New EMU is finally complete (could work on dream with some work)

Why because 96% of code is the same. never told any one to buy itagte easiest thing for us to do would be roll out on itgate box. We could have just kept quiet as there is a working but dirty solution wait for nagra 2 then laugh ma ass of but no we did not. said Linux box deliberately not trying to blow itgates trumpet.


I would of though its actually a higher proprtion than 96% that would be common. At a guess I'd say the only differences are going to be the devices and associated drivers and perhaps the way you interpret certain special files passed from the main box application.

I'm unsure as to what the hardware differences are on the Itgate boxes so I cant really say a lot more. I do suspect that it will involve messing with the source code to make it work on a dreambox though so if none of the authors actually have a dreambox then I cant see it ever getting ported.
 
I would of though its actually a higher proprtion than 96% that would be common. At a guess I'd say the only differences are going to be the devices and associated drivers and perhaps the way you interpret certain special files passed from the main box application.

I'm unsure as to what the hardware differences are on the Itgate boxes so I cant really say a lot more. I do suspect that it will involve messing with the source code to make it work on a dreambox though so if none of the authors actually have a dreambox then I cant see it ever getting ported.

there should be a new thread set up,specifically to talk about getting hypercam to work on the dreambox.
 
oh well i must say this is quite dissappointing, it would appear that the linux comunity is not at all the great fellowship of open source it pretends to be. At a source level you got to believe that a dreambox porting of this emulator would have been barely noticeable in terms of effort.

thus far the itgate team has benefitted from all the help given to the dbox and dreambox comunity, now it seems that there has been an intentional effort to set this itgate linux box apart from such boxes by closing the source and making no effort to port the emulator by those in possesion of the source code.. Infact taking that a step further one is lead to suspect that efforts have been made to make it not possible for aforementioned emu to be ported by way of design.

as for this seemingly endless balls about UK n1 - what exactly is the difference between uk nagra1 and any other nagra1. afaik there is simply one nagra 1 and thats that.
as for nagra2 - it doesn't matter how good your emulator is - without a keyset you aint gonna be using it period. and lets face it there aint no world shortage of nagra2 stuff knocking about.

i wonder is it thus that no other linux cam will run on an itgate box? if not then why would this cam not run on a linux box with help of course from the itgate community.
also i wouldn't bother going to the effort of even trying to get an as yet untested cam to work on other boxes - WHY? simply because it has had no baptism of fire yet - lets face it it autorolls now with a unmodified image there is no evidence that other linux cams will not roll when that time comes for a change. there is also no evidence the Hypercam will, and if you think a rom couldn't be patched before hypercam club even knew what the problems was - well i wouldn't wanna be the one placing that wager.


In short i am dissappointed by the outcome of this thread..


TBC
 
from what i have read it doesnt have the fixed roms it has its own roms,hope this helps
 
@TBC would this emu work without the 'fixed' roms,or is there another set of fixed roms from another party in it?


afaik it has no fixed rom in it, it shouldn't need it after all if one has the source for such a thing then it is fixable and wouldn't take 10 days.

there is always talk of a dirty fix? well i prefer after market repair, nagra do it thats what the bug table for lol.

the patches applied to the dreambox images should not impare its normal emu function. they are made specifically that way as are all patches.
so if something the emu is not unprepared for arises then it will deal with it correctly, if it doesn't then it has to be fixed. make no mistake what is claimed of this hypercam while nice in that the emu could be fixed because the source is held be the developers. what this thread shows is that nothing has changed for the linux comunity. the source is still held by those that think they are the high and mighty and will doubtless if broken need fixing.
so how much better is hypercam?? only time will tell, personally i wouldn't lose and sleep over what will or won't work.
only time this hobby affords any pleasure is when stuff gets broken and needs fixing lol.

tbc
 
If anyone wants to play with Softcams i've uploaded a source for the US rqcamd. This is specifically a Nagra2 emulator but many of the techniques used are very applicable to Nagra1, as is the methods of communicating with hardware etc.

https://www.digitalworldz.co.uk/./index.php?resources/2378/


nb. The file is a gzipped linux tarball. Download the file and then open with WinRar. Dont try to download directly into winrar or winzip.

You can also have to open the inner tarball with Winrar
 
Last edited by a moderator:
If anyone wants to play with Softcams i've uploaded a source for the US rqcamd. This is specifically a Nagra2 emulator but many of the techniques used are very applicable to Nagra1, as is the methods of communicating with hardware etc.

https://www.digitalworldz.co.uk/./index.php?resources/2378/


nb. The file is a gzipped linux tarball. Download the file and then open with WinRar. Dont try to download directly into winrar or winzip.

