TK.

I deleted most of the old releases since I feel a bit sheepish when everybody is able to find the bugs I made in the past... ;-) What is so interesting on the old stuff in particular? Best Regards, Thorsten.

I still don't get it... You are selecting one of the drivers (they don't run in parallel) The clock line is always connected to pin RD5 The data line is always connected to pin RC4 The CS/RCLK line is always connected to pin RC5 in these two configurations. Of course, you can connect the modules to other pins if they are free or if they are allowed to be shared, in this case you only need to change the constant values of the #define's Best Regards, Thorsten.

The PGA2311 provides a serial interface with Chip Select, SDI and SCLK line - so it's very similar to the MAX525, which means, that there are assembly and C based drivers available which you could use as a template. From the MB side you only need a core module... Best Regards, Thorsten.

I don't understand this question - why shouldn't it be possible? Best Regards, Thorsten.

Yes, this DIN mapping works with the first 8 shift registers. Each encoder entry just overrules two DIN pins, so no problem here Best Regards, Thorsten.

After the protection buffer is an offset of ca. 4V, so yes: it's possible to use a LM324, but I'm not sure if it gives you the same results like a LF353 Best Regards, Thorsten.

As Stryd_one mentioned, the purpose of the "Register Clock" is different than the purpose of a "Chip Select" line. At the end it doesn't much matter, to which pins these signals are assigned, you just only have to take care, that a chip select line is active during the whole transfer (AOUT module), and a register clock is triggered after a transfer (AOUT_LC module) You could even call the "RCLK" pin still "CS" pin - the name is not important (it could only confuse others if they read the code) Stryd_One's code is correct, but you have to take care, that only the 16 bits of sr1_value are shifted out Why different defines? Because for different MIDIbox variation, where pins are sometimes allocated by other functions... Best Regards, Thorsten.

1) detented sind mit Rasterung, non-detented ohne - die Rasterung ist hilfreich, wenn man Werte exakt einstellen moechte, aber hinderlich beim sweepen. 2) nein 3) gar nicht Gruss, Thorsten.

Hallo, bei einem Drum Clone kommt es ja nicht so sehr auf Genauigkeit an, oder? Dann koenntest Du alternativ auch einen Spannungsteiler als Referenz hernehmen. Bspw.: [tt] 5V |----/\/\/\/\/\----*----/\/\/\/\/\----| GND 1k | 470 | Vref = ca. 2.35V [/tt] Gruss, Thorsten. P.S.: ich glaube, es ist wirklich Zeit, eine Alternative zum MAX525/MAX6007B zu bieten...

Did you try it? Best Regards, Thorsten.

This is my personal macro to do something if a bit is set or cleared - it improves the readability. The macros are defined in macros.h the difference between IFSET and BIFSET is, that BIFSET works also with BANKED registers >= 0x80 <= 0xf7f It leftshifts the constant n times. e.g., "0x01 << 3" results into 0x08" - this is also to improve the readability yes, I cannot say, where changes have to be made - there might be different places. Better not to change the pin layout of the GP buttons, especially if you want to keep it compatible to MBSEQ V3 You can change the mapping in main.asm, USER_ENC_NotifyChange you can achieve the same (and more) in the song menu, when song mode is turned off (press song button twice). You can predefine pattern sets and chains at different "song offsets" (A1, C1, E1, ...), and select the offsets with the 16 GP buttons during the sequencer is playing. I think that I've described this mechanism in the ChangeLog Best Regards, Thorsten.

Moxi, I don't understand why you are not using the original code of aout.c and doing the changes there? AOUT_Load2SR() has to transfer 16 bits instead of 32, it has to service the RCLK pin instead of the CS pin (compare with aout_lc driver) AOUT_Update() has to select the values which should be uploaded to the shift registers... and thats all! Best Regards, Thorsten. Best Regards, Thorsten.

You have to ensure, that no unused encoder entry overrules the pins, which are connected to buttons (if you are unsure: remove all ENC_ENTRY lines of encoders that are not connected to your box), and you propably have to adapt the DEFAULT_DIN_SR_PIN_* settings, if you have changed them in the meantime Best Regards, Thorsten.

This is not possible with MBSID V1, it will be possible with MBSID V2... Best Regards, Thorsten.

Please read the wishlist, it's everything in there! Best Regards, Thorsten.,

You are using macros, which are not defined, and which do mostly not run on gpasm anyhow. Replace "rgoto" by "bra" IFSET by "btfsc ..." (see macros.h of a common assembly based application) IFCLR by "btfss ..." (see macros.h of a common assembly based application) and if there are still errors, search for other macros I forgot here. In general, the savest thing when using inline assembly functions is not to use macros at all Best Regards, Thorsten.

thats the reason why I will wait at least for one month before ordering samples ;-) Best Regards, Thorsten.

When you are starting the "make" file from the command shell, you will see the errors on screen. In addition, they can be found within the listings, which are located in the _output directory Best Regards, Thorsten.

Really accurate work! :) Best Regards, Thorsten.

R2 is an older revision, so it has to be connected like described at the _old page. There is an interconnection test application available, which you can use to ensure that you made everything right. Best Regards, Thorsten.

All serial connections from MBHP_SID::J2 1:1 to MBHP_CORE::J10, only CS# has to be connected to a seperate pin of the core (propably MBHP_CORE::J14) - thats all. By connecting the CS# line of the second SID together with the CS# line of the first SID, you are already able to play both SIDs the same time (with the same parameters...) Best Regards, Thorsten.

I think that even without automation feature, this is a really useful circuit, because it saves you from buying an expensive mixer just to merge multiple audio outputs! Thanks Altitude for testing this (I haven't commented the circuit and layout, since it just looks good! :) Could you please bring this into the Wiki once you are happy with the result? Best Regards, Thorsten.

Hi Tim, I'm using a detented panasonic encoder for the datawheel of my MBSEQ (therefore it is DETENTED and not DETENTED2 in mios_tables.inc) - it works very stable since two years. Thats all I can say. But Alps are pretty good as well. The detends are very important for the sequencer for exact value adjustments - yes, non-detented are better for MBSID/MBFM (only exception: the datawheel should be detented) Best Regards, Thorsten.

no problems from the programming side, but do you have C skills? Best Regards, Thorsten.

As stated in the wishlist, a first release cannot be expected before next year, some other projects should be finished before. Best Regards, Thorsten.