Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[David Addy]

Hi Alberto,
Startup time with a blank project is 30 seconds here with Kaspersky security
ON.
With Kaspersky turned off, start up time is still 28 seconds - barely any
difference.
Kaspersky is generally known for being a quite  low overhead.



--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Alberto De Luca]

Hi all,

I don't know if this is has already been brought forward: on Windows 10, at
least for me, the antivirus service vastly increases the QGIS loading
times. I did a quick test today (I have got more than 20 plugins installed
and active):

- AV disabled: loading time 23 secs
- AV enabled: loading time 45 secs.

I use the standard Windows 10 antivirus, but I've noticed something similar
on a box running ESET Endopoint AV.

Alberto

On Tue, 18 Dec 2018 at 10:14, David Addy  wrote:

> Hi Nyall,
> In answer to your query, all my files  are held on my desktop PC, with
> specs
> as follows:
> CPU: Intel Core i7-4790K
> GPU: Nvidia Quadro K2200
> SSD: Samsung 840 Evo 250GB
> HDD: Seagate Barracuda 7200.14 2TB
> HDD: Seagate Barracuda 7200.14 2TB
> USB: WD My Book 1230 4TB
> RAM: Corsair CM3X8GA2133C11Y4 2x8GB
> MBD: Asus Z97-K
> I also run Kaspersky Internet Security Suite.
>
> Unfortunately, although opening QGIS with a blank project had speeded up
> the
> load process yesterday, today it is still taking half a minute to go from
> an
> empty outline window to the normal QGIS screen. In the light of other
> comments I tried unchecking the Processing plugin, but there was no
> noticeable improvement. This is all very strange.
>
>
>
>
> --
> Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
> ___
> Qgis-user mailing list
> Qgis-user@lists.osgeo.org
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user



-- 
*Alberto De Luca*
Consulente GIS
albertodelu...@gmail.com
T. 329 427 2205
https://www.adl.cloud 
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[David Addy]

Hi Nyall,
In answer to your query, all my files  are held on my desktop PC, with specs
as follows:
CPU: Intel Core i7-4790K 
GPU: Nvidia Quadro K2200 
SSD: Samsung 840 Evo 250GB 
HDD: Seagate Barracuda 7200.14 2TB 
HDD: Seagate Barracuda 7200.14 2TB 
USB: WD My Book 1230 4TB 
RAM: Corsair CM3X8GA2133C11Y4 2x8GB 
MBD: Asus Z97-K
I also run Kaspersky Internet Security Suite.

Unfortunately, although opening QGIS with a blank project had speeded up the
load process yesterday, today it is still taking half a minute to go from an
empty outline window to the normal QGIS screen. In the light of other
comments I tried unchecking the Processing plugin, but there was no
noticeable improvement. This is all very strange.




--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Bo Victor Thomsen]

Hi Nyall -

You scored a bullseye on that - I've done some testing

   - Startup with all plugins activated: *28* seconds
   - Startup with processing plugin deactivated : *19* seconds
   - Startup with proccessing and "Kortforsyningen" plugins deactivated -
   *11* seconds

(Test done on my 2. work-pc a Lenovo T570, I'll do some further checking,
when I get to my )

So the "Processing" plugin takes a whoppin'  9 seconds to start. But it's
not the only culprit: The "Kortforsyningen" plugin takes 9 seconds too.
Isn't it posibble to pack all this checking  in the various plugins into a
background task ?






Den tir. 18. dec. 2018 kl. 00.25 skrev Phil Wyatt :

> BINGO! That’s a lot faster startup!
>
> > My guess is that the initialization of the python plugins runs very,
> very slow on the Windows pc. So the culprit may be the python interpreter
> on Windows.
>
> This sounds like a known issue since ages (v 2.6?), where at QGIS
> startup the Processing plugin parses all the hundreds of description
> files for the SAGA/GRASS algorithms. You can verify by disabling the
> Processing plugin and relaunching.
>
> Nyall
> ___
> Qgis-user mailing list
> Qgis-user@lists.osgeo.org
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>
>

-- 
Med venlig hilsen

Bo Victor Thomsen
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Patrick Dunford]

Sorry for that, my mistake

Easy one to make, however, when you have four Linux PCs to do the work 
on and one token Windows PC that is unused most of the time these days.


On 14/12/18 10:50 PM, Nyall Dawson wrote:

On Fri, 14 Dec 2018 at 11:36, Patrick Dunford  wrote:

As far as I know, development is done mainly in Linux

The Windows and Mac editions are ports off this. I am sure the
developers and port teams do their best but you can run into issues that
are platform specific and are hard to resolve. Particularly as it is
dependent on third party libraries that again may be less well supported
on all the platforms.

I don't think this is an accurate reflection of the quality of the
Windows builds. While it's true that the majority of developers use
Linux to develop, speaking for myself at least when I'm a QGIS *user*
I'm nearly always in Windows -- and I don't have any major issues with
it.

There is occasional quirks which show up on Windows which don't on
Linux, but there's probably an equal number of quirks which show up
specifically on individual Linux distros and not on Windows!



___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Phil Wyatt]

BINGO! That’s a lot faster startup!

> My guess is that the initialization of the python plugins runs very, very 
> slow on the Windows pc. So the culprit may be the python interpreter on 
> Windows.

This sounds like a known issue since ages (v 2.6?), where at QGIS
startup the Processing plugin parses all the hundreds of description
files for the SAGA/GRASS algorithms. You can verify by disabling the
Processing plugin and relaunching.

Nyall
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Nyall Dawson]

On Mon, 17 Dec 2018 at 18:49, David Addy  wrote:
>
> Hi Nyall,
> Thanks for your additional comment.
> When I was autoloading my most recent project at startup QGIS was taking up
> to a minute to open itself and load the project. Other saves and actions
> like opening a layer attribute table were also very slow.
> After changing my settings to open with a new project (ie a blank project)
> QGIS was opening as normal. Then loading the old project also was as fast as
> it should be.
> However, after editing some polygons,  when I come to save the changes I am
> still getting a long pause when nothing happens, then I get "QGIS is not
> responding",  then it comes back to life again and finishes the save.
> So changing the setting vastly improved startup, but seems to have not
> solved the whole problem.

Question - are your project files local? Or stored on a network drive?
If so, does it make any difference if you use locally stored projects?

