll incorporate the update into the firmware package on my website.

 

Has anyone else had an issue with uFlash where the U1MB is not recognized?

 

I recognize that I might have a lemon U1MB, but thought I'd check.

 

Like for example if there is an alternate path, first flash to a more recent bios before attempting uFlash - or does it work with every U1MB no matter how old? This U1MB might be a few years old.

MetalGuy66 asked me if the newer boards still needed resistors in the MMU cable (which they don't), so I take that to mean the board is comparatively new? What colour is the PCB?

 

A lot of the issues you're encountering on that machine imply system instability which needs fixing before we start wondering if odd behaviours can be repeated elsewhere (this applies to Rapidus as well: we should wait for a firmware update before spending any more time attempting to diagnose misbehaviour with U1MB). Having said that, let's look at some possible software causes.

 

If the U1MB has Candle's BIOS (which I understand it does), then there's no possibility of the flash write lock being enabled. However, if the PBI is turned off and you're booting SDX from the SIDE cart, UFlash has no hope of finding U1MB's flash ROM anyway. You need to boot SDX from the U1MB and then run UFlash. You should also use a recent version of UFlash (the build from the recent U1MB firmware distros is ideal), since older versions used a polling mechanism which wasn't reliable at 20MHz. I can confirm this works, since even my blighted VBXE/Rapidus/U1MB 130XE is able to update the U1MB ROM via UFlash reliably at 20MHz.

[MARIO130XE]

Wohoooo... UPBI181.ROM IS WORKING FINE!!!! Thank you so much, mate.

[flashjazzcat]

Wohoooo... UPBI181.ROM IS WORKING FINE!!!! Thank you so much, mate.

 

Brilliant - thanks for testing and confirming it works. I'm miffed about the bug, but it was perhaps inevitable given the number of changes. Anyway, I'll make this part of the firmware package proper as soon as I get time.

[flashjazzcat]

20 downloads in all and no complaints, so I guess the fix works for everyone.

 

While testing the Spelunker ATR, I found I sometimes had to boot the machine with Option pressed. In the loader, Option is a page down shortcut (which I was asked to add for the XEGS), so it's impossible to hit enter while holding down the Option key. One has to hit enter, then press Option. Is this a problem for anyone? You don't normally have to use Option anyway to disable BASIC (just this awkward ATR required it), but it strikes me as a potential conflict.

[MARIO130XE]

I think ONE single game isn't worth it to expend effort. Maybe someone could patch the game so it can disable basic itself.

[flashjazzcat]

Good point. BASIC is disabled anyway and it's some strange issue peculiar to that game which required the kludgy Option key workaround.

 

Finally getting the chance to do some serious testing on an NTSC 1200XL anyway, and there have been no nasty surprises so far.

[Dmitry]

If the U1MB has Candle's BIOS (which I understand it does), then there's no possibility of the flash write lock being enabled. However, if the PBI is turned off and you're booting SDX from the SIDE cart, UFlash has no hope of finding U1MB's flash ROM anyway. You need to boot SDX from the U1MB and then run UFlash. You should also use a recent version of UFlash (the build from the recent U1MB firmware distros is ideal), since older versions used a polling mechanism which wasn't reliable at 20MHz. I can confirm this works, since even my blighted VBXE/Rapidus/U1MB 130XE is able to update the U1MB ROM via UFlash reliably at 20MHz.

 

 

well then I had no hope of finding it, since I had PBI turned off

 

Frankly I find it fun to ask questions at AtariAge, but don't want to waste anyone's time either.

 

Even though I suspect that a perfectly valid test of flashing the U1MB was done in the past, I want to do it myself, to begin taking a little responsibility for the system.

 

But I do plan to do as you advised, and first stabilize the system. I'll remove all the add-ons and then add them back one by one...until ultimately I've destroyed the machine, and non-working is the most stable state of them all

 

I'm joking a bit....anyway, I will open another thread if I have questions, sorry to hijack your thread.

[flashjazzcat]

PBI update, including fix for Spelunker:

 

v.1.82

 

* Fixed: RAM cleared under BASIC ROM when BASIC disabled (cured Spelunker ATR boot problem)

* Changed: DUNIT boot redirection only applied if DDEVIC = $3x

* Changed: During boot, if drive swap can't be performed (if boot drive or D1: isn't a HDD partition), normal boot redirection performed

* Changed: Drive swap filtering disallows mounted ATRs

 

U1MB:

 

upbi182.zip

 

Incognito:

 

ipbi182.zip

 

Thanks to TheNameOfTheGame for reporting the issue, and to Phaeron for figuring out the cause.

