Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Hello, I got onto flashing a "ZEON 1" SIL3112ACT144 card today and got the following:

Loading BIOS...
Verifying...
Loading failed at address fc00h

Have not yet tested if she still works ...

Any guesses what is happening ?

Thanks, Jan
 
Hello, I got onto flashing a "ZEON 1" SIL3112ACT144 card today and got the following:

Loading BIOS...
Verifying...
Loading failed at address fc00h

Have not yet tested if she still works ...

Any guesses what is happening ?

Thanks, Jan

Jan, it seems that you have a card with a 64K flash memory chip ($FC00 is equal to 64,512 in decimal), whereas the ROM is 128K in size. I'll put together a special smaller 64K ROM for you this evening which should work fine.
 
Is the SeriTek firmware better for compatibility and funcionality than the Weibetech or are they much of a muchness?
 
Is the SeriTek firmware better for compatibility and funcionality than the Weibetech or are they much of a muchness?

The SeriTek firmware is generally more compatible than the Weibetech firmware, it works with OS 9 and doesn't have the problems with certain machines like the Quicksilver G4s. Unfortunately the ROM file is larger than 128K (most cheap cards have a 128K flash ROM chip), and nobody has been able to reduce it - so the only way of using it is to desolder the 128K flash ROM and replace it with a larger one. To complicate matters, it has to be a certain part number to be able to flash it with the DOS utility (unless you have an external programmer).
 
noones been able to reduce it because apparently
it may have some code inside itself that checks its own chiptype + size.. basically it has levels of 'copy protection' written into itself..
or at least, thats whats thought by few that have looked into it.. i would be thrilled to find someone who could backward engineer that one file.. it would be an amazing thing for the powerpc community..

the reason teh seritek firmware is so large is because it actually contains driver KEXT files for mac os x inside of the rom itself that it injects into the os enabling the boot.. there is a bit of seperation between the os9 area of the rom and the osx area of the rom

i had theorized that perhaps it may be possible for someone to seperate the two into 2 seperate smaller roms specifically for os9 + osx booting... this may be an effective way to "reduce the rom"
but it is a careful + precise hackjob that requires someone with a bit of embedded systems technical skill + education, or maybe a magician;)

it may not even be a workable strategy tho because if u reduce the rom and it still contains chipbrand + size check..theres not really much point as it wont work untill those checks are somehow removed and they could only really be removed if it was completely backward engineered + decompiled + recompiled... in whatever embedded systems / assembler language its written in
 
Last edited:
Thanks lads for the detailed replies. I am assuming that none of the cheap PC cards is going to cut it for the Seritek firmware. Now going to scour the other forums for details other than the three chips required.

For what it is worth, the card I flashed earlier with harry's ROM, which stopped my QS from booting, was recognised as a Wiebetech card in my Sawtooth, so it does seem to be specifically a Quicksilver 2002 incompatibility. I also have a QS 2001 I could test it with. Need to haul that one out of the loft, first.
 
Just to add here again. I have not experienced any boot issues with my Quicksilver. Indeed I have no IDE drives connected anymore (except for one DVD drive) and routinely boot from SATA. No idea why this has never affected me.
 
Last edited:
LOL, perhaps. Then the issue is truly specific to the 2002 and not the QS in general?

That seems to be the experience of others. I will haul down my QS 2001 and give it a whirl and report back.

[edit] Nope. My 2001 shows the same symptoms as the 2002. Will not proceed past the chime with the card in.

The card I have has the Silicon chipset: Sil3112ACT144, Q21930.1, 0247, 1.1 for what that is worth. The EEPROM chip is AM29F010, -120JC, 9508MBM, 1991 AMD. so no go for the SeriTek firmware.
 
Last edited:
I have flashed my card with an USB key on a Windows PC (with Freedos) without problem, and i have found a cool thing :

When i plug the card into the PCI 66 MHz slot from my Power Mac G3 B&W, it's faster.

With the card on a classic port, i write at ~35 MB/s and read at ~80 MB/s. With the same cheap card on the 66 MHz slot, i write at 80 MB/s and read at 90 MB/s with a SSD, it's better.

The only problem is that my graphic card (the original ATi Rage 128) is a little slower into a classic slot, and my GeForce FX PCI will not work with Mac OS 9...
 
I have flashed my card with an USB key on a Windows PC (with Freedos) without problem, and i have found a cool thing :

When i plug the card into the PCI 66 MHz slot from my Power Mac G3 B&W, it's faster.

With the card on a classic port, i write at ~35 MB/s and read at ~80 MB/s. With the same cheap card on the 66 MHz slot, i write at 80 MB/s and read at 90 MB/s with a SSD, it's better.

The only problem is that my graphic card (the original ATi Rage 128) is a little slower into a classic slot, and my GeForce FX PCI will not work with Mac OS 9...

CoreImage and OS 9 are mutually exclusive...Apple never bothered to develop drivers for it.

If you want the best OS 9 gfx card for PCI Macs, look for a Radeon 9200. It's actually a pretty decent card, even in a 33mhz PCI slot. The Mac edition has 128mb of VRAM and can drive two displays.

I'm running a flashed FX5200(the somewhat more rare 256mb version) in my G4-upgraded B&W at the moment in the 66mhz gfx slot. I've conceded to that being an OS X machine, and I'm going to get Leopard running on it once I have the time.

My 9200 is going to end up in my G4-upgraded 8600, which is also going to get Leopard(again, once I have the time). A flashed FX5200 won't work in an OWR Mac-otherwise I'd use one, as they're a lot cheaper(at least unflashed) than a Mac edition 9200. I just need to get the IDE hard drives working before I go down the OSX upgrade path.

