I've been adding te books in my library to Calibre, and have only added around one fifth or one quarter (about 1000 books) so far (most downloaded from MR, thanks to all the contributors!) and I find that Calibre is slower than I'd really like (on a Core2Duo). I'm wondering what things will be like as I add more books. I don't feel that my library is particularly large, but I haven't seen too many complaints about Calibre's user interface speed. Is everyone just more tolerant than I am?

I'm at about 700 books right now and it seems fine to me. A small lag when I first start the program, but my macbook is two years old, so every program has a small lag when you load it :)

My calibre library is up near ficbot's, and I haven't noticed any serious lag.

About a thousand or so. MacBook is 2 years old, so like ficbot, speed is same as most programs.

One thing that DOES affect speed dramatically is running with your library on a network share. The latest versions of Calibre have options (via an environment variable setting) to have the metadata.db file locally and the book files on the share. This gets back to about the same speed as running everything locally.

I have about 2700 books in my library. I ran the program and timed a few things. It takes 18 seconds to open and fully populate the library. It takes 4 seconds to compete a keyword search. It takes 20 seconds to add 12 ebooks.

These times seem fast enough to me.

Thanks for the replies. I'm glad Calibre is performing fast enough for you guys.

I was more wondering, though, whether anybody felt that it slowed down significantly over time as more books were added. To me, Calibre is already too slow (run it side by side with something like iTunes for comparison) for a library app so I'm wondering if when I add the rest of my books it will get even slower. Then I should look around for an alternative and keep Calibre for things like conversion and meta-data editing instead.

I have not noticed very much slowdown as the numbers get larger (I have about 8000 in my library). Having said that a lot of people use Calibre from the command Line which is much faster.

Mine is relatively small at only about 300 book but it is nice and snappy on my Mac :)

My calibre library is up near ficbot's, and I haven't noticed any serious lag.

I should add that I'm on a laptop with a 1.7G Pentium M CPU.

If you're having problems even though you have a much better computer, then I have to wonder what else you have running in the background.

I have about 800 or so books in my library and it's pretty snappy for me.

If you're running Linux, I found yesterday that removing gnome-screensaver and gnome-power-management loosened things up a lot. xscreensaver seems to be fine. I'm on Xubuntu, so I don't really need them.

m a r

New one month old 505 and stacking up 155 ,more to come.

1663 books so far (quickly growing!) on a Win7 system, and it's very responsive.

A little over 7200 books. I am seeing reasonable performance. My system is a Quad-core at 2.8GHZ, so your mileage might vary.

I've got a library of just under 5000 books. Load is slow, but I'd imagine that's more because I have the calibre database on an external HD. Actual operation is fine, even loading new books.

I've got a library of just under 5000 books. Load is slow, but I'd imagine that's more because I have the calibre database on an external HD. Actual operation is fine, even loading new books.

I too have Calibre database located on an external HD just so I don't have to keep moving books over to that drive everytime I load a new one. I have about 1000 or so books in the library, I notice that when Calibre starts and my Kindle is connected it drags somewhat as it loads the database, but not anything to dramatic in terms of timing. A few seconds more than when I only had 300 or so books on. Like ssimon2000, actual operation is fine, loading new books is quick.

What? 5000? 7200? jeeze what is that the library of congress? :) :) ;)

I'm just a nobody nothing with my measly 50 books.....

Some of my 'books' are things like recipes and documents for work...

You don't mention what OS you are on.
If you have a Mac and (still) run it in Rosetta mode then it will be very sluggish.

Other then that the only thing I find a bit slow is the coverflow.

What? 5000? 7200? jeeze what is that the library of congress? :) :) ;)

I'm just a nobody nothing with my measly 50 books.....

:rofl:

I've got a lot of sermons (Charles Spurgeon, etc) and early sci-fi/pulp short stories.

:bookworm:

You don't mention what OS you are on.
If you have a Mac and (still) run it in Rosetta mode then it will be very sluggish.

Other then that the only thing I find a bit slow is the coverflow.

I'm on Vista with a 2.9GHz processor.

:rofl:

I've got a lot of sermons (Charles Spurgeon, etc) and early sci-fi/pulp short stories.

:bookworm:


Gotcha. I keep searching for various short story collections but there don't seem to be a lot available but also have a few individual short stories on mine. :)

Well, the Baen Free Library all by itself is well over 1GB of books if I remember correctly, and might even be over two or three... And I've been reading on mobile devices since the original Palm DOC program, so that's a lot of files...

I run Calibre on multiple machines, a Core2Duo WinXP, a Core2Duo Macbook and an AMD Athlon Vista machine. I haven't tried a Linux version yet (tried to build it, but man oh man it has a lot of dependencies).

I find Calibre slow on all the machines that I've tried. Maybe my expectations for user responsiveness are too high, but I tend to think that even if the underlying operations take a long time, the UI should still be snappy because it really doesn't do all that much.

I have 13,000+ books with no performance problems even when running several other heavyweight apps at the same time. Search takes under 2 seconds, adding 10 books less than 5 seconds, converting book formats can take 5 or 6 seconds per book.

Part of the performance answer may lie in system specs. I have a Phenom quad core, 8 Gb ram and 4 Tb hard disks running Xubuntu jaunty.

While pondering the "snappyness" issue, it occurs to me that the overhead is most likley the manipulation of the actual files and embedded metadata, rather than the database itself.

Which makes me wonder if perhaps an option to queue updates to the files until idle / scheduled time / commit command might make sense.

It's the 'loader' app that sucks cpu cycles. Processing news, converting. Database itself is not heavy, I think.

m a r

Runs fine on my system but I am still collecting. On a side note here is a site with lots of short stories..

http://yuletidetreasure.org/archive/quicksearch.html

982 books with a start-up time of about 10 seconds with about 5 seconds per book added on a WinXP 2.0ghz dual core cpu and 2gm ram laptop.