Slashdot Mirror
Performance of OpenOffice.org and MS Office
m85476585 writes "I have used Microsoft Office since I purchased it a year ago. I wrongly assumed that since I paid for it, it must be better, but recently I have noticed that it seems slow, so I decided to try OpenOffice.org to see if it is faster. I compared Writer and Word to see which one is faster and consumes less resources. The results are posted on my website."
And, thanks to totally unscientific nature of the tests, the bloatware really doesn't come out looking any worse than OpenOffice.org!
I wonder what the performance is if Open Office was run on Linux and MS Office on Windows. How does Open Office perform on Windows? Is it slower or faster than on Linux on a similar configuration?
This is my sig. There are thousands more, but this one is mine.
From TFA, opening office takes 12 seconds on average, with first startup being over 30 seconds.
I just rebooted my machine and Word 2000 opened in less than 2 seconds. Oh yea, I'm currently ripping a DVD. My machine is faster than the one tested, but not 15 times faster.
I don't know how the testing is done, but all the quoted speeds seem way, way too high for both apps.
The recent news of Microsoft going to an XML file format in Office in the next release is encouraging, however I don't expect it to be that simple. OO (and AbiWord, for that matter) would probably be considered a real threat to Office/Word if they could simply be on the same level playing ground when it comes to reading/writing compatible files. For all but the simplest docs, I've yet to see a third-party word processor that claims MS Word compatibility to function 100%
I used Openoffice a while back but switched over to abiword. I loved Abiword and used it for some months, but the more I used it I found it to be glitchy and was not as reliable as Openoffice..so I switched back.
Unfortunately, I was recently forced to use MS Office because Openoffice wasn't rendering a DOC file correctly while I was working on a group project for one of my classes. I'm not a big fan of Openoffice's excell equivalent, certain aspects are less intuitive than excell, and I found it impossible to copy graphs from excell to Write. Even if I opened a DOC (created in MS WORD) in Openoffice that contained graphs, things did not turn out right. The graphs were unreadable and didn't render as they do in MS Word.
I am not putting any fault on the part of the folks at Openoffice.org, but the reality is the world is a DOC world and Openoffice has difficulty with those file types. When it comes down to it I'm an Open Source advocate and 90% of the applications are OSS.....the only closed source programs I have 'purchased' are certain games(ie. hl2,doom3,farcry). I don't see any point in purchasing software when you have equal or superior Open Source alternatives.
I will say that Word opens nearly instantly on this platform. It's up in about a second -- perhaps a bit less -- and feels lighter than most of the "minimalist" word processor alternatives I've tried.
My Windows box isn't as muscular as the Mac, but I can't imagine it takes much longer to open Word there. A couple or three seconds, tops.
No doubt that MS Office is bloatware. My Office folder is 486 MB. Outrageous.
But I gotta wonder what is wrong with the reviewer's test computers.
This is my post. There are many others like it. If you don't like what you read here, go try one of the others.
Also, did he make sure that both programs were set to have the same background tasks running (like repagination, automatic spellcheck, automatic hyphenation, etc.)? In one of his tests Word takes a lot longer on a long text file because it's running various automatic tasks on it. Were those tasks run by OO.o as well? I'm pretty sure that all are available, but it may be that some are turned off by default, while with Word it seems that most everything is turned on by default.
I know that when I worked at a Co. that standardized on MS Office, when I got a new PC or they upgraded my version of MS Office, the first thing I had to do was go in and turn off a lot of automatic tasks.
Now that I'm self-employed, I use OO.o. Do I believe it's better than Word? No. Each of them does things the other doesn't and does some things better or worse than the other. Which one is best depends on what your needs are. Right now, my needs are such that OO.o meets them, and it's free.
Start a happiness pandemic
I usually export to pdf from OO.org. It seems to do that pretty flawlessly. Of course, that poses its own challenges if you're emailing a professor, depending on how savvy they are.
