7800 ext5 initial test results

[vickyg2003]

Elizabeth,

I tested out all of this on my test system, long key presses in a macro, toad tog on DiscreteON, on XShift-1, tried cloaking. All these things tested out in my simple environment, so I opened ForVicky and changed the VCR library code to a piece of equipment I have. I tested your macro, it worked. I changed the Shift-1 to DiscreteOn and changed the macro, and it worked. I swapped side of the DiscreteOn and Shift-0, and it still worked.

I need to see one that isn't working.

This has been really cool though. I'd only read about cloaking, and hadn't tested it. I also didn't realize that you could do the LKP inside a macro, which is SO Cool. I had thought the LKP had to be on a physical button. Really cool.

[ElizabethD]

I thought we were to be on break till 4/15 and really appreciate you checking stuff.
Reviewing my list, I see that one problem, spelled out in points 3,4,5 most likely was due to out-of-synch situation. Happens to me often while debugging ToadTogs. Arrrgh!

Legacy issues is a possibility, but I did reload the whole new thing, chekd a bunch of device things in Raw Data and anyway the files match.
Corruption of RAM is a possibility, but 981 resets didn't fix. Things may start close to being ok, and over time deteriorate.

I don't think there are any issues in this extender and your trial proves it..
When I could repeat point 2 results over and over, that was one thing.
When what now comes out varies, that's a whole different story.

Want to read a sad, irrational, story? Here we go:
Things went downhill since my post so much, that it implicates problems here.
I get varying results. Nothing is constant. Shift-1 works, then it doesn't. Xshift works then it doesnt. Base code Discrete works, then it doesn't, and every time different things happpen, rarely correct.

So I fired up Tommy's IRSA gadget and watched things there. Well,
All base code signals I checked are perfect. Some macros are perfect. Some macro steps in the middle are skipped. Some steps at the end are missing. And I even got a totally wrong signal for one of the upgrade discretes not coded as ToadTog. This one is a killer because an unknown PanasonicCombo signal came out when I expected NEC2. By unknown I mean a bit pattern for which I do not have any such function in any of my upgrades. So it made it up. And the Panasonic tuner's panel said "DIMN O". I have no clue what that means. Shouldn't be there.

I have a hunch my one and only C7 might have given up a ghost.

[vickyg2003]

The problem your are experiencing COULD be that the extender, and not your remote. I haven't done any extended testing. I just keep loading and testing things one part at a time. So perhaps over time mine might deteriorate too. I haven't put this to work on my husband's remote, his doesn't have the 6pin head soldered in and its hard to find the sweet spot. I was afraid to do it, because I didn't have any spare Cinema 7's either. I've only soldered once in my life, and I've kind of forgotten how, but I do have detailed instructions somewhere.

I just received an order of remotes from Rob, and when I went to squirel them away so DH wouldn't find out I'd bought MORE remotes, I found 3 cinema 7's. So I'll do my soldering, and if I do it as well as I did it last time I might have an extra.....

So what is Tommy's IRSA gadget?

[unclemiltie]

All right, I'll bite


I've not used LKP yet, although have always wanted to invest the time to make it work but was of the assumption that "long key press" had to have a key that was pressed for a long time.

What does it mean to have a long key press in a macro?

[vickyg2003]

Hi UncleMiltie.

the assumption that "long key press" had to have a key that was pressed for a long time

Well you are right. LKP does have to be on a key that was pressed a long time. Elizabeth has them on the last key in a macro. If you hold down the macro key, the last key is held as long as the button is held down. So the LKP can be the last key in a macro. Actually, I assume it HAS to be the last key in a macro. Intelectually I just didn't expect that to work. If you really want your extender to be beat up, get Elizabeth to do it. After having her test something you've got a pretty good idea that its really working.

That LKP is really a nice protocol. I use it all over the place. I had only used it to avoid shifting. If I hold down the key, I get the xShifted version, otherwise I get the shifted version. But now that I've had a more thorough education I'm going to be using it on my device keys. A regular keypress does the device selection macros, but if you do the LKP it also will also go through the input selection process and selects the appropriate device on my Tv.

LKP is one cool protocol!

[ElizabethD]

SUCCESS! It's stable, reliable, and other than one thing (below), no more troubles. Darn batteries did me in, I forgot to think about that one, so I apologize for all the trouble. Not only were they weak, but they did their usual C7 job of shifting. They're now taped over. It all worked when the lid was closed, but not when batteries faced down. It took a husband to make the connection and IRSA confirmed

There is a problem with issuing last command where punch-through is involved. I confirmed all observations with IRSA down to the bit level. Following examples are about VOL+, but may apply to other punched through functions such as channel or transports, but I haven't tried. Here's what I have,

