Potential problem with DInput code, and solution
|
View:
New views
5 Messages
—
Rating Filter:
Alert me
|
 |
Potential problem with DInput code, and solution
by Megan Fox
::
Rate this Message:
By default, in ndi8server.h, the buffer size is set as follows:
INPUT_BUFFER_SIZE = 128 ... in our testing, this is by far too low, and misses a great deal of input. By setting the value to 512, all of our lag issues were removed entirely. Another issue I corrected at the same time was the addition of the buffer specification to the Keyboard device code (in di8EnumDevicesCallback): case nDI8Device::KEYBOARD: diDev->SetDataFormat(&c_dfDIKeyboard); di8SetDWordProp(diDev, DIPROP_BUFFERSIZE, nDI8Server::INPUT_BUFFER_SIZE); break; ... before, nothing was setting the BUFFERSIZE for the keyboard device, though this could be redundant for all I know. -- -Megan Fox Idyllon, LLC http://shalinor.circustent.us/megan/resume/ ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV *** NOTE: To reply to the list use "reply to all", *** *** to reply direct to the sender use "reply" *** _______________________________________________ Nebuladevice-discuss mailing list Nebuladevice-discuss@... https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss |
|
|
Re: Potential problem with DInput code, and solution
by pangu
::
Rate this Message:
thanks megan, i commited your patch it to svn.
|
|
|
Re: Potential problem with DInput code, and solution
by Megan Fox
::
Rate this Message:
Note - further testing suggests that adding the DIPROP_BUFFERSIZE to
keyboard causes strange crashing. It works, but for instance, pressing the up arrow is causing an assert wherein Nebula is assuming there are only 128 keys (bizarre). More testing is necessary to figure out what the cause of this is, but removing the PROP line bandaid fixes it. On 10/13/06, pangu <johndotsmith@...> wrote: > > thanks megan, i commited your patch it to svn. > > > Megan Fox wrote: > > > > By default, in ndi8server.h, the buffer size is set as follows: > > > > INPUT_BUFFER_SIZE = 128 > > > > > > ... in our testing, this is by far too low, and misses a great deal of > > input. By setting the value to 512, all of our lag issues were > > removed entirely. > > > > Another issue I corrected at the same time was the addition of the > > buffer specification to the Keyboard device code (in > > di8EnumDevicesCallback): > > > > case nDI8Device::KEYBOARD: > > diDev->SetDataFormat(&c_dfDIKeyboard); > > di8SetDWordProp(diDev, DIPROP_BUFFERSIZE, > > nDI8Server::INPUT_BUFFER_SIZE); > > break; > > > > ... before, nothing was setting the BUFFERSIZE for the keyboard > > device, though this could be redundant for all I know. > > > > -- > > -Megan Fox > > Idyllon, LLC > > http://shalinor.circustent.us/megan/resume/ > > > > ------------------------------------------------------------------------- > > Take Surveys. Earn Cash. Influence the Future of IT > > Join SourceForge.net's Techsay panel and you'll get the chance to share > > your > > opinions on IT & business topics through brief surveys -- and earn cash > > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > > > > *** NOTE: To reply to the list use "reply to all", *** > > *** to reply direct to the sender use "reply" *** > > _______________________________________________ > > Nebuladevice-discuss mailing list > > Nebuladevice-discuss@... > > https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss > > > > > > -- > View this message in context: http://www.nabble.com/Potential-problem-with-DInput-code%2C-and-solution-tf2412838.html#a6807178 > Sent from the nebuladevice-discuss mailing list archive at Nabble.com. > > > ------------------------------------------------------------------------- > Using Tomcat but need to do more? Need to support web services, security? > Get stuff done quickly with pre-integrated technology to make your job easier > Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > > *** NOTE: To reply to the list use "reply to all", *** > *** to reply direct to the sender use "reply" *** > _______________________________________________ > Nebuladevice-discuss mailing list > Nebuladevice-discuss@... > https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss > -- -Megan Fox Idyllon, LLC http://shalinor.circustent.us/megan/resume/ ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 *** NOTE: To reply to the list use "reply to all", *** *** to reply direct to the sender use "reply" *** _______________________________________________ Nebuladevice-discuss mailing list Nebuladevice-discuss@... https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss |
