JP1 Remotes

I just purchased the newly released Roku Ultra and using the Roku 1 , 2 , 3 (Official) upgrade.
https://www.hifi-remote.com/forums/dload ... e_id=12733

This upgrade works fine on the Roku 2 and 3 but the new Roku Ultra seems to need a longer delay between repeats. Say I hit the Down button, a lot of times it will overshoot and skip items as though it sent two keypresses.

Not sure if I can just edit the Roku (official) protocol / Protocol Data, Repeat Type or what settings to edit.

I uploaded this file which is several learns from a Harmony Roku Roku Ultra and it works much better than the Roku 1 , 2 , 3 (Official) upgrade.
https://www.hifi-remote.com/forums/dload ... e_id=14241


Thanks!

Jim

I don't know how to tweak the protocol either. But here is a version that uses NEC2 which is closer to your learns. Give it a try and let us know how it works.

https://www.hifi-remote.com/forums/dload ... e_id=14242

mdavej wrote:I don't know how to tweak the protocol either. But here is a version that uses NEC2 which is closer to your learns. Give it a try and let us know how it works.

https://www.hifi-remote.com/forums/dload ... e_id=14242

That one is actually worse.

I'll have to defer to the protocol experts then. I can't figure out how to get it to not send 2 copies.

I moved the other to diagnosis since it's not working.

Based on what I can see, the RCRP05B learns from the Harmony have nearly identical timing to the Roku Official PID 021A executor. BTW, our expectation is that the Roku IR protocol has timing identical to NEC2; the only difference "should" be that commands are sent in pairs with bit 7 of the OBC toggled.

Which remote have you loaded the upgrade into? Is it extended? Have you tried more than one remote?

Folks at remotecentral are also having trouble with repeats on the Ultra. Could be a bug or the new normal.

http://www.remotecentral.com/cgi-bin/mb ... ,last#last

The PID 021A executor sends a single frame of the OBC1 followed by at least one frame of OBC2, which is OBC1 ORd with bit 7. If repeats are requested (e.g. by holding the button) more frames of of OBC2 are sent. So a repeated button sequence is OBC1, OBC2, OBC2, OBC2......
The learns from the Harmony seem to be in an alternating sequence: OBC1, OBC2, OBC1, OBC2.....

However, it isn't clear to me how the learns were done, or if the RCRP05B understood the actual way the repeats are sent from the Harmony. Do the learns work if sent from the RCRP05B instead of the Harmony?

Is this still an open issue?

You say the learns work much better, or is that the harmony works much better?

Do the learns work all the time, or are the buttons still doubling up with the learns?

We can make a non repeating version of this, or we can try to mimic the original protocol with the alternating OBC's.

Looking at this, I see that not only is the pattern repeating differently on the learns

Learn repeat:
OBC1, OBC2, OBC1, OBC2

Upgrade repeat:
OBC1, OBC2, OBC2, OBC2

There is also a leadout time issue.
The leadout time on the official protocol is 19000 units bigger than what I am seeing from the learns. This could certainly account for strange repeat rate.

Jim, do you have IR.exe on your machine? IR is a lot easier when we want to go experimenting with protocols, and we are going to use a lot of trial and error if we want to get this right with repeating working properly. Or we could just eliminate repeating all together.

`````````````````````````````````````
BTW 3FG if you are reading this could you add this JP1.2 version to protocols.ini when you get a chance.

Upgrade protocol 0 = 01 1A (HCS08) Roku (Official) (PB v4.02)
20 16 23 46 21 CF 44 08 08 01 21 03 44 01 21 01
11 D2 DC 11 94 08 CA 00 4E 62 63 33 63 6E 02 AA
CD FF 5F CD FF 92 25 01 81 10 62 11 63 10 A3 CC
FF 5F
End

Your HCS08 executor will be in the next release.

Vicky, shouldn't the Roku PID be 021A? That's what it is in protocols.ini.

Yep is is 02 1A, but I don't have a remote that takes 02 remotes so RM suggested 01 1A as an alternate PID. So when I brought it into Protocol Master and IR to do experimenting, I just kept using 01 1A.


I'm sure Dave will add it to the proper protocol 02 1A in protocols.ini, but I should have specified that in my post.

In the mean time I have looked more closely at the differences between the Harmony and the Roku. I'm not sure how Jim-W did his learns, but this is not what a Harmony Shoots when. The big difference between the two is that even with the shortest push you get frame 1, and frame 2. In the Roku Official protocol the test for repeats required is between frame 1 and frame 2. So quick presses never see frame 2.


On that note, here is a Roku 2 Frame No Repeats version

Upgrade protocol 0 = 01 1A (S3C8+) Roku 2 Frame No Repeats (PB v4.02)
43 8B 21 8B 12 CF 44 08 08 01 21 03 30 01 21 00
FD D2 DC 11 94 08 B6 E4 05 06 60 06 20 11 F6 01
46 46 05 01 56 06 FE 8D 01 46
End

Here is a Roku 2 Frames Before Repeats Test

Upgrade protocol 0 = 01 1A (S3C8+) Roku 2 Frame Before Repeat Test (PB v4.02)
43 8B 21 8B 12 CF 44 08 08 01 21 03 30 01 21 00
FD D2 DC 11 94 08 B6 E4 05 06 60 06 20 11 F6 01
46 46 05 01 56 06 FE 46 29 01 8D 01 46
End

I told Jim I couldn't get to this until next week, which I shouldn't have, but I got down here and my own Roku was giving me fits, because there wasn't a JP1.2 version, so it became a number 1 priority.

Now to get back to the stuff I was supposed to be doing.....

I've tried a 67100 and RCRP05BR both extended and they both do it.

I forgot that I had a Roku 2 IR remote, tried it and it works perfect. This remote is for the newer (4210R) with Faster Processor (2015 Model) Roku 2.

Not sure what I did but this works.

https://www.hifi-remote.com/forums/dload ... e_id=14257

Jim,

Could you check and see if your Roku had a firmware update recently? This is the only thing that I can think of that would explain why the original official Roku protocol didn't work, but now works fine.

I suspect the repeats issue was a bug in the Roku 4 itself that simply got fixed in the past couple of days.