sneakthief

Thanks for all the nice compliments! Couldn't have done it without TK's great work. The orange lcd is super cheap ($6.64 USD including free shipping to most places) and looks good: http://cgi.ebay.com/16x2-LCD-Orange-Characters-Black-backlight-HD44780-New-/120748772099?pt=LH_DefaultDomain_0&hash=item1c1d300b03 However since the LCD cable passes over all the analog circuits I was getting high-pitched whine interference! Luckily I had disassembled some DB-9 cable that had foil shielding which I then used to wrap around the entire length of the LCD's ribbon cable. I connected one end of the shielding to ground and that solved everything. I wonder if anyone else building MIDIboxes had this issue? Re. FS1r - I've had it since 1998 and wouldn't ever sell it. Despite the quirky user interface, Iit's all the FM I need (no offense to SammsichFM or MIDIbox FM)

Altitude - I've got a fat heatsink on my 7805 so all is well on that front. The original cymbal decay pot is 2M while the one that came with the MB-808 batch 1 & 2 is 1M. So, just connect two pins of the cymbal decay pot - and make sure the first pin goes to ground: http://m72.photobucket.com/albumview/albums/wholetone/MB-808/cymbaldecaymod.jpg.html?src=www&action=view%C2%A4t=cymbaldecaymod.jpg&newest=1 If this doesn't solve your problem, then I suggest you check your capacitor and resistor values.

Background: I got one of the second batch MB-808 pcb's and rare parts kits almost 4 years ago. There are a few reasons why it took me so long to finally finish it, namely the desire to cut costs but also do a good job. It simply wasn't a high priority in my studio. 1. I waited patiently for someone to make a case for it but that never really materialized and I didn't feel like paying upwards of 200 or more euro for case machining and silk-screening. Earlier this year I realized I could get a front panel for a good price from the same Dutch guy who did TK's, so I finally pulled the trigger: http://www.frontplaten.net/#/content/FRONTPANELS/3_ACRYLAAT%20PANELEN/mb808-3.JPG/ 2. Sourcing all the parts was a long process. I managed to find compatible switches for a good deal from the US a few years ago, but a big issue was the Mouser BOM. Every week some part would become obsolete or be out of stock! It took a while to find the proper substitutes for everything. 3. Incomplete documentation made this process very slow and tricky. The assembly manual ( http://www.eight-oh-eight.org/Assembly/ ) is either disogranized when it comes to the errata or simply missing a lot of little things: - Cymbal decay http://m72.photobucket.com/albumview/albums/wholetone/MB-808/cymbaldecaymod.jpg.html?src=www&action=view¤t=cymbaldecaymod.jpg&newest=1 - The low tom is missing a resistor which makes it louder than the rest - My congas self-oscillate but I decided I like using them as tones - Some of the DOUT chips were NOT grounded on my pcb and that caused havoc! - I had to cut some pots to make them all the same height because sourcing the proper height Alps pots became almost impossible - There are no proper schematics for the MB-808, just some random pictures of the voice board component layout and a few incomplete signal flow schematics. This is intolerable for a project like this. There are more things that I'm blocking from my memory because of the trauma :P 4. The Mods... what can I say? I only planned a few but then as I started testing them it became evident that it would take this machine into whole new sonic territory. It takes a lot of time to decide which mods are worthwhile and to cross-reference the original 808 schematics with the MB-808 pcb to find the location of each mod. I also had to take spend time experimenting with the clap reverb mod to find out what sounded good. Also, I didn't want to deface the front panel with all sorts of random pots and besides there wasn't room for all 11 so I decided to put them on the very front, minimizing the negative aesthetic impact of ugly mods lol. 5. Machining the case. Coincidentally, I had this steel Hammond case for eleven years already and it's the perfect size for the MB-808 pcb. Laying out all the holes takes time, not to mention all the drilling and sawing. I shouldn't mention this, but my lcd was a bit too large and I actually CUT THE MB-808 PCB for it to fit :S Embarrassing, but nobody will ever see it. All in all, this was one of the most difficult DIY projects I've ever done. Maybe not as complex as my ASM-2 modular, but probably more frustrating because of sourcing the parts, front panel difficulties and documentation issues. Total cost: 360 euro

