What is Digg?412 Comments
- Phocion55, on 10/12/2007, -8/+350The good news here is that Office 2007 is shiny
- trib4lmaniac, on 10/10/2007, -1/+202Of interest is the fact that 65,535 is the largest number that can represented by an unsigned, 16 bit integer (i.e. 0xFFFF).
After a little experimentation in VBA, I have found that 100000 is only returned when accessing the cells "Text" property. When asking for the cells "Value" 65,535 is returned as expected.
I created a quick macro to test this after filling cell A1 with the formula "=850*77.1"
MsgBox (Range("A1").Text) => Alerts 100000
MsgBox (Range("A1").Value) => Alerts 65535 - octophobic, on 10/10/2007, -25/+200I can't believe they missed a bug in a program with millions of lines of code!
- RealmDown, on 10/10/2007, -8/+177Q: How many Microsoft programmers does it take to change a lightbulb?
A: None. They just change the standard to darkness. - Phocion55, on 10/12/2007, -7/+114Steve Balmer once proudly boasted "WE (Microsoft) make the standards!" didn't he?
So it should be official - 850 x 77.1 now equals 100000 from here on out. - yuuko, on 10/10/2007, -38/+114I can't believe they shipped the product to the market with such major bug considering the fact that it's the nth version of excel.
- mithrasinvictus, on 10/10/2007, -3/+77developers!, developers!, developers!, developers!
- Mike89, on 10/12/2007, -4/+73To quote a Slashdot post on the subject:
"The OOXML specs are already 65498 pages long. If Microsoft can give "multiplyLikeExcel2007" a 37 page treatment in the OOXML specs, the total page count for OOXML will reach an amazing 100000 pages (therefore 166 times better than ODF). Sounds like a winning plan to me!"
-- AC - mercurysquad, on 10/10/2007, -9/+76Quite simply, QA/beta testing always involves checking for error conditions, border values and mid-range values, and 65535 *is* a border value. They simply forgot to check for it, this is not missing one source-code line out of millions, this is missing an obvious step in a routine test.
- Sogui, on 10/10/2007, -3/+62My roommate pointed this out to me earlier. I opened Excel and tried it, but before I could respond I collapsed to the floor and lost all sensation in my limbs.
I can testify to title here, it is truly crippling. - tommasz, on 10/10/2007, -9/+64I seriously doubt users are "crippled", or at least most of them aren't. But for those with massive spreadsheets loaded with formulas, this could lead to some headaches.
- Murdats, on 10/10/2007, -2/+53but it comes with a free froghurt
- ImTheChaz, on 10/10/2007, -4/+53This could have some serious consequences for mankind, lets just hope its not as big as the Y2K bug that nearly destroyed planet earth at the beginning of the new millennium.
- JerodSlay, on 10/10/2007, -7/+51not to bash macs (really im not) but didn't something like this happen with the mac calculator a while back?
- schestowitz, on 10/12/2007, -20/+63Well, that ought to go right into OOXML, right? de icaza's "superb standard", where 1900 is a leap year too.
- mozillamonks, on 10/10/2007, -5/+48and thousands of "developers" :P
- dracflamloc, on 10/10/2007, -2/+44From further in the discussion:
There appears to be more of them. For example:
=5.1*12850
=10.2*6425
=20.4*3212.5
=40.8*1606.25
=77.1*850
=154.2*425
=212.5*308.4
=308.4*212.5
=425*154.2
..etc
What's even stranger is this: Suppose the formula is in A1.
=A1+1 returns 100001, which appears to show the formula is in fact 100000
and a very Serious problem.
And if you multiply be say, 2 you get something else:
=A1*2
returns 131070, as if A1 had 65535. (which it should have been)
=A1*1
Keeps it at 100000.
=A1-1 returns 65534
=A1/1 is still 100000
=A1/2 retuns 32767.5
Using MAX() on a range appears not to see 100000. - SeBBBe, on 10/10/2007, -1/+42Actually, yes. Not really the same thing, but similar.
http://digg.com/apple/Mac_Calculator_is_bad_at_mat ... - duke, on 10/10/2007, -2/+43But, =85 * 771 resolves correctly at 65535
So does =8.5 *7710
So does 8500 * 7.71
It's not simply an issue of the product - there's more to it than that. - FizixMan, on 10/12/2007, -7/+46But it's cursed.
- strictnein, on 10/12/2007, -7/+46Hopefully more than just a few diggers recognize that number immediately: (2^16) - 1
- fatas, on 10/12/2007, -11/+50Cool I hope this was explained in the OOXML specs.
- neko, on 10/10/2007, -3/+41The froghurt is also cursed.
