• Please review our updated Terms and Rules here

Chinese 8088-based laptop system with pirated 8088 BIOS

Open source is behind almost all the innovation in the software world over the recent years. Without the folks who spend their personal time, unpaid, creating amazing things, we’d be 100% beholden to the desires of Wall Street. That’s how you end up with Walmart and McDonalds ruling the world.

- Alex
 
NeXT,

With all due respect, this isn't about making money from open source work.

When somebody releases work under an open source license, they are making it available for others to use under the terms of that license. Which for something like GPL means [1] don't remove existing copyrights or credits and [2] share any derivative works under the same terms and conditions. Those are not odious conditions.

Somebody using open source work without providing credit is basically misrepresenting the work as their own, which is basically stealing. Sure, there is no real financial impact because the work was being given away for free, but that's what makes it even more appalling - it would cost nothing to leave the credits in place and acknowledge the people who did the work, so why make an effort to remove it?

This is a risk you take when you publish work under an open license, so nobody is whining about the economics of it. But it is fair to point out when some dirtbag is misrepresenting what they are selling, which is exactly what is going on here. It's just wrong. There isn't much to be done, but it's fair to note it and let potential buyers beware. Who knows what other shortcuts were taken.
 
I don't care. Genuinely don't.
I see innovation. Someone was passionate enough to make their own PC BIOS and someone else was passionate enough to see there being a viable market for ultra-compact DOS machines that they invested into designing one commercially. In fact against your own word, here's two nice things. Here y'all are taking it and ignoring what's been created because you don't want someone's feelings to be hurt. Someone who posted the source code for their work in public.
Every single one of us on this forum do not comply with copyrights, even though it's written in the rules and we'd never copy that floppy and dump that BIOS, right? We have absolutely no reason to respect Sergey's either when we buy this new product which is his work, just without his name on it. We can, but then you are the equivalent to the person who lurks around the mall at Christmas telling children in line at Santa's chair that the dude up there isn't really Santa.

Enjoy the product.
If you wish to appreciate Sergey's efforts you can. You can compliment him for starting the project, donate financially if that's his thing, acknowledge that his work is being used without acting like an absolute loser or there's an amazing idea since this is using his older work: motivate development of a fork with his latest BIOS that you can flash to it. Now you get all the features and bugfixes of the latest BIOS, plus he gets his name back on the screen.
Or, just act like [AN AMIGA FORUM] or [AN APPLE FORUM] where nothing gets done and key innovators are left to stand on high towers, preventing anyone else from making better products.

[drops mic]
What are you even going on about? This isn't a drops mic argument, not even close. There is no difference at all between abandonware and something actively being developed under GPL?

Perhaps you have an eastern view of copyright, IDK.
 
I don't care. Genuinely don't.
I see innovation. Someone was passionate enough to make their own PC BIOS and someone else was passionate enough to see there being a viable market for ultra-compact DOS machines that they invested into designing one commercially. In fact against your own word, here's two nice things. Here y'all are taking it and ignoring what's been created because you don't want someone's feelings to be hurt. Someone who posted the source code for their work in public.
Every single one of us on this forum do not comply with copyrights, even though it's written in the rules and we'd never copy that floppy and dump that BIOS, right? We have absolutely no reason to respect Sergey's either when we buy this new product which is his work, just without his name on it. We can, but then you are the equivalent to the person who lurks around the mall at Christmas telling children in line at Santa's chair that the dude up there isn't really Santa.

Enjoy the product.
If you wish to appreciate Sergey's efforts you can. You can compliment him for starting the project, donate financially if that's his thing, acknowledge that his work is being used without acting like an absolute loser or there's an amazing idea since this is using his older work: motivate development of a fork with his latest BIOS that you can flash to it. Now you get all the features and bugfixes of the latest BIOS, plus he gets his name back on the screen.
Or, just act like [AN AMIGA FORUM] or [AN APPLE FORUM] where nothing gets done and key innovators are left to stand on high towers, preventing anyone else from making better products.

[drops mic]
I have never seen anyone here dump a ROM, change the copyright to their own name, and start selling it. There is a huge difference between archiving and preserving defunct software/firmware vs ripping off active copyright for commercial gain. Surely you can understand this.

Sergey's copyright is not "preventing innovation" in any way. They could have easily complied with his license and it wouldn't have cost them a single cent. See NuXT. They also could have used another BIOS entirely (there happens to be a public domain XT BIOS...). But nope, they decided to be assholes and you've decided to defend that for some reason.

