Daniel Reetz, the founder of the DIY Book Scanner community, has recently started making videos of prototyping and shop tips. If you are tinkering with a book scanner (or any other project) in your home shop, these tips will come in handy. https://www.youtube.com/channel/UCn0gq8 ... g_8K1nfInQ

Scan Tailor "Experimental"

Scan Tailor specific announcements, releases, workflows, tips, etc. NO FEATURE REQUESTS IN THIS FORUM, please.
louiscan
Posts: 14
Joined: 10 Apr 2015, 17:35
Number of books owned: 0
Country: USA

Re: Scan Tailor "Experimental"

Post by louiscan » 04 Apr 2016, 11:40

These are the same build errors I reported above. See Tulon's response - it looks like they should be easy to fix and that we can hope for that soon. Waiting a few days for a real fix is likely better than applying my duct-tape style patch that probably will gum things up and cause unnecessary headaches.

Tulon
Posts: 687
Joined: 03 Oct 2009, 06:13
Number of books owned: 0
Location: London, UK
Contact:

Re: Scan Tailor "Experimental"

Post by Tulon » 04 Apr 2016, 14:00

I pushed a fix for build errors. Also, -DCMAKE_CXX_FLAGS=-std=c++11 should no longer be required.

I am still working on a fix for OpenCL crashes, trying to solve performance degradation issues caused by it.
Scan Tailor experimental doesn't output 96 DPI images. It's just what your software shows when DPI information is missing. Usually what you get is input DPI times the resolution enhancement factor.

Tulon
Posts: 687
Joined: 03 Oct 2009, 06:13
Number of books owned: 0
Location: London, UK
Contact:

Re: Scan Tailor "Experimental"

Post by Tulon » 05 Apr 2016, 15:57

I pushed the fix for OpenCL crashes.
Scan Tailor experimental doesn't output 96 DPI images. It's just what your software shows when DPI information is missing. Usually what you get is input DPI times the resolution enhancement factor.

louiscan
Posts: 14
Joined: 10 Apr 2015, 17:35
Number of books owned: 0
Country: USA

Re: Scan Tailor "Experimental"

Post by louiscan » 08 Apr 2016, 16:05

Hm... FYI, now I get this when running make on the latest commit:

Code: Select all


/Users/admin/Desktop/scantailor-experimental/acceleration/opencl/BinaryRasterOp.cpp:133:15: error: 
      no matching function for call to 'min'
                h_wg_size = std::min(h_wg_size, local_mem_size / sizeof(...
                            ^~~~~~~~
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1/algorithm:2589:1: note: 
      candidate template ignored: deduced conflicting types for parameter '_Tp'
      ('unsigned long' vs. 'unsigned long long')
min(const _Tp& __a, const _Tp& __b)
^
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1/algorithm:2599:1: note: 
      candidate template ignored: could not match
      'initializer_list<type-parameter-0-0>' against 'unsigned long'
min(initializer_list<_Tp> __t, _Compare __comp)
^
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1/algorithm:2581:1: note: 
      candidate function template not viable: requires 3 arguments, but 2 were
      provided
min(const _Tp& __a, const _Tp& __b, _Compare __comp)
^
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1/algorithm:2607:1: note: 
      candidate function template not viable: requires single argument '__t',
      but 2 arguments were provided
min(initializer_list<_Tp> __t)
^
/Users/admin/Desktop/scantailor-experimental/acceleration/opencl/BinaryRasterOp.cpp:139:15: error: 
      no matching function for call to 'min'
                v_wg_size = std::min(v_wg_size, local_mem_size / ((h_wg_...
                            ^~~~~~~~
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1/algorithm:2589:1: note: 
      candidate template ignored: deduced conflicting types for parameter '_Tp'
      ('unsigned long' vs. 'unsigned long long')
min(const _Tp& __a, const _Tp& __b)
^
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1/algorithm:2599:1: note: 
      candidate template ignored: could not match
      'initializer_list<type-parameter-0-0>' against 'unsigned long'
min(initializer_list<_Tp> __t, _Compare __comp)
^
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1/algorithm:2581:1: note: 
      candidate function template not viable: requires 3 arguments, but 2 were
      provided
min(const _Tp& __a, const _Tp& __b, _Compare __comp)
^
/Applications/Xcode.app/Contents/Developer/Toolchains/XcodeDefault.xctoolchain/usr/bin/../include/c++/v1/algorithm:2607:1: note: 
      candidate function template not viable: requires single argument '__t',
      but 2 arguments were provided
min(initializer_list<_Tp> __t)
^
4 warnings and 2 errors generated.
make[2]: *** [acceleration/opencl/CMakeFiles/opencl.dir/BinaryRasterOp.cpp.o] Error 1
make[1]: *** [acceleration/opencl/CMakeFiles/opencl.dir/all] Error 2
make: *** [all] Error 2
TK:scantailor-experimental admin$ 


