Guys,

I got a few questions that might be interesting to others, so here is my reply:

I’ve heard two people that have to hold the button for halve a second, maybe it’s a BD thing, because on my DVD I don’t have this problem. The only thing I can do is remove the forced screen update that locks the remote for a few ms, however this will cause a small delay between the poll and displaying of the result so the time might be 1 second later on the display of the player then on the remote. Now it’s mostly at the same time, it’s only a second later when your polling the player on for example 01h30m03s900ms. When polling at 01h30m03s000ms -01h30m03s800ms the time is right.

The player is quite slow, I ment that using my module should be nearly just as fast as operating normally (IR). The actual code runs within 1ms. The rest of the delays are caused by waiting, TSU, RFX, RS232 and network delay.

It's roughly like this (My debug module logs the time in ms):

Building the command <1ms
Network delay <20ms
RS232 command send >40ms
Player starts executing command >20ms (the actual executing takes more time)
Player sends reply >40ms
RS232 reply send >40ms
Network delay <20ms
Parsing the reply <1ms
ScreenUpdate <4ms
Waiting to execute next command 250 – sum of above

All my actual code runs in less then 1ms. The rest is waiting, network delay and rs232 delay, and player delay. I can only improve the waiting. If I make the scheduling dynamic the commandrate per second will increase from 4 (250ms) to 5(200ms)-6(166ms). If I can manage to send a command while the player is giving an answer it could be increased to 10(100ms)-12(83ms). I'm just not sure if the player will accept commands while it's replying to the previous command, this is something I need to test when 1.39 is send out.
Please note that when using multiple remotes the commandrate will be even lower, 50% for two, 33% for three, etc.

The actual executing of the player after receiving a command is something I can't do anything about.

IP control will be an extra option, so it will not replace rs232.
IP control is very nice, the average response time is around 10ms when using only one remote, which is equal to my ping time from pc to remote. So it's roughly 18 times faster then using rs232. Also there is no need to poll every second, all changes will be send back. So that’s 1 more command per second free. However the actual executing part of the player is equal to rs232.
And when using multiple remotes the socket needs to be closed, reopened and a poll needs to be executed. Then it probably will only be 2 times faster then RS232 single remote, and 4 times faster then rs232 with two remotes.

The reason why a Denon Receiver and Player on the same rs232 extender won’t be supported is because an extender can do only one thing at a time. So let’s show what would happen:

Polling DVD status 166ms
Polling Receiver Volume 166ms
Polling Receiver Input 166ms
Polling Receiver Surround Mode 166ms
Polling Receiver Mute 166ms
Polling Receiver Power 166ms

Sum of above = 996ms. So the actual commandrate left for sending command would be 0.

Actually I’m thinking of dropping receiver rs232. There is almost no point in using it, you would need a dedicated extender and even then it will be slow as hell. Besides that I’m hoping that everyone that has the money for a TSU also has a AVR4306, AVR3808 or higher.