Nyall


>
>
>
> --
> Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
> ___
> Qgis-user mailing list
> Qgis-user@lists.osgeo.org
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Nyall Dawson]

On Mon, 17 Dec 2018 at 20:10, Bo Victor Thomsen
 wrote:
>
> Just my - very late – 2 cents.
>
>
>
> I have 2 computers that is exactly the same type: Lenovo Thinkpad P50. There 
> is no difference in cpu, memory, SSD or graphics. One is running Windows 7 
> and one is running Ubuntu 18.04. Both have QGIS 3.4.2 with the same plugins 
> installed.
>
>
>
> The Linux pc starts QGIS 3.4.2 in just 4 seconds or less. The windows machine 
> is at least 5 times slower starting up. When QGIS is running the Linux pc is 
> slightly faster.
>
>
>
> The crucial difference is the startup time of the plugins. On the Linux pc 
> you hardly notice the "Restoring loaded plugins" text. However, it takes – 
> forever - on the Windows pc.
>
>
>
> My guess is that the initialization of the python plugins runs very, very 
> slow on the Windows pc. So the culprit may be the python interpreter on 
> Windows.

This sounds like a known issue since ages (v 2.6?), where at QGIS
startup the Processing plugin parses all the hundreds of description
files for the SAGA/GRASS algorithms. You can verify by disabling the
Processing plugin and relaunching.

Nyall
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Bo Victor Thomsen]

Just my - very late – 2 cents.



I have 2 computers that is exactly the same type: Lenovo Thinkpad P50.
There is no difference in cpu, memory, SSD or graphics. One is running
Windows 7 and one is running Ubuntu 18.04. Both have QGIS 3.4.2 with the
same plugins installed.



The Linux pc starts QGIS 3.4.2 in just 4 seconds or less. The windows
machine is at least 5 times slower starting up. When QGIS is running the
Linux pc is slightly faster.



The crucial difference is the startup time of the plugins. On the Linux pc
you hardly notice the "Restoring loaded plugins" text. However, it takes –
forever - on the Windows pc.



My guess is that the initialization of the python plugins runs very, very
slow on the Windows pc. So the culprit may be the python interpreter on
Windows.


-- 
Med venlig hilsen

Bo Victor Thomsen
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[David Addy]

Hi Nyall,
Thanks for your additional comment.
When I was autoloading my most recent project at startup QGIS was taking up
to a minute to open itself and load the project. Other saves and actions
like opening a layer attribute table were also very slow.
After changing my settings to open with a new project (ie a blank project)
QGIS was opening as normal. Then loading the old project also was as fast as
it should be.
However, after editing some polygons,  when I come to save the changes I am 
still getting a long pause when nothing happens, then I get "QGIS is not 
responding",  then it comes back to life again and finishes the save. 
So changing the setting vastly improved startup, but seems to have not
solved the whole problem.



--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Nyall Dawson]

On Sun, 16 Dec 2018 at 08:51, David Addy  wrote:
>
> Thanks to everybody, but to Anita and Nyall in particular, for your comments.
> It is very helpful to know that my issues are just minor quirks rather than
> a more serious flaw.

Sorry David -- I wasn't meaning to say that the issues you're
encountering are minor quirks. Rather I was just stating my
experience, and wanted to provide an alternative view to Patrick's in
case anyone else reading this thread may wrongly conclude that the
Windows builds are not recommended.

Back to your original issue ... Am I correct in concluding that you
found it was being caused by autoloading a project at startup? Do you
still see issues with this setting disabled?

Nyall

> QGIS 3 is definitely a big leap forward and I can now plough onwards with
> these reassurances.
> I would rather avoid going to a VM at the moment as that will undoubtedly
> saddle me with a few more issues.
> My learning skills are not what they used to be, (age 70 this month) so
> "keep it simple"  is my motto.
> Thanks again!!
>
>
>
> --
> Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
> ___
> Qgis-user mailing list
> Qgis-user@lists.osgeo.org
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[David Addy]

Thanks to everybody, but to Anita and Nyall in particular, for your comments.
It is very helpful to know that my issues are just minor quirks rather than
a more serious flaw.
QGIS 3 is definitely a big leap forward and I can now plough onwards with
these reassurances.
I would rather avoid going to a VM at the moment as that will undoubtedly
saddle me with a few more issues. 
My learning skills are not what they used to be, (age 70 this month) so
"keep it simple"  is my motto.
Thanks again!!



--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Nyall Dawson]

On Fri, 14 Dec 2018 at 11:36, Patrick Dunford  wrote:
>
> As far as I know, development is done mainly in Linux
>
> The Windows and Mac editions are ports off this. I am sure the
> developers and port teams do their best but you can run into issues that
> are platform specific and are hard to resolve. Particularly as it is
> dependent on third party libraries that again may be less well supported
> on all the platforms.

I don't think this is an accurate reflection of the quality of the
Windows builds. While it's true that the majority of developers use
Linux to develop, speaking for myself at least when I'm a QGIS *user*
I'm nearly always in Windows -- and I don't have any major issues with
it.

There is occasional quirks which show up on Windows which don't on
Linux, but there's probably an equal number of quirks which show up
specifically on individual Linux distros and not on Windows!

Nyall

>
> I haven't used Qgis on Windows in a number of years, this is why I would
> suggest virtualisation of a Linux environment as an option, the virtual
> machines when I ran them could do 99% of what you could expect in a
> native installation and were pretty solid as a working environment. You
> would probably find that to be a more stable way of running Qgis 3 on
> Windows, if that were an option for you.
>
> On 14/12/18 5:40 AM, David Addy wrote:
> > Hi Patrick,
> > Thanks for your comments. It all helps to encourage me.
> > I have now discovered that the User Manual (development version) emphasises
> > the fact that there are changes to the node tool in version 3, and how the
> > dragging of the old method is now replaced by two clicks. Anita's book does
> > not mention this, so I was still trying to move a node in the "old way".
> > Like you I do not want to go back to 2.18, and I believe in learning the new
> > workflows of version 3 ASAP.
> > So now I have edited my polygons, but when I come to save the changes I am
> > still getting a long pause when nothing happens, then I get "QGIS is not
> > responding",  then it comes back to life again and finishes the save.
> > It must still be an issue with the latest QT 5.11.2, n'est ce pas?
> > I have had a look at the QT website and I note that there are a few users
> > complaining about unfixed bugs that go back for ages.
> >
> >
> >
> >
> > --
> > Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
> > ___
> > Qgis-user mailing list
> > Qgis-user@lists.osgeo.org
> > List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
> ___
> Qgis-user mailing list
> Qgis-user@lists.osgeo.org
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Anita Graser]

Hi,

I'm using QGIS on Windows a lot and cannot confirm issues with slow startup
in 3.4. The Windows version is consistently on par with the Linux version.
(Notable exceptions in version 2 were the globe extension and R support
which I couldn't get to work back then.)

