If you go to work next monday and find all computers in the entire building are running Sylenth1 test versions, at least now you know where they came fromHunterKiller wrote:Just bust into people's homes at random and say it's a matter of national security!
Sylenth1 version 2.0 update
-
- KVRian
- Topic Starter
- 1330 posts since 25 Sep, 2006
-
- KVRist
- 433 posts since 26 Sep, 2006 from Aruba
Since the 16bit/32bit colour change resulted in a different behaviour, i think that the specific video driver MIGHT have to do something with it. If you're using a NVIDIA based card, I'd suggest to run the latest detonator driver ( http://www.nvidia.com/Download/index.aspx?lang=en-us ) or if you're using an ATI based card: http://ati.amd.com/support/driver.html
-
- KVRian
- Topic Starter
- 1330 posts since 25 Sep, 2006
I did some testing and found the bug that caused some GUI parts sometimes to be redrawn continuously at idle. It only affects the demo version though, not the full version. I'll fix this asap.cc-purno wrote:- Under 16 bit color, sylenth now also has a high idle CPU usage under savihost(on 32 bit it was zero!) Its around 50% (same as in orion under 32 bit color!)
> Closing the GUI does not help
> As soon as I switch to 32 bit color, idle CPU use jumps again to ~zero
> Stangely, this does not work the other way around. Because when switching again to 16 bit, idle CPU use stays ~zero.
This bug is not the real problem though, the redrawing is very fast on most systems ( < 0.5 % cpu even when redrawing continuously!). The real problem is caused by a windows function that is used for drawing operations, which seems to be extremely slow on some systems. This probably relates to the 16bit vs 32bit issue and possibly the videocard drivers. I'll see if I can create a workaround for it.
Update: I would be interested to know what score you get with this little tool (it's in the Release folder inside the zip):
http://www.stereopsis.com/blttest/blttest.zip
On the system that has the high CPU usage problem I found an average BitBlt of 55 fps. On my pc I get an average of 4500 fps, almost a 100 times faster.
-
- KVRist
- 88 posts since 29 Sep, 2003
Hi Lennard,
I runned the test:
32 bit:
BitBlt:
avg: 405.5 fps [475.1 MB/sec]
max: 461.3 fps [540.6 MB/sec]
min: 61.2 fps [71.7 MB/sec]
ReverseBlt:
avg: 85.1 fps [99.7 MB/sec]
max: 87.5 fps [102.6 MB/sec]
min: 80.3 fps [94.1 MB/sec]
16 bit:
BitBlt:
avg: 226.7 fps [265.7 MB/sec]
max: 334.6 fps [392.1 MB/sec]
min: 36.5 fps [42.8 MB/sec]
ReverseBlt:
avg: 104.1 fps [121.9 MB/sec]
max: 129.2 fps [151.5 MB/sec]
min: 43.3 fps [50.7 MB/sec]
CC
I runned the test:
32 bit:
BitBlt:
avg: 405.5 fps [475.1 MB/sec]
max: 461.3 fps [540.6 MB/sec]
min: 61.2 fps [71.7 MB/sec]
ReverseBlt:
avg: 85.1 fps [99.7 MB/sec]
max: 87.5 fps [102.6 MB/sec]
min: 80.3 fps [94.1 MB/sec]
16 bit:
BitBlt:
avg: 226.7 fps [265.7 MB/sec]
max: 334.6 fps [392.1 MB/sec]
min: 36.5 fps [42.8 MB/sec]
ReverseBlt:
avg: 104.1 fps [121.9 MB/sec]
max: 129.2 fps [151.5 MB/sec]
min: 43.3 fps [50.7 MB/sec]
CC
- KVRAF
- 7795 posts since 20 Jul, 2004 from Clearwater
Lennard, hopefully all is going well! I would like to take that beta seat when it's ready.Lennard wrote:Sure, but a beta version will be ready only about a week or so before official release.djanthonyw wrote:Wow, thought it was going to be sooner than that, but I know you're trying your best to get it out. Lennard, is there any chance of current users being able to beta test?
Wavsen.com - Professional mix delivery platform with client approval, watermarking, and portfolio page builder.
-
- KVRist
- 76 posts since 6 Jan, 2008 from the Netherlands
-
- KVRAF
- 1616 posts since 26 Jun, 2005
Would love to see that! Would bring Sylenth1's sound possibilitys to a hole new level!djones wrote:Will Sylenth1 have some sort of FM mode in the future?

