Freescale fractional clock divider paper

Well that definitely won't help. Try this:

What's the big deal? See...

on the S.E.D/Schematics Page of my website, posted some 14 years ago.

The "big deal" is that these two "engineers" will get H1B visas and come to the US and take your job >:-} ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.

I'm hoping to use a colorburst xtal oscillator and divide it down by

227.5 to get the horizontal NTSC line frequency without the use of a fractional PLL (which I don't have the hardware to implement in the device I have to use)

The linked paper looks like a variant on something I've seen before; have two counters that divde by 227 or 228, run them into a 2 channel multiplexer/3 bit LUT, feed the output into a divide-by-227 and run that back to the multi line-select input. Not sure exactly what improvements/ modifications they're making here

If this is the average quality of professional writing coming out of the subcontinent I'll go there and take 50 of theirs with time to spare. Hope they like shoveling snow, suckas!

227 _and_ 228, rather

From my post: Drive U1A:A from 3.579545... Run U1A:Y thru a DIV455 ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.

Or, look at some old HP gear that squared up a 40 MHz clock with 74H00 and trapped the fifth harmonic... to get 200 MHz so they could divide by N, N.2, N.4, N.6, N.8...

The '74H' dates the scheme, that would have been early-mid 1970s...

Are they still using NTSC and designing new stuff for it?

Who's "they"? Lots of stuff still has NTSC/PAL composite video inputs and as long as that's the case there will probably be a need for stuff with composite video outputs. The fact that with "modern" displays the signal is just fed to an ADC immediately and processed digitally is irrelevant - it still needs to have the correct timing/sync pulses.

It's a while since I played with such things but could one not just feed th e signal to an oscillator running at around 1/2.5th the frequency and have it lock. With the right amplitude it should lock on the voltage extremes bu t not the centre voltage of the osc input.

My memory says no, only whole multiples, but I can't think of any reason wh y.

NT

the signal to an oscillator running at around 1/2.5th the frequency and hav e it lock. With the right amplitude it should lock on the voltage extremes but not the centre voltage of the osc input.

double it then integer divide

m

Which is what I said in...

Message-ID:

From my post,

Message-ID:

Drive U1A:A from 3.579545... Run U1A:Y thru a DIV455

(Only 1st D-flop is needed.)

I guess bitrex needs it built and handed to him on a platter ?>:-} ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 
              
     It's what you learn, after you know it all, that counts.

Well, the same stuff probably has RGB, Composite or digital inputs as well, and will provide a much better picture when those are used.

I really can't see a use of NTSC (or PAL, for that matter) in the modern world. Yes, people clinging on to their old TV set may be using it on cable (transmitters have been shut down over a decade ago here), but using it for equipment-monitor connections is a bit weird.

We use video camera's and monitors to 'see' into the NIR. Fortunately there is still a big consumer market.. (so things are cheap) Back-up video monitors for cars and trucks.

George H.

One use is for closed-circuit television. To make a simple camera-wire-display ensemble.

DVI is short-wire-only (ditto HDMI and DisplayPort) USB options are driver-crippled and short-wire-only IP variants don't function in real-time VGA takes too many wires, and there's no camera so equipped ATSC encoders aren't available in a camera: that digital transmission standard is fast to DECODE, but not to ENCODE.

Find something OTHER than composite video that can drive a 20-meter video link, which doesn't delay to boot an operating system and navigate the internet. Tell me that the cable it uses is as easy to buy as BNC-terminated coax. Tell me that the cameras and monitors are interchangeable with multiple vendors.

Composite video is sometimes the only candidate.

I noticed that it looks like the designers of this FPGA were thoughtful enough to bury an edge-detection functional block way down in the manual that can be configured to fire on the rising and falling edges of a clock simultaneously, thank you for your kind offer, though! ;-)

For reference here's the patent where a logic configuration similar to what it looks like the writers in the original paper (which I still don't follow) are talking about is described:

FPD-Link?

Yep. Good enough is usually good enough.

4X colorburst crystals are easy to find, and cheap. 14.318180 MHz. They were common in a lot of video equipment.

ed the signal to an oscillator running at around 1/2.5th the frequency and have it lock. With the right amplitude it should lock on the voltage extrem es but not the centre voltage of the osc input.

n why.

maybe no-one can answer the question :)

NT

I think "Rob" might be confusing composite with RF-modulated composite, like you'd plug into the antenna jack of an old tube TV set and select channel 3 or channel 4. I'm not modulating anything, just baseband video

• sync. The cable with the yellow plug.