brydling

After some more testing of the two utilities it became obvious to me that none of them allow you to adjust curves anyway :P The only thing you can do is change calibration settings. They should be able to set deadzones too, but it doesn't work on WinXP and later. DirectInput issue it says.. So, anyone knows a utility that actually let you adjust curves? /Niclas

Just tested DIView at my main computer. It worked great there. I have no idea why it didn't work on my laptop. I tried DXTweak2 also, it didn't work as good. When you press "Next" and get to the first virtual joystick of the B240A1310 (the one with eight axes), it crashes. Both 32- and 64-bit versions of the program crashed (tested on Win7/64). Maybe the program can't handle controllers with eight axes? The second virtual joystick shows up fine, and that one only has five axes. Anyone with a BU0836 or other eight-axis controller that would like to try it out? /Niclas

Thanks! Interesting! Thank you for testing that. Would you mind e-mailing your version of DIView to me so that I can make sure we have the same version? I will test at my main computer today where I have my HOTAS:WH and see if that one shows up. My e-mail adress is in my profile. /Niclas

No problem! I know you can adjust curves in FSX, but maybe you need FSUIPC for it. /Niclas

Just tested DIView. The joystick did not appear, so I guess that won't work for setting curves and deadzones. Don't know if it's a Win7 issue. The program is from 2004 so that seems possible. I have no other joystick to test with at this computer. /Niclas

Thank you! One more from Sweden :) So far all but one of the guys that have signed up are from Sweden. Makes shipping a little easier :) /Niclas

Thanks! Thank you! I have thought of adding such things, but I haven't figured out why someone would want to do that. Therefore it hasn't been done. Is it because there are simulators that won't let you adjust curves? If a lot of people is interested in it maybe I could add it to the software, but it is a lot of work for one guy :) Also, I think there is a program called DIView that will let you do it on the PC side. It won't be stored in the controller hovewer, only on that machine. Two things I will probably add is axis inversion and rotary encoder support. /Niclas

Thanks! I calculated a bit before and got 94 USD + shipping as a DIY-kit. I've been told that a colleague at work have a soldering machine at home, so assembled cards may be available. I don't know about that yet though. I am not sure he still has the machine or that it is still working, and I haven't got an answer from him yet. Will probably get it on monday. You have three pictures below. "HID-kompatibel spelenhet" means "HID-compatible gaming device". In the box "Gaming devices" it shows up as two joysticks. Windows only reads the name from one of them and puts it on both. I had hoped one should say "Joystick 1" and the other "Joystick 2", since that is what is programmed in the board. However, you can tell them apart anyway by opening them and see which one has 8 axis and which one has 5. /Niclas

If you ever get rid of that problem, tell me how you did it ;) /Niclas

This is an image of how the pins in the button connectors are connected on the bottom of the PCB (but viewed from the top). As you can see I am using a matrix of rows and columns just like for example Leo's BU0836. The difference is that each row has one pin for every column, simplifying connections. And my matrix is a lot bigger! :D /Niclas

Update: NEW THREAD IN THE FOR SALE SECTION OF THE FORUM! The original post if someone want to follow the story: Just thought I'd post a couple of pictures to show you guys that have shown interest in what my upcoming joystick controller will look like. The photo is of the current controller of which I have only made 2 pcs a long time ago in the PCB prototype milling machine at work. It is called B240A1310 since it has support for 240 buttons and 13 10-bit axis. The new version is called B256A13. I left out the 10 for 10-bit since the same PCB could be used to make a version with 12-bit axis inputs with the swap of an IC and some software changes. It has a little more space between the button connectors compared to the B240A1310, and no power connection up in the left corner since this one draws it from the USB port instead. It is also 10 mm smaller in both directions (150x90). /Niclas EDIT: Sorry, had to remove the 3D render of the new version.

That was a pretty cheap controller with good specs! We want ours to be compatible with SIOC for a quick replace of the present boards. They have big problems with noise on the analog inputs and are expensive if you consider that they only have 6 outputs and 4 (8 bit) inputs. That's why I'm developing my own. Of course SIOC only has support for Opencockpits' USB servo board, so my board shows up in Windows as three USB servo boards to allow that number of I/O's. /Niclas

Nice work! I am developing a servo board that is compatible with opencockpits usb servo board but with 13 10-bit outputs and 8 10-bit inputs that you might be interested in. We are using SIOC at NovelAir but really need to get away from their boards. /Niclas

Hi, On the mechanical side I can't be of much help I'm afraid, but if you have questions about the electronics just send a PM. /Niclas

Thanks for sharing! /Niclas

We are now at eight cards. Just noticed that the company that make the PCB's are having a vacation until 1 september, so the prototype will be delayed more. Expect final delivery around beginning of november if you live outside Sweden. /Niclas

With some reverse engineering and a joystick card you can do it. For example a BU0836. /Niclas

