> It seems that the embedded font is MingLiU (1996-, the 2nd
> one of the checksum list). Do you think it is worse than
> the result for the latest MingLiU (non-tricky version)?
Sorry, maybe too few words. My eyes are already too old to
evaluate the subpixel rendering result is better or worse
with others. So I want to understand how you feel about the
current situation.
a) latest MingLiU shows different result, it is better than old MingLiU.
b) old & latest MingLiU show same results, both worse than other fonts.
c) something else.
Regards,
mpsuzuki
suzuki toshiya wrote:
> It seems that the embedded font is MingLiU (1996-, the 2nd
> one of the checksum list). Do you think it is worse than
> the result for the latest MingLiU (non-tricky version)?
>
> Regards,
> mpsuzuki
>
> P.S.
> Werner, I'm thinking about the insertion of some debug
> messages for tricky font checker for users to help what
> is going on.
> * trace message when the font is recognized as tricky
> * supposed family name when nameless font is recognized as tricky
> How do you think of?
>
> It is possible to write separated code for ft2demos, but
> tricky font functions are private, so separated code
> would include some duplication.
>
> address@hidden wrote:
>> Werner and Toshiya-san，
>> Could you help check this file? the subpixel rendering effect is not very good for this tricky font。
>> I render text use 10 font size， it look very sharp.
>>
>>
>> ------------------ Original ------------------
>> From: "suzuki toshiya"<address@hidden>;
>> Date: Fri, Nov 3, 2017 03:21 PM
>> To: "address@hidden"<address@hidden>; "Werner LEMBERG"<address@hidden>;
>> Cc: "address@hidden"<address@hidden>;
>> Subject: Re: [ft-devel] Text rendering error
>>
>> Just I've updated the tricky font list!
>>
>> By checking the embedded TrueType in PDFs collected
>> from the internet, I found several more tricky fonts.
>>
>> http://greg.seth.free.fr/ftp/Japanese%20Font%20Collection.pdf
>> includes tricky fonts:
>> BBJKGJ+DFGothic-EB
>> BBLENH+DFGyoSho-Lt
>> BBLKIP+DFHSGothic-W5
>> BBLNLM+DFHSMincho-W3
>> BBMBBC+DFHSMincho-W7
>>
>> I added their family names and checksums. About DFHSxxxx
>> families, I would post some request for discussion in later.
>>
>> Regards,
>> mpsuzuki
>>
>> suzuki toshiya wrote:
>>> All data I've checked are PDFs collected from the internet, so
>>> I'm not sure the exact version numbers. However, some subsetted
>>> fonts retain the name table (but no version number) with copyright
>>> notice, so we can guess when the font is changed to non-tricky.
>>>
>>>>> <tricky>
>>>>> DAWJQA+DFGothic-EB-WIN-RKSJ-H
>>>>> tag cvt offset 0x00000112 length 0x00000350 checkSum 0x12c3ebb2
>>>>> tag fpgm offset 0x00000462 length 0x000087a7 checkSum 0xb680ee64
>>>>> tag prep offset 0x0001a017 length 0x00000758 checkSum 0xce939563
>>> "(c) Copyright DynaLab Inc. 1992-1995": tricky (checksums are same with 1.pdf)
>>> tag cvt offset 0x000000bc length 0x00000350 checkSum 0x12c3ebb2
>>> tag fpgm offset 0x0000040c length 0x000087a7 checkSum 0xb680ee64
>>> tag name offset 0x0001b3e8 length 0x000002ad checkSum 0xb7982db8
>>> tag prep offset 0x00011590 length 0x00000758 checkSum 0xce939563
>>>
>>> "(c) Copyright DynaLab Inc. 1992-1996": non-tricky
>>> tag cvt offset 0x0000010a length 0x00000008 checkSum 0x00000000
>>> tag fpgm offset 0x00000112 length 0x00000011 checkSum 0xf40ddf13
>>> tag name offset 0x00011b5a length 0x0000018f checkSum 0xbbce71a9
>>> tag prep offset 0x00011ce9 length 0x00000013 checkSum 0x6286ee0a
>>>
>>> "(c) Copyright DynaComware Corp. 2002": non-tricky
>>> tag cvt offset 0x00000114 length 0x00000008 checkSum 0x00000000
>>> tag fpgm offset 0x0000011c length 0x00000011 checkSum 0xf40ddf13
>>> tag name offset 0x00011ca8 length 0x00000254 checkSum 0x90c0733e
>>> tag prep offset 0x00008988 length 0x00000013 checkSum 0x6286ee0a
>>>
>>> "(c) Copyright DynaComware Corp. 2004": non-tricky
>>> tag cvt offset 0x00000112 length 0x00000350 checkSum 0x12c3ebb2
>>> tag fpgm offset 0x00000462 length 0x000087a7 checkSum 0xb680ee64
>>> tag name offset 0x0001ac62 length 0x000002f1 checkSum 0xbfe60d42
>>> tag prep offset 0x0001af53 length 0x00000758 checkSum 0xce939563
>>>
>>> "(c) Copyright DynaComware Corp. 2007": non-tricky
>>> tag cvt offset 0x00000114 length 0x00000008 checkSum 0x00000000
>>> tag fpgm offset 0x0000011c length 0x00000011 checkSum 0xf40ddf13
>>> tag name offset 0x00011adc length 0x00000329 checkSum 0x75e93e7b
>>> tag prep offset 0x00008b48 length 0x00000013 checkSum 0x6286ee0a
>>>
>>> Regards,
>>> mpsuzuki
>>>
>>>
>>> address@hidden wrote:
>>>> So, DFGothic-EB(version 2.1 (1997 later)) is not tricky font?
