RE: [MSX] Colour loss

2003-08-18 Thread JP Grobler
Hi

  Go back from the analog composite output and see where there is a
 diffence.
  It might be as simple as changing a resistor to get a higher color
(burst)
  level.

 Which resisto have to be changed? Can it be bridged for testing purposes?

 Thanks
 JP




---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.509 / Virus Database: 306 - Release Date: 12/08/2003

___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread Vincent van Dam
Hi,

 I swapped the 9938 back and all the colours is fine again
 Composit and RGB?

It looks like the composit signal is in ntsc and that you are using a tv
that doesn't support ntsc via composit.

Vincent
___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread Jun Sung Kim (???)
Hi,

As far as I know,
No composite signal supported by V9958...


Jun

- Original Message - 
From: JP Grobler [EMAIL PROTECTED]
To: msx [EMAIL PROTECTED]
Sent: Friday, July 25, 2003 1:33 PM
Subject: [MSX] Colour loss


 Hi
 
 I replaced my 9938 with a 9958 in my mns8250
 Now I have colour on RGB, but gray picture on composit
 
 Any ideas why?
 
 JP
 
 
 ---
 Outgoing mail is certified Virus Free.
 Checked by AVG anti-virus system (http://www.grisoft.com).
 Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003
 ___
 MSX mailing list ([EMAIL PROTECTED])
 Info page: http://lists.stack.nl/mailman/listinfo/msx
 
___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: Fw: [MSX] Colour loss

2003-07-25 Thread JP Grobler
HI

The modifications is fine.
Pin 1,4,27 connected GND
Pin 21 and 58 connected +5V

JP
- Original Message - 
From: Saku Taipale [EMAIL PROTECTED]
To: msx [EMAIL PROTECTED]
Sent: Friday, July 25, 2003 7:58 AM
Subject: Re: Fw: [MSX] Colour loss


 
 On Fri, 25 Jul 2003, JP Grobler wrote:
 
  I swapped the 9938 back and all the colours is fine again
  Composit and RGB is fine!
   
  What is wrong with my 9958?
 
 Hmm, iirc you need to do few modifications when using v9958.
 I don't recall now what wires do you need to add, but maybe
 Hans Otten's pages has the answer...
 
 Greetings: Saku Taipale, [EMAIL PROTECTED]
 
 ___
 MSX mailing list ([EMAIL PROTECTED])
 Info page: http://lists.stack.nl/mailman/listinfo/msx
 


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003
___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread JP Grobler
Hi

If I run a basic program chancing the background  border colours in screen
2
I can see the different colours but they are not correct!
In screen0 the colours are basicaly greyscale, I see some red and yellow.

JP
- Original Message -
From: Vincent van Dam [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Sent: Friday, July 25, 2003 8:44 AM
Subject: Re: [MSX] Colour loss


 Hi,

  I swapped the 9938 back and all the colours is fine again
  Composit and RGB?

 It looks like the composit signal is in ntsc and that you are using a tv
 that doesn't support ntsc via composit.

 Vincent
 ___
 MSX mailing list ([EMAIL PROTECTED])
 Info page: http://lists.stack.nl/mailman/listinfo/msx



---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003

___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread Laurens Holst
Jun Sung Kim (???) wrote:
 Hi,

 As far as I know,
 No composite signal supported by V9958...

Indeed. The v9958 outputs an RGB signal, and that signal is used to
construct the composite video signal, but the v9938/58 doesn't do this by
itself, there is seperate hardware for that. At least, I believe that is the
common practice, but the v9938/58 also outputs a composite signal, which is
usually not used but maybe your MSX does... Although, I see it's an 8250
you're talking about, which isn't exactly an obscure type of MSX computer...

Now, the problem suggests the color signal is lost with composite video.
This occurs most commonly when connecting an NTSC output to a device
expecting a PAL input, or vice versa. So uh, are you sure your MSX isn't
running an 60Hz mode (VDP(10)=0)? It always shows grey on composite video in
NTSC mode because the color signals of PAL and NTSC aren't compatible.
However in PAL mode (VDP(10)=2) there should be no problem with getting the
color signal through. Even if the VDP is connected directly to the composite
video output, at 50Hz it *should* output a PAL signal, I actually don't see
how displaying a 50hz signal on NTSC is possible.

So if the signal doesn't get through correctly in both modes, I don't think
it is a PAL vs. NTSC problem. Now what comes to mind is that there are
different versions of PAL (like I believe PAL-M etc), but then again, it
makes no sense if the v9938 had worked before. By the way, I have never
experienced this problem myself, but then again, I never connected my MSX
through composite video since I've got a v9958 installed :).