This list is mostly from http://xlargex.xl.funpic.de/ All resistors #'s refer to the original TR-808 schematic, not the MB-808. 1. Noise level: 22k(B) Potentiometer for TM4 (22kb) 2. SD: On LT decay: "Snappy" Decay C51 (0,47µf) changed to 2,2µf - in parallel a 250k (B) Potentiometer with a 22k Resistor in series x3. "Snappy" Filter: 22k(A) Potentiometer + 1k in parallel for R201 (22k) 4. Clap offset 10kb 5. CP: On HT decay - clap reverb replace r376 (47k) with 100k pot 6. MA Filter: 22K(A) Potentiometer for R339 (3,3k) 7. Rim/CL Filter: 500k 22K(A) pot for R322 (10k) +8. Cowbell Tune 1: 220k(B) Potentiometer for TM1 (220kb) +9. Cowbell Tune 2: 220k(B) Potentiometer for TM2 (220kb) 10. CH Filter: 22K(B) Potentiometer for R153 (2,7k) 11. OH Filter: 22K(B) Potentiometer for R147 (2,7k) For the noise inputs, I simply routed the white and pink noise out to a switched jack. I wanted to do the clap timing mod but Erich Archer never replied to my request for info: http://ericarcher.net/devices/clap-raca/ Upon close examination of his pics, I assume it's some pot between a 4.7k resistor going into what I think is pin 2 of the LM339 comparator. I don't have the energy or patience to explore this myself. I'm just happy it's done!

MB-808!!! only one take. all you hear is 808. the tones are from self-oscillating congas. (ghetto-ass distortion + compressor + reverb, all courtesy a roland m-16dx mixer) mods: - white noise level - snare decay - snare filter - clap offset - clap reverb amount - maraca filter - rimshot filter - cowbell tuning x2 - open hat filter - closed hat filter - white & pink noise input jacks (not used in this demo) - schmitt trigger input jacks (not used in this demo) - self-oscillation of congas when the decay is set all the way better pics coming soon...

Excellent :) Just to be certain, this sequence-change quantization should apply to both switching between A/B/C/D -AND- when loading new patterns (in pattern mode). I have one other request which may not be easy, but definitely interesting: Last four "Assign" options: Repeat note on the 1st next step, 2nd next step, 3rd next step and 4th next step... like a MIDI-delay, eg. Original: x - - - - - - - x - - - - - - - Repeat 1: x X - - - - - - x X - - - - - - - Repeat 2: x - X - - - - - x - X - - - - - Repeat 3: x - - X - - - - x - - X - - - - Repeat 4: x - - - X - - - x - - - X - - -

During playback, how do you configure the MB-808 so it switches A/B/C/D sections or patterns (when selected by hand) at the *end* of a bar and not immediately? I tried various SPM settings with no luck.

Does anybody have the MB-808 errata info? It used to be on the board which has been down for a while.

I have a batch 2 kit and no encoder came with it ...I thought those were ordered separately. Which one do you have? I agree with Nuke's assessment that you should recompile the firmware with different encoder settings.

Ich habe keinen Layout gemacht - ich habe nur die Platte bei http://www.frontplaten.net bestellt. Zusammen kosten die 2 Teilen 75e:

1. Die waren *nicht* geprüft aber ich denke, dass diese .fpd & .dwg Dateien mit Frontplattendesigner funktionieren: http://sneak-thief.com/mb808.zip 2. Vor 2 Wochen habe ich eine MB-808 Frontplatte von http://www.frontplaten.net/ bekommen - die ist genau wie TK's:

That looks great! Such an economic layout - and ergonomic too. How much did those backlit switches cost?

