NextPVR Forums

Guys,

I am having difficulty in getting my machine to wake up from hibernate using my mce 2005 remote. I have a mce 2005 OS running on a asus p5p800 motherboard. Note: I have disabled the mce receiver and scheduling services in MCE to allow GBPVR to run.

The first thing I did was review the motherboard manual (yes the manual) and I changed the jumpers USBPW12 and USBPW34 to +5VSB to enable wake up from S3 and S4 sleep modes (for rear USB ports). The manual showed the jumpers as 2x2 jumper block however I was confused as to the orientation (up and down or left or right). I assumed the jumpers were oriented in the same direction as default setting (+5V).

That is, I configured the jumpers as below

3 | |
2 | |
1 . .

Is my assumption correct?

What is "+5VSB" anyway?

Anyway this didnt seem to work!

I then went to ACPI configuration in BIOS and changed the suspend mode from Auto to S3 only. I also changed ACPI 2.0 support from No to Yes. No luck here.

I then went to the APM configuration page in BIOS. This didnt seem to help as although it allowed you to enable wake from RTC alarm, external modem, pci devices, ps2 keyboard, ps2 mouse, no such option existed for usb. Did I miss something?

I then searched the forums and plenty of guys indicating it does work but none describing how its done. I found one thread that suggested a change registry entries but I was unwilling to do this as I was unable to confirm this through a another source (thread). Is there a thread somewhere that I have missed?

Your help will be greatly appreciated!

Guys I got it working. I found this thread to be helpful http://thegreenbutton.com/forums/thread/29899.aspx

It now appears to work OK in both S3 standby and S4 hibernate modes.

mvandere Wrote:+5vsb is +5 Volts StandBy which is 5 volts which is there even when the PC is powerd off.

I have never been able to get this to work on my computer, and if you think about it it isn't really possible, despite what some people may say.

When you press the button on your remote what does the receiver receive? A series of 0s and 1s. How is that series of 0s and 1s decoded into something meaningful? By the computer. So if the computer is turned off / in hibernate / standby. How can it work out that you have pressed a button? By the changing 1s and 0s. What does the sun and Fluro lights present to an IR decoder all the time? Changing 1s and 0s.

So you can see, even if it was possible, your computer would be coming out of power off / hibernate / standby all the time.

Another reason it can't work is because then remotes that specifically 'press the power button' on your computer wouldn't be needed. eg The Compro remote/card and the Imon remote/card.

Of course I could be wrong

Sorry but ou are totally wrong this time. :eek:

Yes this does work as I use it virtually every day. Also only the code sent by the power key on the remote will wake the computer, no other key will. I don't know the lowest level specifics but is it any harder to think that a remote can wake the computer than it is that yor keyboard or mouse can wake it? They also send a series of 1's and 0's. I even have some a couple motherboards that you can set a password to wake the computer from the keyboard. So, for example, I can type O N and the computer wakes, and it only wakes with O N, nothing else. What you are forgetting is that most of the decoding is done in hardware that is part of the receiver in all three cases. Then that hardware can issue the appropriate command to wake the computer, probably in the form of a NMI (nonmaskable interrupt).

mvandere Wrote:Well, not quite, it doesn't work for me and thousands of others, maybe 50% wrong :p

Thousands of others? ANyway you didn't state that it doesn't work for some people and some hardware you said "if you think about it it isn't really possible" , which is really wrong.

If something doesn't work on your combination of hardware hardly means it is impossible that it works at all. Mine works perfectly with a MCE 2005 USB receiver connected to a USB port on my machine.

mvandere Wrote:My 'receiver' is simply a USB device, there is no inteligence in it it simply passes the data onto the USB bus. When powered down the 'acknowledge' LED does not flash

USB client devices actually have alot of intelligence. They have to be able to identify themselves uniquely as well as negotiate speed and power requirements. Most of this is contained in a single chip that handles the USB part of the interface but it can also be handled by a microcontroller. As far as the LED see below.

mvandere Wrote:As I said previously, Sunlight and CFL (Compact Flourescent Lamps) generate IR 0s and 1s, so with the available inteligence, if it worked, Power up would be random / spontaneous. (and my motherboard says it supports 'Wake on USB', but it does not work from my USB keyboard or mouse either, it does work from my PS2 mouse and keyboard)

No the "available intelligence" is what stops random streams of 1's and 0's from be recognized as anything useful, if there were random strings of 1's and 0's, see below. Otherwise just having sunlight or CFL's in the room would cause all sort of mayhem. When your machine is on and the IR is receiving does it spotanously do things from these random 1's and 0's?