There is only one person here acting like an "absolute loser."
 
Would you be willing to send a message to the seller and request that they re-add Sergey's copyright info to the BIOS? If they modded it, they're entitled to add their own, but not to remove his.
Sure, worth a try, but aliexpress is listing an Estimated delivery date of Jun 5th, so I imagine the production is well under way and it'd be too late to make it into the shipping product. But maybe it can make it into subsequent bios version updates.
 
I heard a comment from someone in Discord who bought one claiming "The seller just sent a copy of the manual and BIOS, including the GPL copyright etc."

I wonder if that (at least in part) fulfills their requirement for source distribution; as I understand it technically they only have to distribute the source to those they send the binary too.

Changing the header still feels a bit sketchy, and I'm the furthest thing from a lawyer, but perhaps they're willing to be at least somewhat cooperative once made aware of the situation.
 
A couple of updates:
  1. A Google Drive location with a ZIP file containing the documentation and the BIOS source for Book8088 has appeared. Several people pointed me at it, apparently, the seller gives the location to the buyers. I am not sure if that happens by default, or on demand. The code indeed is a modified version of 8088 BIOS, version 0.9.4 or so from 2020. I haven't looked at that too thoroughly, but it appears that some comments in Mandarin were added, calls to detect and initialize serial ports, parallel ports, and floppy drives was commented out. I didn't notice any other significant changes yet
  2. Ars Technica published an article on Book8088 system. I contacted the author - Andrew Cunningham, and he was kind enough to add an update, with my comment about this system using a modified 8088 BIOS, without attribution and with GPL license removed. Several other outlets ran similar articles about either Book8088 or the 386SX system mentioning Book8088 too: https://gadgettendency.com/notebook...ixel-screen-book-8088-dos-is-on-sale-for-200/, https://www.extremetech.com/computing/200-mini-laptop-runs-ms-dos-has-640kb-of-memory, https://hackaday.com/2023/05/15/new-dos-pcs-in-2023/, https://www.tomshardware.com/news/hand-386-handheld, https://liliputing.com/past-meets-p...p-with-a-intel-8088-chip-and-640kb-of-memory/, https://www.ixbt.com/news/2023/05/19/intel-8088-640-640-200-book-8088-dos-200.html, and several more (I haven't contacted any of these)
Providing ZIP file is a step in the right direction. As others had mentioned, the system is already in production, or perhaps they are just selling the leftovers ;) So might be a bit too late to update the BIOS, even if they agree to do that. But at least it gives some information about the system, so the proper support can be added in 8088 BIOS or in GLaBIOS ;)
 
Last edited:
The seller responded to a few of my questions via aliexpress buyer's chat translation service. He said that it will absolutely be on sale again, it doesn't seem like he anticipated how popular it would be and he may be assembling them by hand? I'm not sure exactly what's going on in that respect.
 
Last edited:
The same company advertised a similar machine running a 386 (not a clamshell format more like a tablet with a physical keyboard) with VGA but they seem to have vanished from Aliexpress. Just as well.
 
I did a bit of analysis of the source code and the schematic for Book8088

The BIOS code is based on this commit: https://github.com/skiselev/8088_bios/commit/027f3f130fc60dadfac087f334a0b95d02819363 from August 8th, 2022.

I don't see any major changes in the code, such as a new functionality being implemented or any significant rewrites. It is mostly just comments, deletions, and minor changes.

