widdly

I excited too. Adrians cases are the best. I've got a Shruthi and PreenFM with them and they are easily my best looking synths.

I am building a module that takes an analog CV and converts it to a midi message. I am connecting the CV via a 1K resistor to an AIN input with a diode clamp to protect against over/under voltage. I am using Banana cables so when no signal is connected the AIN inputs are floating and will produce randon results. My solution is to connect a 1M resistor to GND on each input. Is this the correct way of doing this? My other question is with regards to the schematic here... http://www.ucapps.de/midibox_ext/lfo/lfo_schematic.pdf The diode clamp is using 1n4148 diodes so the PIC pin will get -0.7 to 5.7. But the datasheet quotes -0.3 to 5.3 as being safe. Is it ok to use 1n4148 with a PIC or would it be better to use a BAT48 or similar that would limit the voltage closer to the datasheet range?

Lemur is out on Android now. :smile:

:phone: I am tuned!

Yep I'd love one. I spent the weekend soldering up the front panel PCB for a midibox seq and was planning on ordering a set of your panels. A wooden case would be pimp! Also nice job on the LXR. I was just looking at it on the Sonic Potions site. The LXR is an awesome machine.

I'd like to join a group buy for a case for an LPC SEQ4. Anyone got an update?

Unfortunately it's in japanese but this site is the one I was thinking of. It's based of the USBhostlite examples. http://www.rjblog.net/archives/2009/e000331.php translated... http://translate.google.com.sg/translate?hl=en&sl=ja&u=http://www.rjblog.net/archives/2009/e000331.php&ei=K5S8TtuDMKWH4gTO_ISABA&sa=X&oi=translate&ct=result&resnum=1&ved=0CB8Q7gEwAA&prev=/search%3Fq%3Dhttp://www.rjblog.net/archives/2009/e000331.php%26hl%3Den%26biw%3D1366%26bih%3D704%26prmd%3Dimvns There is source code included. The source code has a parser for the usb descriptors, checking if they are audio devices with midi streaming. Usbhostlite has an interrupt and a blocking function Host_ProcessTD that fills a buffer from the USB port. Other than the first byte, the buffer is written straight to a serial midi port. The example doesn't have any hotplugging implementation other than exiting if the midi usb is disconnected. There is a description of the first byte content in the USB midi spec, on page 16... http://www.usb.org/developers/devclass_docs/midi10.pdf I'm not sure on the hardware side but perhaps this post might be helpful in setting up the physical USB host port on the LPC1768... http://knowledgebase.nxp.com/showthread.php?t=2064 I would think class compliant USB-midi support would be the way to go. I imagine support for custom stuff would be a nightmare. Although support for the novation launchpad would be a neat way to add a BLM without all the soldering. edit.. Actually looking more closely at the code from the japanese site, it doesn't look quite right. The MIDI_ParseConfiguration function is never called. I'm guessing he must have modified the usbhostlite function Host_EnumDev to call MIDI_ParseConfiguration instead of MS_ParseConfiguration.

Is it possible to add USB host support on the LPC17? I've seen some other LPC based projects on the web that have midi usb host features. I think the seq4lite would be great with one of those usb only mini keyboards like the akai lpk25.

The midiboxFM is a very different synth to the FM8 despite both being FM. The things labelled A-B-C-D-E-F in the FM8 square matrix thing on the advanced page are the operators. It's like an oscillator and an envelope combined. The matrix in FM8 lets you setup any combination of Operators you want where as the OPL3 has only 4 preset routings. FM8 also has a filter and a wave shaper ( G and H on the matrix). FM8 envelopes are very sophisticated, they have unlimited breakpoints, tempo sync and can be used as LFO's as well as operators. Additionally there are two tempo synce LFO's in FM8. The OPL3 built in envelopes are quite limited with only a small range for the adjustments. TK has implemented an additional Envelope and two LFO's in the MidiboxFM which can tempo sync. The MidiboxFM has a wavetable sequencer that is pretty unique and doesn't have an equivalent on the FM8. I like the Midibox FM for bass sounds (similiar to the famous lately bass), percussion clicks and blips and for weird fx. I wouldn't use it for e-piano sounds because of the 4ops and limited op ratios. Pads and soundscapes are do able and they tend to sound lofi and glitchy.

Getting the CV's in the range 0-5 involves using a voltage divider (2 resistors). The problem with the Doepfer is that the ADSR, LFO and the resonance in the filter are not voltage controlled. If I was doing this, I'd implement the modulation sources inside the Midibox and have the modulation matrix in digital. It would be a lot less electronics and would allow you to do cool stuff like midi sync on the LFO's. I'd go with 8 CV's out of the AOut NG controlling Osc1_Pitch, Osc1_PWM, Osc2_Pitch, Osc2_PWM, Osc2_Level, Cut, Res and Amplitude. You could use CMOS analog switches to control other parts. I like the DG series switches ..http://www.intersil.com/products/deviceinfo.asp?pn=DG411 You can use these to switch analogue signals using a DOUT pin. The different wave shapes of VCO's could be selected with one of these chips. The sync signals could be switched with these. The second VCO signal could be switched to mix with the VCO1 or to frequency modulate either VC01 or the VCF. You would need two VCO's, two VCA's, and a VCF. For the VCO's I'd look at the Hearn Morley ones here...http://sites.google.com/site/hearnmorley/synth-diy-resources or the prebuilt pcb's from here..http://mattsonminimodular.com/

You could build a GM-5 interface since you have 5 devices it would be perfect. It's a 5 in/out usb midi interface that is cheaper than anything else out there. You can set it up so the midi port names match the devices they are connected too. So when you are using say Ableton Live you select "Roland JV-2080" instead of "midibox gm-5 port 2". It's pretty ace.

