Petscii Forums "PETSCII.COM"
« Candidate for nIRQ found »

Welcome Guest. Please Login or Register.
Nov 20, 2009, 9:59pm

[Home] [Help] [Search] [Login] [Register]



Petscii Forums "PETSCII.COM" :: General :: DTVHacking :: Candidate for nIRQ found
   [Search This Thread][Send Topic To Friend] [Print]
AuthorTopic: Candidate for nIRQ found (Read 647 times)
gmoon
Wizard's Apprentice
*****
member is offline

[avatar]


[homepage]

Joined: Mar 2006
Gender: Male
Posts: 876
Location: Ohio, USA
Candidate for nIRQ found
« Thread Started on May 17, 2006, 7:14pm »
I think I’ve I.D.ed one of the interrupt lines, for the Hummer at least.

I don’t have a ‘scope, but I do have an old digital logic probe. Poking around the Hummer pcb, the only spot I found that has a low-frequency signal is R48 (left side of resistor, closest to the ASIC, and right beside the IEC DATA tap.)

The logic probe can’t display frequency, so someone with a ‘scope should verify. If it's a 60 Hz signal, it's probably nIRQ.

The schematic does indicate that the nIRQ and nNMI lines breakout next to IEC DATA, so it fits.

I didn’t check anything near the audio or video outputs, since there’s bound to be lots of frequencies present.
Link to Post - Back to Top  IP: Logged
gmoon
Wizard's Apprentice
*****
member is offline

[avatar]


[homepage]

Joined: Mar 2006
Gender: Male
Posts: 876
Location: Ohio, USA
Re: Candidate for nIRQ found
« Reply #1 on May 18, 2006, 3:39pm »
I think it's a safe bet that this is nIRQ.

The dtv programmers doc points out it's bi-directional. Not only does it carry a low-freq signal, if you ground it, the BASIC cursor blinks so fast the screen refresh can't keep pace.

So grounding this trace also generates an interrupt.

[edit-- add image of trace]
(apologies to David Murray for using his pic.)

[image]

The circles, lines and arrows (Officer Obie) around IEC info were drawn by David, and are NOT nIRQ.
« Last Edit: May 18, 2006, 4:20pm by gmoon »Link to Post - Back to Top  IP: Logged
David Murray
Moderator
*****
member is offline

[avatar]


[homepage]

Joined: Dec 2004
Gender: Male
Posts: 1,561
Location: Kennedale, TX
Re: Candidate for nIRQ found
« Reply #2 on May 18, 2006, 4:36pm »
Fascinating. I'll have to experiment with that some. So you can use that line to generate an IRQ as well as sense one externally. I'm going to be finished with my big hacking project and the video that goes with it in a few more days.. I'm like 80% done. At which point I will begin to experiment with the ADC and also the NMI. I flashed the kernal last night (and recorded it happening on video) so I had to hold my breath there for a while.

I suppose the nmi will trigger before the actual IRQ routine, right? So any code that uses that as a clock source would need to setup the data lines the way you want it before exiting the IRQ routine, so that next IRQ, the data gets clocked in.

I took a MUCH better, high-res photo of the PCB. It was tricky. The one you have shown here was one I took a few months ago. The trick is, it was too dark to get a clear shot without the flash. But using the flash means the camera can't get a sharp focus on the object until it is too late.. So I illuminated the board externally with a bright halogen light and took a really clear high-res image. I'll be replacing the ones on my website soon with it. I'm also using it in the videos to zoom into parts and stuff. I'll add the NMI onto it when I add the other normal parts.

Link to Post - Back to Top  IP: Logged
gmoon
Wizard's Apprentice
*****
member is offline

[avatar]


[homepage]

Joined: Mar 2006
Gender: Male
Posts: 876
Location: Ohio, USA
Re: Candidate for nIRQ found
« Reply #3 on May 18, 2006, 6:33pm »
Yeah, this open-collector bi-directional stuff is pretty cool. And I 'probed' and grounded that pin (and others) many times, so it seems that these don't 'brick' easily.

Looking forward to seeing new photos and the rest of your video.

My day job is photography, but sadly I haven't taken any photos of the the pcb, etc. There isn't much point with all the spaghetti I've added--have to get a new, unmodded hummer. A flatbed scanner might also be a good way to capture an image.

And I don't personally do much with macro photography--if I did, I'd have a ring-light and all that jazz. (Other than that, I definitely have some good digital gear.)
Link to Post - Back to Top  IP: Logged
David Murray
Moderator
*****
member is offline

[avatar]


[homepage]

Joined: Dec 2004
Gender: Male
Posts: 1,561
Location: Kennedale, TX
Re: Candidate for nIRQ found
« Reply #4 on May 18, 2006, 10:08pm »
Gmoon... I played around a bit with the ADC. I have noticed a few things. First of all, no matter what I stick on there, I can only get 00 or FF.. nothing in between. What type of thermistor were you working with when you tested that?

Also... I may not be able to sample it as fast as I thought. Using the ML monitoring routine with the hex-readout, when you go between 00 and FF, it actually counts.. you can see it doing it on the screen. It probably takese half a second to change. So I might have to setup my routine to sample one thermistor every second (or 60 IRQs) so that it has time to adjust.
Link to Post - Back to Top  IP: Logged
gmoon
Wizard's Apprentice
*****
member is offline

[avatar]


[homepage]

Joined: Mar 2006
Gender: Male
Posts: 876
Location: Ohio, USA
Re: Candidate for nIRQ found
« Reply #5 on May 19, 2006, 6:13am »
I'm not quite clear here--is the thermistor responding slowly, or is the routine?

Also, I'm not clear on:

"it actually counts.. you can see it doing it on the screen. It probably takes half a second to change"

Is the ADC not responding to a fast voltage change?

Why don't we start a new thread, "modding the ADC" or something, since we're tacking this on to the irq thread...
« Last Edit: May 19, 2006, 6:21am by gmoon »Link to Post - Back to Top  IP: Logged
   [Search This Thread][Send Topic To Friend] [Print]

Google
Webjledger.proboards.com
Click Here To Make This Board Ad-Free


This Board Hosted For FREE By ProBoards
Get Your Own Free Message Boards & Free Forums!