ytsestef

Thanks for the info! I'm just getting started on the project, still learning the NG platform. There is a transpose parameter in EVENT_KB event, so I thought it would be possible in some way, with all the scripting capabilities of the software and whatnot: kb_transpose=<-128..127> Anyway, I could always just send a MIDI CC with the octave shift buttons and let the host software do the transpose.

I hear MIDIBox_NG is 10x slower in scanning the keyboard matrix than MIDIBox_KB. Is this ONLY due to the amount input scanning? If I upload MIDIBox_NG and edit the configuration file so that only 2 in/out shift registers are used, will it match the speed of MIDIBox_KB? I want to build a controller w/ 61-key Fatar keyboard (I've already built it and it works - mostly; more on that later) but I also want to use only a couple of buttons and LEDs (octave up/down keys with indicators etc) in addition to the keyboard. If I chain another DIO_MATRIX module and build a second matrix for LEDs/buttons, then configure MIDIBox_NG to use only 4 in/out shift registers will I achieve something between the performance of MIDIBox_KB and MIDIBox_NG? I don't want to add a second CORE_STM32F4 module just for a couple of buttons and LEDs, but I am not feeling great with the ~3ms scan time of MIDIBox_NG (I also didn't like the quantization in velocity values; 50-1000 "delay value" is a much more detailed range than 5-100, the latter isn't enough, not even for the full 8-bit range of 128 values that MIDI supports (at least not without using velocity mapping - unless I am doing something wrong) EDIT: read some of the .NCG configuration examples of the documentation, the answer is right there: however, delay values are still in the same range, although I'm now able to get less "stepped" velocity values

Even a board for multiple FPGA chips would be cool I wonder if they sound so close I would fail in a blind test and also if they are capable to do the pseudo-sampling thing, would be really sweet given the power of STM32F4 A MIOS32 port of MBSID could also enable sampling features even on real SID chips!

Does this mod work on the original CORE_8_Module (non-MB6582 version) too? I am talking about SIDV2 usage, not another MIDIBox project. Are the hex upload and the bridges still required?

For the Stereo problem, you need to do this: http://ucapps.de/mbhp/mbhp_4xsid_c64_psu_optimized.pdf Bear in mind that (at least on smashTV's PCBs), on the SID Module, only the 4 pins of each row of the J2 are used, yet it is a 5x2 header, so, it's not the outer (last) wire of the ribbon that has to be cut and connected to J14 of the Core Module, but the third from last ;)

Yeap, I wanted to tell you, I think the LEDs are too close to the pots, the skirts of the knobs may be partially covering them. Also, wouldn't you like the labeling a little more symmetrical? Maybe LFO,FILTER,ENVELOPE and OSCILLATOR should be a little smaller and rotated 90 degrees CCW, then placed next to each section?

Beautiful! What are the feedback 1..4 knobs? SID feedback or some kind of filter feedback?

Very good job!!!

Thanks! If I fix it I could use it on my MBFM :)

I tried with the Steiner first, it is quite noisy, it is not so steep (I think it is 12 or 18db/oct), and it doesn't offer CV input for resonance (cutoff only).

You would need -12V to power both the SSM and the AOUT module (at least that's the case for AOUT_LC), so, the C64 PSU isn't really suitable. You can buy the Meanwell which is also very lightweight. When using seppoman's PCB, the filters can be switched on and off via a relay (EXT menu -> SW# depending on which pin of the J5 the SSM2044 bypass pin is connected). This will work only if you have made the appropriate changes in setup*.asm for the J5 configuration and prohibits the use of J5 as DOUT for LEDs or as analog input. The frequency and resonance can be changed in the EXT menu, again, under A#1, A#2 etc, but you can also enable the F2A function on the ensemble menu (EXT submenu) which forwards the SID Cutoff and Resonance values to the analog outputs, therefore you can control the SSM exactly as if you were controlling the SID filter. As you may have already realized, this has some drawbacks if you want to be able to switch between the SID and the SSM filters. When you have enabled F2A you need to be sure to disable the SID filter, plus the two filters have different calibration values if you're using a 8580, so you need to have two different ensembles, one for use with patches made for the SID filter and one for the SSM. This brings extra headaches into the game, because you can have only one ensemble bank (if any - depending on whether you're using multiple banksticks or just one), and you don't want to limit your synth's capacity to 128 sounds, so you have to remember which sound uses the SID and which one uses the SSM and switch between ensembles on each patch change. That said, if you are willing to make some sacrifices in usability and cutoff resolution, you could calibrate the AOUT and the SSM boards to give full sweep range between 000..800 hex values, and with a little more switching, you can use both filters alternatively or even at the same time in series, introducing more original effects (with different resonance/cutoff values or even different routing and mode on the SID filter). Unfortunately, until the firmware supports filter switching/calibration save in the patch, I can't take that route because I'm using the synth a lot in live performances, I don't have time lurking in the menus while I play. :) I think SSM2044 is only 4-pole, low pass. What's more, seppoman's board utilizes two SSMs, so you can have a stereo configuration quite easily, without limiting your synth to mono. All it takes is an extra chip and some caps/resistors/trimmers. http://www.midibox.org/dokuwiki/doku.php?id=ssm2044_pcb

