Pages in topic:   < [1 2 3 4] >
Déjà Vu X3 is here!
Thread poster: Etienne Tessier
David Turner
David Turner  Identity Verified
Local time: 08:46
French to English
+ ...
I must admit, I don't use.... Mar 12, 2014


Switched from Hunspell speller to MSOffice speller.....DVX3 become like a tortoise.


"the check spelling while you type function" but I just turned it on and switched to the MS
Office speller checker as a test. I couldn't see any slowdown at all and am I quite a fast 10-finger typist. Perhaps there's some specific problem with Russian. I did the test with English as the project target language. I'll report it on the beta group anyway.


Even with preview OFF but with all Autofunctions ON DVX3 is slower than DVX2 (It constructs segments from scrap very slowly).


Again, I haven't noticed this, quite the opposite in fact. DVX3 seem noticeably faster. AutoCheck can slow things down a bit so I turn that off and do Q&A checks on the completed project. I don't really use Assemble but if I did, I'd turn off the DeepMiner option which is really too slow at present with big databases.


I see You answer questions about DVX3, but you are not Atril employee, are you?


No, I'm not!


 
mikhailo
mikhailo
Local time: 10:46
English to Russian
+ ...
re Mar 12, 2014

David Turner wrote:
"the check spelling while you type function" but I just turned it on and switched to the MS
Office speller checker as a test. I couldn't see any slowdown at all and am I quite a fast 10-finger typist. Perhaps there's some specific problem with Russian. I did the test with English as the project target language. I'll report it on the beta group anyway.


No, I switched to MSspeller to check spelling additionally.


Again, I haven't noticed this, quite the opposite in fact. DVX3 seem noticeably faster. AutoCheck can slow things down a bit so I turn that off and do Q&A checks on the completed project. I don't really use Assemble but if I did, I'd turn off the DeepMiner option which is really too slow at present with big databases.


I'm started to understand the cause of problem. With small projects (up to 50 MB??) there is no signs of slowdown and DVX3 may be really faster. But my test project was over 200 MB with multiple import/reimport, so this can be a problem (the project 2 IDML and their derivatives xliff and txt)

May be compacting give some speed increasing. (Checkbox «Compact in background» needed?)


No, I'm not!


How should I believe to the company, if questions addressed to, are answered by another user?


 
Grzegorz Gryc
Grzegorz Gryc  Identity Verified
Local time: 08:46
French to Polish
+ ...
Workaround for large projects... compacting... Mar 12, 2014

mikhailo wrote:

I'm started to understand the cause of problem. With small projects (up to 50 MB??) there is no signs of slowdown and DVX3 may be really faster. But my test project was over 200 MB with multiple import/reimport, so this can be a problem (the project 2 IDML and their derivatives xliff and txt)


Indeed, DVX may get slower for large projects with very large files.
I often export EV (Trados bilingual, I have DVX Wrkgrp), then reimport this EV in a new project.
Usually, it permits to reduce the project size and make DVX more responsive.

May be compacting give some speed increasing. (Checkbox «Compact in background» needed?)

It would be great (e.g. for large TMX imports) but I don't think it's easily possible.
The current DVX architecture needs an exclusive access to the DB in order to accomplish this task.

Of course, the compacting after multiple imports/reimports is always a good practice.
I have now a project like that reduced to 50 MB...

BTW, if you have VERY large TMs, you can put 'em on a SSD drive or on a ramdisk.
I use SoftPerfect RAM disk (freeware).
http://www.softperfect.com/products/ramdisk/

Cheers
GG

[Edited at 2014-03-12 09:56 GMT]


 
David Turner
David Turner  Identity Verified
Local time: 08:46
French to English
+ ...
Another workaround... Mar 12, 2014

Increase the cache size to 2000 seconds in Options/General, open the project in a second instance of DVX3 and run an analysis in the background while you're working on the project in the first instance.

Once a segment is in cache, AutoSearch or Assemble response is almost immediate.

This is only a workaround for very big projects and not very satisfactory of course.

Atril really needs to implement some sort of preemptive database lookup. Most projects stan
... See more
Increase the cache size to 2000 seconds in Options/General, open the project in a second instance of DVX3 and run an analysis in the background while you're working on the project in the first instance.

Once a segment is in cache, AutoSearch or Assemble response is almost immediate.

This is only a workaround for very big projects and not very satisfactory of course.