 

well then I had no hope of finding it, since I had PBI turned off

 

Frankly I find it fun to ask questions at AtariAge, but don't want to waste anyone's time either.

Asking questions is rarely a waste of time, and most often, queries prompt solutions. But with regard to Rapidus, I don't want to offer interim solutions to problems which will hopefully go away with the next firmware update. Speculation on the matter has previously attracted derision and some puzzling email diatribes, so I'd encourage anyone with issues to email the vendor/developer.

 

But I do plan to do as you advised, and first stabilize the system. I'll remove all the add-ons and then add them back one by one...until ultimately I've destroyed the machine, and non-working is the most stable state of them all

 

I'm joking a bit....anyway, I will open another thread if I have questions, sorry to hijack your thread.

No problem. If you need the 800XL fixed and can be bothered with another shipping adventure, you know where I am.

[Roydea6]

CIO Z.SYS seems to be erratic from fat32.. Would someone test the attached file in there fat32 with the latest loader build. Just wanting to know if it is my setup or something else.

 

 

 

 

xclock.xex

Edited September 25, 2016 by rdea6

[flashjazzcat]

What's supposed to be happening? Seems OK here providing the Z driver is enabled.

[Roydea6]

The clock second hand does not do a full sweep and will jump back to 12 after about 40 seconds on the face. After the second sweep it will set the time back to the previous minute..

[flashjazzcat]

How do I get the analog clock? It just keeps asking me to enter "M/YY".

 

EDIT: Never mind - figured it out, and you're right. I'll look into it.

 

Nice program, BTW.

Edited September 25, 2016 by flashjazzcat

[flashjazzcat]

Try this:

 

upbi183.zip

 

Hour and second bitmasks were transposed, cutting seconds off at 39.

 

Thanks!

[flashjazzcat]

Here's the same thing for Incognito:

 

ipbi183.zip

 

I've had Roy's analogue clock running most of the day. I really like it, and it's a nice way to test the driver.

 

I'll bundle this into the firmware package and create a new thread for the final, final firmware.

[+remowilliams]

Posting just to note successful upgrades with the 91216 packages and the updated PBIs here of two U1MB (800XL/1200XL), one Incognito, and two SIDE2s.

 

And also to thank FJC again for his fantastic work and dedication

[flashjazzcat]

Pending a complete final U1MB, Incognito and SIDE/SIDE2 firmware release shortly (there has been a main/PBI BIOS waiting for release for some weeks now), here's an update of the U1MB SIDE XEX loader:

 

uload125.zip

 

This is by no means final, and may be considered a test update owing to the amount of code rewritten.

 

New features:

• Search facility totally rewritten and much more powerful. Searching is now recursive (from the current logged folder down), so one may scan an entire FAT partition for files whose names begin with the chosen string. As before, one just needs to start typing alphanumeric characters to enter search mode, but in the new version a new list is built containing search results, and since searches may take somewhat longer than before, you can navigate the list as it is being built.
• Searching can be curtailed with the Break key, and Esc cancels search mode and returns to the folder which was logged prior to search mode being entered. Any standard operation (XEX launch, ATR tagging, etc) may be performed on the search results list.
• In the search results, pressing Ctrl-F takes you out of search mode and to the highlighted file's actual location.
• Lists now have - where appropriate - scroll bars. These provide navigational aid and a unified UI across the main BIOS and loader.
• Metatags: [bASIC] anywhere in a filename causes the ATR/XEX to be booted with BASIC enabled, even if BASIC is turned off in the loader settings.

There are still some things to add before I will call this "final", including named drive slots and the ability to rotate ATRs from the menu. A couple of people also wanted the ability to reboot with SDX enabled, and I'll add that too if there's room left.

 

In the meantime, I'd be interested to hear how the recursive search copes with complex directory trees. There is a hard limit of 255 folders (the rest will simply be ignored), and the maximum number of displayed filenames is now 250 instead of 255 (as an aid to simplifying the scroll bars). Likewise, I would be interested to hear of any spectacular crashes or apparent bugs (neither of which I would be surprised to see).

 

Incognito and SIDE versions will follow later.

Edited November 6, 2016 by flashjazzcat