There are multiple instructions on how to wire a v9958. In my own MSX, the
wires are done pretty cleverly, so that the sockets both supports a v9958
and a v9938, by wiring the different input pins to differing output pins.
However, if you simply connect them to +5v and GND, it is not
interchangeable. And in that case it is logical the v9938 and v9958 behave
differently, because the pin layout is different and one will be getting
incorrect signals. So my guess is that there's something wrong with the
wiring, disrupting the color signal on the composite output. I'd suggest to
try another wiring method.


~Grauw

---
Ushiko-san! Kimi wa doushite, Ushiko-san nan da!!

___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread Laurens Holst
Laurens Holst wrote:
 Jun Sung Kim (???) wrote:
 Hi,

 As far as I know,
 No composite signal supported by V9958...

 Indeed. The v9958 outputs an RGB signal, and that signal is used to
 construct the composite video signal, but the v9938/58 doesn't do
 this by itself, there is seperate hardware for that. At least, I
 believe that is the common practice, but the v9938/58 also outputs a
 composite signal, which is usually not used but maybe your MSX
 does... Although, I see it's an 8250 you're talking about, which
 isn't exactly an obscure type of MSX computer...

Ah, I just looked into it a little, the composite output signal on the v9958
is not present anymore at all. So that won't be the cause of the problem,
since if it were taking the signal from the VDP it would show a black screen
instead of only missing colors (the signal used to be on pin 21 by the way,
so maybe there's a link after all).

As I said, I think the wiring is wrong. Although you're following Hans
Oranje's instructions
(http://www.hansoranje.demon.nl/downloads/v9958inNMS8250.pdf), so those
should be right... weird. Maybe you should pull pins 4, 27 and 21 out of
their sockets?

There's also this description:
http://www.geocities.com/Yosemite/Meadows/7278/MSXKIT/NMS8245/PLUS.HTM
It is about the 8245 but I don't see why it shouldn't also apply to the
8250. Although I must say I personally have more confidence in Hans Oranje's
description.

I don't know the exact details of how it is done inside my MSX. I'll have to
open it up and take a look someday.


~Grauw

---
Ushiko-san! Kimi wa doushite, Ushiko-san nan da!!

___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread JP Grobler
Hi

I removed pins 4,21, 27 from the socket
The colour problem remains

I tried vdp(10)=2 and vdp (10)=2 no visible difference

I did the wiring of pins1,4,21  58,21 on the 9958 itself?

Does anybody have a 9958 running through composit signal?

I have not tested the RFsignal yet, my TV is dead.

JP


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003
___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread Laurens Holst
 I tried vdp(10)=2 and vdp (10)=2 no visible difference

Obviously not :) since they're the same ^_^. Anyways I guess you mean =0 and
=2... Weird. I think there is going something wrong in the RGB - composite
video conversion, but since it does not take the VDP's composite video
signal (the v9958 doesn't have one, after all), I don't understand why. It
must be using more than only the RGB signal I guess... Maybe it uses the
color burst pins? Those were removed aswell. But I don't think those were
used (if the composite isn't, why would they be?).

Well, that's easy enough to find out. Lemme see... (testing on our
vg8235)... If the color burst pins were used, changing the values of the
color burst registers VDP(21), VDP(22) and VDP(23) should make a
difference... and it doesn't.

Maybe if I take a look a which other pins have changed between the v9938 and
the v9958 I could get a clue about why it's doing this...


 Does anybody have a 9958 running through composit signal?

I just tested it on my NMS8245: you are right. The image is remaining grey
on my computer aswell. I wonder what's going on... It don't think the screen
flickered when switching frequencies either. I don't know if it's supposed
to do that, but... it seems to me it should.

Maybe someone with a little more knowledge about this matter. I'll see if I
can ask Hans Oranje on the Bussum fair, if I don't forget and if he's
coming.


~Grauw

---
Ushiko-san! Kimi wa doushite, Ushiko-san nan da!!

___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread Alex Wulms
Op vrijdag 25 juli 2003 19:40, schreef JP Grobler:
 Hi

 I removed pins 4,21, 27 from the socket
 The colour problem remains

Hi,

I have checked the datasheet of the NMS8250, and it show that the NMS8250 
computer has it's own PAL encoding unit. It does not use the composite output 
of the VDP. Only following four signals are used:
csync (is the composite sync signal; hsync and vsync are combined)
red
green
blue

The four signals go directly to the scart (although through a small amplifier 
circuit) and they go to the pal encoder. The output signal of the pal encoder 
goes to the composite video plug and to the RF modulator (for the antenna 
signal).