I'll take 8 of those if there are enough left.

I've also noticed what looks like a bug. The Instrument assign numbers in the AOut menu show the values 5,6,7 and 8 instead of 1,2,3 and 4.

It looks like AOut offset is part of the ensemble not the instrument. I've found it's better to leave the offset at -63 and use the AOut depth in EG5 to sweep the filter.

Couldn't you just add another R resistor between the ground and pin 1 and between the ground and pin 5? That way you could have both equal steps and different configurations.

Looking at different analysis of R-2R DAC's on the web, they all use a 2R for the resistor to ground. I notices in the AOut-LC it uses a 1R. Wouldn't that make the scaling of the bits all wrong?

I'm building a midibox fm. I've got all the circuit and control surface built, working and sounding good but I want to add a pair of AOUT LC and a pair of filters. I'm planning on building the EMS diode ladder filters described here.. http://www.modular.fonik.de/Page37.html They have CV control of cutoff AND resonance so I think they should work nicely. I'm trying to incorporate these filters so the cutoff and resonance settings are completely saved with the patch and don't require any external potentiometers. I want to be able to tweak the cutoff and resonance from the v-pots so I figured I could remap the USER 2 v-pots to control the cutoff and resonance via the AOUT offset parameters. In cs_menu_tables.inc at line 420 I figure I should change it to something like... ;; USER2 Register (00=dummy) |<-->| max print ix, exec ix parameter transfer CS_MENU_ENTRY MBFM_AOUTx_OFFSET, "Cut1", 0x7f, PRINT_AOUTxOFFS, EXEC_SELPAR, R2PP2R_AOUTxOFFS CS_MENU_ENTRY MBFM_AOUTx_OFFSET, "Res1", 0x7f, PRINT_AOUTxOFFS, EXEC_SELPAR, R2PP2R_AOUTxOFFS CS_MENU_ENTRY MBFM_AOUTx_OFFSET, "Cut2", 0x7f, PRINT_AOUTxOFFS, EXEC_SELPAR, R2PP2R_AOUTxOFFS CS_MENU_ENTRY MBFM_AOUTx_OFFSET, "Res2", 0x7f, PRINT_AOUTxOFFS, EXEC_SELPAR, R2PP2R_AOUTxOFFS ...but I'm not sure if this will work as I expect. Will the v-pots control AOUT 1-4 offsets respectively or just control the same offset of whatever AOUT is selected in the AOUT menu? Is there another way of mapping this?

Thanks for the reply. I thought the LED's would draw less current since they are wired in a matrix with the shift registers so they are not actually all turned on at the same time. When it's running I figure only 8 LED's will be turned on at any moment. I suppose the situation will be different at power up before the shift registers are set up by the PIC.

I'm building a monome type 8x8 led button matrix. I'm using a GM5 module, a core, 64 leds, 4 dout and 2 din. There is NO lcd. I'm wondering if it would be feasible to power this from the GM5 5v output and have the whole thing USB port powered.

I don't know how these would go switching the negative cycle of audio signals. Normally when switching audio with a 4066, a DC bias is added to the input so the negative voltage is higher than the negative supply of the 4066. If you wanted to use larger audio signals like in a modular synth I'd try the DG211 chip instead. You would need a bipolar power supply though.

The new ploytech GM5 driver doesn't support hibernation properly on my laptop. The system freezes during resume and I have to hard reset the machine. The original windows drivers didn't have the problem.

No it's not. The FAQ describes how to get Windows to recognize two different GM5 boards by using a different number of ports or changing the PID link. It doesn't describe how to use two different modified .inf files to get 10 custom device names. I'm not sure that would work with the ploytec driver installation since it seems to remove previous instances of the driver. I think I could probably install the two devices then modify the names in the registry. I guess there is only one way to find out if it will work . I'll report back when I finish my second board.

Question on device names with multiple gm5's. If I modify the .inf I can have the names I want for devices. Now lets say I build my second gm5 with 5 ports and I cut the jumper so it uses the ploytec PID. Can I modify the .inf file again to get 5 different names? Will I need to modify PID in the .inf as well? Anyone tried this? Or would I be better off building the second gm5 with four ports? How would that work when installing the driver with the modified .inf file? I would have thought after installing the first gm5, the driver would auto detect when the second one is plugged in.

Awesome, this is great. Multiclient was the one missing feature that was driving me mad. Now I can have my synth editors and Ableton running at once which will totally change my work flow. I'd had no trouble with huge sysex dumps on the stock ms drivers. I'd tried doing it simultaneously on multiple ports without any problems.