HP-15C CE speed improvement over OG 15C

[RPNerd]

It's obvious that the 15C CE was going to be a lot faster than the original 15C but I wanted to get an idea of how much faster, so I decided to let them both loose on a problem that should involve quite a bit of number crunching. I settled on the problem entitled "Using the Solver with Simultaneous Equations" om pp168-169 of "Programming Examples and Techniques" for the HP-42S (see https://literature.hpcalc.org/items/934).

My OG HP-15C chewed through it in approx. 1m15s, which I thought was pretty good for a calculator that was never particularly fast in the first place repeatedly solving a 4x4 system of linear equations until the correct value for one of the coefficients is found.

The HP-15C CE tore through it in a fraction of a second.

Case closed.

[AnnoyedOne]

(08-22-2024 03:25 PM)RPNerd Wrote:  The HP-15C CE tore through it in a fraction of a second.

Yeah. Modern 32-bit (ARM) based models are much faster even when emulating original NUT code. My 2022 HP-12C+ does the same thing :-)

[RPNerd]

(08-22-2024 03:31 PM)AnnoyedOne Wrote:  Yeah. Modern 32-bit (ARM) based models are much faster even when emulating original NUT code.

Even my SwissMicros DM15L (also ARM 32-bit emulating NUT code) running at 48 MHz needed about 4 seconds to get the answer (about 18s @ 12 MHz).

[AnnoyedOne]

(08-22-2024 03:42 PM)RPNerd Wrote:  Even my SwissMicros DM15L...

I considered buying one of those but heard about the HP-15C CE and got one of those instead.

The DM15L uses a NXP LPC1115 with has a ARM Cortex M0 core. The HP-15C CE has a Cortex M4. About 4x faster it seems.

I was actually looking hard at using a LPC111x in a new design in 2009. In the end I stuck with the Cortex M3 chip I'd already used.

A1