Search options (Page 134 of 137)

Heos: Will not work! The controller is a slave device and both the control deck and the USB adapter will try to be master. Will not work (as expected)

Sending this off to the handhelds section.

Oh, that's a nasty heap of wires... If I had designed it, I would have used ribbon cables, and maybe daughterboards for each line of potentiometers.

Ts ts!
I O, you know...

Speaking of the infinity cart... My opinion on this cartridge is that it solves a problem that doesn't exist. The problem I saw in BP's sav is unique to EMS cartridges. It might be a glitch in their CPLD (Memory bank controller emulation) Regardless, the most common glitch I see in savs people send me happens because the cartridge looses its connection to the cartridge slot or that the Gameboy loses power and turns itself off. When thi happens, the Gameboy CPU writes the values 00 39 all over the memory map. As a side effect, up to 8 kB of the external RAM may be overwritten.

This problem is not at all related to the cartridge battery, so the NVRAM in the Infinity cart won't help a single bit in this respect. But, what's even worse... The original Bleepbloop cartridge (non-USB, and USB as well I think) had an MM1134 RAM protection circuit which would disable access to the SRAM if the external power went away, saving it from that type of glitch. The infinity cart cannot use this circuit as it's not battery powered. So, unless Jose has come up with a clever solution to the problem, this cartridge is actually more vulnerable to this type of glitch!

In my opinion, the increased component cost and the doubtful benefit of the NVRAM makes me wonder if it was a wise decision to go for this new design. Especially since the battery holder makes it so easy to swap the battery. (And with my upcoming LittleFM file manager, sorry for the wait, it'll be possible to store up to 8 full 128 kB savs in flash memory, which is truly stable in every respect, on any bleepbloop or smartboy cartridge.)

Low-gain answered most of your questions, but I'll address this one.
1 V/oct is a widely recognized and used standard in modular synthesis because it makes things easier if oscillators (and filters, not to be forgotten) follow a common standard. There are others, less common ones as well, such as Buchla's 1.2 V/oct, which elegantly translates into 200 mV/semitone. (Don Buchla is a famous pioneer in modular synthesis who started his career in 60's. Another name to look out for is Serge)

However, 1 V/oct (1 V/some arbitrary interval) is not a magic intrinsic property of oscillators. The oscillator needs to be designed and tuned for that standard in order to obey it. Often, a high precision oscillator has two trim potentiometers for tuning, of which one is for fine tuning in the higher octaves. This is needed because the oscillator is likely to deviate from its voltage/frequency curve at higher frequencies. Another problem is temperature instability - the oscillator detunes itself once it warms up. This problem is solved by thermocoupling  a heat sensitive resistor  to the critical transistor, to compensate for these effects. Thermocoupling means that the components are placed in such a way that they keep the same temperature, in the case of this resistor, by joinng it to the transistor with a blob of thermal grease.

Now I'm not expecting you to understand all of that immediately (hopefully you will one day ) But rather I'm trying to explain that getting a high precision or even low precision a 1V/oct response is not a trivial thing but requires it to be a part of the design. Specifically, making an APC do 1V/oct would likely require some sort of voltage scoling.