Is the printer port wired and addressed as a Atom 7 bit printer port, or as a BBC 8 bit printer port?
Would be nice to see some photos of the video output
Mark
Another build
Re: Another build
Nice work
Is the printer port wired and addressed as a Atom 7 bit printer port, or as a BBC 8 bit printer port?
Would be nice to see some photos of the video output
Mark
Is the printer port wired and addressed as a Atom 7 bit printer port, or as a BBC 8 bit printer port?
Would be nice to see some photos of the video output
Mark
Standby alert “There are four lights!”
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb
Autumn is here. Bye bye summer 2024...
Not as many MTXs as Dave!
Re: Another build
It's a 7 bit Atom printer port - as that's what the OS will expect - I have tied the normally floating 8th bit low.
Pictures of the colour output as requested, I've a feeling it's the levels on the Y output that are slightly out, or it's the TV not quite accepting the NTSC colour burst.
In text modes, the black background is showing up as a blue - and that, could be why some of the other colours are out. In graphics mode though the black/buff display is black and white, while the black/green set is more bright green on dark green.
According to the VDG data sheet this is supposed to be Green on black text, yellow on black block graphics, black on green text, and red on black block graphics
The alternate colour set should be orange on black, cyan on black, black on orange, and orange on black
The block graphics actually look better - the VDG apparently doesn't do white - it should be "buff". But notice that in this mode the background is black not blue.
Pictures of the colour output as requested, I've a feeling it's the levels on the Y output that are slightly out, or it's the TV not quite accepting the NTSC colour burst.
In text modes, the black background is showing up as a blue - and that, could be why some of the other colours are out. In graphics mode though the black/buff display is black and white, while the black/green set is more bright green on dark green.
According to the VDG data sheet this is supposed to be Green on black text, yellow on black block graphics, black on green text, and red on black block graphics
- Saturation's not quite right ...
- Colour set1.jpg (111.64 KiB) Viewed 50964 times
- Nor here
- Colour set2.jpg (96.26 KiB) Viewed 50964 times
- Buff turns out white
- 128x64 graphics.jpg (120.46 KiB) Viewed 50964 times
Re: Another build
Standby alert “There are four lights!”
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb
Autumn is here. Bye bye summer 2024...
Not as many MTXs as Dave!
Re: Another build
The transistor based setup currently on the board also came from Stardot, page 4 of this thread.
http://stardot.org.uk/forums/viewtopic. ... 4&start=90
The Y output is a straight clone of the original Atom's composite out.
http://stardot.org.uk/forums/viewtopic. ... 4&start=90
The Y output is a straight clone of the original Atom's composite out.
Re: Another build
Version 2 video board completed, the new one is the one on the left with the square chip and no cassette tape ports.
Lez sent me (via Dave) some extra parts to improve on the original design. The big change is the move to dual ported ram. The square chip is 8k of DP ram. With the VDG on one port and the CPU on the other, there's no need for the buffering to keep the VGD and CPU form interfering with each other.
Snow on the screen from the VGD and CPU attempting simultaneous access won't now happen. The RAM can cope with both sides reading the same address at the same time. If the CPU happens to try and write to a byte that the VDG I reading, Ready is activated and the CPU is held until the VDG is done. If the VDG tries to access a byte the CPU is writing, the ram at 35ns is comfortably fast enough to complete the VDG read after the CPU write is complete, so there's no special action required.
With the extra board space freed up from the removal of the memory buffering, I've added an external character rom. Normally the VDG uses an internal 64 character rom, with inverse for 128 displayable characters. However, only having 64 base characters means there's no room for lower case. The other down side is the size of the characters, to save space, they're based on a 5 x 7 matrix, even though the character cells are 8 pixels by 12 rows.
Half of the 8k eeprom holds a replacement character set, using the full size of the character cell to provide descenders etc. Sitting next to that is a 74LS161 ( I didn't have the HC161 in the box) counter which does the row select, and 2 74HC157's to switch the data path between text modes where the rom is active, and graphic modes where it isn't. There is a pair on on-board jumpers that can be set to enable or disable new characters.
That's how the display from a few posts previous appears on the new board with the rom engaged.
To get the lower case to appear correctly, all the data is inverted, as the VDG is still in inverse mode when they're displayed. The inverse number set is currently unchanged, but that could be tweaked for adding custom graphics.