You can also have to open the inner tarball with Winrar

I have the source-code for sasc-ng - another Linux cam that is used by MythTV. It contains code for nagra1 and nagra2. Where should I upload it?

abaddon
 
Last edited by a moderator:
Just for completeness, the other up-to-date opensource emulators are : -

opencam: runs on dreambox, only supports n.a. n2
vdr-sc: runs on linux, supports most systems
emunation: run on windows, supports most systems ( based on vdr-sc)

Personally, I think the rom file fix is better than this new emu since it allow
all the current proven emulators to continue working . The only reason a
new emu would be better for n1 is if the AU keyroll method was more complicated
than a par of 1-bit changes to the key. Any even then it would need to be a lot more complicated.


obwan
 
Put it in the same downloads section that I put the rqcamd source in

https://www.digitalworldz.co.uk/downloads.php?do=cat&id=34

I have uploaded the code at the above location.

I made the comment that the README is interesting. This cam was based on the original mgcam! So, with a bit of effort and understanding we should be able to find a way to put this implementation of nagra1 into mgcam. I have the source code for mgcamd somewhere - it's old source code, but should be a good start point.

The only problem will be time. I think some good time will be required to do this. I will do my best, but cannot dedicate too much time to this excercise. It interests me, so I will make a start. Any input from anybody else would be much appreciated.

abaddon
 
Last edited by a moderator:
I have had a chance to sleep on this and reckon the following is the approach to take:

1. Try and get hypercam (as posted in this thread) working for the dbox (I have a dbox but no dreambox).

2. If the above fails, feedback to trackermut and see if his team are able to rebuild for us.

3. If the above fails, see if we can get hold of the source from trackermuts team and build for ourselves.

4. If the above fails, look at attempting to incorporate the implementation of nagra1 into old mgcamd source.

There is always the option to do nothing and just pray someone can hack the ROM's when we need it done! There's no fun in that though!

Any thoughts anyone?

abaddon
 
or, you could always leave the emus alone, and read stuntguys nagra faq, learn some assembly language and attempt some rom patching when required, as long as you have access to the raw key roll emss logged from the stream.
If the roms can be patched as and when required, there may be no need to mess around with the emus and source code etc..
 
It is usually possible, one way or another to patch the roms BUT I suspect Kudelski could find some ingenious keyroll methods that would be quite difficult to patch against if they were so inclined. The initial keyroll method that was used a few years ago (DES encrypt of the transmitted data using partts of the ROM image itself as the DES key) could easily be extended to make image patching difficult.

The best solution has to always be to update the actual emulator to make it act in a more realistic manner. The better your realism then the more likely your emulator will be immune from the next attack.
 
4. If the above fails, look at attempting to incorporate the implementation of nagra1 into old mgcamd source.

Depends on how old your source is but most of the public sources dont really have a worthwhile emulator. Most of them attempt to handle keyrolls in a very similar method to that used on Funcards/ATMega's.

If you are going to play with this stuff then it really needs a full 6805/ST7 enhanced core emulator capable of fully emulating the Rom7/10/11 card environments.

Code like Emunation does have a core emulator but I suspect it will need considerable rework before it could be used with the kind of keyrolls that have recently been used by VM/UPC
 
Depends on how old your source is but most of the public sources dont really have a worthwhile emulator. Most of them attempt to handle keyrolls in a very similar method to that used on Funcards/ATMega's.

If you are going to play with this stuff then it really needs a full 6805/ST7 enhanced core emulator capable of fully emulating the Rom7/10/11 card environments.

Code like Emunation does have a core emulator but I suspect it will need considerable rework before it could be used with the kind of keyrolls that have recently been used by VM/UPC

I two versions of the mgcamd source code - one from around Feb/Mar 2002 and one that had edits to some files in 2003 specifically for the dbox - it's the latter I am looking at. Are you saying this is a waste of time? Or do you think the combo of the vdr-sc code should be enough to fill in the blanks?

I spent my lunch hour trawling through the old mgcamd source and reckon I know what it is trying to do but now am worried I could be wasting my time! The source I have builds two executables that are relevant to the dbox - "emm" and "mgcamd". I think the "mgcamd" here does not do AU, but the code for "emm" has the stuff that calls Nagra_GetKeys and updates the SoftCam.Key file. Doesn't look like a big stretch to put the emm stuff in mgcamd and get it AU'ing.

One problem I have found with this code is when it is using the contents of AutoRoll.Key - this file has keys for N1, E1, V and N2 - there are two versions of N2, the second being specifically for ROM10. From what I see, this version only ever uses the first "N2" and never uses the N2 key for ROM10 (labelled "N2a" in the file). Is this the flaw with the dbox cams? Or was this fixed years ago and I need something more complex to be incorporated?

Combining emm and mgcamd is do-able for me in a relatively short time. Adding the vdr-sc stuff to this looks like a much more complex task.

I have got the mgcamd source to compile, so will have a play when I get home. See how old/bad this is before I proceed.

Thanks for the feedback. I will be back for more help!

abaddon
 
The general maths routines and interface back to the main program are fine but the key extraction will likely be hard coded so the emulator will be non existant.

Have a look at the 6805 emulation core in Emunation. This is an old version of a Rom7/10/11/102/103 core emulator. It will run card code at the instruction level but would need a fair bit of tweaking in the UK.

If your cam sources dont contain a similar module then they dont contain a true emulator.
 
Back
Top