vetz wrote:

The IDE ports present on the card will boot a harddrive, but it will only function in real mode in Windows, which sucks in terms of performance.

Errr... you're supposed to install Promise drivers under Windows to get direct controller access to the card. It will not work with generic IDE drivers as this isn't a IDE bridge with standard memory ranges/IO ports, but a "complete" (but dumb) HBA card.

vetz wrote:

...just don't expect it to work wonders as a stand-alone IDE controller in your 386/486 as I did!

Ofc... This is a dumb Promise uncached HBA. If you want wonders, you need a smarter HBA, like Tekram IDE cached series.

hyoenmadan wrote:

vetz wrote:

The IDE ports present on the card will boot a harddrive, but it will only function in real mode in Windows, which sucks in terms of performance.

Errr... you're supposed to install Promise drivers under Windows to get direct controller access to the card. It will not work with generic IDE drivers as this isn't a IDE bridge with standard memory ranges/IO ports, but a standard HBA card.

The thing is that the Promise drivers do no work like this. The only thing they do is to have the card show up in device manager so you can change I/O and IRQ. They do not allow you to run the drive(s) in protected 32bit disk access mode.

hyoenmadan wrote:

vetz wrote:

...just don't expect it to work wonders as a stand-alone IDE controller in your 386/486 as I did!

Ofc... This is a dumb Promise uncached HBA. If you want wonders, you need a smarter HBA, like Tekram IDE cached series.

This works even worse than a standard ISA controller card, as you can't get out of MS-DOS compatibility (real) mode in Windows when running a harddrive from it (which you can with any other ISA controller card). It's designed for ATAPI connections. IDE cache controllers have their own issues, like I've started to document in this thread: VLB IDE cache controllers, benchmark

You are running into the brain dead esdi_506.pdr detection routines. It checks the CMOS drive types to determine whether to load or not. If you are running this card as a primary IDE controller with the onboard LBA BIOS enabled, you must set a drive type in the BIOS. The same issue happens when using the XT-IDE Universal BIOS with Windows 9x, a thread describing the problem complete with a patched XT-IDE BIOS can be found below.

http://www.vcfed.org/forum/showthread.php?514 … ndows-95-issues

Also don't expect any performance boost. Its still a plain old ISA card.

NJRoadfan wrote:

You are running into the brain dead esdi_506.pdr detection routines. It checks the CMOS drive types to determine whether to load or not. If you are running this card as a primary IDE controller with the onboard LBA BIOS enabled, you must set a drive type in the BIOS. The same issue happens when using the XT-IDE Universal BIOS with Windows 9x, a thread describing the problem complete with a patched XT-IDE BIOS can be found below.

http://www.vcfed.org/forum/showthread.php?514 … ndows-95-issues

Thanks for the info. I'm going to try if adding a manual dummy drive in the motherboard BIOS will fix it, it actually might. I have to add, there is NO info on this in the manual, documentation or on the packagaing. It's marketed as an BIOS addon card with support for extra ATAPI drives.

Also don't expect any performance boost. Its still a plain old ISA card.

It's the advertising on the packaging that claims this 😀 I benched the card today and there were a 2% performance difference, all within margin of error.

Good to learn a little more about the obscure IDE controller landscape pre-PCI, not a whole lot of good information out there. Thanks for doing some of the legwork lately. 😀

firage wrote:

Good to learn a little more about the obscure IDE controller landscape pre-PCI, not a whole lot of good information out there. Thanks for doing some of the legwork lately. 😀

Funny fact is that this piece of hardware is actually released after PCI controllers became normal, in 1997. Its main feature is to add LBA hard drive support (up to 128GB) on controllers/motherboards that dont support it natively.

I'm going to test this controller with my VLB cache controller from Promise (which does not have LBA support) and see if I can get it going 😀

Promise sold a ROM-only upgrade card called the DriveMAX. The LBA ROM should work with generic IDE cards, just don't expect any sort of enabling of faster DMA or PIO modes by the EIDEMAX ROM (unless Promise added support for a few chip sets). It likely won't work with caching controllers.

The XT-IDE Universal BIOS does have support for a handful of non-caching VLB controllers, so that is always an option.

vetz wrote:

Its main feature is to add LBA hard drive support (up to 128GB) on controllers/motherboards that dont support it natively.

So you just plug this bad boy in next to your existing controller - leaving the HDD connected to your existing controller, and the machine will magically work with LBA HDDs? That would be something I'm interested in for my older machines with CF->IDE thingos installed - I dislike disk overlays.

badmojo wrote:

vetz wrote:

Its main feature is to add LBA hard drive support (up to 128GB) on controllers/motherboards that dont support it natively.

So you just plug this bad boy in next to your existing controller - leaving the HDD connected to your existing controller, and the machine will magically work with LBA HDDs? That would be something I'm interested in for my older machines with CF->IDE thingos installed - I dislike disk overlays.

this will work for a zip 100 + a cdrom????
does it let u boot off a cdrom? or boot from a zip-100?