Critique of hunchbacked's "Apollo mad computer program (part 3)"

[1MtnBoy]

Awesome critique ! I have attempted rational discussions with hunched for some time now hoping someday he may actually demonstrate some knowledge of facts and objective research, here's hoping ! I stink at making video's at the present, yours here is factual and splendid to watch ! only sorry I didn't see your video's sooner.

thanks for taking the time to make these video's.

[rawmonkno1]

@1MtnBoy Nice to here that you liked it 1MtnBoy. I will probably revisit this subject later with the inner working of the AGC but from a pure technical standpoint with no hoaxers involved. :)

[1MtnBoy]

@rawmonkno1 Yes there have been some rather long drawn out comments on his video's, lol. I have never seen anyone make so many utterly false statements while attempting to describe AGC operation, especially Rope programs (ROM). We dissected it all the way back to "Inductive-Coupling" and still could not get him to agree on this simple electrical principle, your video shows more in 8 mins than I did in two weeks of debating with him. I could go on. lol You should do one on Rope ROM !

[hunchbacked]

Tell me one thing: Imagine "TOTO+1" is a valid label, like you suggest it.

Imagine there is a label "TOTO" and a label "TOTO+1".

If there is an instruction "TCF TOTO+1", does that mean that there must be a jump to the label "TOTO+1", or to the program location one byte after the label "TOTO".

Do you see the contradiction, and the problem it may cause?

That's why there are rules for labels and data names too.

[philwebb59]

@hunchbacked It would be a jump to the label "TOTO+1." Next question.

[hunchbacked]

@philwebb59

This is my next question:

And if you meant to do a jump to the byte following the label "TOTO", then how would you do?

[philwebb59]

@hunchbacked Are you familiar with compiler design? Have you ever used lex and yacc to build a compiler or natural language interpreter or an sql front end to a custom database? First you put all your labels and variable names in a name list. The lexicon looks in the name list for a textual match. If TOTO+1 is in the list, then it finds a match. If not, and instead it finds TOTO, then it finds a match. Arbitration rules determine what happens if TOTO and TOTO+1 are both in the list. Next?

[mercatormac]

@hunchbacked I presume "TOTO +1" According to the documentation spaces are significant for delimiting parameters in the AGC source.

[philwebb59]

@hunchbacked If as you say, instructions are one byte followed by a two byte address, then the byte after TOTO would be an address. We both know you were wrong on that one, so I would either do as RM1 described above, TOTO+1(space)+1, or create a new label. Myself, I'd create a new label and give it a descriptor associated with why I'd want to jump there. Next? BTW, where did you get the page numbers?

[philwebb59]

@hunchbacked Oops! I misread your question. I thought you repeated your question to RM1. The byte after label TOTO would be, as mercatormac said, TOTO(space)+1. A modern analogy is BASIC. If I want to append string p onto string s, I should be able to write s=s&p, but that doesn't work because if I put s in the name list, s& goes in automatically, so when the compiler lexically matches the longest string s&, then looks for an operator it sees p and bombs. & is a non-alpha, right?

[rawmonkno1]

I already explained this, if you are going to make an arithmetic expression you need to add white spaces between the label and the expression.

"TOTO+1" is always a label.

"TOTO+1 +1" is the same label with an expression.

"TOTO +1" is another label called "TOTO" with an expression and never the same label as the one above because of the white spaces.

This is the rule and the code follows them. It's idiotic but at the same time a good way to explain a value of a constant.

[hunchbacked]

@rawmonkno1

Yes, you have explained it, but it's rules you invent, because it doesn't work this way on computers.

[rawmonkno1]

@hunchbacked I haven't invented these rules. I've actually read the documentation and looked at the code.

"because it doesn't work this way on computers"

I guess you don't understand that this is just an assembly language, it works the way it's designed to work. It doesn't matter how it works as long as it doesn't break it's own rules.

Do I actually need to continue explaining something as simple as this?

[hunchbacked]

If there are rules to be respected for writing programs, it's for a simple reason: If anything was allowed, it would become impossible for a compiler to compile programs; the analysis would become too difficult, and the chances of misinterpretation too great.

Now, if you want to believe that what you describe is possible, then believe it, but know it will make laugh any serious computer engineer.

[hunchbacked]

It's useless that I discuss with you, because you obviously have no programming knowledge.

[hunchbacked]

No computer uses addresses this way.

There is a reason why not all characters are allowed in labels, and also in data names.

You are implying that the labels are completely free, and that it would explain the weird way the addresses are specified.

You are redefining the way programming works.

And there is still something you have not explained:

A subroutine called by TC cannot call another one, and yet we find examples in the program.

Try to explain this.

[philwebb59]

@hunchbacked Do you understand what compiler directives are? Do you realize you CAN implement a stack in software, probably using any assembler language? Why can't you accept the fact that 1960's computers didn't do things the way they do today? When you call a subroutine or respond to an interrupt, it DIDNT MAKE SENSE to save all your registers, especially if you weren't going to use them. That's wasted clock cycles. Did you even watch this video, or are you just making general comments?

[philwebb59]

Excellent. For an old-time Computer Engineer, especially one who "knew the first microprocessor" and is intimately familiar with computer architecture from the 1960's (which would make him 70+something years old), and for one who has "expert friends" who verified the AGC code didn't compile, nothing Hunchbacked has said on this subject couldn't be answered by "so what?" or "you're wrong and here's why." Great job RM1.

[rawmonkno1]

@philwebb59 Thank you Phil. It's been a learning process to make these videos and I'm still experimenting with a lot of things, but hopefully it's not too hard to understand my broken English at least?

Btw Phil, his latest claim is that the Virtual AGC emulator is not actually running the old AGC code. The virtual AGC project is open sourced so all of these things can be verified. I just don't understand people like HB and how they can continually refuse to accept provable facts.

[hunchbacked]

@philwebb59

It's precisely because I'm an old-time computer engineer that I know what's possible, and what's not.

If you want to believe that what we see in the program makes sense, then do, but don't claim to be an expert yourself, because you are not.

[philwebb59]

@hunchbacked True, I'm an EE not a CE. I'm hardware, not software, but most CE's I know are more savvy about hardware than you appear to be. They understand computer architecture better than you appear to. And a child would realize that when you compile huge libraries of code, labels and variables defined as globals in any section of code are available from any other module. Rawmonkno1 totally destroyed you and you know it. And you still haven't told me where your page numbers come from.

[GoneToPlaid]

7:32 "This is actually a perfect end to this video because your last example proves perfectly well that you have no idea of what you are talking about when it comes to programming." Wow. That really was a very nice way of calling hunchbacked an idiot without actually calling him an idiot! :-)

[markjh2005]

Thanks for the links in the sidebar rawmonko1, when I get the chance I'll be sure to have a look:)