The incorrect description of the current node tool behavior unfortunately
slipped by Andrew and me when updating Learning QGIS. We're very sorry
about that. We're working with the publisher to get a correction out.

Regards
Anita
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Patrick Dunford]

As far as I know, development is done mainly in Linux

The Windows and Mac editions are ports off this. I am sure the 
developers and port teams do their best but you can run into issues that 
are platform specific and are hard to resolve. Particularly as it is 
dependent on third party libraries that again may be less well supported 
on all the platforms.


I haven't used Qgis on Windows in a number of years, this is why I would 
suggest virtualisation of a Linux environment as an option, the virtual 
machines when I ran them could do 99% of what you could expect in a 
native installation and were pretty solid as a working environment. You 
would probably find that to be a more stable way of running Qgis 3 on 
Windows, if that were an option for you.


On 14/12/18 5:40 AM, David Addy wrote:

Hi Patrick,
Thanks for your comments. It all helps to encourage me.
I have now discovered that the User Manual (development version) emphasises
the fact that there are changes to the node tool in version 3, and how the
dragging of the old method is now replaced by two clicks. Anita's book does
not mention this, so I was still trying to move a node in the "old way".
Like you I do not want to go back to 2.18, and I believe in learning the new
workflows of version 3 ASAP.
So now I have edited my polygons, but when I come to save the changes I am
still getting a long pause when nothing happens, then I get "QGIS is not
responding",  then it comes back to life again and finishes the save.
It must still be an issue with the latest QT 5.11.2, n'est ce pas?
I have had a look at the QT website and I note that there are a few users
complaining about unfixed bugs that go back for ages.




--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[David Addy]

Hi Patrick,
Thanks for your comments. It all helps to encourage me.
I have now discovered that the User Manual (development version) emphasises
the fact that there are changes to the node tool in version 3, and how the
dragging of the old method is now replaced by two clicks. Anita's book does
not mention this, so I was still trying to move a node in the "old way".
Like you I do not want to go back to 2.18, and I believe in learning the new
workflows of version 3 ASAP. 
So now I have edited my polygons, but when I come to save the changes I am
still getting a long pause when nothing happens, then I get "QGIS is not
responding",  then it comes back to life again and finishes the save.
It must still be an issue with the latest QT 5.11.2, n'est ce pas?
I have had a look at the QT website and I note that there are a few users
complaining about unfixed bugs that go back for ages.




--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Patrick Dunford]

I run the latest 3.4.x on Linux which has been very stable and reliable.

It will be a while before 3.x is as mature and dependable as 2.x which 
was at 2.18 a very mature product that had been operating a good number 
of years.


Used to run the development version 2.99 but these days I view "stable" 
as effectively development simply because other users (not me) have 
lodged reports for hundreds of bugs that are progressing only slowly.


I wanted the new features in 3.x and would not go back to 2.18 now.

On 11/12/18 6:49 AM, Bernd Vogelgesang wrote:

Hi,

I can't contribute to your particular problems, but have some remarks.

You said you "reverted" to 2.18. On Windows, you can easily run 3.x 
besides 2.x. Do you have the same problems with 2.18?  If so, it's 
really your personal installation that fails, I think.


The node tool was redesigned in 3.x, and there is quite a discussion 
going on, cause not so many people like the new behaviour.


In general, I would like to say, that the promotion of the newest 
release shown in QGIS is quite a bad idea: The inexperienced will not 
hesitate to update and therefore run in every possible bug, being left 
clueless, while the more experienced are more cautious and install it 
only anlongside for testing purposes first.


The developers are in a bad situation: They need lots of testers to 
find bugs, but in my opinion they reach the wrong users with that 
advertisement. (no idea how to improve this)


Furthermore, the new version had a bad start, cause (as I understood) 
last-minute-changes in dependencies caught them unprepared.


Unfortunately, no one gives warnings about the major issues somewhere 
prominently e.g. on the QGIS.org website, so you have to read the 
mailing-list(s) and search in the issue queue yourself. In my opinion, 
in QGIS3 the developers were a little too ambitious, but it seems they 
also had kind of bad luck as well. Lots of features introduced are 
somewhat bleeding-edge and need time to ripe.


As I rarely use Windows, I'm not of big help. I just want to recommend 
you to use the network installer (advanced install!) and install the 
2.x LTR and the 3.x LTR in parallel and update them through the 
installer once in a while, and keep 2.18 for productive work as long 
as you do not trust the 3.x version.


The open source mantra is "release early, release often", but that 
doesn't mean that everyone has to update early and often as well!



___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Patrick Dunford]

If you have enough resources running it in a VM is an option and can be 
a linux version, very stable and very resource efficient


I used to have a bg stack of VMs running all sorts of versions when I 
used to run the development version more than I do now, it was almost 
seamless


On 11/12/18 5:59 AM, David Addy wrote:

I have always loved QGIS and I have extolled its virtues to local historians
and archaeologists in my area.
I have had to keep quiet about it lately.
I was somewhat relieved to hear a comment on a recent video of the Oceania
meeting that QGIS 3.4.1 was a dud release as it caused Windows to crash at
random moments. I had thought that all the long pauses, messages of QGIS not
responding, and crashing when being closed down was some awful feature
unique to my computer. I had even tried reverting to 2.18 for a while.
However, after I upgraded to 3.4.2 I still had problems.
Opening QGIS on Windows 10 can take over half a minute before anything
happens apart from the outline box of the QGIS window full of blank space.
Once a project finally loads up any random operation can still get the "QGIS
not responding" message.
This can result in further issues because I try pressing buttons to get a
response.
I even bought Anita's 4th edition of "Learn QGIS" to see if I was doing
something radically wrong. It did highlight a few wrinkles until I got to
page 55 when I discovered instructions on the edit node tool. This tool I
had used OK in the past, but it now baffles me totally. I cannot get it to
do what the book says.
I am now losing my confidence in 3.4.2 and can only hope that the next point
release will produce improvements.
It would really help if known major issues like this could be posted here,
perhaps as a sticky which stays at the top, to let users know that they are
not alone with an issue.
Sorry to complain.