I add that the visual appearance of the application is a big hurdle for a lot of users. For instance, on three occasions I've installed OO.org on other people's machines (two colleagues and my mother). On all occasions, they judged the underlying functionality by its presentation before they had even typed a word, and found somebody to provide a pirated copy of Office 2003.
It was, no joke, 180mb
I got it home and opened in OOo writer (I was right, KOffice didn't get everything correct, so I thought I'd use that as a conversion step). I verified everything had imported correctly, added a few more graphs and things (finishing it off) and saved as rtf.
1.2mb. Over two orders of magnitude smaller.
The document is here if anyone wants to try and duplicate the result.
I am trolling
"but I assumed that since I paid for MS Office, it must be better"
So, I should then assume you're an idiot? Crappy consumers like you are why companies can get away with charging outrageous prices. Price != Quality.
"It has been over a year since I installed MS Office, but I know it had to be restarted"
I installed MS Office 2003 YESTERDAY on a friends computer. It did not require a restart. You may have had an older version installed or some other application using a resource that the installer needed to replace.
"Opening time in seconds - First run 31.1"
I am assuming first run refers to the first INSTANCE not the first time the application is ever opened...
WHAT?!?! This is Word 2003? Running on a 2.2 GHz machine with 512 ram? You've got to be kidding me. Did you measure this with a sundial? With my AMD64 Mobile throttled to 40% (800mhz) with a gig of ram, I can start Word 2003 in less than a second.
Also, second instances of Word (I don't know about Writer) open and immediately close again. The second instance simply sends a message to the first instance to open another document window or whatever.
"Word takes up more memory total, but Writer uses more in the main process. It is not a big difference."
What the hell is msworks.exe? I don't have it running right now and Word, PowerPoint, and Excel are all open.
I'm really sick of these horrible comparisons that are performed by armatures. He states he hates Microsoft, goes on and on about how OO.o is better, but states he will continue to use Office. If you are going to perform a scientific experiment, please make it scientific. Leave opinion out of it. Show us exact procedures so we can attempt to reproduce your results. etc etc.
Does someone have an article describing proper construction of benchmarks or a guide to proper scientific analysis? We need some sort of rubric before we keep posting this horrible articles.
http://brandonbloom.name
For starters, let's look at what little data was given about the testing method: Hrrm, that doesn't sound like a very clean environment, speaking of, what is the overall state of their computer? Do they have background processes running that may skew results?
Second, the author doesn't tell us the individual results (save the first) and just averages them for us. At the very least, they should give a standard deviation. (note, there is a non-functional link at the bottom which may contain such things, I do not know at this time)
Third, wouldn't it be a nice idea to test on different computers? Not just a single one? I can understand that the author may have limited resources, but it does limit the scope of the "experiment" (Unless everyone has the same computer as the author)
Fourth, in the "closing time" section the author mentions they closed a specific file. What about closing no files at all?
Fifth, the graphs, at which times did things start and complete? I have no idea what is going on here. For all I've been told the ending trails of the graph (which are all pretty low) are where the "experiment" happens.
I can only hope that the results link (which isn't working at the time that this was posted) contains such data. In any event, I'll admit my bias, I dislike using empirical data. And when I am forced (kicking and screaming) to collect it, I try to minimize the number of variables that may vary. This author does not appear to have attempted that. Thus I don't think that the results can be taken seriously. (and to be honest, I can't believe they were posted on
I could be wrong on some of the points I made, or I could have missed something that is obvious, in any event, I just wanted to express my displeasure with calling this an "experiment" in the first place.
Am I open minded towards open source, or closed minded towards closed source?
I have. When I uprgraded from what ever Word was with Win 3.11 to what ever came with Win 95, all my page numbering got fucked up and some other stuff got fucked up too.
All so the file size blew up about 5 times. And office would crash much more with that file.
