[VA] A digression around a digression

[Valentin Albillo]

  
{Note: this reply to C.Ret would surely belong under thread A digression around VA's SRC #012b but as this post is off-topic and quite long, I'd rather create this new thread instead of hijacking J-F's}


Hi, C.Ret, {all highlighting is mine}

(11-27-2022 02:36 AM)C.Ret Wrote:  I don't quite understand the point of further accelerating an emulator that is already over 110x the speed of real hardware.

The point is obvious: the faster the emulator, the less time it takes to solve problems and the faster the sleuthing process goes. In the past I used very extensively J-F Garnier's Emu71/DOS, which is a marvel to behold, and it ran at 240x on my now defunct ~20 yo XP system.

Once it passed away I had to use another emulator (as J-F made it clear to me in no uncertain terms that he didn't plan to release a 32- or 64-bit version of Emu71/DOS any time soon, if ever,) and eventually I was saddled with an insufferably buggy one which was supossed to run at 128x but barely achieved 64x, i.e. 3 times slower than Emu71/DOS and boy, did I miss the 3x-faster speed ! .

Quote:Or, the goal is to demonstrate that the HP-71B is too old and too slow to be used for serious purposes now.

Not at all. In my currect threads SRC #012[a-f] - Then and Now the explicitly stated main goal is, I quote:

"[...] I'll convincingly demonstrate that some advanced vintage HP calcs which were great problem-solvers back THEN in the 80's (some 40 years ago !), are NOW still perfectly capable of solving recently-proposed non-trivial problems intended to be tackled using modern 2020-era personal computers, not ancient pocket calcs."

Quote:I strive to use only real material and I see that the slightest attempt takes several hours while others waste only a few minutes of their free time to obtain more results than I could ever obtain.

It seems you've been trying to run some of the code posted in this thread and have endured run times of a few hours (potentially depleting the batteries) vs. just about 1 min. or so when using an emulator. It might be frustrating but this doesn't mean that the HP-71B can't be used in this case.

First of all, the depletion of the batteries can be avoided by simply using an AC adapter (not charger), which J-F offered to provide to you, and secondly, finding two extremely close real roots of a 10,000 (!!)-degree polynomial to 10-12-digit accuracy in a few hours on a ~40-year-old battery-powered pocket computer is nothing short of miraculous, if you think about it. You can start the program when you get to bed at night and find the solutions already computed by the time you briefly get up to take a leak ...

As for your "waste" qualification, I don't know about you but I don't consider I'm wasting my time when I do something which I find fun and instructive and nobody's asking me to do it. If it takes longer on a physical HP-71B than using an emulator so what ? The emulator also takes much longer than doing it in compiled C#, so does it mean that we must do everything in C# ? Where would be the fun, the challenge to successfully accomplish a goal against allegedly unsurmountable odds ?

Quote:The HP-71B is like me, old and outdated.

Neither the HP-71B nor you are "old and outdated". Myself, I'm much older than you and I don't consider being that either. It is a defeatist thought not supported by evidence.

Also, you say above: "I strive to use only real material" and this raises the question:

What's a real HP-71B, anyway ?

If you take the question at face value, a real HP-71B would be this:

but consider that the object depicted in the image above consists of some hardware (body, display, keyboard, internal circuitry, etc.), part of which you can see, and some software (the System ROMs' content), which you can't see but it's nevertheless there and (unlike the hardware,) it's utterly essential.

You can replace the system software (which includes the BASIC interpreter) by some other completely different, properly developed software which interfaces adequately with the hardware (say a C interpreter) and upon turning it on you'd discover that it's not an HP-71B anymore, despite the looks. It won't behave any longer like an HP-71B nor would you be able to run HP-71B programs on it, at all.

On the other hand, you can place the HP-71B's software (the System ROMs' contents) on a different hardware plus the emulation layer, and it will behave and run like the original HP-71B, so you can use it to run HP-71B programs, the code being unaware that it's running on something else which is not the original hardware.

If the looks were all-important, you might consider that the HP-71B hardware ought to be the physical original device. But I contend that this is not what the HP-71B actually is.

The "soul" of the HP-71B, its essence, what actually makes an HP-71B be an HP-71B, is its original, immaterial software, the contents of its System ROMs, which define everything it does and how it behaves. You can place them in a suitable emulator running on an arbitrary platform (Windows, Mac, Android, iOS, "DM71"), and it will behave and run programs as an HP-71B.

In the future, this might apply to human beings as well. Take your exact body and place on it someone else's brain and the resulting person won't be you, even if it would look identical to everyone just looking. On the other hand, take your brain and install it on some other body and the resulting person would be you, regardless of the body, as everyone dealing with him would attest. So the rethorical question is:  Who'd be the real you, your body with other brain or your brain in another body ?

In short: the software is what matters the most, and any device physical or virtual using the HP-71B's System ROMs' contents to accurately implement the HP-71B functionality is as real an HP-71B as the physical original is, the same way that your mind in another body would be as real you as when residing in your original body.

Hope this helps.
Best regards.

V.