jeudi 29 février 2024

8 outputs MIDI CV/GATE converter : come back from long months of development!

Yes, because when I started this project, I thought I would create a minimal and fairly precise circuit, let's say, easily.

However, during the entire development, testing and modification phase, the appearance of the circuit no longer resembles the original one at all.

The first concept looked like this :



Unfortunately, despite the very interesting characteristics of the digital/analog converter, in fact, the converter used was not at all linear: a real disaster.

So, I redeveloped the system by changing components and also functions. I added the possibility of choosing the possible output values.

  • from -5V to +5V
  • from 0V to +5V
  • from 0V to +10V

So I developed a second prototype corresponding to my new specifications and compatible with the euro rack format :


I wanted the board small so I condensed the component placement and also chose the smallest integrated-circuits possible. Unfortunately, this time again, an operational amplifier supposed to give a rail-to-rail output, started to 'go crazy' as soon as it reached 90% of its power supply. 

So I had to once again modify the concept and implement more 'traditional' circuits. And finally, I enlarged the circuit board because I found that the space between the connectors was not at all convenient.

Which ultimately gives me this:


And finally...

After all these modifications, tests, brain storming, I finally obtained the functional and above all precise outputs : 


The very slight imperfections on this graph come from the change in range of my multimeter and not from the voltage converter. The result is perfect!

So, I was finally able to connect one of my keyboards to this converter and test the small rack modules that I had purchased for the occasion.


Ah, the charm of analog!!! 

I therefore noticed that the VCOne modules do not all respond in the same way, and that the 1V/octave precision that they offer is sometimes completely crazy. 

With certain modules, I manage to have a coherent range over more than two octaves, with others, in particular the '3340 VCO EU' module of Electrosmith, it's total madness. I can't even get a coherent octave. The module would have to be completely recalibrated.

I now understand better why 'small' 'not too expensive' synthesizers often only offer a 3-octave keyboard, and why 61-note analog monsters like the Moog One cost so much!

I can now calmly continue programming this system, and make a front panel allowing me to install it in a eurorack location : Yesss, I see the end of it!

Later...

Yes but... With a correctly calibrated module, in my case a VCO2 from PICO, plus a little VCF and controlled by a VCA triggered by an envelope generator, all passing through a small reverb, that's for sure, despite the 'disadvantages' of analog, sound is still something special!

Later, again...

If everything goes as planned in terms of programming and I have successfully implemented everything, this is what the front panel will look like :




vendredi 16 février 2024

STMicroelectronics? As usual :-(

Almost two years ago now, I decided to leave STMicroelectronics processors to try other suppliers.

The reason? I don't really like the GUI of the STM32CubeIDE. And then also, the update system was starting to crash. And I don't like it at all.

I, 'unfortunately', intend to use a product from STmicro again for an extended version of my MSX-USB cartridge. Certain characteristics of ST microprocessors are interesting for this project.

So I downloaded and installed the latest version of the software for Windows, namely, Version: 1.14.1. The software installed without problem. : 


Obviously, the problems arose immediately, when I wanted to try a 'Hello World' on a processor from the ST32F722X range.



And yes, the system wants to install the latest version of the package specific to this processor. Obviously, this doesn't go well:



Well, as I think it is preferable, when embarking on a new project, to start with up-to-date resources. I used this legitimacy to attempt the upgrade in 'do it yourself' mode, like lets him suggest the message.

And there, welcome to the world of inconsistencies of all kinds with stupid messages, incoherent since emanating from automatic procedures which do not know what you are doing, given that you are leaving the world step by step programmed, and, as you are a human and you have no idea how this automatic procedure works, you do not know in what precise order to carry out the operations and you therefore do anything.

As a result, with each operation, the 'automatic' procedure makes progress, when something has been presented to it at the right time, without telling you, obviously, but the result still crashes. But, you have moved forward without knowing it. So, you try something you've already tried and this time, it works.

In short, all this because of update files or patches that the automatic procedure does not find when downloading, telling you tersely 'network error', while these files can easily be downloaded 'by hand' from the site of ST.

All, of course, with the crappy ergonomics of Eclipse!
ST: borring as desired.

After a... certain time..., in my case, more than two hours, I finally managed to configure two pins of a processor and start generating the source.

Another stupid thing: you can't generate the source from the processor's graphical configuration interface . 



There's no button for that. No, this is done when you save your configuration. As a result, and as all the source is conditioned by tag type comments where you must insert your code under penalty of future inconsistency between the source and the processor configurator (MX), and if after an error you wish to regenerate the source , well you need to modify a pin to make the backup possible and therefore, at the same time, the regeneration of the source.

Potentially understandable in certain conditions, but in reality, it's completely stupid and annoying.

In short: STmicro!




jeudi 15 février 2024

A flash cartridge downloadable from USB, after the first tests in real life : Ready to produce the new prototype

 


I also had to change the three MSX interface integrated circuits which are no longer available for assembly on a production line. 

I can't wait to receive this new prototype and make the few software modifications...


vendredi 2 février 2024

Commander X16, MSX, 8-bit Microcomputers

Since I am of a 'certain' age, I can say that for me, computing stopped somewhere between the end of the 80s and the beginning of the 90s. And yet, I worked in the 'either-called' IT field for almost 40 years. I stuck with boring IT all these years :-(

I obviously started with 8 bits, unfortunately I entered 'professional' mode at the time when 16 bits (Atari, Amiga) were starting to emerge. I found these machines fantastic compared to the IBM 'dung' and other compatible machines of the time.