Happened constantly to me at a previous job. We had thousands of Word documents formatted by RoboHelp to be compiled into Windows Help files. The switch from Office 95 to 97 introduced a great many minor differences. Nothing broken, and all minor things, but we had to manually check everything and change lots of formatting glitches (for example, styles would randomly change font sizes) and indents.
Actually I've seen OpenOffice 2.0's spreadsheet converting an Excel sheet pretty darn good. All sorts of formulas and interactions just kept functioning. A multi-line table had to be reorganized so that the source data were in parallel columns with one single column for all x-values, and i hope that will be improved but if you think of how complicated MS Office applications are it is very well done.
I found one more problem, a locked Excel sheet was no longer locked after conversion.
No, people don't put up with that.
As has been pointed out by others word doesn't open anywhere near that slow. The author's either lying or a moron who's machine is borked up beyond belief.
It amazes that when people here go out of their way to criticize MS products they are basically saying, "I'm a total idiot that can't use a computer". MS has products that totally suck for joe-blow that doesn't know crap. But a power user can and *should* be keeping windows (since 2000 anyways) clean and stable. If you're not, you're a moron. Should MS make it easier to do so and better? Hell yes. But it doesn't take rocket science to keep it up and clean. People that apparently can admin linux boxes in 31 different flavors are too fucking stupid too keep windows up. Or they're lying.
I think I'll safely say the later is the correct choice.
Windows/Word has enough issues, we don't need to resort to FUD and lies though.
From my experience the export to doc function is sketchy due to the fact that Writer doesn't make up its own formatting rules like Word does and has to try to anticipate how Word would do it. Getting Word to put stuff exactly where you want is a trial in patience and, more often than not, futility. A classic example is with outlines. In Word, you try to space it out with just tab stops, letters, and numbers and that damn paperclip pops up (the one you disabled, no less) and says "Hey! You're making an outline! Make it like this!" And before you know it, your outline looks like what Word thinks an outline should look like, not your employer's, your professor's, or your own version of what an outline should look like. Things like that irritate me, so I don't use Word anymore. Same thing applies to lists, paragraphs, block quotes, and probably a whole slew of minor others. Conversely, there's something about Writer that it sometimes has weird random problems with formatting.
Note to Trip MasterMonkey: Exactly how does one "skew' something like memory usage or how much hard disk space is taken up, especially on a statistic that should be relatively constant across multiple machines?
RTFM
Huh? I've *NEVER* seen a case where a later version of Word couldn't open an older version of Word's documents identically to the original.
Then you've never had a Word document with tables or macros in it. My guess is that it is done deliberately to force all users in a company to upgrade. I could cut MS some slack if it were just that an older version couldn't open a doc from a newer version, but it fails both ways.
I know for a fact wordpad.exe saves pictures in .rtf files as uncompressed bitmaps. I wouldn't be surprised if Word did too, just for compatibility. Chances are OO.o is saving them as PNGs or (even better) the same data as it got from the original file.
It always amazes me what a big deal is made of "start up times." Who cares? If you spend more time starting up an application, than sitting there, effectively using it, then you're not a real user of the application, but just toying with it. In many cases, loading more stuff upon startup, will make operation of an application more peppy.
The same people that go on and on about start up times, don't seem to bitch too much (or maybe they do), about modern day games, where one seems to spend most of their time "Loading..." I find some of the best games today almost unusuable because of the loading delays; it really blows the ambiance for me.
Love many, trust a few, do harm to none.
Try loading Word in Wine and compare it with OOo under Linux. Word would load almost instantly, while with OOo would wait... and wait... and wait. Unless Wine contains a hidden built-in Office preloader, I can conclude that Word loads fast by itself.
:)
:)
THis would not surprise me. Back when I used to write file system filter drivers for NT Word was our testing application of choice. Word makes use of every funky trick it can to talk to the file system. Some of these tricks are even documented
Basically our rulle of thumb was, if we could open a long Word doc with lotsa tables and images and what not in it and our filteer driver dint blue screen, it was pretty safe to assume just about anything could be poked at our driver and it wouldnt blue screen