Just in case I am uploading my BIOS, I suspect the image is corrupted and working copy would be nice to compare. Unless of course somebody know how to identify parts from this cryptic ascii arts 😀
Hi! Just wanted to thank you for attaching the BIOS dump. I managed to get one of these boards, but it had some random EEPROM chip inserted (completely unrelated contents) by previous owner. I tried yours and the board works fine! Whatever issue you've had with yours, must be related to something else on the board - the BIOS seems OK!
I'll request the BIOS file gets added to RetroWeb's entry on this rare mobo, unless you want to do it yourself (as you are the real source!)?
Wow what a weird board. Bios is "Phoenix NuBIOS Version 4.04" dated 09/05/96. Thats two years pre Award Software merger. I havent seen any pure Phoenix bioses before, gonna have some fun with this one.
btw Bios is 64KB only, half the dump is empty and redundant.
Apologies for the necro post, but this is such a rare board..
@rasz_pl Your comment regarding redundancy of the BIOS file explains why I see pictures of this board with 512kbit chip used for BIOS. By the way, you might find the contents of the chip that came with this board interesting (see attached, note numerous ascii texts in polish language). I think the previous owner just inserted a random chip into this board.
BIOS itself is quite interesting, albeit very limited in options. UI reminds me of Duke Nukem 3D setup 😁
Weird bios, weird chipset, weird manufacturer, fake cache, whats not to love 😀
When I have some time ill try finding 86box config able to run this bios.
Weird bios, weird chipset, weird manufacturer, fake cache, whats not to love 😀
When I have some time ill try finding 86box config able to run this bios.
Great! Interesting, not all this pop sh@t
Need help? Begin with photo and model of your hardware 😉
I am aroused about any X86 motherboard that has full functional ISA slot. I think i have problem. Not really into that original (Turbo) XT,286,386 and CGA/EGA stuff. So just a DOS nut.
PS. If I upload RAR, it is a 16-bit DOS RAR Version 2.50.
this : https://dfarq.homeip.net/texas-instruments-486sxl-cpu/
"TI also offered the 486SXL in 486SX packaging that plugged into a 168-pin socket"
is the only mention of TI variants being made also in 486 pinout I could find. Newer saw such CPU, not that Im a CPU aficionado or anything.
reposting Stojke image as imgur seem to have deleted his original and might delete hotlinked small version too
Stojke image is in full quality on https://theretroweb.com/motherboards/s/adc-gc10a
NEC did make gate arrays/structured ASICs in the nineties, it would be perfectly reasonable for someone to make a chipset based on such chip and manufacture it at NEC.
Last edited by rasz_pl on 2026-07-21, 21:28. Edited 1 time in total.
Running lspci command in Linux might settle the issue, but based on what I have seen, I believe the "NEC" chipset here was developed by Taiwan's RDC Semiconductor.(https://www.rdc.com.tw/en)
That being said, NEC likely offered standard cell based ASIC foundry service to RDC Semiconductor, and even did the packaging for them in-house.
The silk marking of this chip appears to look like an authentically NEC packaged IC, and historically, Japanese semiconductor companies did in-house packaging in Japan, at least until 1990s.
I think this rather odd Taiwanese company (RDC Semiconductor) emphasizing that the chip was manufactured by Japan's NEC has something to do with Japan being the only Asian country to achieve world class (i.e., Western standards) semiconductor technology at the time (1990s), so this firm wanted to advertise that.
Of course, the world has changed a lot in 30 years or so, and today it is perfectly normal for fabless semiconductor companies to use Taiwanese foundries like TSMC and UMC to get their chip to be manufactured on a contract basis.
Japan's relative importance in the electronics industry has significantly waned over the past three decades, and it is not a total coincidence that Japan now ranks mere 6th place in per capita GDP in Asia ($33K USD) versus Taiwan ranking 4th place ($37K USD).
The top three of the per capita GDP in Asia are all rather fake "city" states (Singapore, Hong Kong SAR, and Macau SAR).
Japan has also earned the dubious distinction of having no economic growth averaged over the past 35 years (this used to be called the "lost" decade, which then turned into two lost decades, three lost decades, etc., and people stopped using this phrase recently), despite having many world class corporations and highly educated populous.
Did the rise of Taiwanese foundries like TSMC and UMC and fabless semiconductor business model contributed to this dubious distinction of having 0% average economic over the past 35 years?
