philetaylor

Sorry I have been absent from the group for a 'long time'. If I can be any help let me know but I haven't played with my DMX code for an even longer time! Cheers Phil

Yes it's technically possible to change the pins but you would need to hack the "universal" app_lcd.c. I will warn you though, that isn't for the faint hearted! Cheers Phil Sorry I just read your message again, I didn't realise you were using the PIC version. That should be easier as all of the pins are defined in app_lcd.inc you should just be able to change the pin assignments. You will need to check that they aren't being used by anything else first though!

No that looks fine to me, it's still 128x64. The large GLCD font is a bit squashed anyway, it looks just like my DOG-L displays! Thanks Phil

Yes that looks exactly the same as the ILI9320 based displays that we bought Shuriken. I did manage to get a MIOS32 driver written and working for it in 8-bit mode but since then I have had a bit of a crash and I think I lost the source code :( If I remember correctly, the key to getting 8-bit mode working was that you place the LE pin high then load the lower 8 bits onto DB08-DB15 and then set LE low and load the upper 8 bits. the transition of low to high latches the current state of the inputs on the 74HC573 (DB08-DB15) and outputs them to DB00-DB07, this has the effect of loading all 16 bits to the display, you can then pulse the clock. Cheers Phil EDIT: Also DO NOT tie the unused DB00-DB07 inputs to ground, they MUST be left floating as otherwise it won't work in 8bit mode!

Hi It's definitely not password protected (I know it is my server!) What message are you getting? Most browsers will just try an anonymous ftp connection (if not try username: ftp and password: <your email address> ) Thanks Phil

Hi. 1. PWM Steps: This defines how fast the H bridge is 'pulsed' to control the fader motor. If you find that your motor is sluggish then you can adjust this. 2. AIN Deadband: All A-to-D (analogue to digital) circuits suffer from a certain amount of 'jitter'. The AIN deadband allows you to set a minimum amount of change in the analogue input before it is reported as a 'move' of the fader. 3: MF Deadband: I guess that this is the reverse of AIN deadband? only engage the motor if the input value has moved significantly. although TK will be able to confirm this??/ If you find that the faders are a bit jittery then you should be able to adjust the deadband values to counter it. Cheers Phil

Hi. What have you tried so far? I am assuming that you have put the next 4 encoders on SR 11? If not then the code needs quite a bit more changing, I would recommend putting them on SR11 and moving the matrices over to SR12-SR15 If you have put the encoders on SR11, the code changes that you will need are: in app.c, change #define NUM_ENCODERS 40 to #define NUM_ENCODERS 44 in mios32_config.h change: #define FIRST_ENC_DIN_SR 1 #define DIN_16x16_L0 11 #define DIN_16x16_R0 12 #define DIN_16x16_L1 13 #define DIN_16x16_R1 14 to: #define FIRST_ENC_DIN_SR 1 #define DIN_16x16_L0 12 #define DIN_16x16_R0 13 #define DIN_16x16_L1 14 #define DIN_16x16_R1 15 I can't see any other changes that will be required but I may be wrong :) EDIT: Having looked at DINs.png, I can see that you have put the matrices on the first DINs and the encoders after them? That is the opposite of the way that the program is written and I would definitely recommend that you reconfigure it so that the encoders are on SR1-11 and the matrices are on SR12-15 otherwise the changes to the program are non-trivial (although still not that complicated!) Phil

Hi TK. If you are using Xcode 4, this doesn't support OS X 10.5 by default (it doesn't have the 10.5 SDK). Apparently the supported thing to do is to install Xcode 3 in a different directory although there is a "hack" http://stackoverflow.com/questions/5378518/how-to-add-base-sdk-for-10-5-in-xcode-4 Cheers Phil

Here is the wiring diagram for connecting 2 x CLCD's to a CORE8: mbhp_2xclcd_mios8.pdf Cheers Phil

Hi. In mios32_config.h you can enable J5a as analogue input: #define MIOS32_AIN_CHANNEL_MASK_0x000f whenever the analogue input changes (pot moved) your APP_AIN_NotifyChange(u32 pin, u32 pin_value) function will be called. You also need to initialise the J5b pins as output using the MIOS32_BOARD_J5_PinInit() function: int pin // Init J5B as output for (pin=4;pin<8;++pin) MIOS32_BOARD_J5_PinInit(pin,MIOS32_BOARD_PIN_MODE_OUTPUT_PP); You can then use the MIOS32_BOARD_J5_PinSet(pin,value) function to turn the LED off and on. If you need more information, look at the tutorials: MIOS32 (for STM32 and LPC17) specifically the J5 Outputs and Scanning 12 analog pots ones! Cheers Phil

