|
Sorry, I think you're going to have to replace this card. It doesn't look like they made any drivers for Windows 2000 or later, and I don't think NT 4.0 drivers were compatible.
In a server I'd go for something cheap and server-oriented. You'd prefer a reliable driver over performance. Our servers have an ATI ES1000 chip, but that's integrated on the motherboard. The cheapest card I can see that would probably suit you is a 64MB Radeon 7000-based card from Sapphire Technologies that costs £17.60, but you have to use ATI's standard unified drivers, possibly including their .NET-based Catalyst Control Centre which is an utter waste of space.
DoEvents: Generating unexpected recursion since 1991
|
|
|
|
|
Ok, I see, thank you very much for your information. I will replace this card. Thank alot for your help.
|
|
|
|
|
http://www.driverfiles.net/Video-Adapters/Trident-Microsystems/Trident-9750-v6.20.35281/download/page,sh,16458,56,5,.html
|
|
|
|
|
Hi,
I am trying to track something down in the TEB I figure If I do a GetThreadContext
The FS Segment register points to it
However like all segment registers it comes back as 16 bit value specfically in my case 0x0038
anyway to convert this to the 32 bit selector value
Thsnkx
|
|
|
|
|
The 16-bit selector value in the register is an index into the global or local descriptor table. Bits 0 and 1 (the lowest two bits) are the Requested Privilege Level; here it's 0. Bit 2 indicates which table to look in, here it's set to 0 meaning the Global Descriptor Table. The remaining 13 bits are the index - the index is multiplied by 8 and added to the base address of the table (the GDTR or LDTR register as appropriate).
The segment descriptor itself then lists the base address of the segment, i.e. the address that segment-relative addresses are relative to.
Windows only uses the FS register. All other segment selectors point to descriptors of the appropriate type whose base is 0 and limit 0xFFFFFFFF, that is, they address the entire contents of memory and addresses are absolute.
x64 effectively disables all segment registers apart from FS and GS. Even then only the base addresses are used; the limits are no longer checked.
You shouldn't rely on anything in the TEB in your code. Use supported structures only (e.g. TlsAlloc for allocating Thread Local Storage, __try/__catch for manipulating exception handlers, CoGetContextToken for COM+ context). If you want to examine the TEB in a debugger, use WinDbg or another debugger from the Debugging Tools for Windows[^] kit and use the !teb command.
DoEvents: Generating unexpected recursion since 1991
|
|
|
|
|
SuspendThread(hThread);
c.ContextFlags = CONTEXT_FULL | CONTEXT_CONTROL;
GetThreadContext(hThread,&c);
GetThreadSelectorEntry(hThread,c.SegFs,(PLDT_ENTRY)&SelEntry);
teb_ptr = ( SelEntry.HighWord.Bits.BaseHi << 24) |
(SelEntry.HighWord.Bits.BaseMid << 16) |
SelEntry.BaseLow; // Fs pointst to TEB
my_ptr = (char *)teb_ptr;
my_ptr = my_ptr + 0x2c; // point to ThreadArray sotrage
my_ptr1 = *my_ptr; // point to first slot
Mike
Think you are right my_ptr1 points to the first threadocalArray
and it 0 Hmmmmmmm ????.
|
|
|
|
|
I'm using winmm.dll to play wmv video. I wanted to find out wmv video Frame Rate and Video Length of any user selected video. I'm wondering what is the API call in winmm will return those values? or any other way to find out?
For avi files (avifil32.dll), AVIFileGetStream will return stream, so stream.CountFrames and stream.FrameRate will return those values.
Thank you in advance.
|
|
|
|
|
I had this incredible idea and i dun know how to implemnt it.
I want to create an external graphics card(i have no idea how to make 1) attached via USB port which can run games(not the current released versions but games made for this graphics card) ppl with sh*tty graphics card would not have to waste money on new cards. Although they will not be able to play new games made to run on Nvidia/Ati graphics card.
