Anyway, you proved once again the the final step in manufacturing a tool is setup and tuning, and that can make all the difference in the world. Good tips on that as well as the safety. Bandsaws really sing if you get them right.

]]>Great write-up as always…

I think you ignored the lowly circular chop saw as a metal cutting alternative. It’s sort of the big brother of the angle grinder (and just as loud) but more portable than a bandsaw.

For non straight cutting, I lust after one of those electric plasma cutters.

Cheers,

Crawford

Alright I’ll play…

Up front let me say that the following comes from someone who was never lucky enough to own an Apple II. As a young kid I could only ever stand in Radio Shack and gaze longingly at wondrous machines which were way out of my financial reach.

The trite answer is that the results are corrupted by floating point arithmetic round off errors and/or bugs. But I thought to try answer with a little more precision.

So donning the Internet goggles I found the following in short order:-

https://archive.org/details/a2_Tip_Disk_1_1982_Beagle_Bros_a

Here we find an archived copy of an Beagle Bros diskette which contains the “Why Does This Do This?” program. More importantly the Archive also provides a virtual Apple II with which to run the software. Very handy for someone like me without any Apple! Interestingly the online version of this program appears to flagged “I” for Integer Basic(?) and won’t load in the emulator. But it was easy enough to type it in from scratch.

https://ia801004.us.archive.org/29/items/Apple_2_Applesoft_BASIC_FP/Apple_2_Applesoft_BASIC_FP.pdf

Here we find an article by Jim Thomas once head of the “Apple Numerics Group” that outlines several deficiencies in Applesoft BASIC floating point arithmetic.

While the “Multiplier Bug” mentioned in this paper stands out as a likely candidate to be honest after peeking and poking about I really wasn’t able to conclusively prove this point.

So sadly I have failed the challenge of providing greater clarity. I wonder if the question from the Beagle Bros in this instance was entirely rhetorical?

I did however have a lot of fun doing my own digital archeology sleuthing. So by way of atonement for not providing an exacting answer. I offer the following Applesoft BASIC program that I cobbled together from multiple sources. The one word output of which would appear to sum the situation up best.

Thanks Quinn for a terrific series of articles!!

Regards, Steve.

—-

10 GOSUB 100

20 OP = 3: REM XOR

52 LET V2=135

55 FOR I = 61597 TO 61588 STEP -1

60 V1 = PEEK (I)

65 GOSUB 200

70 PRINT CHR$ (V1);

75 NEXT I

80 PRINT

90 END

100 DEF FN GE(B) = ((V – ( INT (V / (2 ^ B))) * (2 ^ B)) > = (2 ^

(B – 1)))

110 DEF FN SE(B) = V + ( NOT FN GE(B)) * (2 ^ (B – 1))

120 DEF FN CL(B) = V – ( FN GE(B)) * (2 ^ (B – 1))

130 LET N = 8

190 RETURN

200 FOR B = 1 TO N

210 LET V = V2:B2 = FN GE(B)

220 LET V = V1:B1 = FN GE(B)

230 LET R = B1

240 ON OP GOSUB 271,272,273

250 IF R THEN V1 = FN SE(B)

260 IF R = 0 THEN V1 = FN CL(B)

270 NEXT B: RETURN

271 LET R = B1 AND B2: RETURN

272 LET R = B1 OR B2: RETURN

273 LET R = B1 B2: RETURN

*I mean, “in a timely/spontaneous/organic way.” It’s a lot like winking.

]]>