--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Patrick Dunford]

Plenty of bugs are being discovered and reported but not many bug 
reports are being opened, as it appears to me.


I'm not aware of issues with the new node editor. It works very well 
most of the time.


Qgis 3.4.anything is very stable on Debian, I stopped running it on 
Windows a long time ago.


On 11/12/18 6:49 AM, Bernd Vogelgesang wrote:

Hi,

I can't contribute to your particular problems, but have some remarks.

You said you "reverted" to 2.18. On Windows, you can easily run 3.x 
besides 2.x. Do you have the same problems with 2.18?  If so, it's 
really your personal installation that fails, I think.


The node tool was redesigned in 3.x, and there is quite a discussion 
going on, cause not so many people like the new behaviour.


In general, I would like to say, that the promotion of the newest 
release shown in QGIS is quite a bad idea: The inexperienced will not 
hesitate to update and therefore run in every possible bug, being left 
clueless, while the more experienced are more cautious and install it 
only anlongside for testing purposes first.


The developers are in a bad situation: They need lots of testers to 
find bugs, but in my opinion they reach the wrong users with that 
advertisement. (no idea how to improve this)


Furthermore, the new version had a bad start, cause (as I understood) 
last-minute-changes in dependencies caught them unprepared.


Unfortunately, no one gives warnings about the major issues somewhere 
prominently e.g. on the QGIS.org website, so you have to read the 
mailing-list(s) and search in the issue queue yourself. In my opinion, 
in QGIS3 the developers were a little too ambitious, but it seems they 
also had kind of bad luck as well. Lots of features introduced are 
somewhat bleeding-edge and need time to ripe.


As I rarely use Windows, I'm not of big help. I just want to recommend 
you to use the network installer (advanced install!) and install the 
2.x LTR and the 3.x LTR in parallel and update them through the 
installer once in a while, and keep 2.18 for productive work as long 
as you do not trust the 3.x version.


The open source mantra is "release early, release often", but that 
doesn't mean that everyone has to update early and often as well!


Cheers,

Bernd



___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[David Addy]

Thanks for all your comments. I think things were pretty good using 3.2, and
I had got used to using Version 3 with geopackage etc, and going back to
2.18 was rather painful for me. These hanging up problems for me really
started in 3.4. 
I have just found that changing my Settings to make QGIS start up with a new
project, (rather than loading the last project worked on) seems to be a lot
faster to open this way.



--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Mike Flannigan]

I can report that I do not see the same thing reported
below.  I use QGIS 3.2.0 on Win7 with a pretty good graphics
card and I don't have any problems at all.  I'm thinking it
has perhaps froze up a couple times since the install in late
June 2018, but I don't recall the last time that happened.

I just opened a project with 266,000 point features, and it
takes 18 seconds to open the attribute table, but works
flawlessly.

I am using Intel Zeon CPU E5-1630 v4 @ 3.70GHz - 16GB Ram.

I have never tried QGIS 3.4.1.



Mike


On 12/10/2018 2:00 PM, qgis-user-requ...@lists.osgeo.org wrote:

Hi David,
everything you've detected happens since the transition to 3.x also on
windows 7 on PC with i7 4780 and 32GB ram.
any operation, even the simplest one, such as saving or calculating polygon
areas in a shapefile causes freeze of QGIS. Not to mention the node tool
where a forum should be open apart there is a ticket here but only 8 people
have participated ...
(http://osgeo-org.1560.x6.nabble.com/Node-Tool-proposing-improvements-td5387054.html).
Moreover:
- there is  no currently a native topological verification tool of
shapefiles (the new validator incorporated in editing is almost impossible
to run on shapefile with hundreds of polygon in ...)
It will take years, in my opinion, before the 3.x version can be used
productively.
Unfortunately, the gap between users and developers remains gigantic,
although most of the developers say that users have been asked .. sin until
2.18.x everything was wonderful.

I Hope that somebody from QGIS.org can see our perplexities and adjust as
far as possible this version which for now remains only a container of good
ideas

