krpano 1.16 - Page 10 - News - krpano.com Forum

Administrator

Wednesday, July 17th 2013, 3:46pm

Quoted from "nidrig"

The system I'm using is Mac OSX 10.8.4 Intel i7 with Intel HD graphics 4000 512MB. The screen resolution is 2560x1440.

Thanks for the report!

This is a Mac retina screen or?
I must admit that I don't have a Mac system with a retina screen for testing.

The problem might be related to the system retina scaling and Firefox - it can be the case that Firefox will internally try to render the WebGL graphic in the twice screen resolution, which would be 2 x 2560 = 5120 - which is larger than the 'Max Render Buffer Size' of 4096 - and that probably results in wrong rendering and/or wrongly returned WebGL 'drawing buffer informations' (which would be a bug in Firefox).

Can I send you a special build with some debug/trace informations for testing this special case?
If the 'drawing buffer informations' on Mac Retina Firefox were known to be buggy, it should be possible to ignore them and use the WebGL canvas size instead (which would be technically not fully correct, but should fix that problem in this case).

Additionally - can please run also this WebGL conformance test here - it can also show bugs related to WebGL canvas scaling, but due the small test sizes that were used there, it will probably pass all test successfully:
https://www.khronos.org/registry/webgl/s…d-dpi-test.html

Quoted from "HansNyb"

And this is the 16.4 version it was updated from http://virtualdenmark.dk/dnm/kina/

I have all the licenses in both folders and the update tool does not give me any error.

The tour.js in that example consist of a 1.16.4 embedding script, a inline license, and a 1.17 viewer script.
This combination confuses the krpano Update Tool

.

The krpano Update Tool thinks the 1.17 viewer script part has the license directly embedded, which leads to the removal of the 'inline-license' in the tour.js. I will try to enhance the update tool in the next versions in the let the tool also handle such mixed files correctly.

Try getting a completely fresh tour.js to fix that problem:

drop one image (anyone) once on the MAKE VTOUR (NORMAL) droplet
then copy and take the tour.js from that build tour

Best regards,
Klaus

Go to the top of the page

nidrig

Beginner

Posts: 26

Location: Switzerland

182

Wednesday, July 17th 2013, 7:58pm

Quoted from "klaus.krpano"

Quoted from "nidrig"

The system I'm using is Mac OSX 10.8.4 Intel i7 with Intel HD graphics 4000 512MB. The screen resolution is 2560x1440.

Thanks for the report!

This is a Mac retina screen or?
I must admit that I don't have a Mac system with a retina screen for testing.

The problem might be related to the system retina scaling and Firefox - it can be the case that Firefox will internally try to render the WebGL graphic in the twice screen resolution, which would be 2 x 2560 = 5120 - which is larger than the 'Max Render Buffer Size' of 4096 - and that probably results in wrong rendering and/or wrongly returned WebGL 'drawing buffer informations' (which would be a bug in Firefox).

Can I send you a special build with some debug/trace informations for testing this special case?
If the 'drawing buffer informations' on Mac Retina Firefox were known to be buggy, it should be possible to ignore them and use the WebGL canvas size instead (which would be technically not fully correct, but should fix that problem in this case).

Additionally - can please run also this WebGL conformance test here - it can also show bugs related to WebGL canvas scaling, but due the small test sizes that were used there, it will probably pass all test successfully:
https://www.khronos.org/registry/webgl/s…d-dpi-test.html

hi,

It's not a retina display. It's a Macbook Pro 13', with 1280 x 800 native resolution. The resize issue doesn't happen with smaller resolution (1280x800 in that case). I'm using an external screen as the graphic card supports up to 2560 x 1600 for external display.

Your assumption might well be correct with the scaling and Firefox.

I ran the WebGL conformance test and all tests passed successfully.

Yes, you can send me a special build with trace/debug information.

Thanks!

www.36one.net

Go to the top of the page

nidrig

Beginner

Posts: 26

Location: Switzerland

183

Tuesday, July 23rd 2013, 9:17pm

Hi Klaus,

You wrote this:

"but I'm already working on a replacement maps plugin that works completely without Bing Maps API."

What does it mean? that Bing maps won't be used anymore? or that we will have the choice of maps? I have several clients who have trouble with Bing, because they don't work in their area, but have no problem with Google maps.

Thanks.

www.36one.net

Go to the top of the page

martyhr

Intermediate

Posts: 387

Location: CZ

184

Tuesday, July 23rd 2013, 9:51pm

Yes please :)
when Goole maps for HTML5 ?
can't use BING for this project - please check "wrong" map

Go to the top of the page

klaus.krpano

Administrator

Posts: 13,499

185

Wednesday, July 24th 2013, 7:36pm

Hi,

Quoted from "nidrig"

What does it mean? that Bing maps won't be used anymore?

No, that doesn't mean that - it means the Bing Maps JS-API will not be used anymore when this plugin is ready. Bing Maps provides also a Tile-API - with that API it's possible to load and display the tiles with own code. This way a lot of bugs and issues can be avoided.

Quoted from "martyhr"

when Goole maps for HTML5 ?
can't use BING for this project - please check "wrong" map

There will be a Google Maps HTML5 plugin together with the release of version 1.17.

Best regards,
Klaus

Go to the top of the page

klaus.krpano

Administrator

Posts: 13,499

186

Friday, August 9th 2013, 9:58pm

Hi,

there is another small update for version 1.16:
krpano 1.16.6

Please see here for a list about all changes and fixes:
Release Notes 1.16.6

Best regards,
Klaus

Go to the top of the page

HansNyb

Professional

Posts: 1,063

Location: Denmark

Occupation: Photographer

187

Tuesday, August 13th 2013, 8:58am

Quoted from "klaus.krpano"

Hi,

there is another small update for version 1.16:
krpano 1.16.6

Please see here for a list about all changes and fixes:
Release Notes 1.16.6

Best regards,
Klaus

I am confused about your version numbers. Why do you call it 1.17 when the actual version is 1.16.6
And when is the expiredate for this last update. I can not find it anywhere in the release notes.

