A sample text widget

Etiam pulvinar consectetur dolor sed malesuada. Ut convallis euismod dolor nec pretium. Nunc ut tristique massa.

Nam sodales mi vitae dolor ullamcorper et vulputate enim accumsan. Morbi orci magna, tincidunt vitae molestie nec, molestie at mi. Nulla nulla lorem, suscipit in posuere in, interdum non magna.

TM’s “under the hood” – Trados vs. memoQ

Translation tools use different mechanisms to store data in their translation memories, resulting in TM’s that varies greatly in size and structure. This video shows how a sample Finnish-English TMX with about 70,000 translation units and 48 MB results in a 31 MB TM in Trados Translator’s Workbench 8.3, in a 83-139 MB TM in SDL Trados Studio 2011 and in a 303 MB TM in memoQ translator pro. The Trados Workbench TM consists of 5 files, the Studio 2011 of a single file and the memoQ TM of 20 files.

This constitutes the introduction to a video to be released later on about concordance search in Trados Workbench, SDL Trados Studio and memoQ, focusing on single Finnish words.

Relevant link:
TMX translation memories (fi-en) used in the video

5 comments to TM’s “under the hood” – Trados vs. memoQ

  • István Lengyel

    One important piece of information regarding the memoQ TM size is that reverse lookup was enabled – this almost doubles the size, because it builds an English and a Finnish index as well. If you disable this checkbox, you are only able to use the TM in one direction, but the size (and the import speed) shrinks considerably.

    • Dominique

      Good point, István! I’ll check the impact of reverse lookup vs. no reverse lookup at some point.

  • SafeTex

    Hello Dominique

    I’ve never understood why all the CAT Tools need to convert TMX

    I can do character and word concordances searches in them using Olifant so why do CAT tools insist on converting them?

    Thanks in advance is you find the time to answer this

    • Dominique

      Hi Dave,
      TMX is a convenient (though not perfect) way to exchange translation memories between various tools. Various CAT tools have their own TM format to suit their own needs. For instance, the same TM can be used in memoQ to translate in both directions, whereas in other tools, you typically need to “reverse” it, ie. create a separate TM. It’s possible to perform concordance searches in Olifant (and in Xbench) with TMX TM’s, but it’s rather slow, at least with big TM’s, because these tools don’t use an index together with the main database. Tools like memoQ, Trados, Wordfast etc. usually create an index that speeds things a lot. Some CAT tools accept TMX as a format without any conversion: Swordfish and CafeTran come to mind.
      Hope this helps,

      • Sakina

        Hello Dominique,

        I have been looking for someone with experience with CAT tools to help me, I’m new to using them. First of all, I downloaded Omega T and before I start working on it, it would be very helpful if I could make a translation memory from all my previous work, especially being that I mostly work with an organization that uses lots of repetitive ideas and terminology. I have about 100 translated documents, in the source language and the target language, how can I make a translation memory from them????


Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>