snipe3687 wrote on 2025-07-14, 20:53:

I made the adjustments and uploaded to github. try it again. you shouldn't have any issues now 😀

It seems the vias are still an issue for PCBWay, and black was a no go:

1. Please see pic 1, there are some vias on pads, please confirm should we fill resin for them? If yes, please confirm should we fill for vias in pads or all vias.

2. Please see pic 2, some IC spacing is samller than 0.22 mm, we can not do black soldermask bridge for them. Please confirm if you need soldermask bridge. If yes, please modify file them send me new one. If not need, we will expose the position completely, okay? Or could you accept green, purple, blue, red? then we can do.

EDIT: Looks like the ZIPPED gerber file in the github was older than the extracted version of the files. I am resubmitting with PCBWay. Oops.

thanks for pointing that out! I will adjust that now.

snipe3687 wrote on 2025-09-29, 17:12:

thanks for pointing that out! I will adjust that now.

No problem! And good news, with the updated files and switch to a red PCB, it seems to have passed their review. They are working on a quote now for assembly.

All I did was remove the SOM from the BOM list.

Hi everyone,

I've been trying out old Windows DDKs and made an ATAPI driver for the weecee.

If any of you have time, please test it out and report back.

Duffman wrote on 2025-10-29, 05:46:

Hi everyone,

I've been trying out old Windows DDKs and made an ATAPI driver for the weecee.

If any of you have time, please test it out and report back.

I'm willing to, but I'm immediately lost when I try to use it. There is a long list of choices to select when we force the driver to change to what you have prepared. Any hint on what to select?

@Muon

Start with "Weecee / Pixel86 DX SD-IDE Controller in Native Mode (switchable)" first.
If that doesn't work try with "Weecee / Pixel86 DX SD-IDE Controller in Legacy Mode (switchable)" and see if that behaves differently.

Yeah, I could definitely use some help. I tried a few of the options that had “WeeCee” in the name, but they made my Windows 98 system extremely unstable and slow. Then I tried the very first option (I cannot remember exactly what it was called), and it completely bricked my system. I am willing to try again on a fresh Windows 98 installation, but I will need some guidance this time.

@mbandalauk

Can you post a screenshot of what it looked like in device manager?

Duffman wrote on 2025-10-29, 21:09:

@mbandalauk

Can you post a screenshot of what it looked like in device manager?

No, I cannot at the moment—my system is bricked. I will try setting it up again and see what happens. Initially, I used the “WeeCee / Pixel86 DX SD-IDE Controller in Native Mode (switchable)” option, but that made my system extremely unstable.

Duffman wrote on 2025-10-29, 20:51:

@Muon

Start with "Weecee / Pixel86 DX SD-IDE Controller in Native Mode (switchable)" first.
If that doesn't work try with "Weecee / Pixel86 DX SD-IDE Controller in Legacy Mode (switchable)" and see if that behaves differently.

Here's what my device manager looked like before I tried a swapping with your stuff:

I used WeeCee Legacy switchable and replaced the driver for "Primary IDE controller (single fifo)". Also tried Native - same result.

Here's what it leads to after booting:

Duffman wrote on 2025-10-29, 05:46:

Hi everyone,

I've been trying out old Windows DDKs and made an ATAPI driver for the weecee.

If any of you have time, please test it out and report back.

Wow! How did you do that? 😆
Has it solved the slow transfer speed on your weeCee? And does it involve the bugged DMA on this controller?

@Muon

Seeing "Standard Dual PCI IDE Controller" on that blue screen it looks like it went and rolled back to the default Microsoft driver and didn't install my driver.

I need to figure out a way to stop it from doing that.

Duffman wrote on Yesterday, 12:17:

@Muon

Seeing "Standard Dual PCI IDE Controller" on that blue screen it looks like it went and rolled back to the default Microsoft driver and didn't install my driver.

I need to figure out a way to stop it from doing that.

This is the state of things after my last post after a reboot:

After trying to disable a maximum number of entries in that area of Hard Disk controllers and trying to leave just your driver
+
going into the BIOS and setting the PCI controller to Native instead of legacy
+
having win98se revert to an earlier device configuration without your driver
+
having another go at forcing your driver

here is the state of things:

I've set up a ftp on the win98se side, and started testing a large transfer from my win11 machine.
It starts with a good burst of a few megs, but does the stop for 5s, resume for 5s.
is it faster than before? It's been so long since my last test that I don't remember. But it averages 300k/s, so it seems better than before (yay!), but slower than doing ftp transfers in the ftpsrv suite of mtcp.

@Muon

Can you go into safemode and then into your registry -

go to HKEY_LOCAL_MACHINE\Enum

delete the MF folder entirely

reboot back into normal mode

when it recreates the ATA channels it'll be called either - Primary ATA Channel (dedicated IRQ) or Primary ATA Channel (shared IRQ) - then my INF was used.
Or if it's still called Primary IDE Controller (Single fifo) then it's still defaulting to mshdc.inf