Hans

Go to the top of the page

klaus.krpano

Administrator

Posts: 13,499

188

Tuesday, August 13th 2013, 12:32pm

Hi,

Quoted from "HansNyb"

I am confused about your version numbers. Why do you call it 1.17 when the actual version is 1.16.6
And when is the expiredate for this last update. I can not find it anywhere in the release notes.

1.16.x = the current version
1.17.x = the next version (with new features like html5 multires and more)

The next version, the 1.17 one, is still not finished and therefore currently only available as 'pre-release / preview' version. Once this version is fully ready, it will become the normal/current version (then as full release and without any expire date of course).

The 1.17 preview version will be updated from time to time, sometimes also without forum- or news-announcement. The current expire date is 2013-08-31.

For the current version, the 1.16 one, there will be still bugfixed-versions, like the new 1.16.6, until the 1.17 version is ready.

Best regards,
Klaus

Go to the top of the page

nosferatu

Trainee

Posts: 80

Location: Ukraine

189

Wednesday, August 21st 2013, 10:57am

Bug: Overflow RAM in Chrome, krpano 1.16 (v.4, 5 ,6)

Hi Klaus!
I found some problems in the browser Chrome 28 on Windows XP.
When viewing pano on desktop in forced HTML5-mode initial level of RAM = 800 kb, but after "refresh" RAM = 1100 kb,
again refresh, RAM = 1400 kb, and so infinitely.

Go to the top of the page

klaus.krpano

Administrator

Posts: 13,499

190

Wednesday, August 21st 2013, 9:13pm

Hi,

Quoted from "nosferatu"

found some problems in the browser Chrome 28 on Windows XP.
When viewing pano on desktop in forced HTML5-mode initial level of RAM = 800 kb, but after "refresh" RAM = 1100 kb,
again refresh, RAM = 1400 kb, and so infinitely.

What your are describing can't be a krpano bug, it's not possible for a webpage/Javascript to keep memory after a 'refresh'. So it's either a browser bug (but I can't reproduce it) or just some kind of browser internal caching.

Best regards,
Klaus

Go to the top of the page

nosferatu

Trainee

Posts: 80

Location: Ukraine

191

Wednesday, August 21st 2013, 10:14pm

Quoted

What your are describing can't be a krpano bug, it's not possible for a webpage/Javascript to keep memory after a 'refresh'. So it's either a browser bug (but I can't reproduce it) or just some kind of browser internal caching.

The problem arose in offline mode. Resolved deleting key: --enable-file-cookies
Thanks for the hint Klaus!

Go to the top of the page

nosferatu

Trainee

Posts: 80

Location: Ukraine

192

Saturday, August 31st 2013, 4:51pm

Bug: Overflow RAM in Chrome

Hi Klaus!
Some of my friends have confirmed the memory problem.
I did more tests and found out a few conditions.
1) Use multires.
2) Or use tile size is not 2048 (for example 2000)

This post has been edited 1 times, last edit by "nosferatu" (Sep 3rd 2013, 9:57pm)

Go to the top of the page

klaus.krpano

Administrator

Posts: 13,499

193

Wednesday, September 11th 2013, 4:09pm

Hi,

Quoted from "nosferatu"

Some of my friends have confirmed the memory problem.

Please read above - such can't be a krpano bug. It's impossible for a web-page to keep memory after reloading/after refreshing and also with normal pano viewing it should be not possible to see such increase in memory usage. That means this is probably a specific Chrome bug, maybe related to the operation system...
Please post your problem directly to the Chrome developers (crbug.com), on my side I can't reproduce anything in that direction.

Best regards,
Klaus

Go to the top of the page

klaus.krpano

Administrator

Posts: 13,499

194

Wednesday, September 11th 2013, 8:37pm

Hi,

there is new bugfix release:
krpano 1.16.7
Release Notes 1.16.7

The new release fixes several minor bugs and includes also a workaround for an annoying iOS 7 iPad bug.

In iOS 7 on the iPad the Safari browser sizes the html elemets wrongly so that a part of the html page (and so the pano) is partially off/outside the visible page. The iOS 7 bug itself was reported to Apple, but as it seems it was ignored because the bug is still present in the iOS 7 'Golden-Master' release from today. The new krpano release detects this iOS 7 bug and automatically corrects the wrong sizes.

The 1.17 html5-multiresolution pre-release version has got the same bugfixes and was updated today too:
http://krpano.com/html5multires/#download

Best regards,
Klaus

Go to the top of the page

klaus.krpano

Administrator

Posts: 13,499

195

Friday, September 27th 2013, 11:50am

Hi,

there is a new minor 1.16 update - mainly related to scalechildren and the Flash video problems:
krpano 1.16.8

Please see the release notes here for a list of all changes and fixes:
Release Notes 1.16.8

Best regards,
Klaus

Go to the top of the page

jschrader

Intermediate

Posts: 237

Location: Bavaria, Germany

Occupation: Photographer, Producer

196

Friday, September 27th 2013, 12:50pm

Go to the top of the page

klaus.krpano

Administrator

Posts: 13,499

197

Monday, October 28th 2013, 8:32pm

Hi,

another small 1.16.x update:
krpano 1.16.9
Release Notes 1.16.9

The 1.17 html5-multires pre-release was updated too and has got the same fixes. Additionally the showing of the watermark was delayed and the expire-date extended to 2013-12-31.

Best regards,
Klaus