>>>>
>>>>
>>>> ------------------ Original ------------------
>>>> From: "suzuki toshiya"<address@hidden>;
>>>> Date: Thu, Nov 2, 2017 04:16 PM
>>>> To: "Werner LEMBERG"<address@hidden>;
>>>> Cc: "address@hidden"<address@hidden>; "address@hidden"<address@hidden>;
>>>> Subject: Re: [ft-devel] Text rendering error
>>>>
>>>> Collecting more PDFs, I got a few PDFs including DFGothic-EB
>>>> with same checksums! I think DFGothic-EB in 1.pdf have stable
>>>> checksum.
>>>>
>>>> Regards,
>>>> mpsuzuki
>>>>
>>>> suzuki toshiya wrote:
>>>>> Dear Werner,
>>>>>
>>>>> Werner LEMBERG wrote:
>>>>>>> Werner and Toshiya-san, i found the base name is
>>>>>>> DAWJQA+DFGothic-EB-WIN-RKSJ-H in the pdf file ,
>>>>>> I found a `DFGothic-EB-WIN-RKSJ-H' version 2.1 (from 1997) in the
>>>>>> internet, which is no longer tricky. We thus would need an earlier
>>>>>> version, which might be difficult to find...
>>>>> Yes, I experienced similar. From a few PDFs including embedded
>>>>> DFGothic-EB, they are no longer tricky and the checksums are
>>>>> different. One of the strange points is that the maxp glyph number
>>>>> of tricky DFGothic-EB is greater than non-tricky DFGothic-EB (?_?)
>>>>>
>>>>> <tricky>
>>>>> DAWJQA+DFGothic-EB-WIN-RKSJ-H
>>>>> tag cvt offset 0x00000112 length 0x00000350 checkSum 0x12c3ebb2
>>>>> tag fpgm offset 0x00000462 length 0x000087a7 checkSum 0xb680ee64
>>>>> tag prep offset 0x0001a017 length 0x00000758 checkSum 0xce939563
>>>>>
>>>>> <non-tricky>
>>>>> IMLFFK+DFGothic-EB-WINP-RKSJ-H-90ms-RKSJ-H
>>>>> tag cvt offset 0x00009be4 length 0x00000008 checkSum 0x00000000
>>>>> tag fpgm offset 0x00009bec length 0x00000011 checkSum 0xf40ddf13
>>>>> tag prep offset 0x0001291c length 0x00000013 checkSum 0x6286ee0a
>>>>> JIMJKA+DFGothic-EB-MP-RKSJ-H-90ms-RKSJ-H
>>>>> tag cvt offset 0x00000114 length 0x00000008 checkSum 0x00000000
>>>>> tag fpgm offset 0x0000011c length 0x00000011 checkSum 0xf40ddf13
>>>>> tag prep offset 0x00008988 length 0x00000013 checkSum 0x6286ee0a
>>>>> MIBVGO+DFGothic-EB
>>>>> tag cvt offset 0x0000010a length 0x00000008 checkSum 0x00000000
>>>>> tag fpgm offset 0x00000112 length 0x00000011 checkSum 0xf40ddf13
>>>>> tag prep offset 0x00011ce9 length 0x00000013 checkSum 0x6286ee0a
>>>>>
>>>>>
>>>>> _______________________________________________
>>>>> Freetype-devel mailing list
>>>>> address@hidden
>>>>> https://lists.nongnu.org/mailman/listinfo/freetype-devel
>>>>>
>>> _______________________________________________
>>> Freetype-devel mailing list
>>> address@hidden
>>> https://lists.nongnu.org/mailman/listinfo/freetype-devel
>>>
>
>
> _______________________________________________
> Freetype-devel mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/freetype-devel
>
2.png