Jump to content
IGNORED

In-place Ultimate 1MB / Incognito ROM editor and flash tool


flashjazzcat

Recommended Posts

yes..
-1.loads from SIDE II - empty screen
-2.q-meg - not load
-3.loads from apt - cursor without range..
-4.loads WITHOUT u1mb from APT - that same

maybe STOCK OS required? :) - i reflashed it with Q-MEG 5 :D

Edited by w1k
Link to comment
Share on other sites

ee..
something wrong..

i disconecct SIDE II (hw SIDE II) and loads uflash via SIO2PC with that same config and reflash back STOCK OS.. now o trying loads uflash.xex from APT - and again cursor out of range..

Link to comment
Share on other sites

100%.. trying boot with everything disabled - normaly start BASIC, then switch HI-SPEED - hi-speed sounds and BASIC..

hmmm, i have good load via SIO2PC, Q-MEG 4.4 ..(SDX ENABLED, 1MB, SIDE DISABLED)

-------------

i tryed use first release of this tool - that same problem now..strange..empty screen

-------------

hmm..maybe its conflict with SIDE II.. uflash works with sio2pc (with disconnected side)

Edited by w1k
Link to comment
Share on other sites

I'm using this configuration on a 1200XL as we speak:

 

post-21964-0-93639700-1389907875_thumb.jpg

 

SIO2PC on D2: (uFlash loads fine from there), and APT partition on D3: (uFlash loads fine from there). Tried running some other applications first to test for undefined memory, etc; no problems.

 

SDX is booting with no CONFIG.SYS here, so I don't know whether you have some other conflicting stuff installed. Try a clean SDX boot.

Link to comment
Share on other sites

SIO2PC on D1:, APT on D3 - load fail from d1, d3..
on config.sys flashed on SIDE II i add only this:

DEVICE C:>SDOS>RC_GR8.SYS
SET SC=C:>SDOS>SCMAIN.OVL
SET PROMPT=$L$P\

maybe its conflict with SIDE II.. uflash works with sio2pc (with disconnected side)

Edited by w1k
Link to comment
Share on other sites

SIO2PC on D1:, APT on D3 - load fail from d1, d3..

on config.sys flashed on u1mb i add only this:

DEVICE C:>SDOS>RC_GR8.SYS
SET SC=C:>SDOS>SCMAIN.OVL
SET PROMPT=$L$P\

maybe its conflict with SIDE II.. uflash works with sio2pc (with disconnected side)

 

That's not a "clean" boot, now, is it? Immediately I would suggest removing RC_GR8.SYS. UI in uFlash still supports this, and tries to switch to 80 column mode when the driver is present. However, there'll be a memory conflict since we're leaving the cartridge enabled. I'll remove all 80 column support from this application.

 

So - get rid of DEVICE C:>SDOS>RC_GR8.SYS, and I think your problems will be over.

 

EDIT: Here's a fixed version which ignores RC_GR8 and S_VBXE:

 

uflash.xex

Edited by flashjazzcat
Link to comment
Share on other sites

Hi Jon,

 

It loads fine on my Incognito. One thing I have noticed is that when dumping the Colleen OS ROMs, I get $800 bytes of $20 at the beginning of the file, and a total size of $3000. If I look at a proper 800 type .ROM file, it is $2800 in length, and does not have all those $20s at the start of the file. There is one $20 at the start, but that one belongs there.

 

XL/XE ROMs seem to dump ok.

 

Hope this helps you in debugging.

 

-K

  • Like 1
Link to comment
Share on other sites

It loads fine on my Incognito. One thing I have noticed is that when dumping the Colleen OS ROMs, I get $800 bytes of $20 at the beginning of the file, and a total size of $3000. If I look at a proper 800 type .ROM file, it is $2800 in length, and does not have all those $20s at the start of the file. There is one $20 at the start, but that one belongs there.

 

XL/XE ROMs seem to dump ok.

Thanks Kyle. That'll be correct about the 800 OS ROM dumps. I forgot to strip the 2K padding off them. I'll just adjust the offsets to suit; I'm glad you spotted that.

 

Of course, in the meantime, a dumped 12KB padded 800 OS will flash back to the same slot nicely. Nevertheless, I'll get this fixed. I also want to provide support for PBI ROMs without the 6KB or 2KB padding (for Ultimate and Incognito respectively); there's really no need for the padding to be in the distributed ROM, but that's the way they've typically been issued until now, and that's what The ROM Generator works with.

 

Anyway: I'll fix these issues shortly. Thanks again!

  • Like 1
Link to comment
Share on other sites

Hi.

 

I've just updated my pbi rom by running uflasher from sdx and every time i tried the file selector was showing some garbage in the file list (hearts in most cases). Changing drive from D1: to some other and then back to D1: returned file list to normal and after selecting rom file flashing went fine.

 

Thank you for a great tool!

Link to comment
Share on other sites

Thanks Kyle. That'll be correct about the 800 OS ROM dumps. I forgot to strip the 2K padding off them. I'll just adjust the offsets to suit; I'm glad you spotted that.

 

Of course, in the meantime, a dumped 12KB padded 800 OS will flash back to the same slot nicely. Nevertheless, I'll get this fixed. I also want to provide support for PBI ROMs without the 6KB or 2KB padding (for Ultimate and Incognito respectively); there's really no need for the padding to be in the distributed ROM, but that's the way they've typically been issued until now, and that's what The ROM Generator works with.

 

Anyway: I'll fix these issues shortly. Thanks again!

