LibreOffice 3.5.4 released with up to 100% performance improvements | Hacker News Reader

LibreOffice 3.5.4 released with up to 100% performance improvements

blog.documentfoundation.org

130 pointsGarbage74 comments

I really dislike the term "up to N%" - this reminds me of sleazy commercials reporting "up to 80% discount". 0.2% is also "up to 100%". Can't more accurate figures be reported?

I have just installed LibreOffice 3.5.4 and yes, there is a drastic performance improvement. Documents and application do load much much faster (I am on Windows 7). A real improvement here, but I am not sure 'accurate' figures can be reported. I am impressed by the boost.

It could mean it's 100% faster for some tasks. So it could be 100% faster in loading docx or whatever, and only 20% faster in loading .doc. Are you looking for something like an average speed improvement? That probably wouldn't be very accurate, either. But they could say the new LibreOffice is 20%-100% faster, I suppose.

How can it be 100% faster at anything? This is mathematically impossible. If a task used to take 100 seconds, 1 second (an unlikely improvement!) is just 99% faster.

Throughput can be 100% faster. Eg if it could previously decode docx at 100kb/s and now can do it 200kb/s then it would be 100% faster.

Or wouldn't you say that a car that goes 100 mph is 100% faster than a car that goes 50 mph?

No, I'd say it goes twice as fast.

That's the same thing. "Twice as fast" = "Two times as fast" = "200% as fast" = "100% faster"

Making a 100 sec task taking 1 sec is a 9900% improvement, because in the same amount of time, you can do that task 99 more times.

Surely, a 1 second improvement would mean it is 1% faster. (Put another way, no change would mean a 0% change in speed, right?)

Since somethings are bound to have not changed at all in terms of speed it would probably have to be 0%-100% (Or, well, up to 100%.)

Or if they've used some kind of tradeoff, certain things could be much slower (or use more memory, etc.)

I've always had a problem with this too. If a product is 80% off, you pay 20% of the original price; 100% off and it's free. If a product is 80% faster, it runs in 20% of the original time. If it's 100% faster, it runs in zero time.

Generally 100% is used in terms of an increase, where 50% would be used in terms of the same decrease. Since it's twice as fast, it's 100% faster. Which also means it takes 50% less time to do the same task. If it had gotten slower, it would be 50% slower, which means it takes 100% more time to do the same task.

It's all in how it's worded. Instead of being worded as "100% improvement", it could have been worded "50% decrease in how long it takes to complete the operation". One of these rolls off the tongue nicer, but both are valid and mean the same thing.

In your example, both the sales prices are worded as a negative value. "50% off" could be worded as a positive value as "100% more". In this case, LibreOffice has 100% more than the original amount of speed. Twice as fast.

Your right, they should give more details on how they came to this conclusion.

I rather use average improvement rather than best case improvement - that way users don't feel duped.

Does it import Microsoft document formats 100% better? That' the biggest issue that Open/LibreOffice has and until that one is solved for good, the rest does not matter much...

EDIT: not wanting to contribute further to the fall of HN, I edited M$ into Microsoft

The problem started to haunt Microsoft as well -- there are (simple) cases where their files don't open up correctly in slightly different versions of Office. In fact, I've recently had to open some .doc files with LibreOffice instead of Microsoft Office (on Mac) because of this.

It's hardly news. In fact I switched from MS Office to StarOffice 5 in 2000 precisely because of never ending problems with documents. I even had trouble with doc files that one of my colleague made on the very same version (MS Office 2000) running on the same OS (Win98SE), apparently because we had different default printers.

Coincidentally I am currently building a few template ODT/ODG documents on a up-to-date Mac in LibreOffice 3.5.4 that are subsequently filled and converted to PDFs via JODConverter 2.x on...

    $ cat /etc/lsb-release            
    DISTRIB_ID=Ubuntu
    DISTRIB_RELEASE=7.04
    DISTRIB_CODENAME=feisty
    DISTRIB_DESCRIPTION="Ubuntu 7.04"
    $ dpkg -l | grep "openoffice.org "
    ii  openoffice.org                  2.2.0-1ubuntu3                         OpenOffice.org Office suite

and the output is virtually the same, barring a slight difference in font rendering.

(and I too, ditched Office when it completely failed to open my final home produced Word 97-produced document in Word 97 at the university, while Word 2000 did open it with all images suddenly placed at the top-left corner of the first page. StarOffice on the Sun workstations opened it without a hitch, with few pictures offseted by only millimeters. At that stage I could care less about font substitution.)

I would argue that's a Microsoft problem... I don't think office files are even compatible with their own standard: http://en.wikipedia.org/wiki/Office_Open_XML .

A problem for Microsoft? or one by Microsoft? Yes, sure it's caused by Microsoft, but it has to be fixed on the LibreOffice side, since I am sure Microsoft won't do anything about it.