I will say the answer to be, yes, it did contribute to their lack of economic growth, at least on the margin.
That's my 2 cents.
Apologies for the necro post, but this is such a rare board..
@rasz_pl Your comment regarding redundancy of the BIOS file explains why I see pictures of this board with 512kbit chip used for BIOS. By the way, you might find the contents of the chip that came with this board interesting (see attached, note numerous ascii texts in polish language). I think the previous owner just inserted a random chip into this board.
BIOS itself is quite interesting, albeit very limited in options. UI reminds me of Duke Nukem 3D setup 😁
this bios seem a bios from equipment. eg mg/dm3......
Board data is wrong for all of them, lists cache support instead of warning about fake cache 😀
Looks like whoever was making those chipsets managed to advance all the way to Pentium support :0 Latest chipset revision ADC009B, called the chipset "Power XT" of course 😀and traditionally also doesnt support L2 cache 😁 and does support L2 cache!. Im amazed they managed to support
64bit EDIT: running Pentium on 32bit ram bus with optional 64bit L2 cache 😮
PCI
SDRAM. EDIT: HOW? I think half might be wasted since modules are 64bit and board runs everything on 32bit bus.
two IDE channels!
1 bios mentions PIO Mode 4, no DMA
2 IDE bus is riding on PCI bus pinout like in some Pentium era solutions
RTC - on structured ASIC???
Keyboard controller
EDIT: PS2 Mouse controller
EDIT: Pipelined Burst Cache and Tag chips. This chipset actually does have L2 cache support!
all in same 208 pin package they started with 486.
EDIT: This board is kinda awesome in its Shanzhai'ness. Its running Pentium CPU on a 32bit ram bus 😁 SIMMs data pins are wired in parallel = 32bit only. BUT L2 cache chip footprint is for 64bit 128pin chip like IS61LV6464 or UM61L6464AF 😮 What?!?!?!
can you also add GC10A Re: NEC 486 Chipset bios?
and correct gc9 gc10 gc10a descriptions to add FAKE CACHE warning?
EDIT: now Im not so sure about no L2 cache support on ADC006 without seeing back of PCBs.
Now whats up with manufacturer? Babasha example gc111 https://theretroweb.com/motherboards/s/adc-gc111-gciii has "Eagles" on the bios chip sticker.
I dont think Eagles is "Eagle Computer Technology Co."
All those boards are 'Made in China' and _very_ late for 486, datecodes are 1996-98! There was no "Eagle Computer Technology Co." in China until 1998. There was one in Taiwan, but Taiwanese company would print 'Made in ROC' not 'Made in China'. Most screenshots come from China. Then we have Korean LG Prime 3C IO on every one of them, almost no one used those on mobos (some Zidas).
ps: This led me to compare Prime 3C https://theretroweb.com/chip/documentation/pr … ac445801875.pdf to IT8661F https://ardent-tool.com/datasheets/ITE_IT8661F_RF.pdf present on one of gc111 examples and they are pretty much 1:1 pin compatible (miniscule differences) 😮 didnt know that.
This looks like someone in China in mid to late nineties was learning to manufacture motherboards/chipsets with a low risk low cost low end product. Pretty awesome!
All three available BIOSes are 'Phoenix NuBIOS Version 4.04', but not the same. Looks like some mayor changes inside (havent disassembled yet). Bios stickers dont mention Phoenix making me think maybe even unauthorized copy.
Babasha gc111 "586 System BIOS 1998" "This BIOS, even though completely out of whack will function correctly when programmed to UV ROM 27C010 or flash SST 27EE010" is NOT a standard Award/Phoenix/Ami. Its not like anything I ever saw before 😮 and I fail when trying to disassemble 😐 it starts with
"jmp near ptr unk_F0BFF3" which lands in a sea of 00000 bytes.
Just glancing at the bios Im guessing gc111 theretroweb description "Max RAM size 256MB" is wrong as BIOS goes only to 64MB 😮
Edit: Jumper description document also says max 64MB confirming my finding.
Its a 128KB bios for Pentium board 😮 and so far the code is nightmare of waste, rows of same functions with just few parameters changed instead of one function being called with dynamic data :0, 40KB53.7KB is just 0 filled empty space.
