• Please review our updated Terms and Rules here

XTIDE Universal BIOS - success with other cards?

M

Megatron-uk

Experienced Member
Joined
Oct 3, 2011
Messages
122
Location
Durham, Durham, United Kingdom
Hi all,

I'm hoping to get larger disk access via XTIDE Universal BIOS on my no-name 286 system - I've tried Ontrack Drive Manager and while it detects, partitions and formats my drive to full capacity, I cannot boot anything via the Ontrack 'Insert a floppy and press space to boot an OS' screen.
So, I would like to run the replacement BIOS from either a network or SCSI card that I already own, but was wondering if anyone else had used the same cards to host the XTIDE BIOS?

The cards I've got access to are:

3Com 3C509B
Realtek 8029AS
Adaptec AHA-1542CP

The Adaptec has a 27256 SCSI BIOS chip - but the other two cards don't have anything fitted. I currently have a number of blank 27c128 and 27c256 eproms available that I can program. But is there anything special I need to do with the 8kb AT version of the XTIDE BIOS to get it to run in any of the above cards with either the 16 or 32kb chips I have available?

Cheers,
John
 
I'm using the XT-IDE BIOS in my 386 flashed into a ROM on a NE2000 NIC and a vanilla 16-bit IDE adapter. Just make sure the ROM window on the card you are using is sized to match the ROM you are plugging in. If it's too large, the ROM will be tiled and (in my case) lock up the 2nd time it's executed.
 
Hallo Eeguru,


IMHO your information is not completely correct. If you put a 16 KB EPROM in a socket meant for a 32 KB EPROM, you are correct; there is nothing you can stop the system to run the program twice with all related problems. But John says he has 27c256 EPROMs and those can be placed on the Adaptec card. But he has to take care that he only programs one 8 KB block of the EPROM and leaves the rest filled with $FF. In this way the program is only seen once. Disadvantage: other cards won't be able to use the space occupied by the Adaptec card. Unless you copy the program of this card into the unused area.


Groetjes, Ruud Baltissen
www.Baltissen.org
 
I do not get how would this work isn't there a EPROM in the ISA (ide controller) that set the 520 MB limit, or is it on the motherboard bios(without ide contoller)?
I have an APD50L idecontroller in an old XT, would this also be able to hold the XTIDE bios?

JT
 
Ok, I wrote the 8kb AT rom image to a blank 27c256 eprom and read it back again (included as the attached file).

I have tried the resulting eprom in both the AHA-1542 and the 3C509B, but I get no XT-IDE BIOS messages and no detected drives. The system just acts as if it is using the motherboard IDE support.

Doesn't appear as though either of those cards works as an XT-IDE BIOS host device. Unless anyone has any suggestions? The SCSI BIOS definitely works when I put the Adaptec chip back in again (so it's not as if the BIOS base address is configured crazily on the card).
 

Attachments

  • ide_at-27c256.txt
    32 KB · Views: 7
There's going to be 2 issues to fix here, both are independent from each other, both are required to get drives working.

#1 is that if you're not even seeing a BIOS message from the XTIDE ROM, then the 1542/509B card is not compatible with that eeprom you're using, or some other issue is keeping the ROM from being decoded through the card. I'm pretty sure there was a setup program for the 509B that lets you set the ROM type and size.
#2 is that you have to make sure the IO base address that is programmed in the XTIDE ROM has to match the IO ports your motherboard has for the onboard IDE controller. (typically it'll be 1f0 or 170) We can fix #2 after #1 is working, but without #1, you're not going to get anywhere.
 
Looks like it might be a no-go with the 3C509B. I ran 3C5x9CFG.exe and enabled the BOOT prom, set the rom size to 32kb and chose the same rom base address as I had configured with the SCSI card (0xDC000). Nothing - the system won't even POST. Remove the eprom and it starts again. I tried a few different base addresses and got the same result.

So it seems the 3C509B is not a suitable host.

Can someone tell me a card that is? 8bit would be good, but I can sacrifice a 16bit slot if necessary.
 
Last edited:
Megatron-uk said:
So it seems the 3C509B is not a suitable host.
Click to expand...

I'm using a 3C509B in a IBM 5170 AT to hold a 27C256 with the XT Universal BIOS that manages the IDE port of a Sound Blaster 16 card. So the 3C509B is a suitable host.
 
nestor said:
I'm using a 3C509B in a IBM 5170 AT to hold a 27C256 with the XT Universal BIOS that manages the IDE port of a Sound Blaster 16 card. So the 3C509B is a suitable host.
Click to expand...