Both v9938 and v9958 have the four above mentioned signals. So you would 
expect them to be the same.

However, I have done some search on the internet (via google) and found that 
the exact timing of hsync (the horizontal synchronization) slightly differs 
between pal and ntsc. I know (from the v9938 databook) that the v9938 comes 
in three versions. I have always wondered about the difference. I suspect now 
that at least one of those versions uses the pal hsync timing and that at 
least one of the others will use the ntsc hsync timing. In which case it is 
plausible that the nms8250 (and also other models for the dutch market) will 
contain the version with the pal hsync timing. My suspicion is now that the 
pal encoder in the nms8250 will get confused by the csync signal of the 
v9958, since that one will contain the ntsc hsync timing. This could explain 
the color loss on the composite signal.

Is there any ntsc/pal/video specialist who can confirm the above theory? 
Because I'm not totally sure.


Kind regards,
Alex Wulms

___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread Saku Taipale

On Fri, 25 Jul 2003, Alex Wulms wrote:

 However, I have done some search on the internet (via google) and found that 
 the exact timing of hsync (the horizontal synchronization) slightly differs 
 between pal and ntsc. I know (from the v9938 databook) that the v9938 comes 
 in three versions. I have always wondered about the difference. I suspect now 

Yes, it differs, the fact is that there is lot more differencies in pal
and ntsc system than it's usually thought. First is the 50 and 60Hz, that
can be changed easily in MSX. Then the colour system is actually different
also (this shouldn't cause trouble in RGB system however). And last one is
that there are more lines drawn in PAL than in NTSC. NTSC uses 525 lines
(and 30 FPS) against PAL 625 lines (and 25 FPS). Also colour signal
frequency is different between these systems...

 that at least one of those versions uses the pal hsync timing and that at 
 least one of the others will use the ntsc hsync timing. In which case it is 
 plausible that the nms8250 (and also other models for the dutch market) will 
 contain the version with the pal hsync timing. My suspicion is now that the 
 pal encoder in the nms8250 will get confused by the csync signal of the 
 v9958, since that one will contain the ntsc hsync timing. This could explain 
 the color loss on the composite signal.

 Is there any ntsc/pal/video specialist who can confirm the above theory? 
 Because I'm not totally sure.

Sounds interesting theory, but I wont confirm that. It's far possible but
I would need to get oscilloscope for checking out what happens there
actually :-)

Yeah, I have V9958 and NMS8280 already...

Greetings: Saku Taipale, [EMAIL PROTECTED]

___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread Jon De Schrijder

Hi colourless people,

you could try the following two things:

1. The V9958 requires a *clean* 5V on pin 21 to power the video outputs; If
there is noise on this 5V, then this will result in noise on your video signals
and degrades image quality. On the 5V which is used to power all digital
circuits in the MSX there is a lot of noise, especially where the 5V enters a
large digital chip like the V9958! So it is not a good idea, like described in
the patch of Hans Oranje, to connect pin 21 with pin 58 (digital supply of
V9958). It might be better to connect pin 21 to somewhere else in the MSX where
you have 5V (with less noise). Like in the other patch that Laurens mentioned
(for the 8245). Here they get the 5V from solderhole 26. Probably in the 8245
there is 5V here. This is not the case for the 825x: here there is a 10k pullup
resistor to 5V (R166). So you might try connecting to the side of the resistor
where the 5V is (and remove the 10k resistor because now the V9958 is outputting
a wait signal on pin 26).

The noise can also be seriously reduced by putting a decoupling capacitor (e.g.
100nF) between pin 21 and pin 20 (Analog GND). Keep the length of the leads of
this capacitor as short as possible (5mm).

2. for a NMS82xx the composite video signal is generated with a MC1377 chip. It
uses the RGB info from the VDP, but also the Csync signal. Perhaps the timing
for one of those signals has changed a little bit in the V9958. If this is the
case, then the timing when the colour burst should happen in the cvbs signal,
will also be shifted (and wrong).
You can adjust this by adjusting the variable resistor VR301. It is located on
the separate video board of the NMS825x next to the big black MC1377 chip.

I don't know if it will help, but at least it is worth a try.