Roland M-16DX. http://www.roland.com/products/en/M-16DX/ You can find them used for 250-300 euro or $300-350 US. Nobody knows about this thing - it's got 18 simultaneous analog to USB adc inputs, so you can record every single input AND record the main mixdown. It also works as a midi control surface. The cool/strange part is that you can move the mixer around anywhere you like because it's only connected with one data cable to the I/O box.

Found :D

Anybody have a spare PIC core pcb lying around? It can be with the components or blank - either is fine. Maybe you upgraded to the ARM32 core and don't need the old one anymore? ;) Send me a PM or email at michel (at) sneak-thief (dot) com.

Thanks! :)

Can somebody please post the MB-808 build docs such as MB-808_Manual_Beta.pdf that used to be hosted here: http://webcache.googleusercontent.com/search?q=cache:PbNjjBmIx7cJ:www.eight-oh-eight.org/Assembly/+mb-808+assembly&cd=2&hl=en&ct=clnk&lr=lang_hr|lang_fr|lang_de|lang_en&client=firefox-a&source=www.google.com Thanks! (I checked and there's no archive.org backup of the site)

yeah, i called it: http://eight-oh-eight.org/phpbb2/viewtopic.php?t=583 ;)

Yes, it's that easy :)

Funny you mention the Neutrik one - I just posted about it a couple of days ago in the big GM5 thread. It's awesome!

Got mine and installed it in my MIDIbox sneaquencer: http://sneak-thief.com/sneakyseq.html :) After searching for a nice case-mountable USB-jack, I settled on this reversible Neutrik USB Type-A / Type-B one: Re. Driver: I used the v1.07 of the windows 32 bit driver and all the ports were names were *so* long ("Ploytec USB Midi blah blah" or something) that the port numbers at the end of the name wouldn't show up in the Midi-Ox dialogue boxes ...and I couldn't rename them using the .inf file! I rolled back to v1.06 and everything worked perfect. Just Phil - greetings to Edmonchuck! Lived there for the first 27 years of my life.

Thanks - it works now! I took out "static char fx_status;" and instead used "unsigned char fx_status;" TK - Yes, it fully compiled before and there weren't any more error messages. However, it didn't pass sysex correctly anymore.

I updated my toolchain for the first time in a few years to the latest versions - now the midi sysex forwarding routine doesn't compile anymore http://www.ucapps.de/mios_c_forward_chn1.html This is the error I get: "main.c:236 warning 94: comparison is always true due to limited range of data type" In reference to this line: if( fx_status != 0xff ) ///////////////////////////////////////////////////////////////////////////// // This function is called by MIOS when a MIDI byte has been received ///////////////////////////////////////////////////////////////////////////// void MPROC_NotifyReceivedByte(unsigned char byte) __wparam { static char fx_status; // normal MIDI events are forwarded in MPROC_NotifyReceivedEvnt // this function handles sysex and realtime messages if( byte & 0x80 ) { // Status message if( byte >= 0xf0 ) MIOS_MIDI_TxBufferPut(byte); // transfer byte // determine status if( byte == 0xf0 ) { fx_status = 0xff; // forward until 0xf7 } else if( byte == 0xf7 ) { fx_status = 0; // f7 reached, no forward } else if( byte == 0xf1 || byte == 0xf3 ) { fx_status = 1; // expecting one additional byte } else if( byte == 0xf2 ) { fx_status = 2; // expecting two additional bytes } else { fx_status = 0; // expecting no additional byte } } else { // check if fx status active if( fx_status ) { // forward data byte MIOS_MIDI_TxBufferPut(byte); // decrement counter if required if( fx_status != 0xff ) --fx_status; } } } I even tried starting from scratch with the simple Clockbox v1c and replace the empty "When a MIDI byte has been received" routine and I still get this error.

DETENTED5 funktioniert ganz perfekt :) Super!!! vielen vielen Dank. ich befürchtete, dass ich neue Encoders in meinem 'Sneaquencer' aufzubauen musste ( http://sneak-thief.com/sneakyseq.html )