I noticed the following minor changes:
  1. In some files, including the modified files below, my name and the GPL license were removed. Interestingly some files still contain my name and the GPL license.
  2. bios.asm (the main file) - A lot of code removed, e.g FE2010A chipset support, Xi 8088 specific second interrupt controller support, PS/2 mouse support, checking and printing serial ports, parallel ports, floppy drives configuration (Book8088 doesn't have any?). There is an interesting change regarding the PPI Port B (61h) register. On IBM it is a read-write port, apparently on Book8088 it is a write-only port. So instead reading and modifying it, BIOS simply writes the new value. That's mostly used for speaker control.
  3. keyboard.inc (keyboard BIOS functions) - AT-style keyboard and FE2010A chipset specific code removed
  4. at_kbc.inc (AT-style keyboard controller support) - many comments in Mandarin were added. I don't see any functionality changes. The surprising thing is that Book8088 does not appear to use an AT-style keyboard controller. I don't really know why they went through the trouble of documenting it. Perhaps it was used in an earlier prototype? The commented out "%define AT_KEYBOARD" seems to suggest that, because the rest of unused defines were simply erased
  5. messages.inc (strings that BIOS prints on the screen) - the copyright info replaced and the license info removed. uses "CPU" instead of "Main Processor", "FPU" instead of "Mathematics Co-processor", several changes to the processor names, removed strings related to the BIOS setup (absent in Book8088)
  6. cpu.inc (CPU type detection, FPU detection, turbo mode switching) - FE2010A chipset support deleted
  7. scancode.inc (keyboard scancodes translation table) - several comments in Mandarin were added
  8. sound.inc (boot melody) - code updates for PPI Port B (61h) being read-only
  9. delay.inc (well, delay subroutine) - comments in Mandarin added, along with "SHAN" (the name of the person modifying the code?).
The schematic of Book8088 is quite interesting.
  • In addition to the 8088 CPU and the 8087 FPU, it uses an original 8284 clock generator, 8288 bus controller, 8237 DMA controller, 8259 PIC, and 8253 PIT
  • It looks that they've got most of the discrete logic functionality for an XT motherboard into a CPLD, also the CGA controller functionality is implemented as two CPLDs and the original 6845 CRT controller. According to the pinout, the CPLDs seem to be Altera MAX7000S series or Atmel ATF1500AS series in PLCC84 package.
  • They also use some kind of microcontroller to implement the XT keyboard, that is interfaced directly to the XT motherboard logic CPLD
  • The system uses two EPROMs (why not flash ROMs?!) - one 27C512 for the BIOS, and one 27C256 for CGA font
  • The system memory uses one 512KB SRAM and one 128 KB SRAM chip. CGA uses a 32 KB SRAM for the video RAM
 
Last edited:
It looks that they've got most of the discrete logic functionality for an XT motherboard into a CPLD, also the CGA controller functionality is implemented as two CPLDs and the original 6845 CRT controller. According to the pinout, the CPLDs seem to be Altera MAX7000S series or Atmel ATF1500AS series in PLCC84 package.

Does the schematic detail how the CGA is interfaced to the LCD screen? IE, is the output RGB(I?) and a screen that's able to sync with its output directly connected, or... ? (A couple possibilities that occurred to me is they could possible be generating YPbPr and using a screen that's willing to accept component video at 15khz, or maybe you could get decent clarity with separated luma/chroma (Svideo) output? Since they're using a real 6845 double-scanning up to VGA frequencies seems a little less likely to me than 15khz, since I'd think doing so would break hardware compatibility for anything that programmed video modes the hardware, but I guess some kind of line doubler implemented in the CPLD might not be completely out of the question?)

Honestly, I'd like to believe the guy that did this work just made a *series* of honest mistakes and didn't really understand the terms of the license, but actively purging out the copyrights and credits from the files they touched is certainly... not great. I'm still *very* curious what's going to be preloaded on the CF card when/if these actually get shipped. If it's just a minimal DOS install (which, yes, technically shouldn't be preloaded on anything without a license, but I don't think Microsoft will sell you an OEM license anymore so it at least falls into that fuzzy "Abandonware?" category) and free/demo/shareware versions of games then it'll be easier to buy there wasn't active bad faith involved. Both the demo video and the listing sure liked showing off Planet X3, though, so... hey, maybe it was just free advertising to encourage buying a paid copy, right?

The system memory uses one 512KB SRAM and one 128 KB SRAM chip. CGA uses a 32 KB SRAM for the video RAM

Seems like a bit of a miss to not just use two 512K SRAMs and implement UMB backfill. Downside, I guess, is you might have to build in some DIP switches or something to allow disabling portions of it if you're going to support ISA add-ons.
 
Does the schematic detail how the CGA is interfaced to the LCD screen? IE, is the output RGB(I?

Yes, it's RGB, though analogue RGB, not DRGBI. I've attached the schematic. (For further reference, the BIOS source code is imported here and the original archive received by a friend who bought this can be found here.)

Despite what NeXT says, this is clearly not an "I wanna play DOS games" machine; there's no need to try to use so many original chips in a design for just that. It's much more a "let's make a laptop PC clone" and a really nice idea, actually, though with a few flaws. (I dunno why he uses a microcontroller to scan the keyboard but the doesn't use the original keyboard serial protocol, opening up the possibility of using an external keyboard.)

I've only glanced through the source code, but I believe this is not (much of) an active attempt to hide the source of the BIOS as just, "I'm annoyed by the huge GPL headers at the top of the source files." You'll notice he didn't remove it from all the files, just the ones he appeared to have been editing. I totally understand that myself; when I open a source code file I don't want to see a bunch of licensing cruft, I want just the code and relevant comments in the cleanest possible form.