As far as the wake on USB setting, is it also set in windows to allow the device and any interviening hub to wake the computer? You can set the bios to allow wake on lan but if windows is not configured to allow the network card to wake it the card will not be left in the proper state to allow it to wake the computer. They same thing applies to the USB support, the hardware has to be left in the proper state to allow it to wake the computer.

mvandere Wrote:A keyboard controller can be powered up at all times and 'programmed' to do something different with different codes

So can a USB device. In the case of the MCE ir receiver it doesn't need to be programed to wake the computer, that is built in.

mvandere Wrote:The 0s and 1s from a keyboard/mouse are not subject to noise from sunlight /CFLs

I have IR wireless keyboards with built in pointer sticks that wake the computer just fine, even with sunlight present.

mvandere Wrote:USB supports a protocol, not a specific device, therefore the smarts cannot be put in hardware as the hardware cannot anticipate what the device may or may not 'mean' (unless there is some sort of industry agreed 'code', eg 16 successive 1's, which your remote might generate and my devices might not)

Yes USB is a bus with a specific communications protocol. That is handled by the USB bus interface chip or by code in the client device's microcontroller. As far as what specifically the MCE ir receiver sends the computer to wake it, I have no idea, but I imagine that is was defined for just that purpose, just like there is a keyboard scan code defined that will wake the computer.

Think about how this works for a minute.

My {wireless keyboard} can send a stream of 1's and 0's to its receiver via IR. The receiver, powered from the computers {PS/2} port converts this into the appropriate stream of bits and sends it over the serial {PS/2} port using the defined {PS/2} serial protocol to the {PS/2} interface chip inside the computer that then wakes the computer.

Now change the following:

{wireless keyboard} = MCE remote
{PS/2} = USB

A couple of notes.

First, the problem with too much sunlight or a strong CFL or Plasma screen (which is another florescent light source) is not that they are sending a stream of 1's and 0's it is that they are overloading the receiver's sensor with ir. It is exactly like someone blinking morse code to you with a flashlight and suddenly a car parks behind them and turns on its high beams. You are not going to confuse the car high beams for morse code but you can no longer see the kight from the flashlight so you are stuck. The data sent via ir is encoded on a carrier frequency that lets the receiving sensor separate valid data from background noise for just this purpose. As long as the receiver is not oerloaded to the point that it cannot receeve the signal it will be able to decode the data. That is why the remote works when the computer is on even with sunlight or CFLs present.

Second considering the number of combinations of hardware that can make up a computer it is amazing that as much works as does. Look at how many people have problems with S3 suspend (so called suspend to ram) mode and that has been around longer than USB or wake from a USB device. If you have double checked that the motherboard jumpers apply 5vsb power to the USB port where the receiver is connected and that windows is configured to allow wake from USB, as well as your BIOS, and it still doesn't work complain to your motherboard manufactuer. It might be something a new BIOS version could fix or you may just need a new motherboard.

Just don't go around saying it is impossible since it works for me every day.

PS. Allow for spelling or typos, it is 2:25 am here and I have to get up soon.

Not upset, just didn't want anyone else to think it can't work. Like you said, it all comes down to the hardware and how well things are implemented.

When I first set up my pvr I was using a 1.4 rev of my same board and had several problems including S3 suspend not working. Since I also had a 2.0 board in another machine I switched and everything works great now. You should look for a cheap 2.0 board, it makes a great pvr with gb-pvr.

Mine is a non-X Deluxe, I have always had problems deciphering Asus' suffixes as to what changed. There is also an -E model in Deluxe and non-Deluxe as well as a few others that are usually smaller versions with built in video or such.

This is like I said above, it is a wonder that anything works given all the different permutations of hardware and software the average pc compatible can have.

I think half of the stability always attributed to Mac's had to do with the limited hardware that was allowed. It should be interesting to see if the new intel based Mac's are as stable since they are still strictly controlled as to hardware, unlike pc's.