Vaccinations; recent developments

[sarkeizen]

I am guessing about 10 years or so.  He is an awesome guy and his book  "I WOZ" which I read years ago is now posted on Youtube in an audio version.  I will edit and place a link here for those that want to listen to it.  It is a great book and gives one a lot of insight into what really went into making the world's first pc.

Bill

Cool I wonder if that's the same book I read a few years back.  I've also read "West of Eden", "Odyssey", "Defying Gravity" and "The little kingdom". 

If you haven't read: http://www.folklore.org/ you should.

[SeaMonkey]

That is more than a little romanticized version of history.  Example this rather famous game:

https://www.youtube.com/watch?v=SwM06NbZ2r0

I can tell - just by watching - exactly what their code does as it takes 9s to draw a screen.  I can even tell you why the screen clears to white and why it does in little rectangles.

Just like the 6502 was a compromise so was the software written on it.  Sometimes it was bloated and sometimes it was fast it depended on the outcomes desired.

Also the idea that modern programmers are somehow less concerned with speed or bloat is a bit of a fiction.  ISA's are so complicated that compilers will generally do a better job than assembly programmers.  Not to mention that there is no longer a single target.   For example back when the PIII and P4 co-existed they had mutually exclusive optimal cases.  The PIII was a high IPC target whereas the P4 was a high-clock rate target.  So even something as simple as a memory move had to be coded very differently depending on which machine was running the code.

Familiarity with the routines and the manner in which
certain operations are accomplished within the hardware
does make that sort of analysis possible.

Compilers are able to assemble code reasonably well
and can rival experienced human coding when the
compiler itself is well designed with precise alternatives.

Our experiences have apparently been quite different.
Certain projects have no room for error or malfunction
and must be achieved with the absolute minimum of
code and memory.  Of course those projects are not
games or the types of programs the typical user would
encounter.  Coder knowledge of every detail of the
associated specialized hardware in those cases is essential.

The video was rather interesting and a typical example of
the graphics capabilities of that era.

Games have never been my bag.  I just am not able to find
that sort of entertainment worthy of any investment of time.

Far more interesting is making pyrotechnic devices which
produce spectacular aerial displays.  Now that is art!
Not nearly as spectacular as the military stuff but most enjoyable
none the less...

[sarkeizen]

Compilers are able to assemble code reasonably well and can rival experienced human coding when the compiler itself is well designed with precise alternatives.

Not rival - surpass.  On the Intel architecture you will almost never be able to beat any reasonably mature compiler - the ISA simply has too many options and performance cases.  In fact even given a single Intel target you would be lucky to find someone who has studied the architecture sufficiently to even know what the cases are.  Not even touching the fact that there are mutually exclusive performance cases on multiple targets.

That's not the only instance either.  The late MIPS architecture needed you to keep track of the pipeline state which isn't something you can see in a section of code just by looking at it - the compiler had to insert NOPs to avoid interlocks.

Our experiences have apparently been quite different.

No, yours is very close to the way I felt when I was sixteen.  Since then I've read many books on microprocessor architectures and probably coded more than you ever will.   Your bespoke elitism was common among the self-taught programmers.  Once I realized it was vanity I let it go.

Coder knowledge of every detail of the associated specialized hardware in those cases is essential.

Probably not and outside of some very small edge cases it's simply not feasible.

The video was rather interesting and a typical example of the graphics capabilities of that era.

I was showing it to you because I understand that hardware well enough to understand the compromises that were being made purely by observation. 

Games have never been my bag.

Please.  Your constant digging up nonsense to post here while affecting a dime-store mystic persona is a game.  Not a very challenging one either.   You'd be better served playing Halo.

[Pirate88179]

Cool I wonder if that's the same book I read a few years back.  I've also read "West of Eden", "Odyssey", "Defying Gravity" and "The little kingdom". 

If you haven't read: http://www.folklore.org/ you should.

Thanks for the link.  That looks really cool.  I will have a great time reading all of those posts.

Bill

[SeaMonkey]

Not rival - surpass.  On the Intel architecture you will almost never be able to beat any reasonably mature compiler - the ISA simply has too many options and performance cases.  In fact even given a single Intel target you would be lucky to find someone who has studied the architecture sufficiently to even know what the cases are.  Not even touching the fact that there are mutually exclusive performance cases on multiple targets.

That's not the only instance either.  The late MIPS architecture needed you to keep track of the pipeline state which isn't something you can see in a section of code just by looking at it - the compiler had to insert NOPs to avoid interlocks.

You may be correct in that assessment given the state of
today's education and how it has negatively impacted the
state of affairs technologically in America.  But for the dedicated
individuals who have the motivation, the desire and the ability to
overcome those obstacles we'd be even further behind those
nations who've already overcome the once solid American lead.

No, yours is very close to the way I felt when I was sixteen.  Since then I've read many books on microprocessor architectures and probably coded more than you ever will.   Your bespoke elitism was common among the self-taught programmers.  Once I realized it was vanity I let it go.

In your case there is not doubt at all that it was
pure vanity and you did well to let it go.  That is
a very positive sign.  Are you working on 'letting
go' of more of it?

Suffice it to say that there are processors in use
which bear little resemblance to the advanced Intel
products designed for public consumption.  Truly
Space Age stuff.  The numbers of 'coders' who've
worked with those incredible devices is yet very
small, and, no they're not self taught  They are very
humble men.

Probably not and outside of some very small edge cases it's simply not feasible. I was showing it to you because I understand that hardware well enough to understand the compromises that were being made purely by observation.

Please.  Your constant digging up nonsense to post here while affecting a dime-store mystic persona is a game.  Not a very challenging one either.   You'd be better served playing Halo.

The desperation just never ceases does it?
Sarkey, you do seem to have some talent and
some amount of intelligence.  Unfortunately
(make that Very Unfortunately) your preoccupation
with bolstering your artificial 'supremacist' persona
detracts from what little credibility you infrequently
reveal.  You are very amusing though...