Path: csiph.com!fu-berlin.de!uni-berlin.de!individual.net!not-for-mail From: rbowman Newsgroups: alt.comp.os.windows-10,comp.os.linux.misc Subject: Re: MS Access Date: 20 Aug 2024 04:56:36 GMT Lines: 23 Message-ID: References: <8WUuO.115458$COA2.7128@fx46.iad> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-Trace: individual.net b3vSUUTXmsMDq+2k84SD5gEk/ONqG8cKtYsgBrFOfs6ZYa6BiR Cancel-Lock: sha1:7+hfp1ZbsikRqo0KNB6PFnofDwY= sha256:5YPOsdZQi/KFpzsw/q1atSy5f4mRAhEQoNKQt+hWavM= User-Agent: Pan/0.149 (Bellevue; 4c157ba) Xref: csiph.com alt.comp.os.windows-10:178356 comp.os.linux.misc:58068 On Mon, 19 Aug 2024 23:55:44 -0400, 186282@ud0s4.net wrote: > Assembler CAN be faster, but you have to write good tight assembler. > Some micro-controller experience helps with that. When RAM/storage is > oft measured in BYTES you've gotta do clever work. I worked on a handheld pH meter that used the 8748 and you knew where every byte was. https://www.thomassci.com/nav/cat1/phelectrodes/0 The Ross electrodes were the same but another programmer worked on the ion concentration meter. We could share some code but there weren't enough bytes to do both in the same instrument. The bench models used the Z80 with all the room in the world, or so it seemed. During the interview for my present job the head of engineering posed a problem and started off with 'assume unlimited memory'. Whoa, Dorothy, this ain't the embedded world anymore.' After 25 years of that and being mostly retired I'm back to messing around with Pico Ws, Arduino Nano 33s, and so forth. Even those have luxurious amounts of flash and SRAM.