it's possible only in server version! server release is our new strategic approach, we'll develop a lot of things on it: infact we have now a "single" application, not related with the host, with a single access point: for example we could control it using an iPad, we can save resources (this new version), we can implement new incredible things like a true crosstalk. The beauty: you can live perfectly without it, but if you use nebula A LOT it's better if you have a server running in your computer. Soon we'll release a version able to load as a "service" (this was our first approach), so you'll get a sort of "nebula driver". All these things are not possible using different DLL or DYLIBS: we should use static objects in order to establish a communication among plugins, but it would be a mess, a strange architecture. It would be a simpler approach, but not very scalable. Now think to multiple GPU cards, remote servers, .... you'll save a lot of ram in cuda instances (we are working on it right now) which is VERY important because graphic card are not shipped yet with 16 gigabytes of ram now we are porting the WHOLE engine to CUDA, so it will be terribly fast and optimized...
At maximum latency using asio4all, reaper4Beta neb pro 3 gave me a dozen channels of alex's VCC stero mic input before I hit 100% cpu and 554mb ram.
A new project setup the same from scratch using server clients and Im seeing 7% cpu and 167mb ram. ???
Should I see such a drop in cpu as well as the ram improvements. or is this indicating a problem with the (previous) pro 3 install? Either way, fantastic work. This is nebula as it should be Steve
--2 hours later-- some sort of logging would be good. Is that a feature already? The single disappearance of Reaper in mid-mix normally comes about simply by using all the available ram. be interested to know whether the (so far, only) host crash was during a period of high ram usage. Still great........... !!
Last edited by sneaky on Mon Apr 18, 2011 11:07 am, edited 1 time in total.
13 MLC IL1 instances without the server = 1204MB and 98% CPU and project struggles...
16 MLC IL1 instances with the Server = 303.4MB and 98% CPU but the project runs smoothly!!!!
This is a true revolutionary moment for Nebula. Giancarlo, man, I don't know how you did it but you just broke all paradigms and now CPU power is the real deal breaker heh RAM is not a problem anymore if you're one of those guys (like me) using several instances of the same program (for EQ's).
NOTE: don't read REAPER's performance meter when using the server, read what the server tells you instead since REAPER won't show you what's going on "outside".
i7 3770k :: Asus P8H77-V LE :: 16Gb DDR3 @1600MHz :: Geforce GT 520 :: OCZ-Vertex 128Gb :: WD Black Series 1Tb and Green Series 1Tb :: F*******e Liquid56 :: REAPER 64bit and StudioOne 64bit (both latest versions) :: Win 10 64bit