VileRancour wrote:

Jo22 wrote:

- I've created for you another dump of the CGA ROM, but its content differs significantly from the inital file.
And the size is also different, probably doubled. Hope that's okay.

Thanks, that's interesting actually-- this one seems to have groups of characters mixed from between the different fonts... but not in any consistent way that I can make sense of, heh:

I'm glad it was good for something (I expected worse 😅 ).
The dump was done in an -uhm- special way. I basically convinced the reader to work in a way it wasn't intended (other chip mode, different wiring).
That's also the reason it's twice the size (I didn't expect it to read the data ~8 times, though)..
Plus, the reading was done way sloower than usual this time - instead of seconds, it took several minutes to finish.
Maybe it was a bit too slow? But at least the characters seem to be more clean now.. ^^

VileRancour wrote:

It's also double the size of your previous dump, so I was hoping that at least all the 'pieces' would be there, but in fact the same data seems to be repeated four times (just by eyeballing it... may be even eight with an odd offset).
The strange order/interleave here is mystifying, but having these extra pieces could at least be helpful. I wonder how much of the total font data could be manually reconstructed from this, plus your first dump, plus your composite capture - I might give that a try in my next bout of spare time. 😁

Good luck! 😁 I've got one more screen shot for you. It's taken from a real CGA monitor (my first one ever)!
Quality is far better than the cvbs/composite grab, but the image is also more rounded/curved. 🙁
So I don't know if you can make use of it. At least it shows that the image data is fine so far (which bugs me as well,
because it makes no sense). So what is the card's secret ? Why can it do something all my hardware fails at ?

If anyone has an idea, please tell! I'm honestly happy about any response.

VileRancour wrote:

Once again, I truly appreciate all the effort you're putting into this. If no-one here can tell what's going on with that chip, perhaps the folks at the Vintage Computer Forums may have some insight about why it resists dumping so stubbornly...

Thank's a lot! And I also appreciate your idea! As soon as I reach my limit,
I would be glad to send that chip (or the whole card) to someone with better equipment and more experience.

Jo22 wrote:

VileRancour wrote:

I wonder how much of the total font data could be manually reconstructed from this, plus your first dump, plus your composite capture - I might give that a try in my next bout of spare time. 😁

Good luck! 😁 I've got one more screen shot for you. It's taken from a real CGA monitor (my first one ever)!
Quality is far better than the cvbs/composite grab, but the image is also more rounded/curved. 🙁

Thanks! That actually helped tremendously. Took me a while to get to this (was busy moving), but at least font #3 (8x8 w/thick strokes) is fully recovered now:

I've recovered some more bits and pieces of the other fonts, with the help of your previous dump. As you can see though, there's still a ways to go 😀

Also-
thanks to Ben Garrett of defacto2, you can now use some of these fonts in a free, open-source Chrome / Firefox exension to view TXT/NFO files the way they were meant to be seen:

Source: https://github.com/bengarrett/RetroTxt
Chrome webstore: https://chrome.google.com/webstore/detail/ret … ogcaiojfajanahn
Firefox (48.0+): https://addons.mozilla.org/en-US/firefox/addon/retrotxt/

defacto2.net even has a built-in viewer for the hosted ASCII art exhibits: https://defacto2.net/file/detail/a61c8ea

Check it out - cool stuff.

Firefox 48+ = beta??
It would be cool to use dem fonts for any webpage and or use custom css too..

Yep- FF48 is still beta, but apparently it now supports Chrome extensions which explains the requirement.
You can use Stylish for custom CSS, although the real issue with these fonts on the Web is the aliasing - and the fact that each browser renders fonts differently. As far as web standardization is concerned we're still in 1996...

Maybe this is not the best place to have a rant, but...

I have to say that it's frustrating for a spaniard like me going into most fonts sites. Why?

Because most times it's not clear if a font has all the glyphs for your language. In Spain, the "offending" glyphs would be áéíóúñü (with capital variants). I've found many times that a good font for a job it's hampered because it don't have those glyphs. In some cases (covers) I must reword the job to avoid including words with those characters, but unfortunately both my name and my surname include it.

As a suggestion: some computers (PC1512? C64?) have fonts with national variants. It would be nice to include all those extra characters on their fonts (having a "worldwide" font), or put those variants as separate fonts stating what CP / language they use.

Thanks for reading my rant.

VileRancour wrote:

Yep- FF48 is still beta, but apparently it now supports Chrome extensions which explains the requirement.
You can use Stylish for custom CSS, although the real issue with these fonts on the Web is the aliasing - and the fact that each browser renders fonts differently. As far as web standardization is concerned we're still in 1996...

