Re: 6569 luminances

On 05/13/2012 08:13 PM, silverdr@wfmh.org.pl wrote:
>
> On 2012-05-13, at 19:36, Gerrit Heitsch wrote:
>> I kind of like the 250407 boards. The last board revision with the discrete PLL using the MC4044P and 74LS629. Doesn't look blurry to me
>
> I recall (out of memory) that there was a big difference between the discrete and 8701 revisions in sharpness (or rather blurriness ;-) especially when connected via separate chroma / luma. But also on the composite output the difference was visible. Luma on the latter was tack sharp, and similar to the composite of the latter, on the discrete. Out of my head...

The blurriness, if present, can usually be remedied by removing the RF 
modulator and replacing it with a little circuit that does the signal 
conversion to proper luminance/chrominance signals. It's only 2 
transistors, a 1N4148, some resistors and 2 capacitors.

It's not a matter of the clock generator, the 8701 is not a PLL, its 
output does contain some jitter.


>> and if you're lucky you get the 6569R3 in ceramic.
>
> That's exactly why I plugged-in the 6569R3 some years ago - I was happy to have the c00l jewel: ceramic VIC version - but didn't test it properly back then.

I have one 6569R3 in ceramic and a few 6569R1 in ceramic. They all work, 
if you take the 5 luma level limitation of the R1 into account.

A while ago I did an experiment, using 3 1N4001 diodes I limited Vdd of 
the 6569R5 on my test board to about +10V instead of +12V. The chip 
still worked fine, including the color output, just got a bit less hot.

> Yup - those were IMHO the best boards. With lots of unused space though. Mine (the one I refer to) is 250425. I keep it because it is a) the original first I bought and b) it is customised, with practically everything socketed, etc.

Now you only need to replace 2114 color RAM with a CMOS version and the 
7805 and 7812 with state of the art switching regulators and you're set.

  Gerrit


       Message was sent through the cbm-hackers mailing list