It is currently Wed Dec 13, 2017 1:51 pm

All times are UTC - 7 hours





Post new topic Reply to topic  [ 27 posts ]  Go to page Previous  1, 2
Author Message
PostPosted: Wed Mar 08, 2017 12:35 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 6514
Location: Seattle
bestmmk wrote:
CD0-CD7 should be color data of graphics. This requires a logical analysis. If the palette problem is corrected, it will be of great value
Well, sure. I was curious about the specifics.

krzysiobal wrote:
Maybe you can put some CPLD logic between UM6558/UM6559 on CD0-7 lines, which changes problematic dot colours in the fly to the correct ones, that don't cause troubles?
It looks like it's only 8 bits wide input and output ... I bet you could just use a fast ROM. After all, we can easily get 70ns ROMs, and it's a ~1/(186ns) pixel clock...


Top
 Profile  
 
PostPosted: Thu Apr 27, 2017 7:56 am 
Offline

Joined: Fri Apr 22, 2016 4:28 pm
Posts: 10
I have done exactly this, put a GAL between the colour bus output of UM6558 and the input of UM6559.

In fact it was only necessary to generate new CD4 and CD5 signals. These are the luminance pins of the colour bus output and are in the same format as the colour registers.

CD0 - CD3 are the hue pins of the colour bus output and they are inverted with respect to the colour registers, so software colour 00 (dark grey) is colour 0F in the hardware interface between the two chips.

This inversion makes the hardware colours more intuitive, because hardware colour 00 (software colour 0F) is black and hardware colour 3F (software colour 30) is white.

The hardware palette puts
8 blacks at 00, 10, 20, 30, 01, 11, 21, 31
the darker greys at 02, 12, 22, 32 (02 is software colour 0D which is blacker than black in PAL/NTSC)
the colours at 03-0E, 13-1E, 23-2E, 33-3E
the lighter greys at 0F, 1F, 2F, 3F.


Last edited by vortexion on Fri Apr 28, 2017 2:25 am, edited 1 time in total.

Top
 Profile  
 
PostPosted: Thu Apr 27, 2017 8:08 am 
Offline

Joined: Fri Apr 22, 2016 4:28 pm
Posts: 10
Attachment:
Photo 10-04-2017, 12 58 02.jpg
Photo 10-04-2017, 12 58 02.jpg [ 1.31 MiB | Viewed 1261 times ]


Top
 Profile  
 
PostPosted: Thu Apr 27, 2017 8:13 am 
Offline

Joined: Fri Apr 22, 2016 4:28 pm
Posts: 10
Some pictures of Battle City running on the modified Dendy SECAM. This game is a good test because it uses a dark grey colour (software colour 00, hardware colour 0F) which is rendered as black by the standard Dendy SECAM.

Attachment:
Photo 10-04-2017, 12 44 24.jpg
Photo 10-04-2017, 12 44 24.jpg [ 3.32 MiB | Viewed 1260 times ]


Attachment:
Photo 10-04-2017, 12 45 44.jpg
Photo 10-04-2017, 12 45 44.jpg [ 2.24 MiB | Viewed 1260 times ]


Attachment:
Photo 10-04-2017, 12 55 07.jpg
Photo 10-04-2017, 12 55 07.jpg [ 2.67 MiB | Viewed 1260 times ]


Top
 Profile  
 
PostPosted: Thu Apr 27, 2017 8:35 am 
Offline

Joined: Fri Apr 22, 2016 4:28 pm
Posts: 10
The GAL maps the hardware colours output by the UM6558 onto the following values for driving the UM6559. These are hardware colour numbers, so the first four bits (hue) are inverted with respect to the colour registers.

Blacks
00, 10, 20, 30 -> 00, 00, 00, 00
01, 11, 21, 31 -> 01, 01, 01, 01

Dark greys
02, 12, 22, 32 -> 02, 02, 12, 22

Colours stay the same

Light greys
0F, 1F, 2F, 3F -> 1F, 2F, 3F, 3F


Top
 Profile  
 
PostPosted: Thu Apr 27, 2017 10:51 am 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 6514
Location: Seattle
That's so cool! Thank you for sharing!

Do you know what the top two bits of the 8-bit CD bus carry?


Top
 Profile  
 
PostPosted: Thu Apr 27, 2017 12:00 pm 
Offline

Joined: Sun Sep 19, 2004 11:12 pm
Posts: 19341
Location: NE Indiana, USA (NTSC)
The unknown bits aren't emphasis, are they?


Top
 Profile  
 
PostPosted: Thu Apr 27, 2017 12:11 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 6514
Location: Seattle
Unless they're somehow serialized (perhaps sending only R or B on alternating scanlines) I don't know how...

One of the bits is probably sync, too.


Top
 Profile  
 
PostPosted: Fri Apr 28, 2017 2:18 am 
Offline

Joined: Fri Apr 22, 2016 4:28 pm
Posts: 10
The Dendy SECAM doesn't support colour emphasis bits at all, neither in composite SECAM nor in RGB.

CD6 and CD7 turn out not to be very exciting. CD6, pin 9 of UM6559, is a positive logic version of composite sync (pin 8 is negative logic composite sync) and CD7, pin 10 of UM6559, is a negative logic version of frame sync.


Top
 Profile  
 
PostPosted: Fri Apr 28, 2017 3:18 am 
Offline

Joined: Fri Apr 22, 2016 4:28 pm
Posts: 10
I've made two videos which show the Dendy SECAM composite and RGB palettes after the GAL modification has been applied. There are still some problems with the two palettes.

In RGB, the problem is that the colours do not have equal luminance (brightness) across colours 1 - C. For example, blue (2) is too dark and yellow (8) is too light. This is because the RGB values are not matrixed from a constant Y value and varying R-Y and B-Y values (as they are in PAL/NTSC), but are generated by a simple DAC scheme which does not attempt to preserve constant luminance. Any of the palettes in the NESRGB board look better than the UM6559 RGB colours.

In composite SECAM, the colours do have equal luminance but a different problem occurs with the darkest colours 01 - 0C and the lightest colours 31 - 3C. The UM6559 only generates 4 luminance levels. Colours 01 - 0C have the same luminance as black and 31 - 3C have the same luminance as white.

This means that patterns using the darkest colours on a black background will look very indistinct, because there are no luminance transitions at the colour boundaries. Only chrominance information changes, and this is carried in very low resolution.


Top
 Profile  
 
PostPosted: Mon Jun 05, 2017 3:59 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 6514
Location: Seattle
I just put two and two together and realized that you could replace the UM6559 with a ROM, since all you want is RGB output.

While an 8-bit ROM would only afford R3G3B2, multiple ROMs could be combined to produce a higher-precision "ROMDAC".

Additionally, you could then add an 74'138 and a 3(or more) bit latch to capture writes to PPU $2001, and use those extra latched bits to add emphasis bit support.


Top
 Profile  
 
PostPosted: Fri Jun 16, 2017 5:05 am 
Offline

Joined: Fri Apr 22, 2016 4:28 pm
Posts: 10
I've been working on modifications to improve the SECAM output, because I want to see what a (good) SECAM Dendy would be like.

For RGB, you could use ROMs or possibly the GALs I am using. A GAL16V8 has 8 outputs so two of these will give you 16 outputs. R5G5B5 will give 32768 possible colours, from which the 52 or 54 NES colours can be chosen. Emphasis will require more electronics like you suggest.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 27 posts ]  Go to page Previous  1, 2

All times are UTC - 7 hours


Who is online

Users browsing this forum: Google [Bot] and 10 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group