Commander X16

Hi everyone, my name is Marc, and like many of you, if I'm here it's because I'm very interested in the development project of the Commander X16 computer. Thanks to David Murray (the 8-bit Guy) for this great idea, I know that a dozen 8-bit generation computer nerds have greatly contributed to the development of the prototype, and I firmly believe that the Commander X16 will have a mega dazzling success despite the fact that the 8-bit area is no longer at its climax as was the case in the late 70s and 80s. Indeed, there is still a huge market in the 8-bit round field, and I am even convinced that we are this generation of programmers who, despite the technological evolution of recent years, we are among those who will give back a new breath of life to this ancient period of history and for good reason.

As a programmer, I realize that today new technologies make it much easier for us to design applications and games in 8-bit format. The advantage we have as an 8-bit generation is that we have learned to think and then make the right decisions that apply before coding a program. We were forced to learn to program in assembler for optimum results, although it was a bit difficult, the fact remains that we exercised our logic much more at that time than today.

What fascinates me about the 8-bit generation is that when you've gotten your code to work, there's a certain satisfaction, because you know you've learned something by applying yourself to learning the hardest part. One thing is certain, I will be involved in the development of games for the X16. I recently downloaded KickC which is a C language cross compiler, which in my opinion, its code is more optimized for the 6502 family processor range. Even more so than any other compiler by the way. In fact, the code it produces is very close to code written directly with an assembler.

So don't hesitate to discuss with me, I thank once again the admins for this forum, which I am convinced is full of essential resources that I will certainly need eventually.

On 12/26/2022 at 8:28 PM, the8bitgen said:

We were forced to learn to program in assembler for optimum results,

Due to the 6502 being optimized to fewer transistors and thus better yields. There were a minimum of registers and addressing modes to play with too. Making the challenge even tougher.

Hi Marc!



Welcome to the community. What part of Canada do you hail from? It's a big country.

On 12/26/2022 at 11:28 AM, the8bitgen said:

As a programmer, I realize that today new technologies make it much easier for us to design applications and games in 8-bit format. The advantage we have as an 8-bit generation is that we have learned to think and then make the right decisions that apply before coding a program. We were forced to learn to program in assembler for optimum results, although it was a bit difficult, the fact remains that we exercised our logic much more at that time than today.

I'd suggest expending this thought to how the hardware design of the X16 (and other retro platforms) has come from being a product of thought and hindsight rather than pushed out the door as good enough to generate a buck.The tools available such as schematic CAD and 3D modelling have certainly changed the game; the X16 is a computer designed in someone's home for another hobbyist/enthusiast rather than just to get sold into a household with no computing interest. I believe today the whole social media/communication scene is why most walk around with a device that is based on purpose. Nearly everyone who uses a smart phone perhaps less it as a means to a purpose, rather than being the purpose in itself. I spiral into a form verse function debate....

Perhaps being thoughtful ahead of time was a factor given limited resources and the effort to implement the idea (program) back in the day. Today the effort is much lower and the amount of resources much more available, changing what is done. Consider Youtube - anyone can make a video for others, but not everyone should. The number of channels are infinite, the number worth watching are very finite .? Building a custom 8 bit computer in the 80s was out of reach for me (I was a teen then) but assembling an X16 and customizing it is well within my reach when it finally goes into production. 

Anyway, I hope that you take part in the community and I look forward to trying one of your games someday.





 

 

 

 

On 12/27/2022 at 3:57 AM, neutrino said:

Due to the 6502 being optimized to fewer transistors and thus better yields. There were a minimum of registers and addressing modes to play with too. Making the challenge even tougher.

I'd suggest that the limited instruction set, registers and addressing modes made the 6502 easier to learn than other processors of the time. Those limits do require one to manage the limited resources efficiently, which might be seen as harder to do.  

Well.

I think that what made the 6502 popular was the fact that MOS Technologies had the balls to sell it for 25 bucks when OTHER processors sold, that were no better really, and had instruction sets that were no fancier.. for more than quadruple that. Simply because the "other guys" COULD and were happy with keeping the market limited to keep the price up.

Chuck Peddle went to the people and asked "What features do you want?" And they said, with one voice, "We want a simple processor that WORKS and is cheap. And with that, we can build the damn world" (Ok, I added that last part.)

The 6502's instruction set isn't limited. It's perfect. Oh! Sorry, no Multiply instruction? No Divide instruction? Well guess what, Tex! MATH doesn't have those either!  MATH makes you learn how to multiply and divide through algorithm using simpler steps. You probably learned how to do that when you were 8 to 10 years old. Sure. There's a symbol for multiply... but it's backed up by proof using much simpler rules. Multiplying in binary is the same as multiplying in base 10 like you learned as a kid, it's just in base 2.

The secret to learning to really effectively use tools like the 6502 is to learn how it works, and work with it... not beat it into submission with a fancy compiler that can just multiply and divide like it doesn't care. If you can do that, you can hold up your work and feel like the Kwisatz Haderach while at the same time, discovering the joy of being able do develop complex systems WITHOUT needing a budget of millions and dozens of developers under you.

On 1/2/2023 at 3:11 AM, Daedalus said:

I think that what made the 6502 popular was the fact that MOS Technologies had the balls to sell it for 25 bucks when OTHER processors sold, that were no better really, and had instruction sets that were no fancier.. for more than quadruple that.

Actually it was worse than that. At the Wescon trade show in San Francisco on 1975-09-16 everybody including MOS Technology were forbidden from selling anything so they set up shop nearby at the MacArthur Suite in the St. Francis Hotel. And the price was set to 25 US$, which this invoice from 1975-10-29 shows. https://www.pagetable.com/?p=1295

The Motorola competition reported their previous price in 1975-10-30 at 175 US$ https://commons.wikimedia.org/wiki/File:Motorola_MC6800_microprocessor_ad_1975.jpg

So it wasn't 4x, it was 7x the price. I never really thought much about the 6502 processor during its heydays but I came to appreciate in later years what a monumental change it brought. Given that it smashed the price 7x in 1975 prices. And was capable of addressing 64 kB when so much memory was insanely expensive. Which meant it could handle "all" needs. On top of this it had builtin clock to make running it simpler and essentially free documentation where other corporations held the documentation out of reach for people. It's also a very efficient processor that seemingly is faster than the 7 MHz MC68000. So it made powerful computing available at a decent price.

In 1975 building DRAM most likely meant the MK4096 DRAM chip. Which meant a full 64 kByte would have costed 4224 US$. https://jcmit.net/memoryprice.htm (using the pricing of the Altair 4K DRAM board as a kit). However using a 1975 memory board sales with 4096 bytes the price for 64 kByte comes down to 2080 US$. (Byte_magazine_September1975.pdf - page 62)

Though the much more cost effective 16 kbit Mostek MK4116 DRAM went into the market in 1976. https://www.righto.com/2020/11/reverse-engineering-classic-mk4116-16.html