Best regards,

Jon
___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread Jon De Schrijder


Alex Wulms wrote:

 I know (from the v9938 databook) that the v9938 comes
 in three versions. I have always wondered about the difference.

see page 123 of the V9938 databook: the difference is in the timing of the VRAM
interface. So this won't be the problem...

Bye,
jon
___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread JP Grobler

 A stupid question maybe: if you have a rgb monitor, why looking at the
 composite?
I do not usually use the composit, RGB is much better.
I just got a fright when I replaced the chips and the signal was gone!
Now I am puzzled.

JP


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003
___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread JP Grobler
  Does anybody have a 9958 running through composit signal?

 I just tested it on my NMS8245: you are right. The image is remaining grey
 on my computer aswell. I wonder what's going on... It don't think the
screen
 flickered when switching frequencies either. I don't know if it's supposed
 to do that, but... it seems to me it should.

Glad to hear my msx is not on the blink.

JP


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003

___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-25 Thread JP Grobler

 2. for a NMS82xx the composite video signal is generated with a MC1377
chip. It
 uses the RGB info from the VDP, but also the Csync signal. Perhaps the
timing
 for one of those signals has changed a little bit in the V9958. If this is
the
 case, then the timing when the colour burst should happen in the cvbs
signal,
 will also be shifted (and wrong).
 You can adjust this by adjusting the variable resistor VR301. It is
located on
 the separate video board of the NMS825x next to the big black MC1377 chip.

Tried it. I get color ,4,4 (startup) to go greenish or yellowish, When this
happens Defnite bands of blue, yello  red can be seen as well.
When the background colour is changed, it still shows the correct colour for
a sec then falls back, maybe the signal is weak.

I'll try to clean the power supply.

JP


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003

___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


[MSX] Colour loss

2003-07-24 Thread JP Grobler
Hi

I replaced my 9938 with a 9958 in my mns8250
Now I have colour on RGB, but gray picture on composit

Any ideas why?

JP


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003
___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: [MSX] Colour loss

2003-07-24 Thread JP Grobler
I swapped the 9938 back and all the colours is fine again
Composit and RGB?

What is wrong with my 9958?

JP
- Original Message - 
From: JP Grobler [EMAIL PROTECTED]
To: msx [EMAIL PROTECTED]
Sent: Friday, July 25, 2003 6:33 AM
Subject: [MSX] Colour loss


 Hi
 
 I replaced my 9938 with a 9958 in my mns8250
 Now I have colour on RGB, but gray picture on composit
 
 Any ideas why?
 
 JP
 
 
 ---
 Outgoing mail is certified Virus Free.
 Checked by AVG anti-virus system (http://www.grisoft.com).
 Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003
 ___
 MSX mailing list ([EMAIL PROTECTED])
 Info page: http://lists.stack.nl/mailman/listinfo/msx
 


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003
___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Fw: [MSX] Colour loss

2003-07-24 Thread JP Grobler
I swapped the 9938 back and all the colours is fine again
Composit and RGB is fine!
 
What is wrong with my 9958?
 
 JP
 - Original Message - 
 From: JP Grobler [EMAIL PROTECTED]
 To: msx [EMAIL PROTECTED]
 Sent: Friday, July 25, 2003 6:33 AM
 Subject: [MSX] Colour loss
 
 
  Hi
  
  I replaced my 9938 with a 9958 in my mns8250
  Now I have colour on RGB, but gray picture on composit
  
  Any ideas why?
  
  JP
  
  
  ---
  Outgoing mail is certified Virus Free.
  Checked by AVG anti-virus system (http://www.grisoft.com).
  Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003
  ___
  MSX mailing list ([EMAIL PROTECTED])
  Info page: http://lists.stack.nl/mailman/listinfo/msx
  
 
 
 ---
 Outgoing mail is certified Virus Free.
 Checked by AVG anti-virus system (http://www.grisoft.com).
 Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003


---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.504 / Virus Database: 302 - Release Date: 24/07/2003
___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx


Re: Fw: [MSX] Colour loss

2003-07-24 Thread Saku Taipale

On Fri, 25 Jul 2003, JP Grobler wrote:

 I swapped the 9938 back and all the colours is fine again
 Composit and RGB is fine!
  
 What is wrong with my 9958?

Hmm, iirc you need to do few modifications when using v9958.
I don't recall now what wires do you need to add, but maybe
Hans Otten's pages has the answer...

Greetings: Saku Taipale, [EMAIL PROTECTED]

___
MSX mailing list ([EMAIL PROTECTED])
Info page: http://lists.stack.nl/mailman/listinfo/msx