I have updated the code. There was the "24 bits" missing after "3byte-array 
uint32_t deref", so the random top byte used to mix in with the transparency 
via "bitor".

08.06.2019, 15:03, "Alexander Ilin" <ajs...@yandex.ru>:
> Unfortunately, the posted code doesn't work as expected: the text labels 
> randomly have colored background. I'm looking for a reason.
>
> 08.06.2019, 14:04, "Alexander Ilin" <ajs...@yandex.ru>:
>>  Hello, Norio, and thank you for the benchmark code!
>>
>>    I was able to use it, and here are results from my tests. I took the 
>> typical output values.
>>
>>    Your code from commit a16eb2f6:
>>    Opaque background: 146 ms
>>    Transparent background: 1083 ms
>>
>>    My code from commit 26516f97:
>>    Opaque background: 142 ms
>>    Transparent background: 1800 ms
>>
>>    Then I incorporated the optimizations suggested by Jonh Benediktsson, and 
>> here are the new results from commit fa57d052:
>>    Opaque background: 143 ms
>>    Transparent background: 150 ms
>>
>>    In general, the Transparent test runs about 6-7 ms longer than the Opaque 
>> test, i.e. about 4.6% slower, which is a much better performance 
>> characteristic compared to the previous approaches.
>>
>>    The updated code is posted here: 
>> https://github.com/AlexIljin/factor/tree/win-transparent-text-background
>>
>>  08.06.2019, 10:11, "KUSUMOTO Norio" <kusum...@na.rim.or.jp>:
>>>   Ouch, an e-mail wasn't sent to the mailing list because I sent it direct.
>>>   I rewrite the wrong part and send it again.
>>>
>>>   On 2019/06/05 22:24, John Benediktsson wrote:
>>>>    Or clear the cache each time, so you don’t have to be lower level 
>>>> code...
>>>>>    On Jun 5, 2019, at 4:24 AM, Alexander Ilin <ajs...@yandex.ru>wrote: I 
>>>>> think the pairs of {font, string} are cached and only drawn once by the 
>>>>> GUI framework. I think you need to call the drawing code more directly to 
>>>>> get real performance data.
>>>
>>>   Thank you, Alexander and John.
>>>   I decided to measure only internal processing time.
>>
>>  ---=====---
>>   Александр
>>
>>  _______________________________________________
>>  Factor-talk mailing list
>>  Factor-talk@lists.sourceforge.net
>>  https://lists.sourceforge.net/lists/listinfo/factor-talk
>
> ---=====---
>  Александр
>
> _______________________________________________
> Factor-talk mailing list
> Factor-talk@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/factor-talk

---=====--- 
 Александр



_______________________________________________
Factor-talk mailing list
Factor-talk@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/factor-talk

Reply via email to