|
|
Re: Potential problem with DInput code, and solution
by Megan Fox
::
Rate this Message:
More on this... I'm not sure why, but this has something to do with
making DInput allow for non-instantaneous button presses. Certain commands, in buffered mode, appear to be flagged as instant (like control commands), and they appear with a higher than normal number. Subtracting numButtons from any incoming input > numButtons works in this case, but I suspect there's a better fix. On 10/20/06, Megan Fox <shalinor@...> wrote: > Note - further testing suggests that adding the DIPROP_BUFFERSIZE to > keyboard causes strange crashing. It works, but for instance, > pressing the up arrow is causing an assert wherein Nebula is assuming > there are only 128 keys (bizarre). > > More testing is necessary to figure out what the cause of this is, but > removing the PROP line bandaid fixes it. > > On 10/13/06, pangu <johndotsmith@...> wrote: > > > > thanks megan, i commited your patch it to svn. > > > > > > Megan Fox wrote: > > > > > > By default, in ndi8server.h, the buffer size is set as follows: > > > > > > INPUT_BUFFER_SIZE = 128 > > > > > > > > > ... in our testing, this is by far too low, and misses a great deal of > > > input. By setting the value to 512, all of our lag issues were > > > removed entirely. > > > > > > Another issue I corrected at the same time was the addition of the > > > buffer specification to the Keyboard device code (in > > > di8EnumDevicesCallback): > > > > > > case nDI8Device::KEYBOARD: > > > diDev->SetDataFormat(&c_dfDIKeyboard); > > > di8SetDWordProp(diDev, DIPROP_BUFFERSIZE, > > > nDI8Server::INPUT_BUFFER_SIZE); > > > break; > > > > > > ... before, nothing was setting the BUFFERSIZE for the keyboard > > > device, though this could be redundant for all I know. > > > > > > -- > > > -Megan Fox > > > Idyllon, LLC > > > http://shalinor.circustent.us/megan/resume/ > > > > > > ------------------------------------------------------------------------- > > > Take Surveys. Earn Cash. Influence the Future of IT > > > Join SourceForge.net's Techsay panel and you'll get the chance to share > > > your > > > opinions on IT & business topics through brief surveys -- and earn cash > > > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > > > > > > *** NOTE: To reply to the list use "reply to all", *** > > > *** to reply direct to the sender use "reply" *** > > > _______________________________________________ > > > Nebuladevice-discuss mailing list > > > Nebuladevice-discuss@... > > > https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss > > > > > > > > > > -- > > View this message in context: http://www.nabble.com/Potential-problem-with-DInput-code%2C-and-solution-tf2412838.html#a6807178 > > Sent from the nebuladevice-discuss mailing list archive at Nabble.com. > > > > > > ------------------------------------------------------------------------- > > Using Tomcat but need to do more? Need to support web services, security? > > Get stuff done quickly with pre-integrated technology to make your job easier > > Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > > > > *** NOTE: To reply to the list use "reply to all", *** > > *** to reply direct to the sender use "reply" *** > > _______________________________________________ > > Nebuladevice-discuss mailing list > > Nebuladevice-discuss@... > > https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss > > > > > -- > -Megan Fox > Idyllon, LLC > http://shalinor.circustent.us/megan/resume/ > -- -Megan Fox Idyllon, LLC http://shalinor.circustent.us/megan/resume/ ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 *** NOTE: To reply to the list use "reply to all", *** *** to reply direct to the sender use "reply" *** _______________________________________________ Nebuladevice-discuss mailing list Nebuladevice-discuss@... https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss |
|
|
Re: Potential problem with DInput code, and solution
by Megan Fox
::
Rate this Message:
Oh, wait, I see...
