This is an interesting article/post. But also kind of frustrating for me in a way. I've been a computer programmer for a long time. I started with 8 bit computers. I would program those with Basic...
This is an interesting article/post. But also kind of frustrating for me in a way.
I've been a computer programmer for a long time. I started with 8 bit computers. I would program those with Basic and I had no problem understanding that. I tried a few things in assembly but it is just such a different way of thinking that I never made much progress.
Later I learned many different high level programming language including C, Pascal, and most of the modern ones that are in use now. For years being a programmer has been my job and I do have background in lower level hardware and software including knowledge of boolean logic and and/or gates etc.
But still this article is very difficult for me to understand all at once, and reminds me of the frustration I used to have when looking at large amounts of assembly and only getting it in little bits.
I'm sure it just clicks for some people in a way that doesn't for me. I still enjoy the topic and books like "Racing the Beam" that show how early video games were made.
This is such a great look at how a CPU actually operates. I’ve been asking this question on and off to myself for years. Looks like someone finally did it and now I’ve learned something else new...
This is such a great look at how a CPU actually operates. I’ve been asking this question on and off to myself for years. Looks like someone finally did it and now I’ve learned something else new today!
If you're interested in more, I recommend the game MHRD (MicroHard) and the course NAND2Tetris. The game covers roughly the first half of the material in the course, but both are good ways to...
If you're interested in more, I recommend the game MHRD (MicroHard) and the course NAND2Tetris. The game covers roughly the first half of the material in the course, but both are good ways to learn the next level down.
I keep asking myself why we can't do better than floating point operations. They're compute cycle heavy and imprecise. You really can't even really use them for financial transactions. The...
I keep asking myself why we can't do better than floating point operations.
They're compute cycle heavy and imprecise. You really can't even really use them for financial transactions. The rounding errors and remainder carryover for millions of credit card transactions add up quickly.
So why keep using it? Well, this article does a good job of explaining it. Floating point operations were an early low language solution to a complicated problem. And a pretty good one.
I keep thinking an arithmetic lookup table could do the job way better. But that would require a fundamental shift in processor architecture, and a lot of dedicated cache, and I'm just not smart enough to figure out the engineering.
This is an interesting article/post. But also kind of frustrating for me in a way.
I've been a computer programmer for a long time. I started with 8 bit computers. I would program those with Basic and I had no problem understanding that. I tried a few things in assembly but it is just such a different way of thinking that I never made much progress.
Later I learned many different high level programming language including C, Pascal, and most of the modern ones that are in use now. For years being a programmer has been my job and I do have background in lower level hardware and software including knowledge of boolean logic and and/or gates etc.
But still this article is very difficult for me to understand all at once, and reminds me of the frustration I used to have when looking at large amounts of assembly and only getting it in little bits.
I'm sure it just clicks for some people in a way that doesn't for me. I still enjoy the topic and books like "Racing the Beam" that show how early video games were made.
This is such a great look at how a CPU actually operates. I’ve been asking this question on and off to myself for years. Looks like someone finally did it and now I’ve learned something else new today!
If you're interested in more, I recommend the game MHRD (MicroHard) and the course NAND2Tetris. The game covers roughly the first half of the material in the course, but both are good ways to learn the next level down.
...you might enjoy jon stokes' inside the machine while you're at it...
I keep asking myself why we can't do better than floating point operations.
They're compute cycle heavy and imprecise. You really can't even really use them for financial transactions. The rounding errors and remainder carryover for millions of credit card transactions add up quickly.
So why keep using it? Well, this article does a good job of explaining it. Floating point operations were an early low language solution to a complicated problem. And a pretty good one.
I keep thinking an arithmetic lookup table could do the job way better. But that would require a fundamental shift in processor architecture, and a lot of dedicated cache, and I'm just not smart enough to figure out the engineering.