08-03-2015, 05:09 AM
So about 3 or 4 years ago (yes this is a long post) I passed my RDF trial by making a couple sets of 4 D-Flip-Flops and linking them to a 4 pixel display (which was budded)... the times have changed quite a bit.
I joined PCC (Prestigious Computing Company - a mini 'company' that lived on the server) because they seemed like people who knew what they were doing, and they wanted to do some big stuff. The group had just recently formed (ChucksMoth (now named dcentrics) Nickster258, Greymouth, Alexapplebaum and some other peeps) and had begun to develop CPU's. This was back when people still had hour long flame wars about instructions set sizes and design. I don't really remember how I got the idea, but it probably had to do with MCTP (minecraft transfer protocol, the thing that never really worked). I decided that I was going to make cpu's connect to each other over serial and be able to share programs and workloads. Now, this was a ton of nothing coming from someone who only knew how to make a D-flip-Flop, but I stuck with it and learned about parallel to serial devices.
It took about 3 or 4 days for Connect 1 to be finished, it was a 4 wire dual way serial system that would transfer data by 4 packets of 4 bits each. I made a post on the RDF about it and two days later it was griefed and never functioned again. In the meantime this attracted the attention of Greymouth, who assisted in the construction of Connect 2, which looked a lot better but also never functioned.
I began construction of Connect 3 right around the 'war' between PCC and HCI (Hawk Cut(lass) Industries). They had begun a '1 year' battle to see who could make a better CPU. This is where things got hazy because I never really followed it. Supposedly PCC made a pipelined CPU that pretty much won, but it was finished after the last server backup and the RDF collapsed under its unstable structure, taking the processor from us.
It is unfortunate that I no longer have connect 1 or 2, however I still have connect 3 and further:
![[Image: lDJnPtv.jpg]](http://i.imgur.com/lDJnPtv.jpg)
Connect 3 was a functional success, but I decided to rewire it and ended up breaking it for a long time. I decided to get into CPU's since I knew that I would eventually need one to link my modems to.
With the help of PCC, I created my first CPU:
![[Image: QojoCJX.jpg]](http://i.imgur.com/QojoCJX.jpg)
I then built another 2 by myself for practice, the first of which has been on YouTube for a long time but none of you have probably ever seen it:
![[Image: UhIZvTz.jpg]](http://i.imgur.com/UhIZvTz.jpg)
At this time, since I did have a functional modem, PCC got excited about my ideas and built me a 'server' that I could use to deliver programs to clients. I never used it, and the IS got some hate, but it was pretty neat to have sitting there:
![[Image: SMlwKfg.jpg]](http://i.imgur.com/SMlwKfg.jpg)
Since I had messed up Connect 3, I went on to make Connect 3.1, which was more functional. I also started Connect 4 (not the game!) which would be for 8 bit CPU's instead of 4 bit.
I still had no way to connect a traditional CPU to my modems, so I finally set out to make one myself. I created MCPU, a buss-oriented CPU that allowed the operation of multiple hardware extensions off of the bus, which included Connect 3.1
In my initial testing I realized that Connect 3.1 did not work as intended with the CPU, and it's protocol was a bit underthought. The color scheme was also painful to stare at for long periods of time, so all of that got changed in connect 3.2. The first program MCPU ran was a counting program, which was tradition to me at that point. The second program counted, and sent the current number out through serial, where another MCPU listened for inbound data and displayed the number. It was pretty cool. I then went on to construct a router that would allow up to 3 modems connect without corruption.
I began construction of MCPU 2 which would hold more ROM for program storage, but then I realized that if it was ROM then I would have no way of saving programs.
Thus begins my work on MCPU 3, a Von Neuman architecture that still contains concepts of MCPU but will have everything I need to accomplish my project... It's been a while.
TL:DR - ddthj is slow at finishing projects.
I joined PCC (Prestigious Computing Company - a mini 'company' that lived on the server) because they seemed like people who knew what they were doing, and they wanted to do some big stuff. The group had just recently formed (ChucksMoth (now named dcentrics) Nickster258, Greymouth, Alexapplebaum and some other peeps) and had begun to develop CPU's. This was back when people still had hour long flame wars about instructions set sizes and design. I don't really remember how I got the idea, but it probably had to do with MCTP (minecraft transfer protocol, the thing that never really worked). I decided that I was going to make cpu's connect to each other over serial and be able to share programs and workloads. Now, this was a ton of nothing coming from someone who only knew how to make a D-flip-Flop, but I stuck with it and learned about parallel to serial devices.
It took about 3 or 4 days for Connect 1 to be finished, it was a 4 wire dual way serial system that would transfer data by 4 packets of 4 bits each. I made a post on the RDF about it and two days later it was griefed and never functioned again. In the meantime this attracted the attention of Greymouth, who assisted in the construction of Connect 2, which looked a lot better but also never functioned.
I began construction of Connect 3 right around the 'war' between PCC and HCI (Hawk Cut(lass) Industries). They had begun a '1 year' battle to see who could make a better CPU. This is where things got hazy because I never really followed it. Supposedly PCC made a pipelined CPU that pretty much won, but it was finished after the last server backup and the RDF collapsed under its unstable structure, taking the processor from us.
It is unfortunate that I no longer have connect 1 or 2, however I still have connect 3 and further:
Connect 3 was a functional success, but I decided to rewire it and ended up breaking it for a long time. I decided to get into CPU's since I knew that I would eventually need one to link my modems to.
With the help of PCC, I created my first CPU:
I then built another 2 by myself for practice, the first of which has been on YouTube for a long time but none of you have probably ever seen it:
At this time, since I did have a functional modem, PCC got excited about my ideas and built me a 'server' that I could use to deliver programs to clients. I never used it, and the IS got some hate, but it was pretty neat to have sitting there:
Since I had messed up Connect 3, I went on to make Connect 3.1, which was more functional. I also started Connect 4 (not the game!) which would be for 8 bit CPU's instead of 4 bit.
I still had no way to connect a traditional CPU to my modems, so I finally set out to make one myself. I created MCPU, a buss-oriented CPU that allowed the operation of multiple hardware extensions off of the bus, which included Connect 3.1
In my initial testing I realized that Connect 3.1 did not work as intended with the CPU, and it's protocol was a bit underthought. The color scheme was also painful to stare at for long periods of time, so all of that got changed in connect 3.2. The first program MCPU ran was a counting program, which was tradition to me at that point. The second program counted, and sent the current number out through serial, where another MCPU listened for inbound data and displayed the number. It was pretty cool. I then went on to construct a router that would allow up to 3 modems connect without corruption.
I began construction of MCPU 2 which would hold more ROM for program storage, but then I realized that if it was ROM then I would have no way of saving programs.
Thus begins my work on MCPU 3, a Von Neuman architecture that still contains concepts of MCPU but will have everything I need to accomplish my project... It's been a while.
TL:DR - ddthj is slow at finishing projects.
