Ok, I thought I'd rather not wait to finish this. Pls if you got LPT-reload from the avsim library ... delete the exe file from the pack and replace it with this new one.There were a few BIG issues and while correcting I even noticed that a part of the code was unfinished :$ Not to worry, the part that was unfinished involved LPT2 and LPT3 and I don't think anyone has ever yet tried to use that option anyway.Hope it's 90% bug-free this time! :(I added the option to change timer speed.Timer speed is by default set to read the matrix once every second (1000/1000th of a second). You can increase speed (by making the value in the box smaler !) to as low as 1 (which means 1/1000th of a second) if you want and if your computer will go that fast. I have no idea as to how this will affect FS on your machines.Be advised that THIS option will have to be reset each time you run the program (for now) I kinda forgot to add it to the "save to ini file" section in the code but I will some day.Anyway, the option is in there, you got a visual as to see how fast it goos and the bug is out of the program so no more "toggelin everything once i go in fs" situations. For now this should do fine :)ENJOY !http://users.skynet.be/plipke/PF50HC/LPTsw...PTreload_v2.zip

guys, pls don't miss out on this post aight :(cuz the other one with the issues is more replied to than this one :(make sure people get the update cuz it is only here i posted the url !

philippefirst of all thanks for your great job.I've tested the new release. It works but it still has some problems. Sometimes it doesn't work, when you change the scanning speed it does not have any effect, and the speed of the blinking red square does not change. It still show some kind of memory mishandling: when it starts it use a huge amount of memory, around 16Mb, used memory increase each time you do something (even a click to change folder), sometimes in test mode used memory continuosly increase at a very high rate, finally when you push "RUN" memory usage fall and stabilize at about 4-5Mb. I tried to select highest speed (1) and tried a sort of rotary switch (a hacked rotary selctor): lptswitch misses about 50% of inputs. So I think I should surrender and buy one of those expensive things :( :(Your project however is great for normal push switches!!bye

I had another idea. Why don't you give the user the ability to exclude from the scanning the rows in wich there are no switches connected? This will boost scanning speed. As an example, if I only use 10 switches, there are 3 rows without any switche connected, so why wasting time continuing to scan 3 empty rows? if you exclude them from the scanning the speed will more than doubble!bye ;)

It would seem logical that the people who use this might well use a vast majority of the rows available. But it's a good idea, although it does add to complexity of the code :)Easy if you're using C++, but I gather Philippe is using VB :-erks

>It would seem logical that the people who use this might well>use a vast majority of the rows available. But it's a good>idea, although it does add to complexity of the code :)>>Easy if you're using C++, but I gather Philippe is using VB>:-erksI've created a similar keyboard emulation application in VB and I've scanned 512 inputs every 10ms with no overruns (tons of time to spare). Scanning even a thousand inputs would be fast on my K6-2 500Mhz. The problem lies elsewhere and Philippe might need to streamline the scanning loops perhaps. Hard to say without seeing the code, but I experienced the same problem on my first go at it with my app. It ran like a fat cow and I immediately blamed VB. ;)-Leo

leo and chewI have a fair knowledge o c++, I've studied it at uni, but I was never able to make programs that interfaces with windows, they only teached to me to write programs that run on those dark dos windows! Can you give me some links where I can read about writing programs that interact with windows? I would really try to write something like lpt-switch.I'm now typing from the keyboard I've just finished to hack. I used a flat ribon cable to bring all 8+16 contacts out of the keyboard and you can continue to use it for normal typing. I still don't have knitter switches but I'm tryng to simulate them in a very simple way generating approx 20 on/off cicles in one second. It seems that the keyboard works pretty well and doesn't miss many inputs. I'm just at the beginning however and will keep you informed of any progress.byep.s.leodoes gammaray uses some kind of buffer between rotaries and the interface to avoid losing inputs?

Io Leo !You know VB.Net ? If you got VS.Net and you would like to have a look at my code, you can. Send me a mail to [email protected] and I'll send you a part of the code.

Trallucio,What you need is a good introduction to the Win32 API. Charles Petzold has written what is arguably the best. The book is Programming Windows. It's gone through several editions. I think it's called something like Programing .Net these days. Another option is the book Windows 98 Programming by Herbert Schildt. The basics of talking to Win32 are the same in 98 as in XP.Either of these two will give you a good working introduction. If you want to look at some of the higher level services, try Win32 System Services by Marshall Brain and Ron Reeves. There is also the MSDN library on the Microsoft site. msdn.microsoft.com The information is there but not nearly so nicely organized. It's much better as the definitive reference when you have specific questions.Mikewww.mikesflightdeck.com

Philippe wrote:"You know VB.Net ? If you got VS.Net and you would like to have a look at my code, you can. Send me a mail to [email protected] and I'll send you a part of the code."Sadly I'm an

VC++ does have window generators, but it's much more practical to write code to generate your own. The whole code to make a window only takes a minute to write, and is so much better and streamlined too.For those who know rudimentary C++, try the #define WIN32_LEAN_AND_MEAN to make a compact but efficient window program, after that you can write whatever code you like, but you must write window definition code so Windows knows what to display! I'm happy to give pointers on this if you need help, just ask :)

mike, leo, ceawthank you very much for your advise, yes what I really need is probably a good book to start, possibly the best book, something that is to win32 what the great "kernigam & ritche" is to c! I will give a look to those you suggested.mikeI found this link you may add to yor switch page: http://www.stacoswitch.com/leoyes what I'm looking for is precision in replicating ap inputs. As I said I'm looking at a reliable source for ifr and mcc selftraining rather than good looking mcps and panels. So when I'm flying a sid with a fast jet I really need to be able to insert those heading and altitude with just a coupple of rotary twist.

Here's a quick tip:Most of the programming code in an application based on C++ is merely normal C++ and not Windows propriety code. The only things that are specific to Windows are those that input and output to the Windows interface. Everything else is just plain ordinary C/C++.To give an idea, try writing this (or copy and paste) in your C++ compiler and see what you get :)/////START!/////#define WIN32_LEAN_AND_MEAN#include #include int WINAPI WinMain(HINSTANCE hinstance, HINSTANCE hprevinstance, LPSTR lpcmdline, int ncmdshow){MessageBox(NULL, "Hello, World!", "My first Windows Program!", MB_OK | MB_ICONEXCLAMATION);return(0);}/////END!/////As you can see, instead of using Main, you use WinMain which tells the compiler that the program is for Windows and not DOS. But when making a project from VS, don't select console application if you're writing for Windows, it gets confused :)If you have any questions about this, please ask :) I'll be here all day

This is here to reset the text type :)

Trallucio,Thank you for the link.Mikewww.mikesflightdeck.com