tpaxadpom wrote:cauer29 wrote:tpaxadpom wrote:
A.A. I just found the original remote for Samsung OTA tuner, installed batteries and tested. Long keypress doesn't send multiple commands in sequence(no duplicates). I'm pretty positive that is 1B5 protocol that duplicates those commands.
Well, then there must've been a change in production with respect to the behavior of the remote or STB, since my original remote transmits the identical signal as the 15-100 with the 1B5 protocol. Ideally you need a widget or some other method of recording precisely what your original remote is sending, in order to sort out what is different between the 1B5 protocol and your original remote. Since you're not able to download from the remote with learned cmds, we can't find any clues there either.
If you're happy with the manual settings protocol, then there's no point in pursuing this. I may download your IR file with the manual settings and compare the output between that and the 1B5 protocol, to see what might be different anyway. Perhaps the manual settings protocol doesn't repeat at all? That is, you should be able to press and hold ch+ or the arrow keys and get repeating. If you don't, then that's why the manual settings protocol never causes unintended repeats. The original remote that came with my Samsung STB certainly repeats for intended long key presses and so does the 15-100 with 1B5 protocol.
A.A.
I can press ch+ and arrow keys and I get repeated commands send out with Manual settings protocol. Pressing and holding any numeric keys doesn't do that (just like the original remote).
Dredging up this thread now that I've discovered the reason why the 1B5 Samsung36 protocol behaves differently than the original Samsung remote on the H260F STB. It appears that Samsung have a 60ms leadout cutoff. That is, if the off portion of the leadout is longer than 60ms, then it assumes that any further detected signals that are the same as the previous, are new keypresses. If the off time of the leadout is less than 60ms, then it assumes that it's a long keypress. For keys that are expected to repeat (ch+, arrow keys, etc), the STB acts on the first one and then waits a few hundred milliseconds before repeating, as long as there is never a leadout with off time greater than 60ms. If there is a leadout off time greater than 60ms, then it assumes a new keypress and acts immediately. This frequently leads to repeated actions when not wanted.
I can press the right arrow key with a short tap and get just one right arrow action, but if I just do a normal keypress, the 1B5 Samsung protocol repeats with a leadout off time greater than 60ms and I get 2 or 3 right arrow actions. This is especially frustrating when you're trying to enter numeric values. Instead of '1', you get 11 or 111. Shortening the leadout off time to less than 60ms, fixes it and then you can hold the 1 key down as long as you want and the STB dutifully ignores all the repeats since they're due to a long keypress and nobody is really expecting the 1 key to repeat in action.
I see some descriptions of the Samsung36 protocol saying that the leadout off time is 59154, but this is apparently not what is actually implemented in the 1B5 Samsung protocol, since I always get numbers more like 61136. I suppose that this could be due to some additional overhead. In any case, the 1B5 Samsung protocol as it exists does not work correctly with the Samsung STB. I expect this behavior extends to other Samsung equipment too. I have an ancient Samsung BDP1000 around here somehwere that also uses the same protocol. I'll see if that one is also afflicted in the same way.
I briefly reviewed the assembler listing for the 1B5 protocol and nothing jumped off the page at me as corresponding to the leadout off time. So, I don't know what to change to trim the leadout off time.
Any experts willing to tweak the 1B5 protocol (8910 + 15-100).
A.A.