The solution is to remove the assert from nkeyboarddevice.cc. Instead of: n_assert((btnNum >= 0) && (btnNum < this->numButtons)); Just have: n_assert(btnNum >= 0); I would appreciate someone verifying the safety of this, but it actually works perfectly. No crashes, and all inputs are properly registered - the assert appears to have been in error in this case. On 10/20/06, Megan Fox <shalinor@...> wrote: > More on this... I'm not sure why, but this has something to do with > making DInput allow for non-instantaneous button presses. Certain > commands, in buffered mode, appear to be flagged as instant (like > control commands), and they appear with a higher than normal number. > > Subtracting numButtons from any incoming input > numButtons works in > this case, but I suspect there's a better fix. > > On 10/20/06, Megan Fox <shalinor@...> wrote: > > Note - further testing suggests that adding the DIPROP_BUFFERSIZE to > > keyboard causes strange crashing. It works, but for instance, > > pressing the up arrow is causing an assert wherein Nebula is assuming > > there are only 128 keys (bizarre). > > > > More testing is necessary to figure out what the cause of this is, but > > removing the PROP line bandaid fixes it. > > > > On 10/13/06, pangu <johndotsmith@...> wrote: > > > > > > thanks megan, i commited your patch it to svn. > > > > > > > > > Megan Fox wrote: > > > > > > > > By default, in ndi8server.h, the buffer size is set as follows: > > > > > > > > INPUT_BUFFER_SIZE = 128 > > > > > > > > > > > > ... in our testing, this is by far too low, and misses a great deal of > > > > input. By setting the value to 512, all of our lag issues were > > > > removed entirely. > > > > > > > > Another issue I corrected at the same time was the addition of the > > > > buffer specification to the Keyboard device code (in > > > > di8EnumDevicesCallback): > > > > > > > > case nDI8Device::KEYBOARD: > > > > diDev->SetDataFormat(&c_dfDIKeyboard); > > > > di8SetDWordProp(diDev, DIPROP_BUFFERSIZE, > > > > nDI8Server::INPUT_BUFFER_SIZE); > > > > break; > > > > > > > > ... before, nothing was setting the BUFFERSIZE for the keyboard > > > > device, though this could be redundant for all I know. > > > > > > > > -- > > > > -Megan Fox > > > > Idyllon, LLC > > > > http://shalinor.circustent.us/megan/resume/ > > > > > > > > ------------------------------------------------------------------------- > > > > Take Surveys. Earn Cash. Influence the Future of IT > > > > Join SourceForge.net's Techsay panel and you'll get the chance to share > > > > your > > > > opinions on IT & business topics through brief surveys -- and earn cash > > > > http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV > > > > > > > > *** NOTE: To reply to the list use "reply to all", *** > > > > *** to reply direct to the sender use "reply" *** > > > > _______________________________________________ > > > > Nebuladevice-discuss mailing list > > > > Nebuladevice-discuss@... > > > > https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss > > > > > > > > > > > > > > -- > > > View this message in context: http://www.nabble.com/Potential-problem-with-DInput-code%2C-and-solution-tf2412838.html#a6807178 > > > Sent from the nebuladevice-discuss mailing list archive at Nabble.com. > > > > > > > > > ------------------------------------------------------------------------- > > > Using Tomcat but need to do more? Need to support web services, security? > > > Get stuff done quickly with pre-integrated technology to make your job easier > > > Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo > > > http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 > > > > > > *** NOTE: To reply to the list use "reply to all", *** > > > *** to reply direct to the sender use "reply" *** > > > _______________________________________________ > > > Nebuladevice-discuss mailing list > > > Nebuladevice-discuss@... > > > https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss > > > > > > > > > -- > > -Megan Fox > > Idyllon, LLC > > http://shalinor.circustent.us/megan/resume/ > > > > > -- > -Megan Fox > Idyllon, LLC > http://shalinor.circustent.us/megan/resume/ > -- -Megan Fox Idyllon, LLC http://shalinor.circustent.us/megan/resume/ ------------------------------------------------------------------------- Using Tomcat but need to do more? Need to support web services, security? Get stuff done quickly with pre-integrated technology to make your job easier Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642 *** NOTE: To reply to the list use "reply to all", *** *** to reply direct to the sender use "reply" *** _______________________________________________ Nebuladevice-discuss mailing list Nebuladevice-discuss@... https://lists.sourceforge.net/lists/listinfo/nebuladevice-discuss |
| Free embeddable forum powered by Nabble | Forum Help |
