Installed and registered without problems on a Win 8.1.3 Pro 64 bit system. A clean install.

A company without name and address:
"Recently we’ve reached a milestone of 10,000 likes on Facebook. And we decided to give our users more information about our development process in numbers. Have a look at today's infographic in our Tech Digest. "
http://icecreamapps.com/blog/true-facts-icecream-apps-software-development/

This doesn't give any real information (true facts) about, who they are or the development process...

PDF converter is a difficult field and PDF to ePub even more. They compete with a "Abby PDF transformer" and that's a tough task.

Upon start a small, modern looking, not resizable window opens. You select the PDF to ... or the ... to PDF branch.
The OCR is done with Tesseract, surely the best free OCR software at this moment. Best "free"!, not the best, tesseract cannot compete with ABBY.
I selected a PDF, and ooops, there is no conversion to ePub.

http://i.imgur.com/6msN6LL.png

I consider this as a good sign, they do not try, where they will fail.

Okay, I converted this book to OTF. The conversion takes quite I long time, the result is. Look at yourself:

http://i.imgur.com/dod7mA9.png

This is the result of the Abby program. Much, much faster:

http://i.imgur.com/BGwXnX2.png

It IS the same input and the same OTF output. You see??? Unbelievable!

The program may be useful for very simple tasks, fails completely to do something more specific. Okay, it's a version 1.47 of a young company. But a long, long way to a working program. I hope they will succeed. Instead of concentration on "Likes" they should go into the coding section.

Uninstalled via reboot.

Karl  –  8 years ago  –  このコメントは役に立ちましたか？ はい | いいえ (+176)

翻訳

"Also is OCR a big requirement for everybody?"

If it was there would be more OCR software in the marketplace -- there used to be much more choice & competition. HP developed Tesseract, & released it as open source about a decade after they stopped developing it. http://en.wikipedia.org/wiki/Tesseract_(software)

"It seems difficult to find a good pdf program thsse days"

It depends on what you're after. There is a lot of software that does a good to great job creating PDFs. When you're looking at converters & such, remember that one of the main reasons PDF exists is to protect content, to make it difficult for anyone to get that content out & re-use it. Look at it that way & many companies like Icecream do a reasonable to good job.

<i>"Also is OCR a big requirement for everybody?"</i> If it was there would be more OCR software in the marketplace -- there used to be much more choice &amp; competition. HP developed Tesseract, &amp; released it as open source about a decade after they stopped developing it. http://en.wikipedia.org/wiki/Tesseract_(software) <i>"It seems difficult to find a good pdf program thsse days"</i> It depends on what you're after. There is a lot of software that does a good to great job creating PDFs. When you're looking at converters &amp; such, remember that one of the main reasons PDF exists is to protect content, to make it difficult for anyone to get that content out &amp; re-use it. Look at it that way &amp; many companies like Icecream do a reasonable to good job.
Save | Cancel

mike  –  8 years ago  –  このコメントは役に立ちましたか？ はい | いいえ (+19)

翻訳

This is an honest ethical company, I have used a number of their programs, Also I seldom give a company feed back unless I am VERY impressed, I contacted them and got an almost immediate response, great support from an honest company, and no I do not work for them!

Michael N Hart  –  8 years ago  –  このコメントは役に立ちましたか？ はい | いいえ (+1)

翻訳

A developer producing excellent software can be honest and ethical. But so, too, can a developer producing junk. Comparing the performance of IceCream's program with Abby as Karl has so helpfully done and it's obvious that in this case, the corporate integrity you're applauding and the coding competence Karl is criticising bear no relation to each other. One is apples, t'other is oranges.

A developer producing excellent software can be honest and ethical. But so, too, can a developer producing junk. Comparing the performance of IceCream's program with Abby as Karl has so helpfully done and it's obvious that in this case, the corporate integrity you're applauding and the coding competence Karl is criticising bear no relation to each other. One is apples, t'other is oranges.
Save | Cancel

MikeR  –  8 years ago  –  このコメントは役に立ちましたか？ はい | いいえ (+59)

翻訳

Thanks, Karl, for the very helpful evaluation. I'm paying Adobe (the original creator of the PDF format) $10 per month for their CreatePDF online converter. Of course, it works flawlessly. I do a lot of PDF conversions in my business, so I need software that works without problems.

There seems to be a lot of junk out there, which is disappointing. One thing that many people do not realize is that a conversion program, even Adobe Acrobat, has a very limited number of fonts to convert to. Adobe Acrobat has their "standard 14 fonts," which is actually only five fonts, with three being text fonts (Times, Helvetica and Courier). If an original document uses a font different from these base fonts, the particular font must be embedded in the PDF during conversion. Otherwise, the PDF reader will not be able to recreate the document to match the original. When the particular font is not available, the PDF reader will use whatever font it has that is most equivalent. This often results in the PDF not matching the original document, with some text not fitting on the page, or being jumbled.

So, in many cases, it's an issue with embedding the font (if available) in the PDF during conversion. Sometimes, the original font is simply not available, such as when scanning a document with an OCR (Optical Character Recognition) program. When using an OCR program to convert a printed page, such as a page from a printed book, to a digital document, some sacrifice of format or layout may be necessary.

Karl's test was a good one. One thing the Ice Cream program missed was the column separation. This caused the text from the left column to run into the text in the right column. Also, Ice Cream did not distinguish between the two different fonts used in the original, defaulting all text to Helvetica. And, it missed the diamond-shaped bullet points.

Karl, was this an OCR conversion? That would explain the atrocious results.

The producers make some pretty bold claims as to what this program is capable of. But, the proof is in the pudding (or ice cream). Sorry, but "Ice Cream" should probably be "I SCREAM" because that's what most users are going to do after they see what kind of results this program produces.

Dennis  –  8 years ago  –  このコメントは役に立ちましたか？ はい | いいえ (+30)

翻訳

It is interesting that converting a PDF to an image or presenting it on screen offers no challenges whereas converting directly to a document format appears to be almost impossible.

This means that converting to an image and then doing an OCR of the image is the best route to convert into a document format.
The quality of the OCR program is the most important if this method is chosen and the Abbyy OCR is exceptionally good.

Abbyy OCR was given away with some of the Canon printers in the past and although it couldn't take in PDF directly converting it to an image using PDFfill Tools first made it a viable option.
http://www.pdfill.com/pdf_tools_free.html

Older versions of Abbyy OCR are also available quite cheaply and will give first-class results, well worth a look.

XP-Man  –  8 years ago  –  このコメントは役に立ちましたか？ はい | いいえ (+26)

翻訳