Zup wrote:

Maybe this is not the best place to have a rant, but... […]
Show full quote

Maybe this is not the best place to have a rant, but...

I have to say that it's frustrating for a spaniard like me going into most fonts sites. Why?

Because most times it's not clear if a font has all the glyphs for your language. In Spain, the "offending" glyphs would be áéíóúñü (with capital variants). I've found many times that a good font for a job it's hampered because it don't have those glyphs. In some cases (covers) I must reword the job to avoid including words with those characters, but unfortunately both my name and my surname include it.

As a suggestion: some computers (PC1512? C64?) have fonts with national variants. It would be nice to include all those extra characters on their fonts (having a "worldwide" font), or put those variants as separate fonts stating what CP / language they use.

Thanks for reading my rant.

I agree and or share your rant, coz spanish/castellano is my native language.

Zup wrote:

As a suggestion: some computers (PC1512? C64?) have fonts with national variants. It would be nice to include all those extra characters on their fonts (having a "worldwide" font), or put those variants as separate fonts stating what CP / language they use.

I've already done that in my font pack- several fonts come in 'PxPlus' versions that fully support most Latin-based scripts (including Spanish), as well as Cyrillic, Greek, and Hebrew (sample of the full charset). These fonts include the common IBM ones plus Tandy and PC1512.

On the site you can preview the charset by expanding the font samples, or using the Preview page.. these 'international' versions are also indicated by a 'Plus' next to the font's name. Dunno how I could make it any clearer 😁

VileRancour wrote:

Jo22 wrote:

VileRancour wrote:

I wonder how much of the total font data could be manually reconstructed from this, plus your first dump, plus your composite capture - I might give that a try in my next bout of spare time. 😁

Good luck! 😁 I've got one more screen shot for you. It's taken from a real CGA monitor (my first one ever)!
Quality is far better than the cvbs/composite grab, but the image is also more rounded/curved. 🙁

Thanks! That actually helped tremendously. Took me a while to get to this (was busy moving), but at least font #3 (8x8 w/thick strokes) is fully recovered now:

I've recovered some more bits and pieces of the other fonts, with the help of your previous dump. As you can see though, there's still a ways to go 😀

Cool, it looks very good! 😁 And sorry for the late reply. I recently got myself a Miro video card (miroVideoDC30+) and tinkered with the drivers.
The Win 9x drivers are available on the inofficial miro website, but the XP drivers are missing. Rumours say the guy charges money for these..
Oh, well.. Maybe I can use one of my Power Macs instead (it is Mac compatible). I'll try as soon as possible.
And I'm also still working on the EPROM. I'm waiting for the new programmer I ordered to arrive. Unfortunately, it shipps from china, so I'll have to be patient.

Btw, I've also attached a photo of the composite card for you guys. I heard the Miro products are/were generally quite good in terms of quality.
So this card might be interesting for some of you (haven't tested this one myself yet).

Zup wrote:

Maybe this is not the best place to have a rant, but... […]
Show full quote

Maybe this is not the best place to have a rant, but...

I have to say that it's frustrating for a spaniard like me going into most fonts sites. Why?

Because most times it's not clear if a font has all the glyphs for your language. In Spain, the "offending" glyphs would be áéíóúñü (with capital variants). I've found many times that a good font for a job it's hampered because it don't have those glyphs. In some cases (covers) I must reword the job to avoid including words with those characters, but unfortunately both my name and my surname include it.

As a suggestion: some computers (PC1512? C64?) have fonts with national variants. It would be nice to include all those extra characters on their fonts (having a "worldwide" font), or put those variants as separate fonts stating what CP / language they use.

Thanks for reading my rant.

You're spanish ? Cool! My sister had spanish at school as a subsidiary subject.
I know very little about this, but she always came home with a smile on the days she had this subject.
What she liked the most were the songs they sung.. And the food. Haha. 😁
Another thing she also seemed to like was the grammar of the Spanish language, like for example,
the use of the inverted question mark at the beginning of a question of a sentence. She also mentioned other interesting stuff I forgot about.

Quick update - Haven't forgotten this thread, still waiting for the programmer.
And the digitizer card needs some picture editing software to work on the Mac. I'm looking for it..

