rasz_pl wrote on 2024-06-23, 05:37:

all of them, from all vendors?

The USB 2.0 drivers are all just microsoft generic drivers, there aren't any vendor specific drivers for USB 2.0 as far as I know, USB 3.0 had vendor specific drivers until windows 8.0 came out.

The source code for the USB 2.0 drivers came out with the XP source leak a few years back, as far as I know no one has looked closely at the code though.

Duffman wrote on 2024-06-23, 07:30:

rasz_pl wrote on 2024-06-23, 05:37:

all of them, from all vendors?

The USB 2.0 drivers are all just microsoft generic drivers

aaah, that indeed explains same behavior. I keep wondering what is hammering the CPU if not SOF/uSOFs - it was such a nice theory 😁

there aren't really many dos drivers for ehci

Sweetlow has a modified server 2003 usb 2.0 driver for win9x, no idea if it'll take a difference in slowdowns though.

http://sweetlow.orgfree.com/download/

Well I've used 2 lots of usb 2.0 pci cards both of which stop the pc from booting thr second card wouldn't even post bios, the first would get stuck on wait on the bios boot up page. VIA chipset seems to be the norm with these cards.

Yet another flaw with the gigabyte ga 5aa rev2.2 board the usb header is specific to the motherboard the pin layout is completely different and it's 8pin stead of 10 not that the last 2 do anything both being nc or ground pins.

I had a similar issue with a smaller 8-pin onboard USB header. It was extremely awkward trying to find the right pinpoint, but I finally managed to do it by referring to the manual and a little bit of trial and error. I had to repin a similar connector, since it was actually smaller in size than a standard DuPont header.

Turns out the second card I bought although appearing to be a genuine branded older card it has the same chipset as the modern chinnese knockoff card 🙁

Duffman wrote on 2024-06-24, 01:18:

Sweetlow has a modified server 2003 usb 2.0 driver for win9x, no idea if it'll take a difference in slowdowns though.

http://sweetlow.orgfree.com/download/

site unavailable

Did you discuss "Native USB" (NUSB) drivers by Maximus Decim here before?
Works like a charm in my Win98 build.

https://msfn.org/board/topic/43605-maximus-de … ve-usb-drivers/
https://www.creopard.de/projekte/windows-98-s … er-fuer-usb.htm (german)

crusher wrote on 2024-06-26, 06:09:

Did you discuss "Native USB" (NUSB) drivers by Maximus Decim here before? Works like a charm in my Win98 build. […]
Show full quote

Did you discuss "Native USB" (NUSB) drivers by Maximus Decim here before?
Works like a charm in my Win98 build.

https://msfn.org/board/topic/43605-maximus-de … ve-usb-drivers/
https://www.creopard.de/projekte/windows-98-s … er-fuer-usb.htm (german)

I think so. But they don't revive the dead horses either. 🙁 At least not here.

@schlang

odd, still works fine for me.

It's a http only site (no support for https), probably won't work with browsers with strict security turned on.

jmarsh wrote on 2024-06-26, 18:10:

It's a http only site (no support for https), probably won't work with browsers with strict security turned on.

thank you that explains it

Unearthing this deal I came upon a while ago. For those in the UK, SCAN are selling a PCI USB 2.0 + FireWire card with a 3.5" front bay for £0.49. I bought 5 to get the most out of it cause why not, considering the shipping is almost ten times more than the item itself...

https://www.scan.co.uk/products/scan-2x ... ternal-hub

Even though the pictures show x2 USB, x1 FW and whatever that other little port is on the front bay, mine has x3 USB and x2 FW. Compatible in my GA-BX2000 and SE440BX-2, and Windows 98SE somehow has drivers for them on the disc - not sure how since the OS obviously predates USB2.0.

giantenemycat wrote on 2024-07-04, 07:31:

£0.49

good price even for a 3.5 bay blanking plate 😁

giantenemycat wrote on 2024-07-04, 07:31:

whatever that other little port is on the front bay

4 pin firewire for stuff that didnt require power, like in cameras

digger wrote on 2024-06-22, 14:12:

It looks like there is a small but acceptable slowdown when you install a USB 2.0 PCI card in an older machine, but there will be considerable slowdown once you plug in any high-speed devices, even if you don't actually use to them.

This has been my experience with an NEC based USB 2.0 card in my GA-586T2 (socket 7) machine. More than once I've copied over a new driver or something with a USB stick, installed the driver, run a 3D benchmark to see how I'm looking, and am shocked to find that I've lost 100's of points off my score. Then realise I left the USB stick in 😁

But with the stick out there is no real difference between card installed / no card, and fast USB is so handy on a machine of that era.

I wonder if the decrease in the USB polling penalty is mitigated on NT-class operating systems due to the nature of preemptive multitasking, rather than concurrent multitasking. There is probably some unoptimized code somewhere that's taking up too many CPU cycles where it's not multithreaded.

I had plenty issues personally with VIA based USB cards on older systems (pre-Slot 1). I switched to an ALi based one and that ran without issues.

badmojo wrote on 2024-07-22, 01:07:

But with the stick out there is no real difference between card installed / no card, and fast USB is so handy on a machine of that era.

That's good. But what operating system do you use?

GeorgeMan wrote on 2014-04-27, 10:10:

swaaye wrote:

What cards have you guys used that work reliably on old boards? I have a card with a VIA controller and it isn't even detected by 440BX or AMD 750. It does work on newer chipsets like KT266 though.

You need a card with NEC chip, not VIA. Some old computers refused to boot with the card installed and some others didn't recognise the card at all. 😊
I bought 3 different VIA based pci cards from ebay before finding out...

Thanks for let me know, i have one at my home with old 2000 motherboard 307 socket, and it worked for 4 days, and know it malfunction. 🙁 I`ll guess i`ll have to get one with NEC chip then.