I would looove to know who was behind those ADC chipsets, whats the story, where did company go from there, and who wrote that BIOS 😀
EDIT:
>fail when trying to disassemble 😐 it starts with "jmp near ptr unk_F0BFF3" which lands in a sea of 00000 bytes.
Turns out BIOS is jumbled up. Some address lines must be crossed, either because that was easier to route or as some kind of obfuscation attempt.
Board data is wrong for all of them, lists cache support instead of warning about fake cache 😀
Looks like whoever was making those chipsets managed to advance all the way to Pentium support :0 Latest chipset revision ADC009B, called the chipset "Power XT" of course 😀and traditionally also doesnt support L2 cache 😁 and does support L2 cache!. Im amazed they managed to support
64bit EDIT: running Pentium on 32bit ram bus with optional 64bit L2 cache 😮
PCI
SDRAM. EDIT: HOW? I think half might be wasted since modules are 64bit and board runs everything on 32bit bus.
two IDE channels!
1 bios mentions PIO Mode 4, no DMA
2 IDE bus is riding on PCI bus pinout like in some Pentium era solutions
RTC - on structured ASIC???
Keyboard controller
EDIT: PS2 Mouse controller
EDIT: Pipelined Burst Cache and Tag chips. This chipset actually does have L2 cache support!
all in same 208 pin package they started with 486.
EDIT: This board is kinda awesome in its Shanzhai'ness. Its running Pentium CPU on a 32bit ram bus 😁 SIMMs data pins are wired in parallel = 32bit only. BUT L2 cache chip footprint is for 64bit 128pin chip like IS61LV6464 or UM61L6464AF 😮 What?!?!?!
can you also add GC10A Re: NEC 486 Chipset bios?
and correct gc9 gc10 gc10a descriptions to add FAKE CACHE warning?
EDIT: now Im not so sure about no L2 cache support on ADC006 without seeing back of PCBs.
Now whats up with manufacturer? Babasha example gc111 https://theretroweb.com/motherboards/s/adc-gc111-gciii has "Eagles" on the bios chip sticker.
I dont think Eagles is "Eagle Computer Technology Co."
All those boards are 'Made in China' and _very_ late for 486, datecodes are 1996-98! There was no "Eagle Computer Technology Co." in China until 1998. There was one in Taiwan, but Taiwanese company would print 'Made in ROC' not 'Made in China'. Most screenshots come from China. Then we have Korean LG Prime 3C IO on every one of them, almost no one used those on mobos (some Zidas).
ps: This led me to compare Prime 3C https://theretroweb.com/chip/documentation/pr … ac445801875.pdf to IT8661F https://ardent-tool.com/datasheets/ITE_IT8661F_RF.pdf present on one of gc111 examples and they are pretty much 1:1 pin compatible (miniscule differences) 😮 didnt know that.
This looks like someone in China in mid to late nineties was learning to manufacture motherboards/chipsets with a low risk low cost low end product. Pretty awesome!
All three available BIOSes are 'Phoenix NuBIOS Version 4.04', but not the same. Looks like some mayor changes inside (havent disassembled yet). Bios stickers dont mention Phoenix making me think maybe even unauthorized copy.
Babasha gc111 "586 System BIOS 1998" "This BIOS, even though completely out of whack will function correctly when programmed to UV ROM 27C010 or flash SST 27EE010" is NOT a standard Award/Phoenix/Ami. Its not like anything I ever saw before 😮 and I fail when trying to disassemble 😐 it starts with
"jmp near ptr unk_F0BFF3" which lands in a sea of 00000 bytes.
Just glancing at the bios Im guessing gc111 theretroweb description "Max RAM size 256MB" is wrong as BIOS goes only to 64MB 😮
Edit: Jumper description document also says max 64MB confirming my finding.
Its a 128KB bios for Pentium board 😮 and so far the code is nightmare of waste, rows of same functions with just few parameters changed instead of one function being called with dynamic data :0, 40KB53.7KB is just 0 filled empty space.
I would looove to know who was behind those ADC chipsets, whats the story, where did company go from there, and who wrote that BIOS 😀
EDIT:
>fail when trying to disassemble 😐 it starts with "jmp near ptr unk_F0BFF3" which lands in a sea of 00000 bytes.
Turns out BIOS is jumbled up. Some address lines must be crossed, either because that was easier to route or as some kind of obfuscation attempt.