Regards


-
Antonio


___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Antongeo76]

Saber Razmjooei wrote
> On Tue, 11 Dec 2018 at 10:39, Antongeo76 

> antoviscomi@

>  wrote:
> 
>> berndv. wrote
>> > Hi,
>> >
>> > I can't contribute to your particular problems, but have some remarks.
>> >
>> > You said you "reverted" to 2.18. On Windows, you can easily run 3.x
>> > besides 2.x. Do you have the same problems with 2.18?  If so, it's
>> > really your personal installation that fails, I think.
>> >
>> > The node tool was redesigned in 3.x, and there is quite a discussion
>> > going on, cause not so many people like the new behaviour.
>> >
>> > In general, I would like to say, that the promotion of the newest
>> > release shown in QGIS is quite a bad idea: The inexperienced will not
>> > hesitate to update and therefore run in every possible bug, being left
>> > clueless, while the more experienced are more cautious and install it
>> > only anlongside for testing purposes first.
>> >
>> > The developers are in a bad situation: They need lots of testers to
>> find
>> > bugs, but in my opinion they reach the wrong users with that
>> > advertisement. (no idea how to improve this)
>> >
>> > Furthermore, the new version had a bad start, cause (as I understood)
>> > last-minute-changes in dependencies caught them unprepared.
>> >
>> > Unfortunately, no one gives warnings about the major issues somewhere
>> > prominently e.g. on the QGIS.org website, so you have to read the
>> > mailing-list(s) and search in the issue queue yourself. In my opinion,
>> > in QGIS3 the developers were a little too ambitious, but it seems they
>> > also had kind of bad luck as well. Lots of features introduced are
>> > somewhat bleeding-edge and need time to ripe.
>> >
>> > As I rarely use Windows, I'm not of big help. I just want to recommend
>> > you to use the network installer (advanced install!) and install the
>> 2.x
>> > LTR and the 3.x LTR in parallel and update them through the installer
>> > once in a while, and keep 2.18 for productive work as long as you do
>> not
>> > trust the 3.x version.
>> >
>> > The open source mantra is "release early, release often", but that
>> > doesn't mean that everyone has to update early and often as well!
>> >
>> > Cheers,
>> >
>> > Bernd
>> >
>> > Am 10.12.18 um 17:59 schrieb David Addy:
>> >> I have always loved QGIS and I have extolled its virtues to local
>> >> historians
>> >> and archaeologists in my area.
>> >> I have had to keep quiet about it lately.
>> >> I was somewhat relieved to hear a comment on a recent video of the
>> >> Oceania
>> >> meeting that QGIS 3.4.1 was a dud release as it caused Windows to
>> crash
>> >> at
>> >> random moments. I had thought that all the long pauses, messages of
>> QGIS
>> >> not
>> >> responding, and crashing when being closed down was some awful feature
>> >> unique to my computer. I had even tried reverting to 2.18 for a while.
>> >> However, after I upgraded to 3.4.2 I still had problems.
>> >> Opening QGIS on Windows 10 can take over half a minute before anything
>> >> happens apart from the outline box of the QGIS window full of blank
>> >> space.
>> >> Once a project finally loads up any random operation can still get the
>> >> "QGIS
>> >> not responding" message.
>> >> This can result in further issues because I try pressing buttons to
>> get
>> a
>> >> response.
>> >> I even bought Anita's 4th edition of "Learn QGIS" to see if I was
>> doing
>> >> something radically wrong. It did highlight a few wrinkles until I got
>> to
>> >> page 55 when I discovered instructions on the edit node tool. This
>> tool
>> I
>> >> had used OK in the past, but it now baffles me totally. I cannot get
>> it
>> >> to
>> >> do what the book says.
>> >> I am now losing my confidence in 3.4.2 and can only hope that the next
>> >> point
>> >> release will produce improvements.
>> >> It would really help if known major issues like this could be posted
>> >> here,
>> >> perhaps as a sticky which stays at the top, to let users know that
>> they
>> >> are
>> >> not alone with an issue.
>> >> Sorry to complain.
>> >>
>> >>
>> >>
>> >> --
>> >> Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
>> >> ___
>> >> Qgis-user mailing list
>> >>
>>
>> > Qgis-user@.osgeo
>>
>> >> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> >> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> > ___
>> > Qgis-user mailing list
>>
>> > Qgis-user@.osgeo
>>
>> > List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>>
>> HI Bernd,
>> I fully agree with you,
>> even on the fact of the versions installed and to be used for
>> productivity
>> (I've installed all concomitant versions available for a single machine,
>> and
>> I use them almost every day)...
>> the fact is that every time I write about freezes  or general
>> malfunctions
>> (here or in QGIS issues)
>> I 

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Antongeo76]

-
Antonio
--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Saber Razmjooei]

On Tue, 11 Dec 2018 at 10:39, Antongeo76  wrote:

> berndv. wrote
> > Hi,
> >
> > I can't contribute to your particular problems, but have some remarks.
> >
> > You said you "reverted" to 2.18. On Windows, you can easily run 3.x
> > besides 2.x. Do you have the same problems with 2.18?  If so, it's
> > really your personal installation that fails, I think.
> >
> > The node tool was redesigned in 3.x, and there is quite a discussion
> > going on, cause not so many people like the new behaviour.
> >
> > In general, I would like to say, that the promotion of the newest
> > release shown in QGIS is quite a bad idea: The inexperienced will not
> > hesitate to update and therefore run in every possible bug, being left
> > clueless, while the more experienced are more cautious and install it
> > only anlongside for testing purposes first.
> >
> > The developers are in a bad situation: They need lots of testers to find
> > bugs, but in my opinion they reach the wrong users with that
> > advertisement. (no idea how to improve this)
> >
> > Furthermore, the new version had a bad start, cause (as I understood)
> > last-minute-changes in dependencies caught them unprepared.
> >
> > Unfortunately, no one gives warnings about the major issues somewhere
> > prominently e.g. on the QGIS.org website, so you have to read the
> > mailing-list(s) and search in the issue queue yourself. In my opinion,
> > in QGIS3 the developers were a little too ambitious, but it seems they
> > also had kind of bad luck as well. Lots of features introduced are
> > somewhat bleeding-edge and need time to ripe.
> >
> > As I rarely use Windows, I'm not of big help. I just want to recommend
> > you to use the network installer (advanced install!) and install the 2.x
> > LTR and the 3.x LTR in parallel and update them through the installer
> > once in a while, and keep 2.18 for productive work as long as you do not
> > trust the 3.x version.
> >
> > The open source mantra is "release early, release often", but that
> > doesn't mean that everyone has to update early and often as well!
> >
> > Cheers,
> >
> > Bernd
> >
> > Am 10.12.18 um 17:59 schrieb David Addy:
> >> I have always loved QGIS and I have extolled its virtues to local
> >> historians
> >> and archaeologists in my area.
> >> I have had to keep quiet about it lately.
> >> I was somewhat relieved to hear a comment on a recent video of the
> >> Oceania
> >> meeting that QGIS 3.4.1 was a dud release as it caused Windows to crash
> >> at
> >> random moments. I had thought that all the long pauses, messages of QGIS
> >> not
> >> responding, and crashing when being closed down was some awful feature
> >> unique to my computer. I had even tried reverting to 2.18 for a while.
> >> However, after I upgraded to 3.4.2 I still had problems.
> >> Opening QGIS on Windows 10 can take over half a minute before anything
> >> happens apart from the outline box of the QGIS window full of blank
> >> space.
> >> Once a project finally loads up any random operation can still get the
> >> "QGIS
> >> not responding" message.
> >> This can result in further issues because I try pressing buttons to get
> a
> >> response.
> >> I even bought Anita's 4th edition of "Learn QGIS" to see if I was doing
> >> something radically wrong. It did highlight a few wrinkles until I got
> to
> >> page 55 when I discovered instructions on the edit node tool. This tool
> I
> >> had used OK in the past, but it now baffles me totally. I cannot get it
> >> to
> >> do what the book says.
> >> I am now losing my confidence in 3.4.2 and can only hope that the next
> >> point
> >> release will produce improvements.
> >> It would really help if known major issues like this could be posted
> >> here,
> >> perhaps as a sticky which stays at the top, to let users know that they
> >> are
> >> not alone with an issue.
> >> Sorry to complain.
> >>
> >>
> >>
> >> --
> >> Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
> >> ___
> >> Qgis-user mailing list
> >>
>
> > Qgis-user@.osgeo
>
> >> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> >> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
> > ___
> > Qgis-user mailing list
>
> > Qgis-user@.osgeo
>
> > List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>
> HI Bernd,
> I fully agree with you,
> even on the fact of the versions installed and to be used for productivity
> (I've installed all concomitant versions available for a single machine,
> and
> I use them almost every day)...
> the fact is that every time I write about freezes  or general malfunctions
> (here or in QGIS issues)
> I seem to be the only one to be affected (or we are in very few users to
> observe and report these anomalies) ...
>
> Greetings
> unfortunately it is often difficult to communicate remotely ...
> we 