The biggest problem i'm facing now is how to display all these 3d calculations of images or models onto the desktop.
|
|
|
|
|
There're USB gfx adapters out there currently. The problem is that USB bandwidth is so low they're limited to IIRC 1024x768x16x60hz and still stutter anytime the USB bus is being hit hard by anything else or the CPU is busy doing something else. Firewire800 might have enough bandwidth to make a semi decent gfx adapter but you'd only be able to sell it to the 3 people who have firewire800 ports. USB3 with it's gigabit optical connection might work (if it doesn't strangle the host CPU) but is several years away.
You know, every time I tried to win a bar-bet about being able to count to 1000 using my fingers I always got punched out when I reached 4....
-- El Corazon
|
|
|
|
|
wow ur knowledge about this stuff amazes me. I want to create a device like PSP with my own graphics i have no idea where to get the components. I know it uses the MIPS microprocessor with ATi/nivida GPU(not sure which one).
Is it possible for an individual to make their own GPU card.
|
|
|
|
|
actually, I just looked at a few reviews of USB gfx adaptors recently hoping I could add a 3rd display to my laptop.
You're on your own with the PSP type device.
You know, every time I tried to win a bar-bet about being able to count to 1000 using my fingers I always got punched out when I reached 4....
-- El Corazon
|
|
|
|
|
dan neely wrote: actually, I just looked at a few reviews of USB gfx adaptors recently hoping I could add a 3rd display to my laptop.
I never even considered looking for such a thing, for second display for my laptop. Its own display is broken, and it costs half the price of a new one to repair, plus ten days minimum downtime. I won't be buying another one for a few months yet, so a USB gfx adapter could be an option.
Semicolons: The number one seller of ostomy bags world wide. - dan neely
|
|
|
|
|
unfortunately they (and thier slightly better performing PC card brethren) have rotten performance and cost a few hundred bucks.
You know, every time I tried to win a bar-bet about being able to count to 1000 using my fingers I always got punched out when I reached 4....
-- El Corazon
|
|
|
|
|
Try this[^] and add your own software.
|
|
|
|
|
OK, dont forget to lie down afterwards.
Morality is indistinguishable from social proscription
|
|
|
|
|
I have a working app I wrote that uses software breakpoints but read a great article on using hardware breakpoints, advantages - non intrusive and speed. Since both aspects were appealing I undertook a conversion. I began porting the dll I had written to asm and implemented the proper routines. However, I am having some very strange problems that have taken me hours to try and figure out and I'm stumped.
Here's the shorthanded basics...
#define MB MessageBox etc etc
while TRUE
{
ContextFlags = CONTEXT_DEBUG_REGISTERS
GetThreadConext
Dr0 = Addey1
Dr1 = Addey 2
Dr7 = 00000000000000000010001100001111b
SetThreadContext
WaitForDebugEvent
if ( dwDebugEventCode == EXIT_PROCESS_DEBUG_EVENT )
{
MB __T("Process exiting")
break
}
if ( dwDebugEventCode == EXCEPTION_DEBUG_EVENT )
{
if ( ExceptionCode == EXCEPTION_BREAKPOINT )
{
if ( ExceptionAddress == Addey1 )
MB __T("Hello from addey 1")
elseif ExceptionAddress == Addey2 )
MB __T("Hello from addey 2")
// here we clear the Dr regs enable single step in FLAGS, back up EIP
// continue & wait for debug event and then xor FLAGS and continue
// pretty much the same as the working dll that uses 0CCh breaks
}
else
{
ContinueDebugEvent - DBG_CONTINUE
continue
}
} // end EXCEPTION_DEBUG_EVENT
ContinueDebugEvent - DBG_EXCEPTION_NOT_HANDLED
} // end while
Now - the results are. If the app is ran the MB saying "Process exiting" pops up. Load it up in Ollydbg and run it, same result. If I set a breakpoint in Olly at the if EXCEPTION_BREAKPOINT, it hits and single stepping in results in trace into kernel32 which then traces into ntdll which returns back to the app and the proper MB "Hello from Addey x" msg as it is designed to do.