Atril really needs to implement some sort of preemptive database lookup. Most projects stand almost idle most of the time (sometimes for several days) while the translator is thinking about the next sentence so there's plenty of time for this.
Collapse


Alberto
 
Grzegorz Gryc
Grzegorz Gryc  Identity Verified
Local time: 08:46
French to Polish
+ ...
Poor man multithreaded DVX :) Mar 12, 2014

David Turner wrote:

Increase the cache size to 2000 seconds in Options/General, open the project in a second instance of DVX3 and run an analysis in the background while you're working on the project in the first instance.

Or, probably better, pretranslation with assemble
This "poor man multithreaded DVX" is a terrible weapon if one likes to work with assembled segments.

Of course, the cache size set to 2000 seconds will help a lot.
Frankly speaking, I don't understand why this value is not set to 2000 by default...

Cheers
GG


 
mikhailo
mikhailo
Local time: 10:46
English to Russian
+ ...
re Mar 21, 2014

Assembling is very slow even in small projects.

Another strange thing - having pasted bolded part of segment from Word to DVX3 it's impossible to unbold text in DVX3. Neither «B»-button, nor Ctrl-B works.


 
Grzegorz Gryc
Grzegorz Gryc  Identity Verified
Local time: 08:46
French to Polish
+ ...
Poor man multithreaded DVX revisited :) Mar 21, 2014

mikhailo wrote:

Assembling is very slow even in small projects.

Generally, depending of a lot of factors (TM/TB size, disk/LAN speed etc.), assembling may be a slow operation.
So, I almost never assemble interactively i.e.:
- I open the same project twice on the same machine (AFAIK no other CAT can do it properly),
- I translate interactively in one instance,
- In the other instance, I run pretranslation in the background for one file or for the entire project(almost continuously, if necessary).

Another strange thing - having pasted bolded part of segment from Word to DVX3 it's impossible to unbold text in DVX3. Neither «B»-button, nor Ctrl-B works.

The same occurs for PDF but (e.g.) not for HTML.
It's a bug but it should be easy to fix it in some future build.

Cheers
GG


 
mikhailo
mikhailo
Local time: 10:46
English to Russian
+ ...
re Mar 23, 2014

Preview of Docx in DV3 in projects, created in DVX2 does not work. You will see the pure xml code that You see at F6. So forget about this feature if You got projects from supplier .

Is there any way to preview normally such files?

For all
I.
On Proz TGB DVX3 Pro with 1 year Proz membership — 295 Eur.
II.
To work with MT without API
1. Export file as RTF without codes
2. Use Qtranslate (Ggl will help; it is free) to translate sets o
... See more
Preview of Docx in DV3 in projects, created in DVX2 does not work. You will see the pure xml code that You see at F6. So forget about this feature if You got projects from supplier .

Is there any way to preview normally such files?

For all
I.
On Proz TGB DVX3 Pro with 1 year Proz membership — 295 Eur.
II.
To work with MT without API
1. Export file as RTF without codes
2. Use Qtranslate (Ggl will help; it is free) to translate sets of segments
3. Paste in translation column
4. Copy to sentence status column code 32 (Machine Translated)
5. Import RTF back to DV.
6. correct and put back codes in DVX3
This way allow to bypass autoassemble brakes but reqiures to use CTRL+ALT+Down.

Mr D'Acosta - You were quick to respond on my message about bad russian translation. May be You give us here the timeline for patches to correct already discovered bugs?
(IDML filter, Interface colors, direct working with TRD, MQ, TRANSIT projects)
Will PRO version have got XLIFF EX/IM support for external QA tools, for example?

Proposed feature - Import of revised and corrected documents instead of existing translation.
(in reality aligning two file in one language)

may be it's my last post about DVX3 and it's bugs till the new version. Bye.
Collapse


 
Enlandtrans
Enlandtrans
China
Basically satisfied with DVX3 Apr 7, 2014

Despite all and any defects as mentioned above, as a user following DVX through DVX2 to DVX3 now, I still have to say that DVX3 is the most significant and exciting update so far. Office-alike interface looks cool and comfortable, and more fashionable than DVX2, even though you have to spend time getting used to it, but it really deserves it.
Some changes such as in shortcut or menu items arrangement may take your time to adapt yourself, but it really make me work more efficiently than wi
... See more
Despite all and any defects as mentioned above, as a user following DVX through DVX2 to DVX3 now, I still have to say that DVX3 is the most significant and exciting update so far. Office-alike interface looks cool and comfortable, and more fashionable than DVX2, even though you have to spend time getting used to it, but it really deserves it.
Some changes such as in shortcut or menu items arrangement may take your time to adapt yourself, but it really make me work more efficiently than with DVX2, I still have to say. As for speed, in my machine (AMD X4 3.0G, 4G RAM, 128G SSD), I see no slowness than ever or even a little faster than in DVX2.
The most problem I encounter may be the one with Wubi input (Chinese character input method). When I use my favorite Wubi inputs to work in DVX3, the characters which I typed are actually input in duplicates! So now I have to change to another Wubi input tool, not so bad but it is more or less a pity.
Further improvements are still expected and so is the Workgroup edition~~