In any case, though, I'm not surprised that your SATA card is faster in the gfx slot. Remember that bandwidth is shared across the PCI slots, and the slowest card can bottleneck the entire bus. The 66mhz slot on the B&W/Yikes is on its own bus, and of course also runs faster than the main 33mhz bus. Intell has said that the gfx slot in these computers can more or less be thought of as a 1x AGP slot.
 
Last edited:
After a careful reading of this thread, I have some questions:

1) I'm curious why the WiebeSATA3112.zip file is not mentioned in the original post, although it is alluded to in the later pages of the thread. This seems to me to be the most expedient way to flash the card since it can be done on the Mac. Could you enlighten me?

2) I have a SIIG SC-SAT212 card, identified as V 2.0, with the Sil3112ACT144 chip on it. Since I couldn't think of a reason not to, and I don't know that I can hustle up a PC to follow the OP instructions, I tried the WiebeSATA3112 Flash procedure. This is what I get in terminal:
FLASHROM.command requires Administrator privileges.
When prompted, please enter your Administrator password at the prompt.

******************************************************************************
WELCOME TO THE PCI CARD FIRMWARE/BIOS UPDATE UTILITY

This is the update program for Mac 10.2.8 & higher.
The following card(s) are supported.
1. Vendor ID : 1095 (Silicon Image)
2. Device ID : 3112
Please make sure that the above mentioned card(s) is present in the system.
*** NOTE: You must login as 'root' or 'administrator' to run this utility.
Copyright (c) WiebeTech 2006. All Rights Reserved.

Press C to continue; any other key to quit (followed by RETURN key):c

******************************************************************************
/Volumes/G4SEA-JAG/PCI Cards/Sil3112 Card/3112 Flash/WiebeSATA3112/FLASHROM.command: line 18: 258 Bus error ./FlashTool

logout
[Process completed]

Note that while the last line says "Process completed", I also see a "258 Bus error." I am trying to understand what this all means. I don't see the card in System Profiler, neither before nor after "flashing" the card. BTW, I'm running a G4 MDD 2003 @ 1.25 Mhz and Tiger 10.4.11.

3) Should I expect to see the card represented in some way in System Profiler even before it has been flashed?

4) Could the OP's "firmware.bin" file be substituted into the WiebeSATA flashware to complete a correct process on the Mac?

Your help greatly appreciated. This was an $8 goodwill pickup and the clock is running if I need to return it.

Best regards,

Mark Harrison
 
After a careful reading of this thread, I have some questions:

1) I'm curious why the WiebeSATA3112.zip file is not mentioned in the original post, although it is alluded to in the later pages of the thread. This seems to me to be the most expedient way to flash the card since it can be done on the Mac. Could you enlighten me?

2) I have a SIIG SC-SAT212 card, identified as V 2.0, with the Sil3112ACT144 chip on it. Since I couldn't think of a reason not to, and I don't know that I can hustle up a PC to follow the OP instructions, I tried the WiebeSATA3112 Flash procedure. This is what I get in terminal:


Note that while the last line says "Process completed", I also see a "258 Bus error." I am trying to understand what this all means. I don't see the card in System Profiler, neither before nor after "flashing" the card. BTW, I'm running a G4 MDD 2003 @ 1.25 Mhz and Tiger 10.4.11.

3) Should I expect to see the card represented in some way in System Profiler even before it has been flashed?

4) Could the OP's "firmware.bin" file be substituted into the WiebeSATA flashware to complete a correct process on the Mac?

Your help greatly appreciated. This was an $8 goodwill pickup and the clock is running if I need to return it.

Best regards,

Mark Harrison

The bus error means the flashing process fails.

You can flash it on the Mac only when:

The EEPROM chip (see my post #120) on the card is from these manufacturers: SST, ATMEL, PMC, or WINBOND. WiebieTech's flasher won't flash if it's from other brand. That's why the OP uses the Silicon Image's flasher, but it's PC only.

I also recommend you to boot into root safe mode to flash. I did that in 10.5.8. But I think Tiger should be also fine.
 
It seems SIIG SC-SAT212 V 2.0 uses SST39SF010 which is a 5v chip. I'm not sure the WiebeTech flasher would flash it even if it's the brand it supports. Because all the flashings I've done with the flasher has been with 3.3v chips.

When it's successfully flashed, the card will be shown up with the name WiebeTech in it.
 
for this, Thanks for the replies. The EEPROM chip on this card reads:

WINBOND
W29EE011P-90
202669501
0390NWA-

Another chip on the card has a swirl logo and:

33
APL1084
GG71U

I tried to make out a period between the two 3s, but to no avail. I'm optimistic that this card passes your criteria to flash on the Mac. Still haven't tried from root safe mode yet.

I reviewed your post #120 and followed onto your 68kmla.org link, but only got through page one so far. I'm concerned that the card doesn't show up in System Profiler at all. I've put other PCI cards into the slots I've tried and they are working fine, so I don't think the slots are the problem, but why wouldn't the unflashed card show up as "Other Mass Storage Controller" or some such, as suggested on the other site?

Could the card be a bust already?

Best Regards,

Mark Harrison
 
W29EE011P-90 is a 5v EEPROM chip.
APL1084 is a voltage regulator.

The card could be a bust then. Because an unflashed card should still show up in the System Profiler.
 
Thanks for this, I returned the card. Attended my first computer swapmeet over the weekend but only found two sata cards, both with via chips. So I passed, based on the advice I've seen here.

Need more help but will move it to a new topic.

Best Regards,

Mark Harrison
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.