Edit: I've found and installed the MacOS X driver for the MiroVideo.
Now I need to figure out how to make it work with my copy of Premiere..
Edit2: Premiere 4.0 installed just fine, but doesn't recognize my av card yet.
Perhaps that's because it runs as a classic application on X.
It may take a few day, but I'll find a solution for this.. 😉
Edit3: Phew! I've got the card finally working! I've attached another snapshot for your.
It's still blurry, though. I had to use a video to BMP converter and a lo-fi RCA cable..
This capture was done in sqare-pixel mode, but I've also got a capture in CCIR 601 if necessary.

Jo22,

Just remembered this page which explains the differences between CGA/MDA-style character ROM pinouts and 'normal' 27xx chips. Not sure what type of ROM your Epson card has, but perhaps the last sentence in that paragraph may be helpful in dumping it.

Thanks a lot! 😀 I'll see what I can do.

Btw, my new programmer should arrive about next week, too. Perhaps it is more lucky in reading this chip.
If not, I'll seek for a skilled fortuneteller with a good glass sphere. 😉

Still waintig for that programmer. Not sure if it will ever arrive. 🙁
So I've ordered another programmer, a TL866CS. I've choosen this model, because of the positive feedback of other Vogons users.
I already have got the confirmation of payment and it is marked as "shipped". It ships from my country so it should arrive in a few days.
Hopefully it will be able to read that chip. If not, I'll built an adapter with the schematics shown in the link mentioned earlier. 😉
If this still won't work, I'll try to do an RGB capture of that (partial) character set.. Anyway, sorry for the delay and thank you for your patience.

Edit: Good news! The waiting is over, my second order arrived today. 😁
Edit2: Today, I have unboxed my new programmer. It is smaller than my old one, albeit a bit more heavy.
Build quality is fine, too. The software is okay - someone can define pulse delay and voltages.
It was also nice that the seller included a chip puller and some PLCC adapters..
I've attached a small photo for you.

Good News!

The new programmer was able to read that chip. Well, at least it claims it did..

Note that its chip's database doesn't know of the M5L2764K -2, so I choosed a generic CMOS version instead (AMD).
It worked fine after I unchecked the "Check ID" option.

To my relief, the verify check didn't complain and passed without any errors several times.
So I hope that this dump is a bit better than the previous ones, but I still have my doubts.

Actually that's just about perfect. You rock! 😁

That first font just needs some slight rearrangement. It's 8x14 (probably meant to display at 9x14, Monochrome-like) so those bottom two scanlines of each character are unused, although they actually seem to contain the character's 8-bit index number (red):

Will definitely convert this for my next update and credit you. Seriously, you're a hero for going through the trouble - I hope you have further uses for that programmer 😀
Hope to finish that update some time next month really, it's been too long...

Woah, thank you! 😁 I'm glad it worked this time! Yes, I'm sure that programmer will come in handy in the future.
Someone can never have enough programmers.. 😉

Time to bump this with great news (well... for those than need GR chars that is)
A friend gave me a portable 5155 pc to check out why it does not power on. This was a machine purchased in Greece and thus has the official Greek character rom - designed by Memotek a very popular/important company of the era in Greece.
There is some addon PCB holding a 27C256 eprom - for some reason there are connections to the LPT port - maybe for bank switching? There are 2 ICs on the PCB that are obscured so atm I don't know what they are. One is under glue , the other has had it's markings erased. Is the rom scrambled? No idea either 🤣

Here are some pics:

for some reason there is a soldered extra 9pin output on the cga, it leads to a bracket with a 9pin connector - no idea why the extra output 🤣

I am attaching the ROM dump , really curious to see what's the deal here. 😊

Very cool find! The ROM looks just fine - in fact you have five(!) different Greek encodings in there, and each one comes in 'thin' and 'thick' versions (except the last one which is a 'thick' version repeated twice, heh).

Wouldn't expect a 32K ROM on a CGA card, but I guess that's why they rigged all that extra fluff (to remap a selectable 2K window to the address lines expected by the card) 😉

This could probably replace/expand the Greek range I'd already included in the "Plus" (international) version of the IBM CGA charset:

Great news then , glad this worked out! You can't get more official than that when it comes to GR CGA characters IMHO 🤣
I wonder why there are cables going to the lpt port though, some kind of change bank mechanism?

And another thing that pops in mind, how can I make a ROM for my IBM cga that I bought from the US? It would be great to actually use some old GR programs or magazine disks but I obviously don't care to have so many font variations... is there a tool or something to split this ROM and burn to a smaller drop-in replacement eprom ?

edit
now that I think about it, the "correct" (or the one everyone used back then) encoding would be the one found in the ACER 500+ clone I used to own, it displayed magazine disks OK and it was used in a university so it must be the encoding to have... I attach the rom image for it here , I wonder which of the IBM ones match that 😎