White Flame,

Thank you for your observations.

I have in mind a perhaps small stand-alone, real-time computer (not a microcontroller) that is entirely interrupt driven and has no poling, despite the number and nature of the peripheral devices (USB ports included).

Such a device would be especially appropriate for robotics, sound CAD, music composition, music performance, and a seed platform for games (add your own DSPs). I have no doubt that it would find many other applications.

I don't believe in disks. A flash drive can currently hold two terrabytes of data. That amounts to the data in 425 DVDs. The cost is on the way down. Besides, we are inclined to dump any mass storage that requires moving parts. This will happen eventually.

Also eventually, robots will be closed-loop devices, like all animals are. This means that the cost of profitable robots will go to where people who are only rich enough to buy a house and a car can afford automation for small operations.

See https://www.futurebeacon.com/EffectiveRobotics.htm

Jim Adrian
https://www.futurebeacon.com/jamesadrian.htm

You can get a Roomba (a closed loop, mass market, utility robot) for $375. (They go up from there.)

If we've learned anything, the cost of things like robots and the like will not be the electronics. We've managed to make the task of converting raw materials in to finished goods pretty cheap.

Most anything is running on flash now a days anyway since they make them backwards compatible with disks. I only have one drive left in one server at home, all our phones, tablets, laptops, etc are all flash drives.

Now robotics I go a different route. It was around 1990 that I saw MIT's robot insects. Small walking robots was not new but they were running small neural networks with a feed back sensor to close the loop there. The bots would power up just flailing legs at random but once the sensor detected forward movement it would start varying the randomness to try and trigger more forward movement, in essence learning to walk.

I started later when we got a PIC programmer at school with the 16c54 and 16c71 to make each PIC into a 'neuron' where they could all perform small tasks and communicated with each other and had the ability to perform reflex actions as well as let the higher up the chain send simple commands like walk and the downlevel ones would do all the coordination to walk which freed up the higher level to do other tasks. Never did finish it though.

I read an article recently about how the "robots" that are communicating with each other, are doing so with languages that they made up themselves.

It was more a discussion of, like your insects, we "don't know what's going on" inside of these devices, since many are self trained networks. Now they're communicating in languages we don't understand.

Skynet is pretty close

I graduated high school in 1991 so around 1990/91 my robots designs were 6502 based of course. I bought my Amiga in college so my designs went to 68k then later the PICs came out I wrapped the pic based neural network around the 68k core. I ended up buying Parallax's PIC programmer though and moved to MSDOS/Win31 on a 286 to use it and got rid of all my 6502, 8085, and 68k stuff later on. Fast forward 20 years Now I've bought a pair of 6502's and 8085's again and have an (non-working) Amiga in the closet and haven't used Microsoft stuff at home now in over a decade.

Not being able to know what the computers are saying is a little scary. I'd hope there was a way to extract at least some of the language out of them. That way we wouldn't be totally clueless.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

I use USB and bluetooth a lot now a days but use bluetooth to serial or usb to serial adapters

The main USB problems are on the win/mac side, you have to find the driver versions to don't try to brick counterfit chips but falsely detect genuine chips as counterfit and prevent them from working. Since I run linux as the host/dev side I don't have those issues, usb to whatever just works.

I 've done a couple handheld designs as well. My 8085 we had to make for college is in a handheld formfactor and I was starting with PIC's to make data acquisition modules for my 48sx.

Let's just hope Skynet doesn't use Siri as their liaison to humans.

_________________
Cat; the other white meat.

I tend to stay as far away from anything apple as I do anything microsoft, they both seem to have an equal amount of issues.

EugeneNine,

The USB ports on a 6502 series device do not need to be related to or compatible with any other computer. An open platform could very well help change what people use. Lead, don't follow. They can write translation software if they want to follow us. They had their chance to stand for service to the customer. They are just trying to dominate and monopolize the market for every kind of computer use.

Jim Adrian

https://www.futurebeacon.com/jamesadrian.htm

_________________
Jim Adrian

https://www.futurebeacon.com/jamesadrian.htm

I'm talking about the other side. When you have a windows/mac computer then use a usb to serial adapter then talk to a uart on your 6502 system. I see the same issue over and over, doesn't matter weather its a 6502 on the other side or a HAM radio or FlashForth on a PIC or an arduino or an HP calculator. There are clones of the ftdi/prolithic USB to Serial chips and the win/mac drivers are regularly updated to try and block those clones so the person using trying to talk to the 6502/radio/etc goes on the support forum/list and says its not working and the forum says roll back to driver version x. I don't run into any of those problems and use USB to serial all the time because I'm running Linux on the dev side so there are no political updates to the drivers.
It makes an impact going back to the original question of someone needing to spend $ on a 'PC' to develop for the 6502. You can get free 'PC's from the trash, load a free open source OS and then download free open source development tools and start developing for your 6502, there is no need to spend $ on an OS, tools, etc. The OP was eluding to the fact that he didn't want to invest in a development 'PC' and would rather have a display and keyboard and storage on his 6502 system, but there really isn't a need to do it that way.

_________________
http://WilsonMinesCo.com/ lots of 6502 resources
The "second front page" is http://wilsonminesco.com/links.html .
What's an additional VIA among friends, anyhow?

Does anybody here know how to write into a flash drive?

Jim Adrian

What nonsense is this? Of course they have to be compatible, that's why you call them USB -- so that they meet the standard for the communications and can interoperate. USB is an open industry standard. If you just want to use it for its connector, then, be my guest, but it's a disservice to the public, as they may rightly expect some kind of common USB connectivity when they see a USB port. Not quite sure what sinister cabal you're talking about here, being as USB is, you know, ubiquitous.

USB was a novelty back in '97 when Apple jumped on it for keyboards and mice and dumped the floppy drive. Today it's lingua franca.


What kind? They're all different. Generic flash chips have their own unique protocols. There's IDE adapters for Compact Flash, SD Flash Cards support both SPI and a more complicated parallel interface. USB sticks are, well, USB Mass Storage Class devices.

I think someone recently was struggling getting SPI to work with some random cards, someone else has an IDE CF reader board.

All of those are "block level" access to the raw device.

If I was going to add a flash drive capability, I'd look to mount something like the CH376, which gives high level SPI access to DOS formatted USB sticks. That way I don't have to implement a file system myself. There's breakout boards for this thing everywhere, but I can't find the actual chip. Otherwise I'd look to an SPI interface for an SD Card.