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Antongeo76]

berndv. wrote
> Hi,
> 
> I can't contribute to your particular problems, but have some remarks.
> 
> You said you "reverted" to 2.18. On Windows, you can easily run 3.x 
> besides 2.x. Do you have the same problems with 2.18?  If so, it's 
> really your personal installation that fails, I think.
> 
> The node tool was redesigned in 3.x, and there is quite a discussion 
> going on, cause not so many people like the new behaviour.
> 
> In general, I would like to say, that the promotion of the newest 
> release shown in QGIS is quite a bad idea: The inexperienced will not 
> hesitate to update and therefore run in every possible bug, being left 
> clueless, while the more experienced are more cautious and install it 
> only anlongside for testing purposes first.
> 
> The developers are in a bad situation: They need lots of testers to find 
> bugs, but in my opinion they reach the wrong users with that 
> advertisement. (no idea how to improve this)
> 
> Furthermore, the new version had a bad start, cause (as I understood) 
> last-minute-changes in dependencies caught them unprepared.
> 
> Unfortunately, no one gives warnings about the major issues somewhere 
> prominently e.g. on the QGIS.org website, so you have to read the 
> mailing-list(s) and search in the issue queue yourself. In my opinion, 
> in QGIS3 the developers were a little too ambitious, but it seems they 
> also had kind of bad luck as well. Lots of features introduced are 
> somewhat bleeding-edge and need time to ripe.
> 
> As I rarely use Windows, I'm not of big help. I just want to recommend 
> you to use the network installer (advanced install!) and install the 2.x 
> LTR and the 3.x LTR in parallel and update them through the installer 
> once in a while, and keep 2.18 for productive work as long as you do not 
> trust the 3.x version.
> 
> The open source mantra is "release early, release often", but that 
> doesn't mean that everyone has to update early and often as well!
> 
> Cheers,
> 
> Bernd
> 
> Am 10.12.18 um 17:59 schrieb David Addy:
>> I have always loved QGIS and I have extolled its virtues to local
>> historians
>> and archaeologists in my area.
>> I have had to keep quiet about it lately.
>> I was somewhat relieved to hear a comment on a recent video of the
>> Oceania
>> meeting that QGIS 3.4.1 was a dud release as it caused Windows to crash
>> at
>> random moments. I had thought that all the long pauses, messages of QGIS
>> not
>> responding, and crashing when being closed down was some awful feature
>> unique to my computer. I had even tried reverting to 2.18 for a while.
>> However, after I upgraded to 3.4.2 I still had problems.
>> Opening QGIS on Windows 10 can take over half a minute before anything
>> happens apart from the outline box of the QGIS window full of blank
>> space.
>> Once a project finally loads up any random operation can still get the
>> "QGIS
>> not responding" message.
>> This can result in further issues because I try pressing buttons to get a
>> response.
>> I even bought Anita's 4th edition of "Learn QGIS" to see if I was doing
>> something radically wrong. It did highlight a few wrinkles until I got to
>> page 55 when I discovered instructions on the edit node tool. This tool I
>> had used OK in the past, but it now baffles me totally. I cannot get it
>> to
>> do what the book says.
>> I am now losing my confidence in 3.4.2 and can only hope that the next
>> point
>> release will produce improvements.
>> It would really help if known major issues like this could be posted
>> here,
>> perhaps as a sticky which stays at the top, to let users know that they
>> are
>> not alone with an issue.
>> Sorry to complain.
>>
>>
>>
>> --
>> Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
>> ___
>> Qgis-user mailing list
>> 

> Qgis-user@.osgeo

>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
> ___
> Qgis-user mailing list

> Qgis-user@.osgeo

> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

HI Bernd,
I fully agree with you, 
even on the fact of the versions installed and to be used for productivity 
(I've installed all concomitant versions available for a single machine, and
I use them almost every day)... 
the fact is that every time I write about freezes  or general malfunctions
(here or in QGIS issues) 
I seem to be the only one to be affected (or we are in very few users to
observe and report these anomalies) ...

Greetings
unfortunately it is often difficult to communicate remotely ... 
we should have a developer sitting next to prove that the software often 
does not run as it should and you would expect



-
Antonio
--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Antongeo76]

Saber Razmjooei wrote
> It's been said a lot and often gets dismissed by users (similar to have
> you
> turn off and on your computer), but most of the issues can be resolved by:
> 1- Start your QGIS 3, with a clean profile.
> 2- Remove/disable all 3rd party plugins
> 
> If all fails, report the bug with a sample data and step-by-step to
> reproduce the issue.
> 
> Regards
> Saber
> 
> 
> On Mon, 10 Dec 2018 at 17:49, Bernd Vogelgesang 