[Edited at 2014-04-07 07:28 GMT]

[Edited at 2014-04-07 12:27 GMT]
Collapse


 
Etienne Tessier
Etienne Tessier
France
Local time: 08:46
TOPIC STARTER
New Déjà Vu X3 build Apr 14, 2014

In response to feedback we have received from our customers since the initial release of Déjà Vu X3, we are releasing build 9.0.698 of Déjà Vu X3 to address most of the issues reported. You can download it by going to www.atril.com and clicking on Downloads.

This build includes the following new functionality:

- Support for joining segments across paragraph breaks in Offce Live
... See more
In response to feedback we have received from our customers since the initial release of Déjà Vu X3, we are releasing build 9.0.698 of Déjà Vu X3 to address most of the issues reported. You can download it by going to www.atril.com and clicking on Downloads.

This build includes the following new functionality:

- Support for joining segments across paragraph breaks in Offce Live filter.
- Support for exporting segment status highlighting and comments to the Office Live filter.
- Support for extracting index entries in Word files in the Office Live filter.

It also includes the following fixes:

- Updated visual appearance of various wizards.
- Performance improvements for Live Preview.
- Fix for loss of inline formatting when sending an entire project to the translation memory.
- Improved handling of control characters in the Office Live filter (only applies to newly imported files).
- Fix for issues with formatted numbered lists in the Office Live filter (only applies to newly imported files).
- Fix for issues with incorrectly escaped characters in hidden runs in the Office Live filter (only applies to newly imported files).
- Fix for extraneous codes caused by proofing language changes in Office Live filter.
- Fix for inconsistent embedded HTML and regular expression support in the Office Live filter.
- Fix for inconsistent support for importing comments in Office Live filter.
- Fix for issue with revision tracking in Word documents in the Office Live filter.
- Fix for incorrect detection of tag-only segments in the aligner.
- Fix for character encoding issues in hidden text in RTF and DOC filters.
- Fix for issues with hidden text in the Office 2007 Open XML filter.
- Fix for issue with screen flicker when AutoSearch is enabled.
- Fix for issue with docking pane state not being saved in some cases.
- Fix for issues with disabled dockable panes.
- Fix for disabled button in FILE > Deliver and Export window when entering non-existent path.
- Fix for issues with removing items from the default ribbon or context menu layouts.
- Fix for issue with inconsistent text wrapping and missing lines in the translation grid.
- Fix for certain File tab actions not working when added to customized ribbon.
- Removed overwrite prompt when selecting an existing translation memory in the Alignment wizard.
- Restored support for exporting files from full satellites in the Professional version.
- Reduced docking pane caption size.
- Removed upper case conversion for user and file names in status bar.
- Status bar labels and messages no longer appear in upper case.
- Fix for project tab set not being shown when opening empty projects.
- Fix for error in Batch QA when using invalid SQL filter.
- Fix for various issues with customizing the Quick Access Toolbar
- Fix for incorrect option names in the Export options window.
- Fix for issues with exporting translation memories to Excel.
- Fix for issue with duplicate detection in translation memories with orphaned translations.
- Fix for missing display of unknown client codes when browsing the translation memory.
- Fix for issue with missing termbase matches in AutoSearch when using COM+.
- Fix for incorrect SQL when selecting Confirmed status in the Build Status Expression window.
- Fix for issue with recent file list showing items that no longer exist.

Regards,
Etienne
Collapse


 
Etienne Tessier
Etienne Tessier
France
Local time: 08:46
TOPIC STARTER
In case of missing icons after the last update Apr 14, 2014

Hello again everybody,

Some users reported that some icons were missing and random crashes after installing the last update. This is a random behaviour that happens to some users.

If it happens to you, please install the pat
... See more
Hello again everybody,

Some users reported that some icons were missing and random crashes after installing the last update. This is a random behaviour that happens to some users.

If it happens to you, please install the patch that is to be found here:
http://helpdesk.atril.com/index.php?pid=knowledgebase&cmd=viewentclient&id=193

Thanks,
Etienne
Collapse


 
Pavel Tsvetkov
Pavel Tsvetkov  Identity Verified
Bulgaria
Local time: 09:46
Member (2008)
English to Bulgarian
+ ...