Think I'll change The ROM Generator so that the 800/Colleen OS Slots only saves the proper 10k block so that the 2k padding and 1/4th (4k) SIDE/Fat32 Loader is stripped out of it. This should avoid some confusion...

Link to comment
Share on other sites

What happens if I try to flash a proper OS ROM file (without padding)?

The OS will be misaligned by 2K. Best wait until I fix it up.

 

Also, may I flash only one slot without fear of erasing the others? Example, suppose I only wanted to change the BASIC image, may I load and flash just that one?

Absolutely - this was the primary aim of uFlash: to be able to change the contents of any slot without disturbing the others, even when using 64KB sector flash ROMs (although only earlier Ultimates have those). One of the reasons I never changed the contents of any of the BASIC or OS slots until I developed this tool was that I couldn't be bothered with the rigmarole of preparing and then laboriously flashing (either via SIO, or by pulling the PLCC out and putting it into the programmer) the entire 512KB.

 

I've just updated my pbi rom by running uflasher from sdx and every time i tried the file selector was showing some garbage in the file list (hearts in most cases). Changing drive from D1: to some other and then back to D1: returned file list to normal and after selecting rom file flashing went fine.

 

Thank you for a great tool!

Many thanks! As mentioned, the file selector (which I originally developed for a stalled SDX text editor project) is somewhat unfinished and not very robust. However, once the the flashing function is proven 100 per cent reliable, it'll be pleasant work to tidy up some of the other aspects of the program. Certainly at the moment, the file selector produces a list of hearts when it fails to open the target drive, when of course it should pop up an error message. Thanks for pointing this out, though. ;)

 

Think I'll change The ROM Generator so that the 800/Colleen OS Slots only saves the proper 10k block so that the 2k padding and 1/4th (4k) SIDE/Fat32 Loader is stripped out of it. This should avoid some confusion...

Yeah that would make sense. It's a shame that these "special cases" exist (i.e. the PBI BIOS, Incognito SIDE loader and Colleen OS ROM slots), since everything else can be neatly described in terms of 4KB banks. In particular, the Incognito's SIDE loader is interleaved with the Colleen OS slots, which necessitated a special "sector list", rather than calculated linear sectors. But at least that lines up on sector boundaries.

 

Since the Colleen OS slots and unpadded PBI BIOSes (the latter for both Ultimate and Incognito) require further offsets which don't line up on sector boundaries, I'll either write special code which intervenes and adds padding during the load process, or create a further byte-aligned offset table for the actual load offset. This will allow for unpadded PBI BIOSes (which currently come in at only 6KB) and Colleen OS ROMs. The padded PBI BIOSes are 32KB, so I think this is a change worth making. The padding was really only needed when we had only simplistic, bank-bound flashing tools.

 

An additional question of compatibility which I raised earlier is whether the main BIOS and SIDE loader for each interface should be delivered as a single 32KB package, or as two discreet 16KB files. Right now, uFlash expects the Ultimate BIOS and SIDE loader in a single contiguous 32KB chunk, which makes sense because they are contiguous in the ROM. Conversely, the Incognito SIDE loader (as mentioned previously) is interleaved with the Colleen OS ROMs, and uFlash expects the Incognito BIOS and SIDE loader as two separate 16KB files. It's very easy to choose one scheme and apply it to both interfaces. I suppose it depends on how things tend to be delivered. I would imagine SIDE loader changes would happen more frequently than BIOS changes, but it might be best to simply consult with Candle on whether BIOS and SIDE loader should be considered a single entity.

 

Just a quick query will modified os roms like the ones with Hias hisio every be able to run the UFLASH.XEX file.. I had to drop back to using QMEG4.4 to correct a blunder I did testing this file..

I'm sure uFlash works fine with the High-Speed OS, but I can certainly test again to be sure. The application makes no undocumented calls to the OS, of course. I know little to nothing about Q-Meg, meanwhile; I'd rather stipulate the use of a stock OS if there are major incompatibilities, but then again I can't see why there would be any.

Edited by flashjazzcat
Link to comment
Share on other sites

The OS will be misaligned by 2K. Best wait until I fix it up.

 

 

Absolutely - this was the primary aim of uFlash: to be able to change the contents of any slot without disturbing the others, even when using 64KB sector flash ROMs (although only earlier Ultimates have those). One of the reasons I never changed the contents of any of the BASIC or OS slots until I developed this tool was that I couldn't be bothered with the rigmarole of preparing and then laboriously flashing (either via SIO, or by pulling the PLCC out and putting it into the programmer) the entire 512KB.

I'll be eagerly waiting :)

 

This is absolutely awesome! Thanks for this incredible tool!

  • Like 1
Link to comment
Share on other sites

Sanity checks are a good safeguard against insane users :)

 

One other thing, could the default drive in the flasher be set to the default SDX drive? It always starts on D2: now, even if I load it from C:\ROMUTIL. Once I change to D3: in the drop down list, it shows D3:\ROMUTIL> at the top of the file selector box. This is a VERY minor detail.

 

-K

Link to comment
Share on other sites

Sanity checks are a good safeguard against insane users :)

 

And developers: I'd have already bricked one of my boards were it not for the magic byte and file size checks. ;) Speaking of which: I'll add code to ensure image files aren't too long as well.

 

One other thing, could the default drive in the flasher be set to the default SDX drive? It always starts on D2: now, even if I load it from C:\ROMUTIL. Once I change to D3: in the drop down list, it shows D3:\ROMUTIL> at the top of the file selector box. This is a VERY minor detail.

 

Not so minor. The default is D2: only because that's the drive I used for testing. As I said, I'll set it to the SDX current drive in the next update. ;)

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
  • Recently Browsing   0 members

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