Ray,thank you very much for setting up this forum. This is the one I was really looking for.It seems that the "Analog settings" of EPICenter 0.1.1.38 does not do any thing. All settings won't be saved as well.I know it surely did at 0.1.1.36.Does anyone know what's wrong? I cann't live with 0.1.1.36 since there is a display problem.Dan

Dan~Sometimes I've found it beneficial to remove all the registry settings that a new EPIC EEPROM can create.Try removing the registry settings, then re-installing the EEPROM load, and let USB Config do it's thing.You might talk with Ralph or Steve to get the registry hives to remove.Ray

Ray,thanks for your suggestion.I gave it a try with clearing up the registry but my first try didn't any help. Will try to get a tool from R&R as you suggested.Have you ever tried 38? Did it work OK?BTW, I happend to find out EXEC(140) magic in your code. It really help me a lot in speeding up my engine. How did you know thay? Is there any docs which describe such magic?Dan

It's not really magic... :-) In one of the early EEPROM releases, there was some code specifically for the 18Mhz EPIC board that was left in by mistake. As a result, all the 6Mhz cards slowed down to a crawl.That exec(140) merely resets the EPIC back to 6Mhz operation.Ray:-wave

Ray,it's still magic to me. Not a problem.I have ran the EPICenter 0.1.1.38 on the very clean, no EPIC piece at all Win2K Pro system but the rusult was same. Hope it will be fixed at next release. Until then I need to find out any alternatives.Dan

There are still a few bugs in the Analog Settings. I am going to get with Steve today to see if we can resolve them.As of V41- you can set the rates, but they are not saved to the card. You can use the format device.analogname.rate = 1; to set an analog to a rate. See the analog section of SYNTAXUSB.DOCI tried to attach but nothing happened. Email me an I will send the latest DOCS.ZIP Attempting to attach a label to analogs, I found that this did not work either. I will discuss this today also.--Ralph

Ralph,Thanks for the test result. Now I understand that the v41 still has problems with Analogs. Will wait for further release.PS:Today I received the v41 files from Flight Link Tech Support. Thanks to Flight Link Tech Support.Dan

>Ralph, >>Thanks for the test result. Now I understand that the v41 >still has problems with Analogs. Will wait for further >release. >>PS:Today I received the v41 files from Flight Link Tech >Support. > Thanks to Flight Link Tech Support. >>Dan Dan how did you get the new v41 files they are not posted. did you talk to some one at flightlink? and what is the v41 files? is it newer than .38? and do you have any news on the new epicmain? Ron.

Analog settings now work in version 42. The rate curves can be built and saved now. Analog linking to keystrokes and labels now work. This takes a little getting used to since any movement of the analog above the position shown will cause a scrolling to that position.This should only be used for input rates, keys, and linking. To map input channels to device analogs (output analogs) define these in the device descriptors. To change the source of an output analog use:deviceName.analogName.source = ; and device.analogName.type = ANALOG8; **Ralph

TVRON,Flight Link Tech Support sent me a set of v41 code along with their answers to my EPIC USB related questions.You'd better ask v42 now.Dan

Ralph,thanks for good news. I'm now asking Flight Link Tech Support to send me a new code.Dan

Uh, it wasn't accidental. The enhanced version of the SPROM(the 8pin IC in the socket) had to work with all speeds. This adds a hardware divide, capability to read/write up to 8 bytes at a time from the expansion bus while allowing the processor to do other things, and a keyboard 2 byte sequencer. The hardware divide speeds up divides by about 50 times. The keyboard fifo was needed because the release codes are 2 bytes. The two byte sequence was "release" wait 1ms for the byte to get out, 1ms gap, then send the second byte (keycode). The bus sequencer will allow a future bus request stack which will free up more processor time for more complex math. The pushc(x); exec(140); calls a routine that sets up base level timing, 18mhz allow full speed bit, etc. This also writes to the enhanced logic to divide by 3 if 18mhz to slow down the logic clock to be the same speed as a 6mhz. This also has a divide by 2,in case the 18s are flaky.

There is a bug in the auto unplug/replug settings. You need to have a 2 second wait (OS system wait time) and the setting does not stick. This is corrected in V43 (may be out today, will be checking it out later, hopefully). The initial setting is for 1 sec and for my system this is a touch too fast. This is needed since Windows has to figure out the device that unplugged, teardown registry setting, drivers, etc. then when plugged back in, figure out who just plugged in, get the report descriptors, build the registry, put the drivers in, etc.

I can send you the current EPICMAIN project, or you can get it from Ray. This has some bug corrections and additions from my trip to the motion cockpit in Georgia. EPICMAIN was intended to be a temporary workaround until Dowson gets his EPICINFO.DLL communicating with EPICIO.DLL. He sent me the beta release of this and I will check this out when I get time.This will make all pigeon holes, Qprocs, etc work as per the Dowson docs. We have a couple of issues to work on, then the new version of EPICINFO will be posted at the normal Dowson download points.