I've had compatibility problems between different versions of Microsoft Office.

This is also Microsoft's problem.

Not until they have competition able to break their lock-in, till then the craziness of their format acts as a barrier to entry. So again, it's a problem created by Microsoft, but it's a problem for their users and competitors. And that's why it's not going to be fixed.

It's not nearly as big a problem for Microsoft. First of all, the compatibility problems are much bigger between LibreOffice and MS Office than between different versions of MS Office. Secondly, if you tell your colleague, "Oh, the margins are screwed up because I upgraded to MS Office 2051", they'll just shrug it off. If you tell them, "Oh, I'm using this new software called LibreOffice because blah blah blah", they'll probably yell at you for being a moron and wasting their time.

If my colleagues ever said I'm a moron and that I'm waisting their time for using LibreOffice ... I would tell them to either buy a Windows+Office license for me, or otherwise go fuck themselves.

And if you're in a position to tell your colleagues to go fuck themselves, then you can do what you want. The rest of us aren't so lucky.

Also, the cost of a Windows+Office license is so small that pretty much any company will happily buy it for an employee.

> Yes, sure it's caused by Microsoft, but it has to be fixed on the LibreOffice side

You can't really fix something when the spec is incomplete, ambiguous, and refers to non-disclosed information, all the while multiple official implementations exist that contradict both themselves and the spec about the behavior to have.

Now what about Microsoft implementing ODT correctly?

One by Microsoft for everyone, including themselves. Not respecting standards is just retarded on too many levels.

EDIT: Unless, of course, they have a private internal standard...

It IS the user's problem as he will be forced to use MS Office for most documents where there are other people involved.

Note: Both google docs and iWorks handle this significantly better.

I just downloaded it because its been bogging down my machine. Its running notably smoother and quicker but I've got a regression on how its rendering the document I'm working from. Whats really stinks, its an internal document so I can't send it upstream so they can fix the issue.

The rendering bugs with Microsoft formats are understandable but still extremely frustrating.

I love Open/LibreOffice but you are 100% correct.

Solution: email who sent you the file "hi, my Microsoft word 2011 can't open the file. Can you send it back saved (insert format you can open)"

I see that as a workaround rather than a solution. Solutions solve problems.

It will draw more attention to the core of the problem: people using Microsoft products deliberately breaking standards for lockin reasons

http://www.gnu.org/philosophy/no-word-attachments.html

M$? Wow, HN has fallen.

Does his comment contribute something? I think it does. Does yours? I don't think so.

So which one you think lowers the standards of HN?

By your own metric, yours is the most useless comment in this thread.

To make mine better, I leave you with this very relevant gem from 10 years ago.

http://www.penny-arcade.com/comic/2002/7/22/

Yours.

I thought about it, but no... When we're talking about office file formats, it was Micro$oft at its best. the last big case of them acting like compete jerks with lots of money that I can remember.

Great. Now how about getting around to outline view, which just celebrated its 10th birthday in the OO bug database?

Whether it's 100% or 40%, I'm very happy with this. The problem with OpenOffice for many years was that it was very slow, and it's even a problem today (e.g. if your machine is heavily loaded and you find yourself having to start an office suite, you usually regret it).

Not just when under load. Up until now, it was a pain watching the UI match up the information around caret position (e.g toolbar like bold, italic, font, etc... and in the style window) full seconds after it moved there.

So how is Apache's new OpenOffice since they got custodianship?

Most of the community developers have jumped the bandwagon to work with LibreOffice. It seems that the developers still with OpenOffice.org are mostly working for IBM and Oracle. The project just did the first release (http://www.openoffice.org/news/aoo34.html) after two years of silence, but there's still a lot to do before they are featurewise competing with LibreOffice. You can see a feature comparison here: http://people.gnome.org/~michael/blog/2012-04-26-ooo-compari...

It's going very well, actually. There was just a release a couple of weeks ago, and IBM just donated the source code to their Symphony product - which was derived from OOo. Symphony apparently has some really nice UI features which will hopefully make it into the mainline of Apache OO.

There is talk on oo-dev about starting the process of graduating from the Apache Incubator, which is an encouraging sign.

I can run new libre from /opt/libreoffice3.5/program only, without Gnome or KDE. Old versions add softlinks to runnable paths. This new version doesn't.

use gnome-do

This is good news actually, we always want improvement in every new software version.

But, i need a better, truly far better, presentation software. If we compare between Ms Office, iWork, and Libre, the only thing i don't like with Libre is the Impress. Impress do presentation like 10 years ago. No keynote-like (just like MS do in Office 2k10) 3D animation, no simple color picker, no shadow effects, no gradient,..

I won't get into the nitty gritty of how much faster it is, or how that kind of speed improvement should be calibrated. All I can say is it loads a heck-of-a-lot faster than it did previously - and also add a big thank you to the folks at LibreOffice.