Tulon
Posts: 687
Joined: 03 Oct 2009, 06:13
Number of books owned: 0
Location: London, UK
Contact:

Re: Scan Tailor "Experimental"

Post by Tulon » 08 Apr 2016, 18:36

louiscan wrote:Hm... FYI, now I get this when running make on the latest commit:
Should be fixed now.
Scan Tailor experimental doesn't output 96 DPI images. It's just what your software shows when DPI information is missing. Usually what you get is input DPI times the resolution enhancement factor.

foler
Posts: 9
Joined: 15 Apr 2016, 06:24
Number of books owned: 0
Country: Croatia

Re: Scan Tailor "Experimental"

Post by foler » 18 Apr 2016, 04:06

How to use experimental cli 64bit with multithreading? Cant see parameter in -h.

thanks

tuf
Posts: 20
Joined: 25 Oct 2015, 08:41
E-book readers owned: Kindle Touch
Number of books owned: 100
Country: Romania

Re: Scan Tailor "Experimental"

Post by tuf » 05 May 2016, 03:14

Is there a link to the page that ALLWAYS shows the latest version of STE?

Is this the one?
https://github.com/Tulon/scantailor/releases
Is the latest version from 22 february 2016?

I have some problems but i must be sure that i use the last version of the program.

Tulon
Posts: 687
Joined: 03 Oct 2009, 06:13
Number of books owned: 0
Location: London, UK
Contact:

Re: Scan Tailor "Experimental"

Post by Tulon » 05 May 2016, 09:52

tuf wrote:Is there a link to the page that ALLWAYS shows the latest version of STE?

Is this the one?
https://github.com/Tulon/scantailor/releases
Yes, the latest release should always be at the top of that page. The latest version is indeed the one from 22 february 2016.
Scan Tailor experimental doesn't output 96 DPI images. It's just what your software shows when DPI information is missing. Usually what you get is input DPI times the resolution enhancement factor.

tuf
Posts: 20
Joined: 25 Oct 2015, 08:41
E-book readers owned: Kindle Touch
Number of books owned: 100
Country: Romania

Re: Scan Tailor "Experimental"

Post by tuf » 07 May 2016, 05:11

This is an image that crashes the program:
http://www.mediafire.com/download/44s7e ... ge_003.zip

1. It happens if you manually split the image in 2 pages.
In order to crash the program, load the image and then go straight to the splitting menu, and select the double page.

2. Another error is the out of memory issue (i use the 64 bit version on 6 GB RAM, without swap file).
In order to crash the program: run the Fix Orientation option, the jump to the Output option and run it.
The CPU stays at 26%, but the RAM usage went through the roof: 3300 MB.

Tulon
Posts: 687
Joined: 03 Oct 2009, 06:13
Number of books owned: 0
Location: London, UK
Contact:

Re: Scan Tailor "Experimental"

Post by Tulon » 07 May 2016, 06:44

I could not reproduce the crash or the out-of-memory situation so far. Do you have OpenGL / OpenCL enabled? Does it still crash with those disabled?

The out-of-memory situation on 64-bit systems is usually caused by Qt's limitation of 2GB per image that's in place even in 64bit builds. Your image is huge (18000x18000 px), yet it doesn't break the 2GB limit, even with 2x resolution enhancement factor. However, if the image in question was color rather than grayscale, it would have broken that limit.
Scan Tailor experimental doesn't output 96 DPI images. It's just what your software shows when DPI information is missing. Usually what you get is input DPI times the resolution enhancement factor.

Post Reply

Who is online

Users browsing this forum: No registered users and 2 guests