PDF (Un-) Optimizer ... :-(

[guebert]

The results are not good. See examples attached.

Michael

[Will - Tracker Supp]

Hi Michael,

Thanks for the post - could you please zip and upload a screen-shot of the settings that were used? I actually found, while testing, that the optimization gave me a dramatic reduction in file size (1MB to 211 KB).

Cheers,

[guebert]

Hi Michael,

Thanks for the post - could you please zip and upload a screen-shot of the settings that were used? I actually found, while testing, that the optimization gave me a dramatic reduction in file size (1MB to 211 KB).

Cheers,

Will,

which file did you use as input? "Test original.pdf" is about 2 MB, not 1 MB.

I tried nearly all *default* settings. The file name represents the settings.

Michael

[Will - Tracker Supp]

Hi Michael,

I apologize for the lack of response here, I'm not sure how this one slipped by. I'm going to need to re-familiarize myself with the issue and will post back my results & procedure again.

Again, apologies for the delayed response.

[Will - Tracker Supp]

Hi Michael,

I have been able to reproduce this now and have created an internal Development ticket to help track the progress of the issue. The ticket number is RT-2741.

We appreciate your patience and understanding here!

[guebert]

Hallo!

Anything new regarding this issue?

Michael

[Tracker Supp-Stefan]

Hi Michael,

No movement in the ticket for now. We will post again here when there are any further news.

Regards,
Stefan

[John - Tracker Supp]

Improvements to the Optimizer are coming in build 310 due in the next 7-14 days if all goes to plan.

[guebert]

BTW, the current build of the editor still produces big files when using downsizing feature of OCR function instead of doing OCR and then "file - save optimized" ... any reason for this?

Michael

[Will - Tracker Supp]

Hi guebert,

I'm not sure that I understand the feature that you're referring to. The OCR does not have any ability to optimize the file or downsample images. Can you please clarify what you are referring to?

Thanks,

[guebert]

I mean using the downsample feature of OCR - scanning a document at 400 dpi, running OCR and set 100 dpi as output (Create New Searchable PDF, Quality 100) produces poor quality but much bigger files than running OCR without downsampling (Preserve Original Content and Add Text Layer) but saving as optimized PDF with 100 dpi (eg presetting "mobile").

--> https://help.pdf-xchange.com/EUM/de ... :ocr_pages___

Michael

[Will - Tracker Supp]

Hi Michael,

Thanks for the info. - what is likely happening, is that the document is black and white, and using the Create New Searchable PDF option will likely change the image layer to greyscale, resulting in an increased file size. For these kinds of documents, I'd recommend that you simply use the ...Add Text Layer option to leave the original image untouched.

If you do not want to alter the original scan, simply use the File --> Save As Copy feature. The number of steps involved in the process will be the same.

Cheers,

[guebert]

Will,

but why does the OCR feature this change if it is not necessary? Shouldn't it do the same as OCR "Add Text Layer " and "save optimized" with the selected dpi-value?

Michael

[Will - Tracker Supp]

Hi Michael,

The current iteration of the OCR is a little old now and hasn't been modified, to any large degree, for a while. I'm not 100% sure why this was done initially, as the Dev. responsible for creating it is not longer with the company, but I suspect that we could probably change it safely, in a future release. I'll ask to see if this can be done.

Thanks,