Reply 140 of 341, by retrocanada76
wrote:FPGA output has 3-bit DAC, resistor values 2k, 1k and 500, IO voltage 3.3 volts, right?
the version 2.0 has 4 bits: 4K, 2K, 1K and 500
I never tried tri-stating them. May worth trying.
wrote:FPGA output has 3-bit DAC, resistor values 2k, 1k and 500, IO voltage 3.3 volts, right?
the version 2.0 has 4 bits: 4K, 2K, 1K and 500
I never tried tri-stating them. May worth trying.
I just pushed another firmware with the colors calibrated to CGACAL.
Just check the color before and after. Look, my before looks like keropi's dreamblaster board
before
after:
although my cellphone photo might change it a little but at eyeball I can see the light blue and light red aren't correct as well light magenta.
the reason why the blue, red green and magenta aren't correct is if you pick the image from the cgacal website and pick the color on paint or whatever you will see there is a little of red and green for the light blue, the same for all bright color except the yellow. So I am adding the little white for all the colors on the bottom except the yellow.
most of the people know the brown color is logic tweaked on the cga monitor but I can tell from my observations that yellow is also tweaked, It does not have bright for the blue channel, otherwise it would be a little washed out.
I flashed the latest SOF file with the "Calibrate the CGA colors according to CGACAL" tag but I don't see any difference...?
new cgacal pic above is a little blurry but it looks the same as before to me , an old pic is under this message
is there a chance the SOF file on gihub remained an older version? 😕
well you can always recompile the code. But are you sure you are sending the correct sof ? because if you check out in a different folder, the altera programmer keeps the old path as the sof, you have to delete from the programmer and add the correct one again
nah I have no idea how to compile this ...
I just put the .SOF in the root of D:\ and open it - I have deleted all other *.sof files from there and programmed it 4-5 times in a row, I get the success message on the quartus message area but no change in color behavior.
I have updated the board twice when you added the OSD, one at the first osd style and the last time when the bar was added for the values - these updates worked fine.
I will mess with the paths now on the software and see what happens.
edit:
Nope did not work. Newly downloaded and renamed .SOF at desktop, loaded and programmed OK but the colors are still the same as before. I can't think of anything else to do at this point.
try programming the .jic
I tried that in the past I get an error about the setup, let me put a shot for you to see
wrote:most of the people know the brown color is logic tweaked on the cga monitor but I can tell from my observations that yellow is also tweaked, It does not have bright for the blue channel, otherwise it would be a little washed out.
Where did you read this? Yellow is the same as other "bright" colors so actually the yellow is not tweaked any more than the other bright colors like blue, red, magenta, cyan etc.
For bright colors, all the low intensity rgb bits are always on, and the high intensity RGB bits determine the color.
www.minuszerodegrees.net/oa/OA%20-%20IB ... (5154).pdf (page 8 in the PDF)
https://en.wikipedia.org/wiki/Enhanced_Graphi … r#Color_palette
Also, the RGBI to RrGgBb mapping happens in the monitor, so some EGA cards might send out unused color bits when in lo-res mode.
Therefore, if the mode is detected to be lo-res CGA RGBI mode, it should ignore the extra two color bits and just remap the 4-bit RGBI to 6-bit RrGgBb for the DAC or the colors might be wrong.
wrote:try programming the .jic
YES!
this worked , in my case I had to manually tick the "Program / Configure" checkbox and just hit Start leaving the devices diagram as-is. Auto-Detect screws things over with device errors , maybe because I use a USB-Blaster 1 clone and not a II or a genuine one. But it works so it's good 😊
edit:
YES, finally I am getting this sweet color action 😊 😊 😊
thanks retrocanada76!
wrote:Where did you read this? Yellow is the same as other "bright" colors so actually the yellow is not tweaked any more than the oth […]
wrote:most of the people know the brown color is logic tweaked on the cga monitor but I can tell from my observations that yellow is also tweaked, It does not have bright for the blue channel, otherwise it would be a little washed out.
Where did you read this? Yellow is the same as other "bright" colors so actually the yellow is not tweaked any more than the other bright colors like blue, red, magenta, cyan etc.
For bright colors, all the low intensity rgb bits are always on, and the high intensity RGB bits determine the color.www.minuszerodegrees.net/oa/OA%20-%20IB ... (5154).pdf (page 8 in the PDF)
https://en.wikipedia.org/wiki/Enhanced_Graphi … r#Color_paletteAlso, the RGBI to RrGgBb mapping happens in the monitor, so some EGA cards might send out unused color bits when in lo-res mode.
