Having PC drive an IR emitter based on JP1 upgrade data?

[semuther]

If you want to run long wires from your printer port all over the room, I guess it isn't harder for the PC to support up to a dozen instead of one.

I was assuming one IR output aimed at several devices across the room.

When I did such things long ago on much slower computers without Windows (so latency was entirely under my control) I found it much easier to use a pair of IR LEDs wired in series rather than just one. That reduced aiming problems, and the hardware person helping me at the time said the higher voltage across two series IR LEDs made the current control simpler. Just currious if you have an opinion on that one. I have no idea now how you do current limiting for an IR LED powered by the printer port.

You'd only be able to draw about 2.5mA from the printer port. Typically a resistor in series with the LED sets the current. Find out the operating voltage of the LED, subtract it from the 5V available on the port and divide by the LED's specifide curret requirments and you have the value of resistor needed. If you want to drive more than one LED, you'd need an external buffering circuit with it's own power source. I prefer running LEDs in parallel. LED's are not exactly the same in terms of current draw so running them with their own limiting resistor is preferred.

My PC and all the AV equipment are mounted in a rack. I have one master IR emitter with two IR LEDs sitting about 3 ft in front of the rack and are, at present fed by a wired repeater for control from other rooms. Works great. I just want to throw the PC into the picture as an additional control device.

Steve

[underquark]

For hardware guys who don't like too much software, what about a rather retro idea: Output from PC to an array of switches (either virtual, electronic, relays, STAMP chip, whatever). Hardwire said array grid-fashion into the contact pads (or any convenient part connected to each pad) of your chosen remote. Only 2 switches need to be selected from a panel of about 12 to activate a given button. Manually fidly (but some prefer that to coding), not elegant, not all that flexible but timing errors of even many milliseconds ought not bother it.

Code: Select all

  1    2     3     4     5     6
A Mute Power Menu  Guide Info  Text
B Num0 Num1  Num2  Num3  Num4  Num5
C etc. etc.
D 
E 
F 

[semuther]

For hardware guys who don't like too much software, what about a rather retro idea: Output from PC to an array of switches (either virtual, electronic, relays, STAMP chip, whatever). Hardwire said array grid-fashion into the contact pads (or any convenient part connected to each pad) of your chosen remote.

Perfectly valid idea. The one limitation is that you're limited by the capabilities of the remote. Having the complete protocol and upgrade files on a PC means the ability to fully emulate a given remote with virtually unlimited macros etc. It is software intensive however.

Steve

[johnsfine]

BTW, do you have any of dual CPU, dual core, or hyperthreading?

Those are all another way of improving latency issues. During IR transmit have the thread doing that run at higher priority and the OS will mostly use the other apparent CPU for everything else.

With hyperthreading the use of the imaginary "other" CPU cuts out almost half of the cpu speed of the high priority one. But that doesn't matter at all. The operation doesn't need a fast CPU. Ten times slower would still be plenty. It just needs a continuous CPU.

With a little more understanding of Windows CPU scheduling than I have, one could probably adjust things so all hardware interrupts go to a single CPU and the IR activity always runs on the other. Then you should have perfect CPU resource even for super fast IR protocols.

[underquark]

The one limitation is that you're limited by the capabilities of the remote. Having the complete protocol and upgrade files on a PC means the ability to fully emulate a given remote with virtually unlimited macros etc.

If this is for home use (moving your Sat dish, scheduling VCR recorder, powering on and off devices) then I can't see that being too much of a problem as you would only expect to be using a finite number of target devices and the remote can be set up to cope. If it's for a mobile, commercial application where you intend to, say, take a laptop around to set up devices for various clients then that's a different matter.

[semuther]

The one limitation is that you're limited by the capabilities of the remote. Having the complete protocol and upgrade files on a PC means the ability to fully emulate a given remote with virtually unlimited macros etc.

If this is for home use (moving your Sat dish, scheduling VCR recorder, powering on and off devices) then I can't see that being too much of a problem as you would only expect to be using a finite number of target devices and the remote can be set up to cope. If it's for a mobile, commercial application where you intend to, say, take a laptop around to set up devices for various clients then that's a different matter.

My VCR, a JVC D-VHS, has a remote with more buttons than will fit on a universal remote even if shift functions are used (not enough memory) as well as a very cumbersome method of inputting programming data. There are 8 up down arrows for inputting time scheduling for example. I've made a full upgrade file for it but have never gotten it loaded into my 8910 remote in any maner that makes sense or is useful so this is one typical example of the PC blaster method that would take care of all that.

Steve

[underquark]

My VCR, a JVC D-VHS, has a remote with more buttons than will fit on a universal remote even if shift functions are used

Must be one beast of a remote.

[The Robman]

My VCR, a JVC D-VHS, has a remote with more buttons than will fit on a universal remote even if shift functions are used (not enough memory) as well as a very cumbersome method of inputting programming data. There are 8 up down arrows for inputting time scheduling for example. I've made a full upgrade file for it but have never gotten it loaded into my 8910 remote in any maner that makes sense or is useful

This sort of situation calls for some imaginative programming. We have a special protocol called the "Device Multiplexor" which lets you stack multiple setup codes onto a single device button. You could create several upgrades for your VCR and then use the multiplexor to alternate between them. The first upgrade would contain all the buttons that you use on a regular basis (eg, power, play, etc). Then you could have a second upgrade that contains all the buttons that you would need to do the programming (ie, those arrow buttons, etc). If needed you could also create a 3rd and 4th upgrade, etc.

Next, you could program a "Device Specific Macro" onto the MENU button in VCR mode, which when pressed would not only bring up the VCR's programming menu screen, but would also change the code assigned to the VCR button from the "regular" code to the "programming" code, this would make the up/down, left/right, VOL +/-, CH +/- buttons switch to being the buttons that you need to do the programming. If you use the EXIT button to clear the on-screen-menu, you could program another "device specific macro" to the EXIT button that not only sends the EXIT signal, but also switches the code back to the regular code.

That's just an example, you wouldn't have to use any of the buttons that I used, but hopefully you get the idea.

[semuther]

Thanks Rob, will look into that for local control. My interest in using the PC for control still remains though. I have my PC set up with VNC such that I can log in and control it from anywhere. It's great for setting up my OTA and satellite tuner cards while having lunch at the office or traveling. The added capabilities of being able to manipulate my IR devices remotely also has great appeal.

Steve