• Please review our updated Terms and Rules here

Running ega card with feature adapter in ibm5155

That is really cool!
I suppose you have to set the EGA card to CGA monitor mode though, so only 200-line modes are supported?
 
Hugo Holden said:
Hello all,

I have just posted an article, it is provisional (any corrections please email me)

http://worldphaco.com/uploads/FITTING_AN_EGA_CARD_TO_AN_IBM_5155.pdf

This will interest anyone wanting to put an EGA card in a 5155, it shows how a home made feature adapter does the job and a few notes about the screen options in BASIC and remarks about the 5153 monitor.
Click to expand...

Really cool! The EGA feature connector is really flexible, I use mine to switch between EGA and Tandy video depending on the external video bit.
 
Scali said:
That is really cool!
I suppose you have to set the EGA card to CGA monitor mode though, so only 200-line modes are supported?
Click to expand...

Yes that is right, as noted in the article the switches on the EGA card select it for the CGA (5153 monitor), but the 200 line mode is just fine and the main reason to do it is to get the expanded color palette of the EGA card to make better graphics. In fact the 640 x 200 graphic mode (screen 8 in BASIC) has a look to it that is vintage enough to conjure up images of the charm of vintage computer graphics, but yet detailed enough to be quite impressive looking too, so I like the proportions of it.
 
If I read your article correctly, your feature board would be sufficient to generate monochrome composite output from an EGA card. However, it would not be sufficient to generate a color composite output signal, you have only generated a Y/luma signal, you need a C/chroma signal for color.

The photo of the 5155's amber screen seems to show that each set of four colors almost repeat themselves, so colors 0-3 are almost indistinguishable from 4-7 and the same goes for 8-11 and 12-15.

I always thought that the 5153's contrast having no effect on the intense colors was just my monitor acting funny due to age, but it is apparently a CGA monitor feature because my Tandy CM-5 also has the contrast work in the same way, although the screen does not get nearly as dim.
 
Yes, there is no need to generate the chroma sub-carrier for a monochrome monitor and as mentioned in the article, it could be done, say to drive external color composite monitors but it would require extra IC's.

Also the degree of intensification one gets on individual colors (as seen as a monochrome image) is the same as IBM made for the composite video (luminance component) generated by their CGA card, because I used the same resistor matrix and the same value resistor they used for the intensity channel. I think it is a little extreme though and ideally the circuit would increase the intensity of individual colors by the same proportion, for a composite signal, to match up with the apparent intensity increase seen for each color on the actual CGA screen. I may design a circuit variant to do this, but for now it simply conforms to what IBM did with the generation of the luminance component of the composite video on their CGA card.

If you look at the schematic of the 5153, you will see how they use the intensity signal to disable the manual contrast control.

(I will also take a photo of the amber screen running CGACAL with a CGA card, in theory it should be very close)
 
Last edited:
Hugo Holden said:
Also the degree of intensification one gets on individual colors (as seen as a monochrome image) is the same as IBM made for the composite video (luminance component) generated by their CGA card, because I used the same resistor matrix and the same value resistor they used for the intensity channel.
Click to expand...

I take it you used a 'new CGA' card to base this on? The 'old CGA' card suffered from lack of difference in luminance in its colours.
This was improved with the 'new CGA' model, but perhaps still not a perfect distribution of 16 luminance values.
 
Hi Scali,

Yes I used the resistor values from the newer CGA card that also match up with those on the schematic for the CGA card in the manual that is on minuszerodegrees.

I have just updated the article, so have another look at it near the end (you may have to clear your browser history to view the new article) & ignore the occasional typo... I re-fitted a CGA card and photographed the amber monitor using CGACAL which is an identical color bar like test pattern to the one I wrote for the other demo. The most striking feature is the fact that the chroma sub-carrier that the amber monitor is needlessly fed with in the standard 5155 setup from the CGA card, is detrimental to the image formation.

IBM could have avoided this but as I noted in the article text, they hijacked the RF modulator connector and they would have had to have added an extra buffer transistor and a duplicate resistive mixer circuit (sans the sub-carrier input) just for the amber monitor feed.

So using the EGA card and the home made feature adapter gives a better amber monitor image than the CGA card and allows the expanded palette on the 5153 monitor so its a win-win situation.

The amber monitor in the 5155 is a real darling though, it has the best video high frequency response and best contrast/brightness and scanning linearity of any small CRT monitor I have ever seen.
 
Hugo Holden said:
IBM could have avoided this but as I noted in the article text, they hijacked the RF modulator connector and they would have had to have added an extra buffer transistor and a duplicate resistive mixer circuit (sans the sub-carrier input) just for the amber monitor feed.
Click to expand...

Could they not have made a jumper to select between mono and colour composite output?
The onboard Paradise PVC4 circuit on my Commodore PC20-III motherboard has such an option (see http://dostalgie.de/downloads/pc10III-20III/PC10III_OM_COMMODORE_EN_DE.pdf JMP614 on page 2-2) I suppose it would be quite simple to bypass the chroma part of the composite circuit that way.
 
yes, it would be easy to have a jumper to just disable/disconnect the chroma signal into the resistor matrix to get rid of it , but to get two simultaneous composite outputs, one with chroma for an external monitor and one without for the internal monitor say, it would require a double up on the transistor emitter follower buffer and most of the resistors.

I am working on the article and will add more detail about the construction of the pcb..
 
Last edited:
I have now updated the article:

http://worldphaco.com/uploads/FITTING_AN_EGA_CARD_TO_AN_IBM_5155.pdf

The update includes a pcb design for the feature adapter. Also an analysis of the color processing inside the 5153 monitor. As noted, there was little attempt by IBM to make a composite signal for a TV as a monitor resemble the color intensification effects of the 5153. In addition I found the cause of a subtle vertical scan linearity defect in both of my 5153 monitors and I have documented the cause and the fix for that.
 
You must log in or register to reply here.
Back
Top