Hi John. To be honest, web browsers make pretty useless ftp clients! I would suggest that you download Filezilla which is a 'proper' ftp client from http://filezilla-project.org/ I it possible that the download was timing-out due to the slow connection. That server is on a dedicated 4Mb leased line which very little bandwidth use so you should be able to download quicker than that! If the slow connection is the problem then filezilla supports resuming the download... I just downloaded it from home (to verify that the file is OK) and I got it in about 360KB/s which took 80 seconds. Thanks Phil

I bought mine from Mouser who while not exactly 'in' the UK, provide cheap shipping and duty paid. 100 cost about £14 and if you need other bits, you can get the delivery price down quite a bit as well. http://uk.mouser.com/ProductDetail/ALPS/SKHHDTA010/?qs=fMKjfF2mFohbx9AfkVFXKg%3d%3d Phil

That looks like some bizarre kind of spam? They have posted the same question to a number of irrelevant forums: http://www.hitachicordlesstools.net/hitachi-cordless-battery/can-i-upgrade-to-a-quad-core-with-my-laptop/ Their other two questions on here are also a bit generic and not particularly MIDIbox related? Phil

Hi Jef. Sorry I have been away for a few days... Yes the DOGM-128 display should work fine. Phil

Ah yes I missed the MIDI bit from your post TK! I'm not sure how USB Host MIDI would work though as it would need to support a number of different USB MIDI devices to be useful? I seem to remember this was discussed a few months ago as somebody was looking to do a commercial project with USB Host support.... Phil

Not exactly free but there is the USBHostLite for LPC17xx http://ics.nxp.com/support/software/usb.host.msc/ The license seems to be a bit ambiguous. It says free to NXP customers for unlimited use with NXP LPC2000 and LPC3000 families but includes an LPX1768 example.... Phil

Hi. Does the file: $MIOS32_PATH/include/makefile/common.mk exist? If not, you have not successfully checked-out the MIOS32 source. Try the checkout again: svn co svn://svnmios.midibox.org/mios32 Thanks Phil

Hi. Even large chips like the PIC (40 pin) shouldn't take more than a couple of minutes to solder. If each pin is taking more than a second or two then there is either a problem with your technique or your equipment :) A soldering iron that is too cold will not melt the solder quickly enough and you will really struggle to get a proper joint. One that is too hot can also cause problems as it can damage sensitive components and even cause the copper traces on the board to come unstuck so I would definately recommend a decent temperature controlled iron. I have had the same Weller TCP soldering station for over 20 years (I 'borrowed' it from college...) If you are struggling with technique then there are lots of tutorials around the 'net. Cheers Phil

I purposely didn't quote a number as they are the defaults but by no means are they limits! I know that people have successfully tested 32 DIN shift registers. If some method of 'boosting' the SPI signal was used then (in theory) many more DIN/DOUT modules could be supported (with minor code changes)

Hi, In general AIN/AOUT DIN/DOUT is achieved with external modules, these are daisy-chained to achieve large numbers of inputs/outputs using shift registers. Take a look at http://www.ucapps.de under the MB Hardware Platform you will see the different modules that can be connected to the core module. Cheers Phil

As Albert Einstein (allegedly) said "the only stupid question is the one left unasked" As a general rule, as long as somebody has at least tried to look at the masses of docs on www.ucapps.de, wiki.midibox.org and this forum and still can't work it out then you will never get a smart/offensive response. Good luck with your MB6582, you will love it :) Phil

The biggest problem with this display is that in 8 bit mode it requires 2 way communication with the CPU to be useful as the CPU does not have enough RAM to maintain a reasonable screen buffer. Unfortunately both the CORE32 and LPC_CORE32 use a 74HC595 to interface with the LCD which doesn't allow bi-directional comms. I have 2 of these displays and for the last few months I have been trying to work out a way to interface them to the CORE32 but I haven't had much time recently. If your version includes the SPI interface option then this is probably the best method of tallking to it but for some reason they left SPI off the ones that I bought even though the ILI9320 supports it :( Cheers Phil

IC2 is just a 74HC595 which is available from pretty-much anywhere Mouser has them for less than $0.50 http://www2.mouser.com/ProductDetail/Toshiba/TC74HC595APFM/?qs=sGAEpiMZZMvtrnhC60i%252bOn8CARDmBV7K Cheers Phil

Modesty has always been one of my strong points...... As there is already a telnetd (and httpd) implementation in the ethernet examples section of svn (I modified the uIP samples) using this for debugging would be a fairly simple exercise although the memory requirements of uIP mean that it really doesn't add much to the debugging capabilities already offered by MIOS Studio (over USB or MIDI I/O) at the cost of a few KB of precious RAM. I would also like to make the point that the contribution to MIDIbox made by some of the previous posters is immeasurable whereas to date your contribution is definitely measurable! Phil

OK so I had to look up stochastic :)