Talk:MOS Technology 8563

Reason for revert of renaming

Since there are two MOS chips commonly called "VDC"; this one and the 8568 [D]VDC (the latter is more often than not referred to as the 8568 VDC, i.e. without the "D"); I reverted the recent renaming. --Wernher 03:12, 14 December 2005 (UTC)

Update: see the next thread. --Wernher 04:33, 14 December 2005 (UTC)

Merge 8568 article into this one?

Since the MOS Technology 8568 article is nothing but a very incomplete telling of the information in this one, we might be better off merging the two, and just note the differences between the chips in the common article---whose title should probably be "MOS Technology VDC", which I recently reverted to "8563" before thinking things through and coming up with this proposal. *sigh* --Wernher 04:33, 14 December 2005 (UTC)

Absolutely! We don't have multiple articles on the VIC-II or the SID, even though these also had multiple numbered revisions. Why not just merge both this and MOS Technology 8568 into MOS Technology VDC? We can add a short blurb about the 8568 revision and what (if any?) differences it had with the 8563 in the article. The 8568 article just doesn't have enough to be said about it to stand on its own. Crotalus horridus 05:52, 14 December 2005 (UTC)

I have found a reference [1] with 8568 data. It explains the similarities and differences between the two VDC versions, and quite confirms our comments above. From earlier discussions about merging or not of chip articles, however, I know that some editors favor keeping such articles separate if the differences include pinouts. I'm not 100% sure where I stand on this issue, really. A combined article showing both chips' pinout diagrams would make for quicker/easier comparison of the chips, though. --Wernher 08:42, 14 December 2005 (UTC)

If we want to include the pinouts, I don't see why we can't do it in one article. In fact, we could even do it in one image - a side by side comparison. Crotalus horridus 09:00, 14 December 2005 (UTC)

After checking my C128/DCR Maintenance & Repair Manual I'm suddenly not so sure anymore; turns out the 8568 is quite different after all, HW-wise. Therefore I say we wait a little before we decide on this---I'll try and effect a 'total makeover' of the 8568 article so that it becomes complementary to the 8563 one. There are also several other names used for the two chips, in addition to (D)VDC, which might also play a part in settling the article title issue.

Nevertheless, I was a good thing we started discussion this, since it instigated my rummaging about on the 'net and in my own pile of old docs to find out more. :-) --Wernher 06:39, 15 December 2005 (UTC)

Binary prefixes

Recently changes have been made to this article to use binary prefixes (KiB, MiB, kibibyte, mebibyte etc). The majority of reliable sources for this article do not use binary prefixes. If you have any thoughts/opinions then this specific topic is being discussed on the following talk page Manual of Style (dates and numbers) in the sections to do with "binary prefixes". Fnagaton 10:41, 25 April 2007 (UTC)

8563 could do a lot more

The main article needs to be corrected.

With a good 80 column monitor, the 8563 in a Commodore 128 could generate a full-color text mode of 30 rows by 80 columns, and that's without adding any extra ram. I.e., using just the original 16K of ram on the 8563. (That's if you used third party software written in assembly language, and were willing to do without BASIC, that's clearly five more rows' worth of text than you would find using the software ("firmware") that came pre-packaged with your Commodore 128.) 198.177.27.16 (talk) 02:13, 26 January 2009 (UTC)

Re-suggest merging

I've looked over these two articles, and the majority of the content appears to be redundant even today, years after the first comments on the topic, above. I can't see why the D article can't be a single "compare and contrast" section in this one, and some sidebar material. Maury Markowitz (talk) 15:48, 2 November 2013 (UTC)

Significant misunderstanding re capabilities from (and in) source...?

(Copying over original topic from 8568 talk page as it's probably at least as relevant here)

OK, the claim of 720x700 seemed extremely suspect, as regular 15khz displays just don't have enough lines to do that even in interlace, and there was nothing else mentioned of multisync etc (even though the controller looks like it could be reprogrammed to give different line frequencies and counts, your pixels per line would be somewhat limited as the controller was most likely only using a single clock of around 14 to 16MHz). You might be able to get away with MDA or even something a bit like EGA, and interlace those up to actually reach some of the stated resolutions, in software, if it was a full 16MHz, but that would be your limit, and the others mentioned in the article such as "Atari ST standard" 640x400 would be out of reach because that requires a 32MHz clock instead, or pixel rates somewhere in-between. And was the chip even made to do interlace natively, when that would be no use for the text mode it was mainly employed for? (And if it was actually 14MHz instead, as used by CGA, you would maybe scrape 640 pixels on an MDA and certainly not get that many on EGA)

Additionally there's mention in the attached article of 65,000 colours and suchlike, with attached images that look definitely more like the output of an analogue system rather than dithered/flickered RGBI, and you're just not doing that with the C128's standard hardware.

But, most crucially, one of the scanned articles that the linked source relies on shows and describes a HARDWARE graphics booster that provides those expanded capabilities. It's an add-in graphics card, essentially. Either it's overclocking something, or adding its own RAMDAC and faster memory and pixel clock circuit, or a bit of both. I wouldn't be entirely surprised to find that it's got some relation to the Commodore C900 workstation computer, FWIW, as apparently it had some version of the C128 chip in, but running at upto 640x400 (and probably noninterlaced in that application, thus probably a higher frequency than in the C128 and with at least 32 if not 64k memory, as it's unlikely that having to lace to achieve that relatively "low" resolution would be accepted in a workstation application). The underlying hardware may well still be good for that level of performance, but locked down to more modest output by the rest of the C128 circuitry and need the booster to achieve its full potential.

Anyway... In which case that booster is its own thing and nothing to do with either the 8568 or the 8563, any more than other hardware graphics enhancements are with the native graphics hardware of other machines they might be found in.

Is there any way of properly certifying what the chip could do *by itself* within the computer as sold, without any add-ins, and maybe only with simple programming tricks that could be employed by the everyday user at their own risk of issuing a killer poke? 92.10.68.40 (talk) 11:55, 3 March 2023 (UTC)

The English translation of the Commodore Welt article on the Graphics Booster 128 says the hardware add-on is just a 64K VDC RAM upgrade (the original German doesn't mention 64K, but it does say that the add-on is no longer needed with the newer 128D units, the ones with the Datassette port on the left). The manual also says that only the EPROM is needed in the DCR. (No explanation as to what the wire is for, though.)

For the resolution, the same article says you might have to tweak the vertical sync knob for the higher resolutions (could be as low as 40 Hz laced, I reckon). (And speaking of interlace, COMPUTE!'s Mapping the Commodore 128 describes the registers (8 for interlace, pp. 439–441). There's even a six-line BASIC program to demonstrate an 80×50 text-mode display. COMPUTE!'s Gazette published a more elaborate program in the December 1989 issue, bundled with a SpeedScript patch.)

Some of the information in the manual for that product is definitely off, though: it mentions Epson graphics at 1920 dpi, but Epson cites that mode as 240 dpi (p. 353 in the PDF).

Maybe one of these days I'll get the DCR out and try the software clone (and if it works, hook up the 'scope and check the refresh rate :-) .—dah31 (talk) 07:02, 21 November 2023 (UTC)