I must say that for me, the computer cannot be seen as an object of mass manipulation as is the case today, but as an object allowing the expression of a certain art, and must therefore be to this service, and not the other way around.

OK, this being written...

I've long thought that 16-bit recreations would eventually create an alternative market. Actually no. A few FPGA creations from Atari STs or Amigas and then, nothing very concrete.

On the importance of community! A few years ago, I became interested in the MSX standard. Because it was the only architecture resembling a sort of 'standard', used in the world and open enough to allow a single person to create thanks to and with this type of machine. 

So I made a copy of Sergey Kiselev's MSX Omega computer. And, one thing led to another, I found myself creating material around this machine.


I also watched with great interest the videos of The 8-Bit Guy aka David Murray and his concept of a dream computer.

Wikipedia

After several years of questions, controversies and prototypes, an electronic board is finally available. At the time of these lines, this is the second version :


So I received a small package containing an order placed at the end of last year.

The content contains the bare minimum:


I must admit that I did not expect to order a PS2 keyboard and mouse more than 30 years later!

As for the motherboard, I must say that it is absolutely professional quality. I remember the first PC and PC-XT motherboard. This has nothing to do with. And I love the three ISA type connectors allowing the construction of various expansion boards.

In my opinion, this is what is missing from the MSX standard. But that can change ;-)

The Commander X16 is a machine in the spirit of the Commodore 64/128 based on 6502. The MSX is based on Z80. I think I'm going to stop at these two machines, it already represents a lot of work developing on these two platforms!

There is also the AgonLight machine, but I must say that it interests me less. It doesn't really have any means of extension. I find it too closed, but that's my personal opinion. If a new version is developed with what I consider to be the bare minimum, I might be interested in it.


In this post, I have only touched on the subject. These three machines involve the efforts of many people. I thank them here for their inspiring 'work'.

GoWin FPGA, why?

I discovered GoWin FPGAs a few years ago. I have nothing to say about these FPGAs. They work well, the software also works correctly. Everything is for the best in the best of all possible worlds then?

Hmm, not really! I ordered these FPGAs at the beginning of 2021. And, thanks or because of the shortage of components at that time, the price increased by 50%. Which makes these circuits, how to say... less interesting to use now.

Indeed, I design electronic systems for the 'enlightened' amateur domain, and not for the professional domain. The spectrum of manufacturing prices cannot be high. I therefore have to use components with the best quality/price ratio.

This is the reason why I decided to test a new family of FPGAs. I chose the Efinix brand and the FPGA TRION range. I know absolutely nothing about these components. Before ordering a development kit, I tested the IDE and, despite a small compilation problem, quickly corrected with a patch provided by Efinix, I was able to compile a previous Gowin FPGA project without problem, namely the FPGA implementation of the µPF1. 

Although for the moment I have not produced anything concrete with these new components, I believe I was able to successfully complete the 'feasibility' tests.

So, I ordered an Efinix kit:


The adventure goes on ;-)

A flash cartridge downloadable from USB : after the first tests in real life.

I had the opportunity to provide a prototype cartridge to Aoineko, the maintainer of the MSXgl graphics library : https://aoineko.org/msxgl/index.php?title=Main_Page

The cartridge works almost as expected. I have planned an improvement in the management of the USB bus because the cartridge tends to disconnect from time to time to his PC. 

I also added the possibility of declaring the cartridge in 64KB mode automatically. In fact, it is the cartridge itself which adapts according to the size of the file received from the PC.

When I studied the cartridge, I did it from the MSX computer I own, which is an OMEGA. So I placed the connector far enough away from the surface of the OMEGA. But this is not enough!

Oups!

So I placed the USB connector higher and took the opportunity to modify the placement of the processor which manages the USB bus. I wasn't totally happy with the USB signal tracks on the previous board. It's now much better. I also placed a precise 8MHz oscillator on this processor. This should make the USB bus work better.


On the other hand I also have cartridge cases. All I have to do now is have this cartridge version made.



lundi 15 janvier 2024

Happy new year everyone.

I hope 2024 brings many good things to as many people as possible in these difficult times, and especially to me because I think this year will be decisive. 

After leaving my crappy job in national education, I now also have to leave France, a country which is falling apart more and more quickly.

And to start 2024 off right, here is a little photography of the finished universal clock, finally equipped with a very beautiful red filter, which makes the displays readable in all circumstances :

As I explained in a previous article, this clock is based on the operation of the Texas Instrument TMS1122 processor.

It allows the programming of four outputs to the nearest second, over a week.

This can obviously enable the operation of any type of device with great flexibility. But it can also, and above all, allow load shedding from the domestic electricity network by programming large energy consumers so that they operate one after the other and not at the same time (when possible of course), in order to avoid strong current draws.

This makes it possible to subscribe to a lower electricity subscription and also to start certain devices during periods of the day when the billing for the KW consumed is the lowest. So that there is possible financial savings on the subscription AND on consumption.

In order to carry out some possible sales tests, I put this clock for sale on Etsy. I have had a good number of visits over the last few weeks. The presentation of the object on the Etsy site is not good at all. But it was for a first try. I will have to review my copy on this site now.

Creating a well-thought-out, well-finished, precise and reliable product is not an easy task. It takes a lot of time but the result is there...