Re: Cinematronics > Atari Adapter???

Tom McClintock wrote:

>Well get right on that Rodger! :)
>
>Seriously, I was under the impression that the settling time for the DAC was
>determined more by the DAC than subsequent Op-Amps. The DAC7541 seems to have a
>settling time of 30 m sec, while the DAC-80s (or DAC7580) have settling times
>around 4 m sec.
>
>

Not sure where you got the 30 mS spec, the only settling times listed on
the DAC7541 spec sheet are .6 uS typical for the output CURRENT to
settle to .5 LSB. The opamp that's doing the current to voltage
conversion will be the speed limiter. VOLTAGE output DACs are
notoriously slower than current output DACs.

>Unfortunately, even if the DAC was $11 each, you need two, plus the supporting
>hardware so your cost is running about $35 in parts alone (gotta include those
>connectors). A finished PCB would add another $10 or so. This gets into the
>realm of what a Cine monitor board costs (well, usually on the high end as most
>operators don't want these monitors at all).
>
>But I agree a 'new' Cine->WG conversion board would be really cool. What's the
>cheapest way to go with this?
>

The cheapest way? Build your own R-2R array and don't even buy the
DACs. The way Cinematronics draws vectors (capacitor integration) you
really don't need an accurate DAC. Probably don't even need 12 bits.
 All you're generating are start points and "end points" (not really end
points, more like "it goes thisaway" points). A bumpy DAC won't matter
because it doesn't draw the line in between the points.

>
>tm
>
>Rodger Boots wrote:
>
>
>
>>You could use something like the DAC7541 which is under $11 even in
>>single lots. The op-amp following the DAC will determine the settling
>>time. Like Tom says, you can have the DAC only give a positive output
>>(in this case by using a negative reference for the DAC) to allow cheap
>>switches and subtract out the midpoint (using the same reference voltage
>>for best results) to get the desired output swing.
>>
>>Really nothing much to it.
>>
>>Tom McClintock wrote:
>>
>>
>>
>>>Wait, here are some notes from Zonn.
>>>
>>>------------------
>>>Your biggest problem with doing a Cine->WG board is
>>>that you are forced to use 12 bit parallel DACs, which
>>>are expensive. The Cine design is pretty much going to
>>>force you to find ~1 to 2 us settling time parts. (To
>>>keep the frame rate up, we had to do the same.)
>>>
>>>Though you might be able to save some money by using
>>>single supply 12 bit DACs and shifting the voltage
>>>after the analog switch / RC stuff. This is the
>>>approach we took in the ZVG. This also allows the use
>>>of the much more available and less expensive 74HC4066
>>>switches which are only single polarity switches. The
>>>LF13331 were bi-polar, and also expensive (when they
>>>were available).
>>>
>>>You can't use serial DACs. They would require a
>>>parallel to serial conversion and there is no time
>>>allocated in the hardware to allow this. The DACs are
>>>latched and the Z-axis is turned on for the draw,
>>>there's no "clock data into DACs" time, which at
>>>minimum would take 16 clock cycles, if both DACs were
>>>simultaneously clocked.
>>>-------------------------------
>>>
>>>
>>>Beyond that, lemme know if you come up with something :)
>>>
>>>
>>>tm
>>>
>>>
>>>

---------------------------------------------------------------------------
** To UNSUBSCRIBE from vectorlist, send a message with "UNSUBSCRIBE" in the
** message body to vectorlist-request@synthcom.com. Please direct other
** questions, comments, or problems to vectorlist-owner@synthcom.com.
Received on Mon Oct 14 09:54:58 2002