I mean applying the settings by running with /autoload

Ok, good to know, I should do extra test with emm386. It's probably the way it allocate memory or something I don't know.

8-30-26 : Updated to 2.1.1
- Removed "shift" from Setting data structure (Thanks assembler bsf command)
- Fixed a few shift oops from data structure (there gone, free fix)
- Fixed some other issues the clanker that purged the ~1500 shift parameters found (confirmed the fix with the docs, and non, it's not 100% vibeclanked, it's just that I did not feel deleting 1500 things manually)
That include one or more SHIFT issue in the following : ICH2, ali1531, i1440lx, i82815, piix, piix3, piix4, sis471, sis5511, sis5591, um8886
That also include one issue that could have made a GPF if a user tried to change one of four specific IDE setting on ICH2 and 1 of 3 specific memory settings in sis5581
- Reverted back to CauseWay DOS extender instead of PMODE/W, While I like PMODE/W because it take like 35KB less space (and can be compressed with it's own tools), it also induce a loading delay when the user is running EMM386

Pr1ha wrote on 2026-08-20, 08:27:

I mean applying the settings by running with /autoload

"Fixed", Sadly it was the DOS Extender, I did switch to PMODE/W lately and of course I tested with no memory manager at all, himem, jemmx and xmgr... Of course I did not test with EMM386.
Returned to CauseWay.... Launch delay is now fixed.

Hi, @Rav,
And thank you for keeping on the work on this handy DOS tool!

I want to ask, if you can later add an option for enabling Distributed DMA (DDMA) through the BIOS, if the chipset supports DDMA?

Rav wrote on 2026-08-30, 07:06:

"Fixed", Sadly it was the DOS Extender, I did switch to PMODE/W lately and of course I tested with no memory manager at all, himem, jemmx and xmgr... Of course I did not test with EMM386.
Returned to CauseWay.... Launch delay is now fixed.

Does this mean, that CauseWay DOS extender is more compatible than PMODE/W?

aVd wrote on 2026-08-30, 07:29:

Hi, @Rav, And thank you for keeping on the work on this handy DOS tool! […]
Show full quote

Hi, @Rav,
And thank you for keeping on the work on this handy DOS tool!

I want to ask, if you can later add an option for enabling Distributed DMA (DDMA) through the BIOS, if the chipset supports DDMA?

Rav wrote on 2026-08-30, 07:06:

"Fixed", Sadly it was the DOS Extender, I did switch to PMODE/W lately and of course I tested with no memory manager at all, himem, jemmx and xmgr... Of course I did not test with EMM386.
Returned to CauseWay.... Launch delay is now fixed.

Does this mean, that CauseWay DOS extender is more compatible than PMODE/W?

Hi yeah, I already began a more complete implementation but so far only ali 15{2,3,4}1/3 are done. (don't remember if they have DDMA).
What chipset you want support? I can take a look

Rav wrote on 2026-08-30, 18:04:

Hi yeah, I already began a more complete implementation but so far only ali 15{2,3,4}1/3 are done. (don't remember if they have DDMA).
What chipset you want support? I can take a look

I want all of them 😀 I had a three hour chat session with the stupid free chatGPT bot, just to "convince" the intentionally broken search algorithm to find out for me all the intel chipset with officially documented DDMA in the available documentation (all internet and forum speculations excluded). If you want, I can send you the intel DDMS chipsets table, which I managed to populate during these mostly lost hours. I know it will be even more difficult to do such thing for VIA and ALi/ULi chipsets, but I may try another day.

As for my question about DOS extenders - silence means acceptance, so I'll take "default yes" as an answer.

aVd wrote on 2026-08-30, 18:27:

Rav wrote on 2026-08-30, 18:04:

Hi yeah, I already began a more complete implementation but so far only ali 15{2,3,4}1/3 are done. (don't remember if they have DDMA).
What chipset you want support? I can take a look

I want all of them 😀 I had a three hour chat session with the stupid free chatGPT bot, just to "convince" the intentionally broken search algorithm to find out for me all the intel chipset with officially documented DDMA in the available documentation (all internet and forum speculations excluded). If you want, I can send you the intel DDMS chipsets table, which I managed to populate during these mostly lost hours. I know it will be even more difficult to do such thing for VIA and ALi/ULi chipsets, but I may try another day.

As for my question about DOS extenders - silence means acceptance, so I'll take "default yes" as an answer.

Sure, if you don't mind sharing, I can take a look
As for the extender, I don't really know, it's not *really* a compatibility issue, it work, it's just that apparently pmode/w is trying to map all the available memory for itself and emm386 may be just mapping it over EMS for some reason, which is indeed slow

Rav wrote on 2026-08-30, 19:58:

Sure, if you don't mind sharing, I can take a look
As for the extender, I don't really know, it's not *really* a compatibility issue, it work, it's just that apparently pmode/w is trying to map all the available memory for itself and emm386 may be just mapping it over EMS for some reason, which is indeed slow

Ok, but since this intel DDMA capable chipsets information is based on official documentation (pdf) "ai" searches, I will not publish it here, because I don't trust the "ai" after what I saw many times from it as fake information and I don't want to feed the "ai" in future with its own possibly not 100% correct info. I'll send you a PM with the table attached in txt-file, so you could decide for yourself, if it could be used as a base for DDMA intel capable chipsets human research based on the official intel's documentation for the given chipsets.