Did the rise of Taiwanese foundries like TSMC and UMC and fabless semiconductor business model contributed to this dubious distinction of having 0% average economic over the past 35 years?
I will say the answer to be, yes, it did contribute to their lack of economic growth, at least on the margin.
That's my 2 cents.
I think many Japanese believe that their economy was intentionally gimped by Ronnie Raygun and his friends to keep them in their place. They were starting to make the US look bad.
Thanks for engaging me on what I posted about this rather odd looking "NEC" 486 chipset and the fate of the Japanese semiconductor industry.
I do not wish to hijack the thread, but I will write a little more to get my point across.
Around mid-1980s, US started to threaten Japan with tariffs on semiconductors, color TVs, VCRs, etc.
Sounds eerily similar to what the current occupant of 1600 Pennsylvania Avenue has been doing since early 2025.
Shortly after, a rather one sided US-Japan Semiconductor Agreement was inked, and boy, Japan made pretty staggering concessions on how to conduct their semiconductor business.
1) US got to set minimum prices on Japanese DRAM sold in the world market
2) Japanese manufacturers were forced to disclose their DRAM cost (pretty much a top secret at any firm)
3) Japan had to agree to allocating 20% of domestic Japanese market to "foreign" (read: American) semiconductors
These, along with Intel's suit against NEC's V20 and V30 processors (Intel 8088 and 8086 pin compatibles with 80186 instruction support) sent a message to the Japanese side that US will not tolerate Japanese manufacturers from getting into the more lucrative microprocessor business.
South Korean DRAM manufacturers (Samsung, Gold Star, and Hyundai) were not impacted, and eventually pushed most Japanese DRAM manufacturers out by early 2000s.
What is relatively underreported is the stealthy Taiwanese pure-play foundries (i.e., TSMC, UMC, etc.) taking Silicon Valley fabless firms's foundry service away from Japan quietly around 1995 to 1998.
I noticed back then that most PC chipsets now say Korea (Intel) or Taiwan (OPTi, VIA, ALi) instead of Japan.
If I recall, Matrox G400 was pretty much the last PC related major chip fabricated by a Japanese foundry (not coincidentally by NEC), and after that, I cannot recall any high volume chip other than NEC USB 2.0 and 3.0 chips that was PC industry related.
Having lost DRAM and logic foundry services, Japanese semiconductor industry no longer had the volume necessary to keep their fans profitable, yet they still did build fair number of expensive 300 mm fabs in the early 2000s.
Alas, it purportedly ran around 20 to 30% capacity, and around 2007, 4 manufacturers (Renesas, NEC, Fujitsu, and Toshiba) all gave up on process technology beyond 40 / 45 nm process.
Regarding your comment on making US side looking bad, I think there was some what I will call racial overtone in the background that influenced the policy implemented by the US side.
Not surprisingly, today's controversy regarding AI and semiconductors with regards to China is reminiscent of what happened to Japan's semiconductor industry in the 1980s.
Did the rise of Taiwanese foundries like TSMC and UMC and fabless semiconductor business model contributed to this dubious distinction of having 0% average economic over the past 35 years?
I will say the answer to be, yes, it did contribute to their lack of economic growth, at least on the margin.
That's my 2 cents.
I think many Japanese believe that their economy was intentionally gimped by Ronnie Raygun and his friends to keep them in their place. They were starting to make the US look bad.
Around mid-1980s, US started to threaten Japan with tariffs on semiconductors, color TVs, VCRs, etc.
Sounds eerily similar to what the current occupant of 1600 Pennsylvania Avenue has been doing since early 2025.
Not similar at all. Japan was dumping electronics on US market below cost, some of Japanese CEOs did it to take over the market, others (was it Matsushita the founder of Panasonic?) as a revenge for losing the war 😐
Frontline: Coming From Japan [The Fall Of The US Television Industry] (1992) https://www.youtube.com/watch?v=aesJTsZqm6c
Mastering the Market: Japanese Government Targeting of the Computer Industry https://www.jstor.org/stable/2706788
MITI'S SUCCESSES AND FAILURES IN CONTROLLING JAPAN'S TECHNOLOGY IMPORTS December 1994 https://www.jstor.org/stable/43294946
TLDR: Modern Chinese tactics like foreign currency quotas, ownership limits (minority shares, joint ventures), dumping while being propped by locked in internal market.