1. Volume is punched through to the tuner in all device macros
2. Macro on 6 sends Vol+ to the tuner and repeats when 6 is held:
--- X_DVD;VOL+
3. Macro on SHIFT-DVD/LD does not send the final vol+ to the tuner at all (it does in 6131)
--- X_RCV;SHIFT-1;XSHIFT-1;XSHIFT-1;SHIFT-VOL+;X_DVD;DiscreteOn;SHIFT-2L;VOL+
4. Macro on SHIFT-TV sends vol+ to TV (Panasonic device 128) instead of the tuner (device 160)
--- X_RCV;SHIFT-1;SHIFT-VOL-;X_TV;SHIFT-1;SHIFT-2L;VOL+
5. DVD upgrade does not have any functions assigned to the Volume keys
6. TV upgrade has functions assigned to itself, which is then ignored because of V_RCVR in all device modes.
7. Shift-vol+ is likewise ignored, and I believe it should not be (I had to move shift-vol+ from the tail end of the device macros to an earlier position, but I don't care about this one since it can be so moved)
Based on above, my theory is that when buttons are not mapped AND macro runs on a device button, Volume punch through is lost. I've ruled out everything else I could think of and can't think of anything else.

Wish list: Could you preset the ToadTog register with zero when the extender is initialized. I think it sometimes comes in with unknown bits already set. And while you're at it, how about also zero out RBF (if unused), and possibly find another one or two unused registers that we might want to use for other things.

I forgot to answer about what IRSA is - it's a infrared signal analyzer, a very useful gadget in my opinion - accurate and can record long macros. So if you know what to expect, you can nail down what comes out to the most basic level without even having to sit in front of your gear.
https://www.hifi-remote.com/forums/viewt ... light=irsa

[vickyg2003]

Liz,

For your registers, I'm pretty sure we can pick up 4 just because the 7800 doesn't have as many features as the 1994, but if you need more registers, you can always shorten the Macro buffer. We allow "no more than 32 pending keys" right now, but if you want more registers, we can shorten that. I can't imagine macros that complex, but YOU probably can .

I've got some really nice foam that is mushy enough to not bend the battery cover, but really holds the batteries in place. Loose batteries and the extender REALLY don't mix.

When I get some time I'll do some tests with the extender to see if the toadtogs are random. If so we can clear the register when the extender is loaded. I wouldn't want to add any unnecessary bulk to the extender. The original writers really did a good job of compacting these.

Remind me after April 15th to take a look at the 6131 and see how they make the volume work. Ive got 6131's too.

[ElizabethD]

Thanks. No I do not imagine macro of 32 steps. In fact those 32 can be infinite length. 2 registers is what would be cool to have for some other protocols. 4 would be a luxury. I just think it's cool to know what's free to use and zero at start.

Foam sounds fancy. I normally use one layer of paper towel when the lid is on. But when you upload as often as testing requires, you don't close the lid. I already broke one latch on 8910, not about to do it to C7.

ToadTog register - I don't know if it's random. But I'm darn sure that it's not zero when extender gets enabled. At least bits 0 and 5 aren't clear. Discrete on won't work until I first use Discrete off on two items. I've been wrong before. Could be wrong now. No big deal. See ya in April Oh, and thanks again for the extender.

[vickyg2003]

Elizabeth

Send me the IR for the 6131 that works correctly for the volume punch tly and the 7800.IR that doesn't and I'll take a look to see what is happening.

I'm going on vacation and hope to be able to bring along the laptop so I have something to do in the evenings while DH watches sports. Last time having the extender to work on was fun.

I have a protocol I want to try, and then maybe look at JP1.2.

[ElizabethD]

Vicky, files you asked for
https://www.hifi-remote.com/forums/dload ... le_id=4425
https://www.hifi-remote.com/forums/dload ... le_id=4426
I presume you have 6131 extender code or 8910 where the vol repeats work.
Have a nice vacation. Done your taxes yet?

[vickyg2003]

Ah, seeing this in IR form, made this clear. I should have read this in your post and been able to diagnose this, but I didn't get the big picture.

Macro on SHIFT-DVD/LD does not send the final vol+ to the tuner at all (it does in 6131)
--- X_RCV;SHIFT-1;XSHIFT-1;XSHIFT-1;SHIFT-VOL+;X_DVD;DiscreteOn;SHIFT-2L;VOL+

Where you have the X_DVD you want an X_CANCEL. Same goes for the Shift_TV, and the Shift_VCR. Those X_ commands force all the indexes to be 1,2,3,4,5.... So since there is no volume on the DVD there will be no volume sent. Since the SHIFT-DVD is called from a lkp in the DVD macro, the base device is already DVD.

[ElizabethD]

Arrghhh! it was the xcancel thing. Nothing beats a second pair of eyes.
Initially I held on to shiftdevices so husband wouldn't be too upset by LKP (C7 is his favorite). Putting x_cancel there would defeat it, so I did what I intended to do in the first place, move it all into LKP, the standard way, works fine. Passed HAF
I also checked out multiplex (just 2 things, no room for more), it works fine.
So this is the current, functioning state. I'll be changing things, but there is no question all the extender 5 stuff is FINE.
https://www.hifi-remote.com/forums/dload ... le_id=4428
I think the only way to not ever mess up is to write an extender in the first place