As the character set is 4k in size, there's also an option to add a 3rd jumper to open up the other half of the rom for a 2nd set of defined characters.
- mk2 video board.jpg (122.5 KiB) Viewed 50831 times
Snow on the screen from the VGD and CPU attempting simultaneous access won't now happen. The RAM can cope with both sides reading the same address at the same time. If the CPU happens to try and write to a byte that the VDG I reading, Ready is activated and the CPU is held until the VDG is done. If the VDG tries to access a byte the CPU is writing, the ram at 35ns is comfortably fast enough to complete the VDG read after the CPU write is complete, so there's no special action required.
With the extra board space freed up from the removal of the memory buffering, I've added an external character rom. Normally the VDG uses an internal 64 character rom, with inverse for 128 displayable characters. However, only having 64 base characters means there's no room for lower case. The other down side is the size of the characters, to save space, they're based on a 5 x 7 matrix, even though the character cells are 8 pixels by 12 rows.
Half of the 8k eeprom holds a replacement character set, using the full size of the character cell to provide descenders etc. Sitting next to that is a 74LS161 ( I didn't have the HC161 in the box) counter which does the row select, and 2 74HC157's to switch the data path between text modes where the rom is active, and graphic modes where it isn't. There is a pair on on-board jumpers that can be set to enable or disable new characters.
- character rom active.jpg (101.68 KiB) Viewed 50831 times
To get the lower case to appear correctly, all the data is inverted, as the VDG is still in inverse mode when they're displayed. The inverse number set is currently unchanged, but that could be tweaked for adding custom graphics.
As the character set is 4k in size, there's also an option to add a 3rd jumper to open up the other half of the rom for a 2nd set of defined characters.
Re: Another build
@Martin - Yet more great work there 
Mark
Mark
Standby alert “There are four lights!”
Step up to red alert. Sir, are you absolutely sure? It does mean changing the bulb
Autumn is here. Bye bye summer 2024...
Not as many MTXs as Dave!
-
lezanderson
- Posts: 186
- Joined: 14 Aug 2012 15:31
Re: Another build
Martin,
I've got other 6502 type chips you may find useful ?
W65C816 Enhanced 65C02
R6522 IC MPU DIP
MB8877 IC DIP Floppy disc controller FDC DIP
6845/68B45 DIP video Chip
6821/68B21 DIP Peripheral Interface Adapter (PIA)
68B50 DIP ASYNCHRONOUS COMMUNICATIONS INTERFACE ADAPTER (ACIA) - MC6850
68B54 DIP Advanced Data Link Controller (ADLC)
146818 DIP RTC : similar to DS12887
SAA5050 DIP Teletext chip
D7002C DIP ADC
SN76489AN DIP
Just let me know if you want any ..
I've got other 6502 type chips you may find useful ?
W65C816 Enhanced 65C02
R6522 IC MPU DIP
MB8877 IC DIP Floppy disc controller FDC DIP
6845/68B45 DIP video Chip
6821/68B21 DIP Peripheral Interface Adapter (PIA)
68B50 DIP ASYNCHRONOUS COMMUNICATIONS INTERFACE ADAPTER (ACIA) - MC6850
68B54 DIP Advanced Data Link Controller (ADLC)
146818 DIP RTC : similar to DS12887
SAA5050 DIP Teletext chip
D7002C DIP ADC
SN76489AN DIP
Just let me know if you want any ..
Re: Another build
Lez,
A lot of those you've sent already, and they're sitting in the spares drawer awaiting projects. Or more accurately waiting for me to have the time to start the planned projects
You may be interested to know, that because you had sent "samples" of the 65816, when I was starting on this design I allocated space on the "Atom" backplane to allow for the extended address bus of the enhanced chip "just in case". So it should be possible to design a 16bit CPU card to take advantage of the extra memory space available.
I just need the time .....
A lot of those you've sent already, and they're sitting in the spares drawer awaiting projects. Or more accurately waiting for me to have the time to start the planned projects
You may be interested to know, that because you had sent "samples" of the 65816, when I was starting on this design I allocated space on the "Atom" backplane to allow for the extended address bus of the enhanced chip "just in case". So it should be possible to design a 16bit CPU card to take advantage of the extra memory space available.
I just need the time .....