> bernd.vogelgesang@

> 
> wrote:
> 
>> Hi,
>>
>> I can't contribute to your particular problems, but have some remarks.
>>
>> You said you "reverted" to 2.18. On Windows, you can easily run 3.x
>> besides 2.x. Do you have the same problems with 2.18?  If so, it's
>> really your personal installation that fails, I think.
>>
>> The node tool was redesigned in 3.x, and there is quite a discussion
>> going on, cause not so many people like the new behaviour.
>>
>> In general, I would like to say, that the promotion of the newest
>> release shown in QGIS is quite a bad idea: The inexperienced will not
>> hesitate to update and therefore run in every possible bug, being left
>> clueless, while the more experienced are more cautious and install it
>> only anlongside for testing purposes first.
>>
>> The developers are in a bad situation: They need lots of testers to find
>> bugs, but in my opinion they reach the wrong users with that
>> advertisement. (no idea how to improve this)
>>
>> Furthermore, the new version had a bad start, cause (as I understood)
>> last-minute-changes in dependencies caught them unprepared.
>>
>> Unfortunately, no one gives warnings about the major issues somewhere
>> prominently e.g. on the QGIS.org website, so you have to read the
>> mailing-list(s) and search in the issue queue yourself. In my opinion,
>> in QGIS3 the developers were a little too ambitious, but it seems they
>> also had kind of bad luck as well. Lots of features introduced are
>> somewhat bleeding-edge and need time to ripe.
>>
>> As I rarely use Windows, I'm not of big help. I just want to recommend
>> you to use the network installer (advanced install!) and install the 2.x
>> LTR and the 3.x LTR in parallel and update them through the installer
>> once in a while, and keep 2.18 for productive work as long as you do not
>> trust the 3.x version.
>>
>> The open source mantra is "release early, release often", but that
>> doesn't mean that everyone has to update early and often as well!
>>
>> Cheers,
>>
>> Bernd
>>
>> Am 10.12.18 um 17:59 schrieb David Addy:
>> > I have always loved QGIS and I have extolled its virtues to local
>> historians
>> > and archaeologists in my area.
>> > I have had to keep quiet about it lately.
>> > I was somewhat relieved to hear a comment on a recent video of the
>> Oceania
>> > meeting that QGIS 3.4.1 was a dud release as it caused Windows to crash
>> at
>> > random moments. I had thought that all the long pauses, messages of
>> QGIS
>> not
>> > responding, and crashing when being closed down was some awful feature
>> > unique to my computer. I had even tried reverting to 2.18 for a while.
>> > However, after I upgraded to 3.4.2 I still had problems.
>> > Opening QGIS on Windows 10 can take over half a minute before anything
>> > happens apart from the outline box of the QGIS window full of blank
>> space.
>> > Once a project finally loads up any random operation can still get the
>> "QGIS
>> > not responding" message.
>> > This can result in further issues because I try pressing buttons to get
>> a
>> > response.
>> > I even bought Anita's 4th edition of "Learn QGIS" to see if I was doing
>> > something radically wrong. It did highlight a few wrinkles until I got
>> to
>> > page 55 when I discovered instructions on the edit node tool. This tool
>> I
>> > had used OK in the past, but it now baffles me totally. I cannot get it
>> to
>> > do what the book says.
>> > I am now losing my confidence in 3.4.2 and can only hope that the next
>> point
>> > release will produce improvements.
>> > It would really help if known major issues like this could be posted
>> here,
>> > perhaps as a sticky which stays at the top, to let users know that they
>> are
>> > not alone with an issue.
>> > Sorry to complain.
>> >
>> >
>> >
>> > --
>> > Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
>> > ___
>> > Qgis-user mailing list
>> > 

> Qgis-user@.osgeo

>> > List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> ___
>> Qgis-user mailing list
>> 

> Qgis-user@.osgeo

>> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
>> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
> 
> 
> 
> -- 
> Saber Razmjooei
> www.lutraconsulting.co.uk
> 
> ___
> Qgis-user mailing list

> Qgis-user@.osgeo

> List info: 

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Saber Razmjooei]

It's been said a lot and often gets dismissed by users (similar to have you
turn off and on your computer), but most of the issues can be resolved by:
1- Start your QGIS 3, with a clean profile.
2- Remove/disable all 3rd party plugins

If all fails, report the bug with a sample data and step-by-step to
reproduce the issue.

Regards
Saber


On Mon, 10 Dec 2018 at 17:49, Bernd Vogelgesang 
wrote:

> Hi,
>
> I can't contribute to your particular problems, but have some remarks.
>
> You said you "reverted" to 2.18. On Windows, you can easily run 3.x
> besides 2.x. Do you have the same problems with 2.18?  If so, it's
> really your personal installation that fails, I think.
>
> The node tool was redesigned in 3.x, and there is quite a discussion
> going on, cause not so many people like the new behaviour.
>
> In general, I would like to say, that the promotion of the newest
> release shown in QGIS is quite a bad idea: The inexperienced will not
> hesitate to update and therefore run in every possible bug, being left
> clueless, while the more experienced are more cautious and install it
> only anlongside for testing purposes first.
>
> The developers are in a bad situation: They need lots of testers to find
> bugs, but in my opinion they reach the wrong users with that
> advertisement. (no idea how to improve this)
>
> Furthermore, the new version had a bad start, cause (as I understood)
> last-minute-changes in dependencies caught them unprepared.
>
> Unfortunately, no one gives warnings about the major issues somewhere
> prominently e.g. on the QGIS.org website, so you have to read the
> mailing-list(s) and search in the issue queue yourself. In my opinion,
> in QGIS3 the developers were a little too ambitious, but it seems they
> also had kind of bad luck as well. Lots of features introduced are
> somewhat bleeding-edge and need time to ripe.
>
> As I rarely use Windows, I'm not of big help. I just want to recommend
> you to use the network installer (advanced install!) and install the 2.x
> LTR and the 3.x LTR in parallel and update them through the installer
> once in a while, and keep 2.18 for productive work as long as you do not
> trust the 3.x version.
>
> The open source mantra is "release early, release often", but that
> doesn't mean that everyone has to update early and often as well!
>
> Cheers,
>
> Bernd
>
> Am 10.12.18 um 17:59 schrieb David Addy:
> > I have always loved QGIS and I have extolled its virtues to local
> historians
> > and archaeologists in my area.
> > I have had to keep quiet about it lately.
> > I was somewhat relieved to hear a comment on a recent video of the
> Oceania
> > meeting that QGIS 3.4.1 was a dud release as it caused Windows to crash
> at
> > random moments. I had thought that all the long pauses, messages of QGIS
> not
> > responding, and crashing when being closed down was some awful feature
> > unique to my computer. I had even tried reverting to 2.18 for a while.
> > However, after I upgraded to 3.4.2 I still had problems.
> > Opening QGIS on Windows 10 can take over half a minute before anything
> > happens apart from the outline box of the QGIS window full of blank
> space.
> > Once a project finally loads up any random operation can still get the
> "QGIS
> > not responding" message.
> > This can result in further issues because I try pressing buttons to get a
> > response.
> > I even bought Anita's 4th edition of "Learn QGIS" to see if I was doing
> > something radically wrong. It did highlight a few wrinkles until I got to
> > page 55 when I discovered instructions on the edit node tool. This tool I
> > had used OK in the past, but it now baffles me totally. I cannot get it
> to
> > do what the book says.
> > I am now losing my confidence in 3.4.2 and can only hope that the next
> point
> > release will produce improvements.
> > It would really help if known major issues like this could be posted
> here,
> > perhaps as a sticky which stays at the top, to let users know that they
> are
> > not alone with an issue.
> > Sorry to complain.
> >
> >
> >
> > --
> > Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
> > ___
> > Qgis-user mailing list
> > Qgis-user@lists.osgeo.org
> > List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> > Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user
> ___
> Qgis-user mailing list
> Qgis-user@lists.osgeo.org
> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user



-- 
Saber Razmjooei
www.lutraconsulting.co.uk
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Bernd Vogelgesang]

Hi,

I can't contribute to your particular problems, but have some remarks.