Therefore, if the mode is detected to be lo-res CGA RGBI mode, it should ignore the extra two color bits and just remap the 4-bit RGBI to 6-bit RrGgBb for the DAC or the colors might be wrong.
Empirically. I had to tweak the yellow otherwise it would look unsaturated. Look, the CGA is RGBI, but the internals of the monitor it converts that for analog R G B lines. There is no I anymore, the I intensifies the voltage for R, G and B. But here is the catch: when I get yellow wich is RGBI 1101 I cannot intensify the blue line. But ligh blue by the other hand needs little red, little green and full blue. Just look the pictures, the brigh blue, green, red and magenta are only achievable by have small values for RGB except the Yellow. Doing this on yellow you get a pale yellow.
wrote:I just pushed another firmware with the colors calibrated to CGACAL. […]
I just pushed another firmware with the colors calibrated to CGACAL.
Just check the color before and after. Look, my before looks like keropi's dreamblaster board
before
after:
That is WAY better.
Holy crap i need this in my life!!! When is this going to be available on serdashop?
wrote:Empirically. I had to tweak the yellow otherwise it would look unsaturated. Look, the CGA is RGBI, but the internals of the monitor it converts that for analog R G B lines. There is no I anymore, the I intensifies the voltage for R, G and B. But here is the catch: when I get yellow wich is RGBI 1101 I cannot intensify the blue line. But ligh blue by the other hand needs little red, little green and full blue. Just look the pictures, the brigh blue, green, red and magenta are only achievable by have small values for RGB except the Yellow. Doing this on yellow you get a pale yellow.
This is flawed thinking. You should convert the colors the way they should be, not how you think they should look. The user can tweak their monitor's controls if they don't like the result.
I have a real IBM 5153. Do we need reference photos of the original CGA monitor to resolve this?
Your colors look good keropi. Now just turn down your monitor brightness 😜
wrote:Holy crap i need this in my life!!! When is this going to be available on serdashop?
a batch of 10 will be hand assembled soon (ready for shipping 1st half of march).
The price, including preprogrammed fpga board, would be 82 euro. (+ shipping)
4 are already reserved, pm me with your address, if interested, to get a shipping estimate for the amount of boards you want.
maybe later (probably june), I may do a larger batch factory assembled
Visit http://www.serdashop.com for retro sound cards, video converters, ...
DreamBlaster X2, S2, S2P, HDD Clicker, ... many projects !
New X2GS SE & X16GS sound card : https://www.serdashop.com/X2GS-SE ,
Thanks for your support !
wrote:wrote:Empirically. I had to tweak the yellow otherwise it would look unsaturated. Look, the CGA is RGBI, but the internals of the monitor it converts that for analog R G B lines. There is no I anymore, the I intensifies the voltage for R, G and B. But here is the catch: when I get yellow wich is RGBI 1101 I cannot intensify the blue line. But ligh blue by the other hand needs little red, little green and full blue. Just look the pictures, the brigh blue, green, red and magenta are only achievable by have small values for RGB except the Yellow. Doing this on yellow you get a pale yellow.
This is flawed thinking. You should convert the colors the way they should be, not how you think they should look. The user can tweak their monitor's controls if they don't like the result.
I have a real IBM 5153. Do we need reference photos of the original CGA monitor to resolve this?
then tell me then how can you tweak a washed out yellow, making it more yellow without affecting the other colors ? What kind of independent control do you have on your monitor ??? The yellow needed a tweak to look like the reference pictures, including the CGACAL. Why is so hard to understand this ? sending RRGGBB: 111100 instead of 111101. But the light blue in RRGGBB is: 010111, light green: 011101, light magenta: 110111, light cyan: 011111, etc. Sending a yellow like 111101 won't look good it doesnt matter how you change the brightness, i am talking about color saturation, You don't have a color knob for VGA output. Specially a make more yellow please knob...
let's see if you understand by the color theory. Is simple: look the CGACAL reference picture. Can you spot the light blue ? Look the light blue is not saturated. It is not deep bright blue, because it has some red and some green on it, shifting the color towards the white. If you keep adding brightness you end up with white color instead. But now look at the yellow: it's a deep saturated yellow. There is no white on it. That's fine because it's not bright yellow but yellow instead. But the yellow is on the "bright yellow" side of the RGBI. The dark yellow was teaked to brown. If I treaded bright yellow it would look unsaturated. This is why I can tell the CGA monitors tweaks the Yellow removing the blue Intensity from it.
You can't fix that in your monitor. The voltage must be correct. Simple as that. color theory...