Moderator of this forum
. Apr 28, 2014

Dear Etienne,

1. Has the issue with duplicate termbase recognition error (reported earlier) been fixed?
2. Also, please, advise: is there a shortcut to copy source to target segment?

Best regards,
Pavel Tsvetkov


 
Alexandre De Sena Viegas
Alexandre De Sena Viegas  Identity Verified
Local time: 08:46
Member
English to Portuguese
+ ...
SQLITE3 issues Apr 29, 2014

Hello Etienne,



I recently got a DejaVu license and wanted to import a db3 SQLITE database, but I am getting an error message instead.

There is also some issues with MS Office: no support for MS Office 2010/2013 (import/export)?

Furthermore, when importing TMX files from TRADOS or MetaTexis I first need to open the files with Olifant and change their names.

But I love the XLS import function.

[Bearbeitet am 2014-04-29 23:13
... See more
Hello Etienne,



I recently got a DejaVu license and wanted to import a db3 SQLITE database, but I am getting an error message instead.

There is also some issues with MS Office: no support for MS Office 2010/2013 (import/export)?

Furthermore, when importing TMX files from TRADOS or MetaTexis I first need to open the files with Olifant and change their names.

But I love the XLS import function.

[Bearbeitet am 2014-04-29 23:13 GMT]
Collapse


 
Etienne Tessier
Etienne Tessier
France
Local time: 08:46
TOPIC STARTER
Various comments Apr 29, 2014

Hi everybody,

@Pavel:
1- This willl be fixed shortly, in the next public release hopefully.
2- F5 will do.

@De Sena Viegas

1- Would you mind sending the SQL database to the support ([email protected])?
2- Ms Office 2010/2013 is supported. What kind of issues have you encountered ? A problem has been identified with certain powperpoint presentations, but it has been fixed. If you have issues with such files, import them with the previous Mi
... See more
Hi everybody,

@Pavel:
1- This willl be fixed shortly, in the next public release hopefully.
2- F5 will do.

@De Sena Viegas

1- Would you mind sending the SQL database to the support ([email protected])?
2- Ms Office 2010/2013 is supported. What kind of issues have you encountered ? A problem has been identified with certain powperpoint presentations, but it has been fixed. If you have issues with such files, import them with the previous Microsoft Office 2010 Open XML filter.
3- Again, do not hesitate to send those to support.

Best regards,
Etienne
Collapse


 
Alexandre De Sena Viegas
Alexandre De Sena Viegas  Identity Verified
Local time: 08:46
Member
English to Portuguese
+ ...
XLSX file Apr 29, 2014

Hello Etienne,

I mean following:

If you import/export data from XLS into DejaVu or viceversa, you can only import/export a maximum of 65,000 segments.
If I try to do the same with a XLSX file, DejaVu shows an error message, saying that the file is not a valid EXCEL file.

And it is not a SQL database. It is a SQLITE/SQLITE3 (db3) database.

When using the function EXTERN DATA/ SQLITE to import a .db3 to the DejaVu TM it simply crashes.
... See more
Hello Etienne,

I mean following:

If you import/export data from XLS into DejaVu or viceversa, you can only import/export a maximum of 65,000 segments.
If I try to do the same with a XLSX file, DejaVu shows an error message, saying that the file is not a valid EXCEL file.

And it is not a SQL database. It is a SQLITE/SQLITE3 (db3) database.

When using the function EXTERN DATA/ SQLITE to import a .db3 to the DejaVu TM it simply crashes.
The point is DejaVu doesn't show a field for the upload of the database, it only shows a window with
2 fields, but one is not able to choose anything there. Or do I need first to install a connector?


http://www.ch-werner.de/sqliteodbc/




It still doesn't work after driver installation.

[Bearbeitet am 2014-04-30 09:12 GMT]
Collapse


 
Pages in topic:   < [1 2 3 4] >


To report site rules violations or get help, contact a site moderator:

Moderator(s) of this forum
Pavel Tsvetkov[Call to this topic]

You can also contact site staff by submitting a support request »

Déjà Vu X3 is here!






Protemos translation business management system
Create your account in minutes, and start working! 3-month trial for agencies, and free for freelancers!

The system lets you keep client/vendor database, with contacts and rates, manage projects and assign jobs to vendors, issue invoices, track payments, store and manage project files, generate business reports on turnover profit per client/manager etc.

More info »
Anycount & Translation Office 3000
Translation Office 3000

Translation Office 3000 is an advanced accounting tool for freelance translators and small agencies. TO3000 easily and seamlessly integrates with the business life of professional freelance translators.

More info »