meghas@hpcvra.CV.HP.COM (Megha Shyam) (05/05/90)
/ hpcvra:comp.sys.handhelds / rhelps@yoda.byu.edu / 6:00 pm Apr 22, 1990 / >I am looking for the pinouts for the serial interface on the HP48sx >The suppserial cables from HP seems no more than a trickle at present. For >those adventurous souls that want to do it themselves here is a repeat of my >earlier message on pinouts. >Pin numbering. Lay the calculator on the table, look into the port and number .1-4 from left to right. 4 is closest to the IR leds. .IBM AT 9 pin plug HP plug >2 (Rx) 2 >3 (Tx) 3 >5 (Gnd) 4 >--- 1 (Chassis ground ??) >You may also want to short out pins 1-6-7-8 on the IBM side to keep its >handshaking satisfied. See other notes on this net for suggestions for a plug >to go into the HP. >Request to HP: We are all eagerly looking out for accessories such as the >Programmers Reference Manual and the engineering formula ROM card. Don't keep >us hanging on too long. RAM cards have arrived but the price is daunting. >With the current price of RAM chips for PCs does your special packaging really >justify $250 for 128K? >- Richard Helps - >BYU, Provo, Utah. All the usual disclaimers and please be careful when >handling your calculator hardware. Comparing RAM prices for PCs and deducing the price for the HP 48SX RAM card is like comparing apples and oranges; PC almost exclusively use dynamic RAMS which are of course very cheap; HP 48SX needs and uses static RAMS which are much more expensive. Part of the reason is one bit of a static RAM has 6 transistors while i bit of a dynamic RAM has one transistor. Static RAM density has not reached the DRAM density level and will probably never compete in price. The other issue to consider is that the RAM card should be considered as a system which meets HP 48 electrical specifications which is quite different that the standard 5 +/- .5 V specification. All these add to cost of the manufacture of a card. I hope that this clarifies the situation. Megha Shyam Hewlett-Packard
madler@piglet.caltech.edu (Mark Adler) (05/05/90)
In article <21580088@hpcvra.CV.HP.COM> meghas@hpcvra.CV.HP.COM (Megha Shyam) writes: >>us hanging on too long. RAM cards have arrived but the price is daunting. >>With the current price of RAM chips for PCs does your special packaging really >>justify $250 for 128K? > >Comparing RAM prices for PCs and deducing the price for the HP 48SX RAM card >is like comparing apples and oranges; PC almost exclusively use dynamic RAMS >which are of course very cheap; HP 48SX needs and uses static RAMS which are >much more expensive. Part of the reason is one bit of a static RAM has 6 >transistors while i bit of a dynamic RAM has one transistor. Based on current mail order prices (just to be fair), a bit of dynamic ram costs about 7 u$ (micro-dollars), while a bit of HP-48SX memory costs about 200 u$. This factor of almost 30 is only partly accounted for by the increased chip area of static RAMs (and that factor is not six, but closer to four since the dynamic ram is a transistor and a capacitor per bit). The remaining factor of seven can be partially explained by the packaging of the 128K cards, but I suspect that most of that factor is due to a combination of lower volume production than 1Mx9 SIMM's and higher profit margins, also due to the lower volume. HP can have a higher profit margin since they have essentially a captive audience for HP-48SX add-ons and no real competition. I do not begrudge them this profit, since how else can they keep their employees so happy. However, I have not bought a 128K card yet, because of the price---especially the ratio of the add-on price to the calculator itself, which is nearly one. The purpose of this note is to suggest to HP that they might realize greater profits on the 128K cards if they lower the price. I suspect that the market research that led to the current price is, perhaps, wrong. Or, maybe they're smarter than me and getting the current price from those that will pay it, but once that storehouse of die-hards is used up, THEN they'll lower the price to seduce the next level of calculator junkies. Mark Adler madler@tybalt.caltech.edu