- mzwaterski, on 10/10/2007, -3/+40"and any other calculation for which the result is 65535"
I tested this myself. 850*77.1 indeed resulted in 10,000
but 85*771 resulted in 65535
Maybe it has something to do with one of the numbers including a decimal - Koldkompress, on 10/10/2007, -1/+32That's not a bug, it's a feature. Look:
OPENOFFICE RC 2.01
[TWEAK] Increased 'Reflection time' during between starting Open Office. Allows time to consider how to achieve project. - Elranzer, on 10/10/2007, -0/+30The toppings contain potassium benzoate...... that's bad.
- init100, on 10/10/2007, -1/+31"you would probably never find it unless you were looking for it."
Is it a good thing that it would silently corrupt data without anyone noticing? - macbwizard, on 10/10/2007, -5/+34If scientists are using Excel for their calculations, we're in trouble.
- jonnyeh, on 10/10/2007, -0/+28Which would make it a floating point operation instead of an integer operation
- rossinio, on 10/10/2007, -3/+30The froghurt comes with a choice of toppings
- Ranneko, on 10/10/2007, -0/+27Yeah, this is not a simple boundary value issue, it seems to only break in specific circumstances. Unlike what the description says.
- myak, on 10/12/2007, -0/+27Probably that's why they chose 100000 over 65535. It just looks prettier, at least for people who aren't IT geeks.
- Andir, on 10/10/2007, -2/+28Congratulations. Thanks for reading.
- zadatak, on 10/10/2007, -5/+30I'll attempt to translate for you. He's trying to say that as this bug only affects certain multiplication calculations where the result is 65535 and that you would probably never find it unless you were looking for it. Saying that it "cripples users" is a gross overstatement.
- insomniac8400, on 10/10/2007, -8/+30It could be worse, if it was an Apple product Apple would say it's a feature and doing anything to fix the problem will void your warranty.
- jub0r, on 10/10/2007, -0/+22I beg to differ. Any time a calculator fails to calculate correctly, it is a huge bug. The bare minimum I expect Excel to do is to calculate correctly! People pay Microsoft good money so they have a professional level spreadsheet who's results they can count on. This bug could easily invalidate a lot of people's hard work. I would venture to assume that many scientists, economists, and accountants use Excel to calculate spreadsheets. The data in those spreadsheets is important, and errors in calculation matter.
- renesisx, on 10/10/2007, -1/+22Critical spreadsheet and financial bugs like these were the source behind the original invention of the EULA. Thereby making companies not responsible if their bad coding causes you to go bankrupt.
- aaronm67, on 10/10/2007, -11/+32OpenOffice's main bug: I could open 15 copies of Excel in the time it takes to open up OpenOffice.
- luchid, on 10/10/2007, -3/+24I shall only buy it if it uses propietary, non open formats, is riddled with DRM, and has fading transitions....
- ToadLeg, on 10/12/2007, -6/+26Tell that to the bank customer who ends up with $10,000 instead of $65,535 in his bank. Or the billion dollar satellite that launches with 10,000 pounds of fuel instead of 65,535 pounds.
- scribby, on 10/10/2007, -0/+20lloydmedley, there are 10 types of people in this world....
=:-) - init100, on 10/12/2007, -1/+21Yeah, who cares about bugs when it has the marvelous ribbon. :)
- jambarama, on 10/10/2007, -14/+33Nice sensational headline. I hate me some Microsoft as much as the next *nix fan, but lets be honest: this isn't exactly a critical bug, and it doesn't cripple users. It is an embarrassing oversight but given the length of time Office 2007 has been out, and given the length of time it took to find this bug, it isn't that common to multiply to 65535. If this had been an OO.o bug, would we have seen this outrage? No, we'd have seen the microsofties comlpaining about qa in OSS, but no one would care much and it'd be fixed quickly.
- fatlip, on 10/10/2007, -2/+21not only that, but its not ALL calculations... i did 655350/10 and it works fine -- not marking as inaccurate though because theres still a really ***** bug.
- fredrated, on 10/10/2007, -5/+24"Cripples" users? OMG, will it break my leg, or something worse? Will I get to use the handicapped parking space after I use Excel 2007?
- inactive, on 10/10/2007, -3/+22we do.
- xeomage, on 10/10/2007, -6/+25Wow. I need to upgrade to the latest version right away. Excel 2007 features new maths unlike any previous release! Excel is way ahead of their competitors too. Neither Google nor Open Office provides mathematical answers even _close_ to those Excel 2007 provides.
- ShavedRegressor, on 10/23/2007, -1/+19I checked in Excel 2007, and it does indeed report 580*77.1 is 100000.
If you sum that bad result with 1, it reports 100001, but if you sum the bad result with 2, it correctly reports 65537. This looks like a bizarre display bug.
(And you sir are an "asshat" for throwing about insults after screwing up your attempt at duplication.) - EmileVictor, on 10/10/2007, -3/+20What?
- zimsters, on 10/10/2007, -2/+19ok, remove aluminum foil hat....
now. -
Show 51 - 100 of 410 discussions
Check out the new & improved 
© Digg Inc. 2009
— Content created and posted by Digg users is