I wonder what on earth is the matter with my particular setup then? Did you do anything specific when writing the image - perhaps concatenate 4 x 8kb roms to fill the full 32kb eprom? (I'm clutching at straws here!)
 
JT64 said:
Could i replace the ROM on the ADP50L(Uses 27256 or 27C256 ROM Chip) with a ROM with XTIDE or is there more to it?
Click to expand...
No. the ADP50L is a nonstandard IDE controller and doesn't use IO ports to access the drive, it's all memory mapped. That is an oddball card.
 
I'm trying to understand, so please be patient with me. As I understand it, you're trying to use the XTIDE universal BIOS on the IDE interface that's built onto your motherboard. If this isn't right, skip the rest of my comment. As Emily Latella said, "Never mind".

But if it's true, it's simply not possible using the XTIDE version of the Universal BIOS. Now, on the other hand, if you re-assemble/compile the universal BIOS to use the code for the standard 16-bit IDE interface (that is, change the conditionals and re-compile), you may have a lot more luck. As far as taking a larger EPROM and using it in a 64Kbit EPROM socket; there's no problem--just bend the extra address pins from the EPROM socket and tie them to ground.
 
Chuck(G) said:
I'm trying to understand, so please be patient with me. As I understand it, you're trying to use the XTIDE universal BIOS on the IDE interface that's built onto your motherboard. If this isn't right, skip the rest of my comment. As Emily Latella said, "Never mind".
Click to expand...
That is correct. My 286/AT motherboard has onboard BIOS with type 1-47 support and will not support the native capacity of any of my CF disks. I am attempting to program an eprom and load it via either my AHA-1542 SCSI controller, or the 3Com 3C509B network card. Both of which support 32k eproms.

edit: Sorry, not IDE on the motherboard, but via a standard multi-io card with no onboard BIOS of it's own (uses motherboard BIOS). Effectively the same thing of course.

But if it's true, it's simply not possible using the XTIDE version of the Universal BIOS. Now, on the other hand, if you re-assemble/compile the universal BIOS to use the code for the standard 16-bit IDE interface (that is, change the conditionals and re-compile), you may have a lot more luck. As far as taking a larger EPROM and using it in a 64Kbit EPROM socket; there's no problem--just bend the extra address pins from the EPROM socket and tie them to ground.
Click to expand...

I'm now confused... the XT-IDE manual, as found on googlecode states the following:

The most convenient way to use XTIDE Universal BIOS is to use XTIDE card. It can be used on any PC with free 8-bit ISA slot. You might not want to connect any drives to it in 16- or 32-bit systems since 8-bit transfer will be very slow. Using XTIDE card allows EEPROM flashing so it is really easy to update XTIDE Universal BIOS.

Another option is to use any card with free ROM socket for 8 kiB or larger ROMs. Official XTIDE builds are meant for 8 kiB ROMs but you can burn it on larger ROM if you append enough zeroes to the end (only append zeroes so checksum does not change). Many network cards have unused ROM sockets but there are also few multi I/O cards and IDE controllers with ROM sockets.

Is it not the case that the IDE_AT.BIN as found on the googlecode download area can be programmed onto an eprom and used from a NIC/SCSI/IDE card with onboard BIOS?
 
Last edited:
The key here is that you're using the AT version of the XTIDE universal BIOS. All is good, assuming you are using a 286 or higher (you are) and that your IDE controller is AT standard, using standard IO ports, which we will determine at some point. You have other problems, so hopefully we won't get sidetracked here.
 
Ok. So that confirms what I am trying to do should be possible. And nestor's comment about using a 3C509B to do the same thing as me means that at least one other person has gotten it working.

So the problem is either in the setup (or BIOS) of my system (which at the time I tried it only had a VGA card and a multi-io card)...
The process by which I'm burning the XT-IDE Universal BIOS...
Or the 3Com card I'm attempting to run the BIOS from
 
Okay, just making sure that the obvious has been covered.

The first question after this to me is "Does this card use the standard set of ports and interrupt"?

Make a DOS boot disk with the attached program on it and run the program. It should report the drive attached to your controller. If it doesn't, then there's something about your card that we haven't figured out yet.
 

Attachments

  • IDESDI.ZIP
    7.5 KB · Views: 8
Yep. It detects no problem:

P1020556.JPG
 
Good! That should settle the matter of port assignments.

Next, have you used DEBUG to actually locate the copy of your new IDE BIOS PROM in memory? Is it all there?
 
You must log in or register to reply here.
Back
Top