100%? so now instead of taking X amount of time to do something, it takes 0 time (X-X*100) to do something? Or have I completely misunderstood what a % performance improvement is?

I think you're looking for an inverse relationship.

e.g. My car has 200hp and can do a quarter mile in 20 seconds. If I then increase my car to 400hp (a 100% increase), let's pretend I can now do a quarter mile in 10 seconds.

Thus, a 100% increase in performance drops cuts my quarter mile (or in this case processing) time in half.

Or you can interpret it the other way: before the code could complete X units of work (whatever the metric), now it can complete X + X in the same time (2X).

Now it's taking half, because it's twice as fast.

If we assume "100% improvement" means "100% increase," the most common definition of the latter is simply 2x the former amount. So a 100% improvement in performance likely means 2x the former speed.

In general, I think vendors should be more transparent about performance and reliability:

Microsoft Office 2011....Now using up to 30% less bugs, 20% faster load times, and you get X amount of features.

At least performance/reliability would be considered a feature. It would be better than a lot of the bloatware that comes out these days. Users should be conditioned to ask about performance and reliability of a product rather than how many features they added. You clearly see this from Apple but not Microsoft.

On Ubuntu 10.04 compared to Open Office Writer 3.2, storage of a 209 pages document with Libre Office Writer 3.5.4.2 is even slower.

Program startup of Libre Office Writer is also slower.

Minor detail: after unpacking "LibO_3.5.4_Linux_x86_install-deb_en-US.tar.gz" I got a directory "LibO_3.5.4rc2_Linux_x86_install-deb_en-US", the "rc2" lets me doubt, that this is the release version.

100% performance increase!! thats a bit too much of a statement.

I am wondering if a product can have an 100% performance increase, then it must be pretty badly built the first time around. isnt it? I like openoffice, and didnt really think it needed a 100% performance upgrade, but hey more is better.

So what makes it so much faster? Did they fix a major bug causing it to slow down?

If Libre Office or Open Office really want to get ahead, they need to innovate on their user interface. Being libre/free isn't enough of a reason for some of us, what we need is ease of use and productivity. The Ribbon has been a great step forward IMO and I'd love to see the Libre team challenge it with something even better. Traditional menu bars feel so... 90s.

You are literally the first person (not counting Microsoft press releases) that I have ever heard praise the ribbon UI. If you like it, great, but I'm pretty sure you're in the minority.

Literally every person I've talked to in the corporate world enjoys the ribbon. Only geeks dislike it. It's like saying GUIs are stupid because a geek can be proficient on a command line.

I like the ribbon.

Ribbon is just stealing screen estate.

Just fished helping the wife with a paper on libreoffice... Heck latex is the ultimate user friendliness. But mostly because i can write in vim.

openoffice is a joke. Don't know if it's because i still can't find how to fine tune key shortcuts but all the "speed" problems were me hunting down the same 3 items in a menu over and over.

They should move all the focus on maintaining a non-word clone interface.

Yes exactly. I use LO because I use Linux and don't want to pay for Office on Windows until/unless I have a requirement. But I detest using LO, and your menu example characterizes my overall experience.

Chasing Office compatibility is a losing bet. I think they'd do much better tracking Office at somwhere between 50 and 80%, with sensible default/fallbacks, and spending the bulk of their time making LO the absolute best office solution on Earth.

tl;dr If you try your hardest to be like Office, you're going to suck worse than Office.

oh really, can they open docx files on ubuntu 11.10 now? or do we just have to watch libreoffice crash everytime trying opening one of those?

openoffice/libreoffice family can open docx for a long time.

i know, but on a fresh ubuntu 11.10, it chooses to crash, now i updated to 3.5.4, it's finally OK with it. i hope at least.

100% performance improvement? That seems a little ridiculous (and impossible). I can't find the data that this claim comes from, so I have a feeling they might mean 50% performance improvement (i.e. 1s => 0.5s). Is there more detail anywhere?

(Edit: would it be possible to have an explanation for why this was downvoted? If I've said something stupid/wrong, I'd like to be corrected please.)

Going from 1s to 0.5s is not a 100% improvement, it's a 50% decrease (in time).

i.e.: 100% increase = 50% decrease.

Percents are weird to calculate.

I know that going from 1 to 0.5 is a 50% decrease rather than a 100% improvement. (I actually said this in the parent comment.)

But it's alright, I've got my definitions sorted out now.

To begin with, it's perfectly possible to increase several orders of magnitude the performance of a piece of code, you can Google for plenty of examples. In addition, going from 1 sec to 0.5 sec is indeed 100% improvement.

Given 0.5 sec = 1 ; 1 sec = x, then x = 2

100% performance improvement = 2x as fast as before

Thanks! I just had the terminology backwards: performance is number of ops in a given time, not the amount of time for some number of ops.

(It seems every time performance comes up I manage to get it wrong.)

Reply on news.ycombinator.com