hecht@unx.sas.com (Mike Hecht) (10/25/90)
Now that Apple has stabilized their product line and brought some order to their
pricing structure, I've gotten curious as to how they set their prices. So,
armed with my trusty stat package, JMP, I set about trying to derive a formula.
I created a data table containing information about each mac model. My table has
the price, RAM, hard disk size, processor, clock speed, whether an FPU or an MMU
is included, number of ADB ports, number of nuBus slots, whether sound input and
stereo output is available, and whether the machine has a processor direct slot,
an external drive connector, built-in video, a cache slot, or a built-in
monitor. I left out things that were common to all the macs (like FDHD floppy
drives). I also left out the Macintosh LC, since it is not yet in full
production and I don't have an "official" price for it.
I created a model predicting the price from all the other factors and came up
with the following formula:
1) Apple credits you 2564.30
2) Apple charges 166.58 per MB of RAM
3) Apple charges 11.91 per MB of Hard Disk
4) Processors are cheap. If you have a 68000, it's free.
If you have a 68030, Apple pays *you* 2619.31
5) Clock speed is expensive, however. Apple charges 300.32 per Mhz.
6) FPU's and MMU's cost nothing (ie: they're not significant in the model).
7) ADB ports are a shocker -- they cost 1713.02 per port!
8) NuBus slots however, give you a credit of 454.85 per slot.
9) Sound input is free (gee, why isn't it on my IIfx?).
10) Stereo output costs you 790.36, but if you don't have it Apple *credits*
you by the same amount.
11) The rest of the factors are all free.
This gives me the following predicted prices (numbers are MB RAM-MB Disk):
Model Retail Predicted
Price Price
===== ======= =========
Macintosh Classic 1-0 999.00 927.50
Macintosh Classic 2-40 1499.00 1570.50
Macintosh SE/30 1-40 4369.00 4480.92
Macintosh SE/30 4-80 5569.00 5457.08
Macintosh Portable 1-0 4799.00 4910.79
Macintosh Portable 1-40 5499.00 5387.21
Macintosh IIsi 2-40 3769.00 3680.92
Macintosh IIsi 5-80 4569.00 4657.08
Macintosh IIci 4-0 5969.00 5842.59
Macintosh IIci 4-80 6669.00 6795.41
Macintosh IIfx 4-0 8969.00 8982.84
Macintosh IIfx 4-80 9869.00 9935.67
Macintosh IIfx 4-160 10969.00 10888.49
Not too shabby a correlation.
Ok. So now that I have a formula, what's the most powerful Mac that Apple could
make at the lowest cost? Well, it would have to have an '030, which includes the
MMU. Since FPU's are free, we'll take one. Let's give it 5 MB of RAM and no hard
disk (we'll buy a 3rd party one). Let's run the clock speed at 20 Mhz. We ought
to have at least one ADB port even though its a money-pit. However, we could
more than pay for the ADB port by adding six nuBus slots. We'll take sound input
(it's free) but no stereo output. A processor direct slot, an external drive
connector, built-in video, a cache slot, and a built-in monitor are all free, so
let's take them all.
I fed in my requirements into JMP and it spit out a predicted price for the
"Dream Mac": -150.73!! Apple owes me money!
So, Apple, if you're listening, please send me the requested machine and a
cashier's check for the above amount. I urge other readers to do the same.
:-) :-) :-) (please insert about 10,000 smiley-faces at this point) :-) :-)
Aren't statistics wonderful?
==============================================================================
Michael Hecht Internet: hecht@unx.sas.com
SAS Institute Inc. AppleLink: SAS.HECHT
Cary, NC USAogil@quads.uchicago.edu (Brian W. Ogilvie) (10/25/90)
In article <1990Oct24.233058.14408@unx.sas.com> hecht@unx.sas.com (Mike Hecht) writes: >Ok. So now that I have a formula, what's the most powerful Mac that Apple could >make at the lowest cost? But Mr. Hecht's figures are for retail price, not cost. And, as everyone knows, Apple's retail prices seem to have nothing to do with how much it costs them to make a machine! (Remember, the SE cost Apple less than the Plus in manufacturing cost per unit!) :-) :-] ;-/ (-::-) [Siamese twins...] -- Brian W. Ogilvie "Find a need and fool it!" ogil@midway.uchicago.edu --Zippy the Pinhead
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) (10/26/90)
In article <1990Oct24.233058.14408@unx.sas.com>, hecht@unx.sas.com (Mike Hecht) writes: >Now that Apple has stabilized their product line and brought some order to their >pricing structure, I've gotten curious as to how they set their prices. So, >armed with my trusty stat package, JMP, I set about trying to derive a formula. >I created a data table containing information about each mac model. [cut randomly generated drivel....] >I fed in my requirements into JMP and it spit out a predicted price for the >"Dream Mac": -150.73!! Apple owes me money! Do they have a drug-testing policy at your place of employment? If so, how have you managed to avoid detection? >Aren't statistics wonderful? 97.36% of all statistics are made up anyways. >============================================================================== >Michael Hecht Internet: hecht@unx.sas.com >SAS Institute Inc. AppleLink: SAS.HECHT I thought SAS was a statistical package....
geoff@pmafire.UUCP (Geoff Allen) (10/27/90)
sysmgr@KING.ENG.UMD.EDU (Doug Mohney) comments on an article written by hecht@unx.sas.com (Mike Hecht): |[cut randomly generated drivel....] | |>I fed in my requirements into JMP and it spit out a predicted price for the |>"Dream Mac": -150.73!! Apple owes me money! | |Do they have a drug-testing policy at your place of employment? If so, how |have you managed to avoid detection? | |>Aren't statistics wonderful? | |97.36% of all statistics are made up anyways. Hey, give the guy a break. After all he *did* say... > :-) :-) :-) (please insert about 10,000 smiley-faces at this point) :-) :-) near the end of his message. -- Geoff Allen \ Computers are useless. uunet!pmafire!geoff \ They can only give you answers. bigtex!pmafire!geoff \ -- Pablo Picasso