You said you "reverted" to 2.18. On Windows, you can easily run 3.x 
besides 2.x. Do you have the same problems with 2.18?  If so, it's 
really your personal installation that fails, I think.


The node tool was redesigned in 3.x, and there is quite a discussion 
going on, cause not so many people like the new behaviour.


In general, I would like to say, that the promotion of the newest 
release shown in QGIS is quite a bad idea: The inexperienced will not 
hesitate to update and therefore run in every possible bug, being left 
clueless, while the more experienced are more cautious and install it 
only anlongside for testing purposes first.


The developers are in a bad situation: They need lots of testers to find 
bugs, but in my opinion they reach the wrong users with that 
advertisement. (no idea how to improve this)


Furthermore, the new version had a bad start, cause (as I understood) 
last-minute-changes in dependencies caught them unprepared.


Unfortunately, no one gives warnings about the major issues somewhere 
prominently e.g. on the QGIS.org website, so you have to read the 
mailing-list(s) and search in the issue queue yourself. In my opinion, 
in QGIS3 the developers were a little too ambitious, but it seems they 
also had kind of bad luck as well. Lots of features introduced are 
somewhat bleeding-edge and need time to ripe.


As I rarely use Windows, I'm not of big help. I just want to recommend 
you to use the network installer (advanced install!) and install the 2.x 
LTR and the 3.x LTR in parallel and update them through the installer 
once in a while, and keep 2.18 for productive work as long as you do not 
trust the 3.x version.


The open source mantra is "release early, release often", but that 
doesn't mean that everyone has to update early and often as well!


Cheers,

Bernd

Am 10.12.18 um 17:59 schrieb David Addy:

I have always loved QGIS and I have extolled its virtues to local historians
and archaeologists in my area.
I have had to keep quiet about it lately.
I was somewhat relieved to hear a comment on a recent video of the Oceania
meeting that QGIS 3.4.1 was a dud release as it caused Windows to crash at
random moments. I had thought that all the long pauses, messages of QGIS not
responding, and crashing when being closed down was some awful feature
unique to my computer. I had even tried reverting to 2.18 for a while.
However, after I upgraded to 3.4.2 I still had problems.
Opening QGIS on Windows 10 can take over half a minute before anything
happens apart from the outline box of the QGIS window full of blank space.
Once a project finally loads up any random operation can still get the "QGIS
not responding" message.
This can result in further issues because I try pressing buttons to get a
response.
I even bought Anita's 4th edition of "Learn QGIS" to see if I was doing
something radically wrong. It did highlight a few wrinkles until I got to
page 55 when I discovered instructions on the edit node tool. This tool I
had used OK in the past, but it now baffles me totally. I cannot get it to
do what the book says.
I am now losing my confidence in 3.4.2 and can only hope that the next point
release will produce improvements.
It would really help if known major issues like this could be posted here,
perhaps as a sticky which stays at the top, to let users know that they are
not alone with an issue.
Sorry to complain.



--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Re: [Qgis-user] QGIS 3.4.2 still very slow to work in Windows 10

[Antongeo76]

David Addy wrote
> I have always loved QGIS and I have extolled its virtues to local
> historians
> and archaeologists in my area.
> I have had to keep quiet about it lately.
> I was somewhat relieved to hear a comment on a recent video of the Oceania
> meeting that QGIS 3.4.1 was a dud release as it caused Windows to crash at
> random moments. I had thought that all the long pauses, messages of QGIS
> not
> responding, and crashing when being closed down was some awful feature
> unique to my computer. I had even tried reverting to 2.18 for a while.
> However, after I upgraded to 3.4.2 I still had problems.
> Opening QGIS on Windows 10 can take over half a minute before anything
> happens apart from the outline box of the QGIS window full of blank space.
> Once a project finally loads up any random operation can still get the
> "QGIS
> not responding" message.
> This can result in further issues because I try pressing buttons to get a
> response.
> I even bought Anita's 4th edition of "Learn QGIS" to see if I was doing
> something radically wrong. It did highlight a few wrinkles until I got to
> page 55 when I discovered instructions on the edit node tool. This tool I
> had used OK in the past, but it now baffles me totally. I cannot get it to
> do what the book says.
> I am now losing my confidence in 3.4.2 and can only hope that the next
> point
> release will produce improvements.
> It would really help if known major issues like this could be posted here,
> perhaps as a sticky which stays at the top, to let users know that they
> are
> not alone with an issue.
> Sorry to complain.
> 
> 
> 
> --
> Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
> ___
> Qgis-user mailing list

> Qgis-user@.osgeo

> List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
> Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user

Hi David,
everything you've detected happens since the transition to 3.x also on
windows 7 on PC with i7 4780 and 32GB ram.
any operation, even the simplest one, such as saving or calculating polygon
areas in a shapefile causes freeze of QGIS. Not to mention the node tool
where a forum should be open apart there is a ticket here but only 8 people
have participated ...
(http://osgeo-org.1560.x6.nabble.com/Node-Tool-proposing-improvements-td5387054.html).
Moreover:
- there is  no currently a native topological verification tool of
shapefiles (the new validator incorporated in editing is almost impossible
to run on shapefile with hundreds of polygon in ...)
It will take years, in my opinion, before the 3.x version can be used
productively.
Unfortunately, the gap between users and developers remains gigantic,
although most of the developers say that users have been asked .. sin until
2.18.x everything was wonderful.

I Hope that somebody from QGIS.org can see our perplexities and adjust as
far as possible this version which for now remains only a container of good
ideas

Regards



-
Antonio
--
Sent from: http://osgeo-org.1560.x6.nabble.com/QGIS-User-f4125267.html
___
Qgis-user mailing list
Qgis-user@lists.osgeo.org
List info: https://lists.osgeo.org/mailman/listinfo/qgis-user
Unsubscribe: https://lists.osgeo.org/mailman/listinfo/qgis-user