That said, I think that even all copyright issues aside it's definitely Not Nice to have removed attribution to the original author, and it would look cooler anyway to have one's clone come up with a "Sergey BIOS" notice at boot, just as my old clone back in the day told me it was running a Phoenix BIOS.
 

Attachments

  • BOOK8088图纸.pdf
    4.8 MB · Views: 21
Does the schematic detail how the CGA is interfaced to the LCD screen? IE, is the output RGB(I?) and a screen that's able to sync with its output directly connected, or... ? (A couple possibilities that occurred to me is they could possible be generating YPbPr and using a screen that's willing to accept component video at 15khz, or maybe you could get decent clarity with separated luma/chroma (Svideo) output? Since they're using a real 6845 double-scanning up to VGA frequencies seems a little less likely to me than 15khz, since I'd think doing so would break hardware compatibility for anything that programmed video modes the hardware, but I guess some kind of line doubler implemented in the CPLD might not be completely out of the question?)
It uses a simple resistor based DAC to convert digital RGBI to analog RGB. More precisely, the it uses two signals per color and gray signal... so 6 bit color? HSYNC and VSYNC are produced by 6845 CRTC.
Perhaps the display module itself does the double scan?
Honestly, I'd like to believe the guy that did this work just made a *series* of honest mistakes and didn't really understand the terms of the license, but actively purging out the copyrights and credits from the files they touched is certainly... not great. I'm still *very* curious what's going to be preloaded on the CF card when/if these actually get shipped. If it's just a minimal DOS install (which, yes, technically shouldn't be preloaded on anything without a license, but I don't think Microsoft will sell you an OEM license anymore so it at least falls into that fuzzy "Abandonware?" category) and free/demo/shareware versions of games then it'll be easier to buy there wasn't active bad faith involved. Both the demo video and the listing sure liked showing off Planet X3, though, so... hey, maybe it was just free advertising to encourage buying a paid copy, right?
Curious to see what is on that CF card too. I wouldn't be surprised if it would be an unlicensed copy of Planet X3 on it.
Seems like a bit of a miss to not just use two 512K SRAMs and implement UMB backfill. Downside, I guess, is you might have to build in some DIP switches or something to allow disabling portions of it if you're going to support ISA add-ons.
I was going to mention that... But I decided to keep my judgement to myself. Perhaps extra jumpers is the reason for not implementing the UMB support.

Another technical detail, the data bus of all 82xx controllers, the XT logic CPLD, and the memory are connected directly to the CPU data transceiver. This probably will reduce reliability when using external ISA cards. Typically, these things sit behind an additional transceiver...
 
LCDs that take 15kHz RGB aren't that unusual

Maybe it was just bad luck that the random scaler board I tried didn't support it. (What's a little frustrating is the datasheet for the realtek chip driving it *does* claim that 31khz is the lowest it goes on the RGB input despite also sporting 15khz composite, but reading between the lines it seems like the sheet isn't clear about the differences between what the hardware can actually do and what the default capabilities of the reference firmware are.)

(I dunno why he uses a microcontroller to scan the keyboard but the doesn't use the original keyboard serial protocol, opening up the possibility of using an external keyboard.)

Took a look at the schematics, I would guess it's just to save a few parts; supporting the serial protocol would have required adding a number of extra parts that would have been redundant for the laptop. (Shift register and whatnot.)
 
... credit where credit is due, I guess, the multi-layer PCB is nicely laid out. To fit all those discrete parts on a hobby double-sided board it'd probably have to be around twice as large.
 
My friend's machine just arrived.

Curious to see what is on that CF card too. I wouldn't be surprised if it would be an unlicensed copy of Planet X3 on it.

Nope. There are unlicensed copies of MS-DOS and GWBASIC, unsurprisingly, and even some version of Windows, but the C:\GAME dir includes only ARAKNOID, ABCGA, and DUCK. I don't recognise any of those (except that I know the arcade version of Araknoid) or much else on the drive, but I'm sure the folks more into this newfangled 16-bit stuff will know more. I've attached a listing of everything on the card.
 

Attachments

  • FILES.TXT
    33.1 KB · Views: 26
I suspect "ABCGA" is "Afterburner".

Still might have been better to stick with shareware, there's plenty of reasonably impressive titles you could stick on there (Commander Keen, Monuments of Mars, etc), but at least it's unlikely the inclusion of what's on there is costing many sales of CGA Afterburner and Araknoid.
 
Back
Top