No it wouldn't. It would probably be some software that takes the joystick button presses and convert them to keyboard presses. They joystick would probably still work as normal if you map it directly in-game without using the keyboard emulation. I think his solution seems a little bit hard to use. Let's say you have a card with 3 inputs and you do the same thing, a board that let's you make all three different combinations of them. B1-B2, B1-B3 and B2-B3. Let's say now that you want to send the command "ctrl+g", so you set B1 to send "ctrl" and B2 to send 'g'. Now the next combination (B1-B3) is locked to having 'ctrl' as one of the buttons, and the last combination (B2-B3) is locked to having 'g' as one of the buttons. It just seems like it would be hard to fully utilize all the inputs. Of course you could change all the keyboard mappings in-game, then it would be no problem. If the game supports combinations like "a+f" that is.. I don't think it will. If the game only supports ctrl, shift and alt as modifiers then you can put them on the three first inputs, and then utilize all the combinations that include one of these. But not the rest of the combinations. I hope someone was able to follow.. :) /Niclas

We are at six cards now. I have just finished tuning the CAD. It now has 256 inputs instead of 240. There are 16 inputs in each group and 8 groups. Since the extra input in each group needed a pin, the common in each group is now only multiplied to four pins instead of five. The size have also been reduced to 90x150 mm. I have ordered a prototype that probably will arrive in 2-3 weeks. When it arrives, I will update the firmware to support the extra 16 inputs. Then I will order your cards when I have finished and verified my design. Therefore the time before they are ready for shipping will be delayed by 3-4 weeks (if the shipping company is as fast as they say). The card is firmware-upgradeable over USB, so I will maybe extend the feature list in the future. The first thing I am thinking about is rotary encoder support. The second thing is filtering of the analog axis inputs. /Niclas

When a game asks for an interface to a game controller with DirectX, it get's 32 buttons by default if it doesn't specify anything (or so I've read somewhere sometime). So that is why some games are limited to 32 buttons from a single game controller. DCS can handle 128 buttons from a single game controller, probably because it specifies that it want to read 128 buttons. I have made a joystick card that has 256 buttons and 13 analog 10-bit inputs. In Windows it shows up as two joysticks, no special drivers needed. DCS can handle all the buttons. I don't know if 8 analog axis is the limit, since my card doesn't violate that limit (8 axis on the first "virtual" joystick and 5 axis on the second)

I have calculated a little bit to make up two examples. If we get to a total of 5 cards the cost for each would be 120 USD + shipping. If we get to a total of 10 cards the cost for each would be 94 USD + shipping. This is including the female connectors with 10 cm of flat cable attached to each one of them (about 11 USD in total). Not including connectors for the analog inputs (regular 0.1" pin headers). The current state is: wilcox has shown interest in one card and will ask some of his friends too. I would happily buy 3-5 myself depending on the final cost. I can probably sell them later anyway if we don't need them at NovelAir. It's always easier to sell when you already have them and know exactly what they cost. So, we're currently at 4. Also, expect up to 5 weeks lead time. It will take me a week to finish the CAD (want to change it a bit if the cards are going to be sold), it will take three weeks for the producer of the PCB's and it will probably take at least a week to ship around the world. The cards are currently 100x160 mm in size. I might shrink them one or a few cm in both directions when tuning the CAD. /Niclas

I don't want to limit my interface card to DCS:WH. I want it to be flexible for the future. And like I said, this may be used for custom HOTAS too. In that case it will absolutely be an issue. But in other aircraft I know of situations where it will occur without the HOTAS being involved too. Yes, I think that is the best option too. That is what I have been saying all the time :-) I will use such a card. The meaning of my original question was "Is there any way to get around these issues with a keyboard emulator?", not "Will this be an issue?", because I already know it will :-) But like I said, it doesn't really matter since I already have an interface card without these issues. It was just a question out of curiosity, and I thought that if there is a way to get around them that could maybe give me some new possibilities when developing new interfaces in the future. I have already disrupted this thread enough. Thanks for the answers! /Niclas

No, you don't sound confusing. I understood what you meant the first time, maybe I expressed myself poorly. I am an electronics engineer and have developed both joystick controllers and keyboard emulators, so I know exactly what you mean :-) I agree with you that probably no switches HAVE to be held at the same time. However, since I am a little perfectionist I do not want these flaws. I want to be sure that I can hold a switch for raising my seat or starting an engine or whatever at the same time as pushing a pushbutton elsewhere in the cockpit without the ejection seat going off or any other undesired action :-) Since I have developed my own joystick card I am going to use that. This was more out of curiosity, if someone knows if it is at all possible to get around these issues with a keyboard emulator. /Niclas

Thanks for the info. I don't want that kind of limitations in my interface card, since I don't know in advance what I am going to need. But if you use it the way you say, it won't be a problem. However we make both HOTAS and panels where you need to hold stuff in. For all the (on)-off switches for example. You usually hold them in the (on)-position for some reason. Then there are gyro alignment buttons and such things too. And canopy/seat up-down, I bet there is more too. For me, it's all about the small details :-) Little bit of thread hijacking from my part here.. Sorry about that. Your discussion may now continue undisturbed :-)

But how does it handle buttons that you need to hold down, like the trigger when bursting? Let's say the trigger has ctrl+space keyboard command. Wouldn't you screw all other button presses that are being made while you hold the trigger? Ctrl affects the whole keyboard. Or can xkeys bypass these problems in some way? For example by pretending to be several keyboards? /Niclas