Sorry your browser is not supported!

You are using an outdated browser that does not support modern web technologies, in order to use this site please update to a new browser.

Browsers supported include Chrome, FireFox, Safari, Opera, Internet Explorer 10+ or Microsoft Edge.

DarkBASIC Professional Discussion / Bug: I was not right...

Author
Message
Viktor
22
Years of Service
User Offline
Joined: 7th Oct 2002
Location: Austria
Posted: 16th Feb 2003 15:33
Link
That´s true, I was not right about the bug in BOX command (If you don´t remember: Flashget causes a slowdown by 100 times). Now LDIR rewrote his zoom code for screen editor in order to speed it up and found out that not the BOX was slowed down, but the POINT function...
So test it please, make 500 pixel POINTs readout without Flashget, and then start flashget (no need to download anything with flashget, just start it), and read 500 pixel POINTs again.
PC: AMD 1200/512 Mb SDR/GF2MX Dual Display/Win2000, 19" Monitor
ZX: Spectrum 128+ Customized version.
Current Project: BMP2SCR Pro (with LDIR)
Back to top
Profile PM
IanM
Retired Moderator
22
Years of Service
User Offline
Joined: 11th Sep 2002
Location: In my moon base
Posted: 16th Feb 2003 17:10
Link
Flashget? What's that.

BTW. Things will run a little faster with point() if you try this -



I got a massive speedup to a fill routine by doing this.
Back to top
Profile PM Email Website
Viktor
22
Years of Service
User Offline
Joined: 7th Oct 2002
Location: Austria
Posted: 19th Feb 2003 02:41
Link
Flashget is a download manager like Getright.
And about the Lock Pixels: LDIR shown me how he improved the code (it is used for a 8x8 zoom):
He did GET IMAGE, then MAKE MEMBLOCK FROM IMAGE, and finally he readed the data with MEMBLOCK BYTE.
Result:
Zoom of 56x56 pixel area:
Reading with POINT and output with BOX: 70 ms
COPY BITMAP with zooming: 180 ms
Reading IMAGE BYTES directly: 10 ms
Reading with POINT and Output with BOX if Flashget is running: ca 6800 ms

PC: AMD 1200/512 Mb SDR/GF2MX Dual Display/Win2000, 19" Monitor
ZX: Spectrum 128+ Customized version.
Current Project: BMP2SCR Pro (with LDIR)
Back to top
Profile PM
IanM
Retired Moderator
22
Years of Service
User Offline
Joined: 11th Sep 2002
Location: In my moon base
Posted: 19th Feb 2003 14:30
Link
I generally try and avoid using GET IMAGE in my main loop because of it's lack of speed, but that seems to work quite well for you.
Back to top
Profile PM Email Website
Viktor
22
Years of Service
User Offline
Joined: 7th Oct 2002
Location: Austria
Posted: 20th Feb 2003 10:51
Link
Yes, it was worth the change because it ich much faster than POINT() function. I still hope that GET IMAGE will be faster with later patches.

PC: AMD 1200/512 Mb SDR/GF2MX Dual Display/Win2000, 19" Monitor
ZX: Spectrum 128+ Customized version.
Current Project: BMP2SCR Pro (with LDIR)
Back to top
Profile PM
IanM
Retired Moderator
22
Years of Service
User Offline
Joined: 11th Sep 2002
Location: In my moon base
Posted: 20th Feb 2003 14:04
Link
The reason that GET IMAGE is so slow is that it has to retrieve the image from the display memory on the card. Even at 8xAGP that is much slower than if it was in memory.
Back to top
Profile PM Email Website
Viktor
22
Years of Service
User Offline
Joined: 7th Oct 2002
Location: Austria
Posted: 20th Feb 2003 18:12
Link
Anyway, it is still much faster than Point, so I have no other choice than using this.
What is faster:
GET IMAGE and MAKE MEMBLOCK FROM IMAGE
or
COPY BITMAP and MAKE MEMBLOCK FROM BITMAP
?
I guess, the second method is faster.

PC: AMD 1200/512 Mb SDR/GF2MX Dual Display/Win2000, 19" Monitor
ZX: Spectrum 128+ Customized version.
Current Project: BMP2SCR Pro (with LDIR)
Back to top
Profile PM
Back to top

Login to post a reply

Server time is: 2025-05-18 10:40:45
Your offset time is: 2025-05-18 10:40:45