What gives? I have thought possibly that Olly is making a Ring0 call therefore can properly handle the DR reg calculations, but that would mean I would have to create a wdm or sys file runnning in Ring0 just to look at hardware breakpoints? Creating a Callgate to ring0 or manipulate SEH isn't feasible as MS could break such code with a patch, not to mention it's just not reliable from machine to machine.
I really do need the speed of the hardware breakpoint as the current dll is fast but not as fast as is desired, not to mention ReadProc and WriteProc 0cch is costly. I would be prepared to write a device driver but it has to also communicate with Ring3 code i reckon using MemMapping would be best but I'm a newb at the current device driver implementations.
Any help on why this code runs when traced in a debugger using single stepping but just exits when run would be most appreciated.
|
|
|
|
|
Wow, no answer on this yet?
Ok the 2 addresses are the program's send and receive addresses. I listen for the incoming/outgoing packet and read a byte sig and filter that by packet type. I send the data from the packet to another window alerting the user of an event in the case they may be semi-afk or in another window if it's the byte sig we want i.e. "Global announcement: there will be a server reboot in 1 hour!". It's been very useful but I want to speed it up by using hardware breakpoints which I also read is recommended over software operation.
I reworked the code and it is ok on recv but on send the program closes immediately. I have read and read and googled and googled and still cannot find a clear answer. Do I have to be in Ring 0 to manipulate the Dx regs and set them for the remote thread, or can this be done in userland?
|
|
|
|
|
Can anyone please point me to an offline one of these? Something that actually detects the configuration instead of having to capture it would be nice icing on that cake.
Semicolons. the number one seller of ostomy bags world wide. - dan neely
|
|
|
|
|
I must start by saying that am very impressed with the good work potrayed by the administrators and contributors of this solution blessing community  . I've gained and learnt a lot from people's problems and solutions just from watching behind. Now its time to join in the fun by posting my own encounter. I BELIEVE in you guys and thats why I am takng this step.
I am a C# developer(starting on C#1.1) and my skill level is between Beginner and Intermediate(though I have been building Client/Server applications recently) but now, I have just been given an advanced real life project to tackle.
A Filling Station manager wants me to build an application that will track/record the sales of fuel(preferably, as entered by the staff) from the electronic fuel meter pump machines. He wants the tracking to be indicated LIVE in the application which is hosted by a computer remotely(in the same station).
1. Is this possible?
2. If it is, what should be the design architecture & development strategy, deployment process and security considerations?
Please, this is urgent! can someone help me... anybody. 
|
|
|
|
|
Hi friends,
Can anybody help me out in getting good documents on blueTooth and EEPROM API list?
I have to do coding for these devices. Kindly Help me out with good documents which helps me in getting the API list and Code samples for the above said other than MSDN.
Thanks in advance..
poojaaa
|
|
|
|
|
i am assuming this is on a custom piece of hardware?
is the EEPROM IIC or SPI interface?
|
|
|
|
|
Is there some way I could measure the total load on the PSU now featuring as chief suspect in my shut-down problem described below? Maybe if it's an overload, and the unit is delivering close to rated current with low load, I could assume it would be subjected to an above rate current draw when the load increases?
Semicolons. the number one seller of ostomy bags world wide. - dan neely
|
|
|
|
|
Well you could use a plug in power meter which sits between the wall outlet and the PSU. Like the Kill A Watt[^]
//Johannes
|
|
|
|
|
If you have access to a variable transformer or a UPS where you can set the output voltage, you can lower the voltage to the PSU until the computer fails and shuts down.
If the shutdown voltage is above the standard -15% of nominal voltage, you want a new PSU.
Keep in mind that this might be a destructive test of the PSU.
|
|
|
|
|
Make sure you read the specsheet on the PSU before making a judgement. Better quality PSUs are designed tolerate larger deviations than cheap ones. Better quality new models can take Japanese 100v, us 110v, European 220V, or anything in between without having to flip a hardware switch.
You know, every time I tried to win a bar-bet about being able to count to 1000 using my fingers I always got punched out when I reached 4....
-- El Corazon
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