I am very satisfied with how my MBSID runs on this PSU. I could have another one for my upcoming FM synth :)

Good Luck! Just so you know, remapping the register pins is very easy. :)

I solved it! I'm so happy, after weeks of headaches! The Keyboard Controller had a midi volume control pot (actually slider) and I had it thrown away. But I forgot to tie the input to the ground. As soon as I put the slider back in and turned it all the way up, the problem disappeared. :D Thanks everyone for your help!

Found it, guys. It was the bridge. On mike's PCBs, there is a bridge for the last pin of the last Shift Register that should be soldered underneath the PCB. It had come off and the SR voltage never reached the resistor. I soldered it again and all is good :) Thanks

Pin #7 of the Shift Register #4 of a DOUTX4 module is outputting -0.73V (with LED connected) when it should light up. All the other pins of all the other shift register work correctly (about 2.5V with LED lit) I already checked LED polarity and resistors...

OMG, I didn't know about that. I'm so angry now. :getlost: Plus, it's FUGLY!

I have no idea about the sammichSID in particular, but I know that 1*** has nothing to do with it, this actually displays available cores (stereo SID pairs), so even with two SIDs (in stereo setup), 1*** should still be displayed.

I'm currently working on altering the firmware so that it always routes AOUT to channels 1/2 anyway, regardless of the SID that is triggered by the dynamic voice allocator (ie even if the right SID is playing, the curoff/resonance values are forwarded to the left AOUT pair). I can't really guarantee that I will succeed, but you should hear from me during the next week, I am currently trying. :)

Unfortunately there's no room left in my build :( Oh well ... Thanks for your time anyway :)

Damn, I needed 6 voices so badly! Another workaround would be via wavetables, but they behave strangely when they reset before the envelope ends (ie when you play fast) Anyone has any idea how hard it can be to alter the firmware to route both filter channels in one aout channel?

I found a problem with your implementation, imp (or I don't know how to do it properly :P ) I created a patch using the ML engine for 6-voice polyphony, then made the envelope control the cutoff frequency. The thing is, for the first 3 voices (left channel) the filter follows the envelope normally, but when it's time for the right channel to play, the envelope fails to trigger the filter. I guess the firmware assumes that 4 AOUT_LC are connected and the right channel CV is routed to the second AOUT_LC which doesn't exist. I have to mention that I have edited the .asm file as follows: #define DEFAULT_F2A_CUTOFF_L_AOUT 1 #define DEFAULT_F2A_RESONANCE_L_AOUT 2 #define DEFAULT_F2A_CUTOFF_R_AOUT 1 #define DEFAULT_F2A_RESONANCE_R_AOUT 2 ...in order to try to force the firmware to send both CVs through the same channel, but it didn't work, so, I guess each CV output is exclusive to one AOUT_LC only. Can you check if you can recreate the problem? I'm sure this can be rectified by modifying the firmware code, but I'm pretty much a noob at programming. It's a pity because I can't use the filter properly in ML engine, half notes are being modulated and half are static.

I predict a masterpiece! I love keyboard versions of MidiBoxes, as soon as I finish my MBSID, I'll start a keyboard FM :D

Progress has been made :D I went hardcore-DIY and drilled the panel myself, labeled it using a sharpie :P Now I just have to calibrate the SSM2044 properly, build the rest of the control surface and connect the DINs and DOUTs with each other and with the buttons/encoders/LEDs. Then I have to fix this terrible disappearing sound problem... everything else works like a charm

Of course that would be it and I'd do that if I knew how to do that :P Anyway, I wasn't trying to get anyone make me something, I just want to throw 3-4 ideas on the table, and if he likes some of them he can include them in a future release and we could all benefit ;)