[ChamFan]

Let's just talk about value here.
If I need a spreadsheet program and the choice is between these two, one will involve an outlay of £55 (GBP) and the other £349.95 (GBP). Not only is the microsoft product more than six times as much as the apple one, it also works out as £15.71 per macworld mouse for Numbers and a whopping £116.65 per macworld mouse for Excel - nearly seven and a half times as much!
Given that one has evolved over decades and the other is effectively 'new' it's a no-brainer which one I'm going to spend (my own) money on.
Or, to put it another way which would prefer to spend your £350 on, just Office 2008, or iWork 08 and an iPhone and a nice lunch with your missus?
Like I say... No Brainer

[PIrwin]

In order to use Numbers.app for scientific purposes one needs to have all of the old Microsoft (2004) Excel?s various functions (e.g., the solver add-on and/or the various array-based matrix functions: MMULT(array1,array2), TRANSPOSE(array), MINVERSE(array)). This is particularly important since Office 2008 is almost unusable for scientific purposes.

There are two options: (1) stay with Excel 2004 or (2) buy the Windows version (Excel 2007; Excel 2003 with XP is a better choice). The first option penalizes Intel-based Mac users with somewhat reduced performance and more than normal crashing. The second option requires Vista or XP and VMware Fusion (excellent product) or other virtualization program using either the Excel 2007 (ribbon interface is awful) or Excel (2003 - more like Excel 2004).

[hayesk]

Plrwin, I'm not sure why you'd use either for scientific purposes. There are purpose-built statistical and graphing applications that do a far better job than Excel on Mac or Windows.

[rei_vilo]

Bonjour,

Congratulations for the great comparison!

iWork is slow on my iMac G5 20.1 MHz. Two key features are missing: target value and pivot table.

Target value provides the answer to "" how to" get that result. Pivot table is the most efficient way to sum up raw data.

So I'm still waiting for iWork' 09!

Best regards from France,

[Jason Snell]

ChamFan said:

Let's just talk about value here.

If I need a spreadsheet program and the choice is between these two, one will involve an outlay of £55 (GBP) and the other £349.95 (GBP).

Office sells for 87.48 GBP. iLife for 51.98 GBP. According to amazon.co.uk.

It's easy to bash office when you quote the full price, but the Home and Student edition is the one almost anyone who doesn't need Exchange support would buy. It's not license restricted like the old Student and Teacher edition -- anyone can buy it, and it's quite a good deal.

For us Americans with our extremely weak currency, that's $130 for Office and $70 for iLife.

[ChamFan]

Ah! There you go! It's the "Home and..." bit I didn't see.
I had the student edition once upon a time and had to jump through the hoops taking my matriculation card to the local apple reseller before they'd let me buy it. Guess I just assumed this was much the same. Thanks for the enlightenment and apologies for my error. Guess I'll just have to work up another justification to get an iPhone now. (Hmm... apps? #100 off? 32gig? 3G???)

[CheeseHead]

Jason Snell said:

For us Americans with our extremely weak currency, that's $130 for Office and $70 for iLife.

Office Home/Student edition also allows you to install it on three machines. iWork is for a single machine unless you get the $99 family pack.

[bowser]

"Plrwin, I'm not sure why you'd use either for scientific purposes. There are purpose-built statistical and graphing applications that do a far better job than Excel on Mac or Windows."

Agreed. Niether is built for the purpose, and there are far better alternatives as I mentioned previously.

For serious computation Matlab blows Excel away, Excel NEVER produces valid statistical analyses so SPSS should be used instead, and if you need to do good looking graphs, DeltaGraph or KaleidaGraph kick Excel's a.

There's no reason at all to use Excel unless you're doing book keeping.

[PIrwin]

Why use excel? Simple. Everyone has it on their computer and most statistical analyses can be trivially performed with the various matrix or other built-in functions. (see: "Curve Fitting in Nuclear Magnetic Resonance Spectroscopy: Illustrative Examples Using a Spreadsheet and Microcomputer". Concepts in Magnetic Resonance 6:57-67. 1994) The statistical packages you mention are useful but expensive AND they are "black boxes" and require little understanding of the processes behind the statistics. With the old Excel you perform the raw calculations (transposing the derivative matrix, multiplying the transpose by the original matrices, taking the inverse of this result, multiplying the previous by the original matrix, multiplying the latter by the observed minus the calculated vector; see also: "The modified Gauss-Newton method for the fitting of non-linear regression functions by least squares" Technometrics 3:269-280) and have control over how fast or slow the results converge. Much more useful than any statistical package which tend to be completely closed. Of course, one must understand such things.

[mdennis74]

As an educator, I must say there is one fatal flaw to Numbers: the inability to orient text vertically within a cell. This oversight makes the program cumbersome for even simple tasks.

For example, look at the grade-books most teachers use. You will see that they are set up to allow student names down the side (1 per row), and assignments across the top (1 per column). To make this work, grade-books give room for you to write the assignment vertically in the column. That way, you can fit a maximum number of columns on each page, allowing you to view a larger snapshot of your students' work over time.
In numbers, however, we are forced to enter all text horizontally, wasting a massive amount of screen real estate and making it necessary to scroll horizontally across a page or flip between too many separate tables. It is impossible to print all the data you would want to see on a single page.

The only work-around provided is to use text boxes. However, the data in text boxes cannot be sorted the same way as data in cells, they not resize properly when the column width is adjusted, and they wreak havoc if you have to change printer margins.

This is not a major engineering feat and should not have been overlooked by a company that wants to lay claim to the education market. I am also surprised that so many reviews have neglected to mention it, since viewing and printing a simple page of data is a major function of any spreadsheet program at any market level.

[c218]

jscottk said:

Oh, I forgot. While it's no where near as powerful as Excel's "freeze panes" abilities, when you use Header rows, they will show up on every page when a table gets too big to fit on a single page. (Of course, you can alway use the scaling slider in "print preview mode" to fit it all on one page if your want too!)

Scott

How do you do this? I use my mom's Macbook to help her at her art gallery. I've been working on her painting inventory as a spreadsheet in numbers. I can't for the life of me get the header row to repeat as I scroll down the page. And I have "repeat header row selected" I'm a big Excel (PC) person, so numbers is a bit frustrating.
Thanks!