real amiga vs winuae

[ElPolloDiabl]

If WinUAE has it any shortcomings the fact that its free (at no cost to you) might have something to do with that. Also the fact that they are not allowed to copy Amiga patents and copyrighted code (not sure if they got permission for at least some of it) could be a contributing factor.

     Any argument against WinUAE is pretty petty. Something better to argue about would be: Why was Shapeshifter on Amiga faster than a real Mac? (I think hardware and the price of said hardware was the reason).

[Anding]

I want to assemble a working amiga that is actually usable, and is not too
difficult to use (I want my kids to be able to load games too

Why not look at the Amiga Forever package?  The Player format is ideal for kids.  They now have C64 Forever available too

[amigaksi]

If WinUAE is cycle exact for a specific 680x0 chip, then this is from the point of view of the emulated environment exactly what happens.

Now, obviously, if you access stuff outside of the emulated environment that is timing critical this may fail due to a variety of factors.

But, if you use timing critical software on an Amiga that is not the same spec as another Amiga this can (and often will) also fail. There are lots of different Amiga's out there, all subtly (or not so subtly) different. Some hardware for A1200's for instance doesn't work on all models, some expansions for other Amiga's actually require specific motherboard revisions to work reliably, etc.

You violated the law of physics.

[amigaksi]

Somebody needs to read up on Pentium Pro/Pentium II's handling of self-modify code and cache.

Part1

Part2

Your first link states that this book cannot be used.  The point is caching did affect self-modifying code on later than 8088 processors (not just Pentium I/II) but they are still considered backward compatible.

[amigaksi]

Again with confusing cycles and timing (seriously, are you doing this deliberately?).

As has been stated, repeatedly, by multiple folks on here: Timing, especially on the Amiga is an inexact thing. It is different between any two Amigas since the timing comes not from the processor, but by the crystal oscillator. Those Crystal oscillators are far from exact. It gets even worse when you consider that PAL and NTSC Amigas have different timings altogether.

None of which invalidates anything that Karlos has said.

Learn to quote properly.  I made a valid point and it's there in post #202.  I am being consistent with the timing/frequency: T=1/f.

[amigaksi]

If WinUAE has it any shortcomings the fact that its free (at no cost to you) might have something to do with that. Also the fact that they are not allowed to copy Amiga patents and copyrighted code (not sure if they got permission for at least some of it) could be a contributing factor.

     Any argument against WinUAE is pretty petty. Something better to argue about would be: Why was Shapeshifter on Amiga faster than a real Mac? (I think hardware and the price of said hardware was the reason).

It's not petty.  It's not a real amiga just like a fake diamond is different from a real diamond.  Now subjectively, whether it makes some difference to you or not is another matter.  You cannot establish it's a real amiga by trying out a few applications (inductively).

[the_leander]

Learn to quote properly.

You first sunshine.

I made a valid point and it's there in post #202.  I am being consistent with the timing/frequency: T=1/f.

No, you're mixing cycle precision with timing precision. Timing on an Amiga is not exact due to a whole range of differences within the various Amiga models and revisions, as well as the relative imprecision of the type of crystals used to supply the timings. Ergo, any argument against emulation on the basis of timing is eroneous at best.

[the_leander]

You violated the law of physics.

Tell that to the owners of those accellerator cards with those motherboards that didnt work together.

[bloodline]

It's not petty.  It's not a real amiga just like a fake diamond is different from a real diamond.  Now subjectively, whether it makes some difference to you or not is another matter.  You cannot establish it's a real amiga by trying out a few applications (inductively).

First define a "Fake Diamond". If you mean a piece of glass cut to look like a diamond... then in some situations it will function as effectively as a real diamond, i.e. cosmetic uses. But it isn't a diamond since it can't be used where the physical and chemical properties of diamond are required (Diamond being carbon based, and glass being silicon based).
In computing terms, it would look like a the real thing to the "user"... but any "software" would fail.



If you mean synthetic, that is to say manufactured rather than mined, then the "fake diamond" will function perfectly. In computing terms, it might not look quite so good to the "user" (synthetic diamonds tend to be made  smaller than natural ones, and so require careful configuration to look good), but any "software" would run perfectly.


Remember, the point of an Emulator is to run software designed for one hardware platform, on another unrelated hardware platform.

[bloodline]

Learn to quote properly.  I made a valid point and it's there in post #202.  I am being consistent with the timing/frequency: T=1/f.

No, the_leander is perfectly correct here.

The computer has no concept of time, the quantum (that is to say unit of time) is based on the clock cycle. It actually doesn't matter to the computer what the period (in real world seconds) of that cycle is. What does matter is that all devices on the system work within the specified number of clock cycles as defined by the specification.

I stated earlier that I can use my ActionReplay MkIII to slow the system clock of my A500 down... the software all works fine, because the software has no concept of the real time of the clock cycle... from the computers Point Of View, the real world has just become faster...

[amigaksi]

You first sunshine.



No, you're mixing cycle precision with timing precision. Timing on an Amiga is not exact due to a whole range of differences within the various Amiga models and revisions, as well as the relative imprecision of the type of crystals used to supply the timings. Ergo, any argument against emulation on the basis of timing is eroneous at best.

They are related.  I can prove it if I write application that actually uses it for timing things.

[amigaksi]

Tell that to the owners of those accellerator cards with those motherboards that didnt work together.

I already answered this idea of timing many times in the other thread and you never replied to it.  The NTSC crystal timing is precisely defined.  As per spec, it's doing what it's supposed to do.

[motrucker]

After reading ALL of the comments, I've decided to edit this one out completely. You guys can have at it...

[the_leander]

They are related.  I can prove it if I write application that actually uses it for timing things.

The only way to test it would be to have two different amigas run any timing dependant code (such as a basic drum beat, for instance) side by side and check for differences in speed over time. This is demonstrable and even I understand this.

The fact is that the NTSC and PAL specs have a reasonable amount of leaway due to the fact that at the time of their inception, the equipment and componants used were (by todays standards) imprecise. This imprecision can be seen too in the crystals used by the Amiga to produce the clock frequency.

You are deliberately trying to confuse cycle and timing precision, both of which have very specific meanings that have been spelled out to you. Stop it.

[amigaksi]

The only way to test it would be to have two different amigas run any timing dependant code (such as a basic drum beat, for instance) side by side and check for differences in speed over time. This is demonstrable and even I understand this.

The fact is that the NTSC and PAL specs have a reasonable amount of leaway due to the fact that at the time of their inception, the equipment and componants used were (by todays standards) imprecise. This imprecision can be seen too in the crystals used by the Amiga to produce the clock frequency.

You are deliberately trying to confuse cycle and timing precision, both of which have very specific meanings that have been spelled out to you. Stop it.

Huh.  If I send out a bit through some I/O port using the Copper to time it, it's going to be the same across all amigas because the frequency of operation is the same.  Even if there's some variance 1/100000000 across machines, it's still considered performing per spec.  They are related-- timing and frequency.

[Karlos]

Even if there's some variance 1/100000000 across machines, it's still considered performing per spec.  They are related-- timing and frequency.

The variance is significantly more than that. The crystals used in the old miggies had tolerances in the tens of ppm.