pouët.net

Blueberry information 1388 glöps

• general:
• level: user
• personal:
• first name: Aske Simon
• last name: Christensen
• cdcs:
• cdc #1: Nexus-7 by Andromeda [web]
• cdc #2: Tint by The Black Lotus [web]
• cdc #3: Gift by Potion
• cdc #4: STS-02: Electric Kool-Aid by Synesthetics
• cdc #5: TBC Realtime Experience by Too Bloody Cheesy [web]

• demo Amiga OCS/ECS Black Light by Wanted Team [web] & Lamers
• I like it.
• rulezadded on the 2015-11-21 19:13:55
• 32k Windows XX by Onslaught [web]
• Mmmm, let us do that again some time. :)
• rulezadded on the 2015-11-20 17:23:09
• 4k Windows Light Is Strange by Eos
• Beautiful! Very nice idea.
• rulezadded on the 2015-11-11 00:17:21
• demotool Windows MacOSX Intel Clinkster by Loonies [web]
• There is a new version available!
  
  It adds the buffer padding in the player code mentioned in my reply to the comment by TheT.
  
  It also includes a 64-bit build of the VST as suggested by Weyland Yutani. This actually seems to be significantly faster (tested in 32-bit Renoise), so that was an easy win. :)
  
  In addition to this, I have included a multithreaded version of the player source, computing the left and right channels in separate threads. Assuming at least two CPU cores, this computes the music about twice as fast, at the expense of somewhat bigger code. If you can spare the space, use it!
  
  Both versions also contain a small addition to the index decay computation which avoids running into denormals if the index decays down to very small values. This also improves the speed considerably in some cases.
• isokadded on the 2015-10-31 21:15:38
• demotool Windows MacOSX Intel Clinkster by Loonies [web]
• Quote:

  Also, I think I might have found a mostly irrelevant bug. In the loop that converts from the MixingBuffer to the C_MusicBuffer, you're using ECX*4 as an offset relative to the start of both buffers; however, ECX is in the 1..TOTAL_SAMPLES range, so the first value is unused and more "importantly" the write to C_MusicBuffer will overwrite what's right after it.

  
  Well spotted. Thanks!
  
  However, the fix you suggest will impair the compressed size of the player code, as the otherwise nice (24 bit aligned) label constants will then not be as nice. An alternative fix, starting the counter from TOTAL_SAMPLES-1, would have the same problem.
  
  I think the best fix would be to simply pad the MusicBuffer to avoid overwriting anything important. The one missing sample at the beginning of the music is probably not a big deal, unless you have a REALLY sharp start to your music. :)
• isokadded on the 2015-10-31 18:17:47
• demo SAM Coupé Back To The Past by Speccy.pl [web]
• Great celebration.
• rulezadded on the 2015-10-21 07:58:08
• 64k Windows A+ by Inque [web]
• Well put together.
• isokadded on the 2015-10-20 19:09:00
• 64k Amiga OCS/ECS refocus by Pacific
• What 4pLaY said. ;)
• rulezadded on the 2015-10-20 00:06:19
• 64k Amiga OCS/ECS Fari Bars by Rift & Slipstream [web]
• Cute one.
• rulezadded on the 2015-10-20 00:02:01
• demo Windows Function Over Fame by Excess [web]
• I'm usually not much for the Excess style, but I enjoyed this one.
• rulezadded on the 2015-10-19 23:57:50