Hi guys Of late, there have been a lot of complaints and a number of serious debates over the speed of Bryce5's render engine. Bryce5 is indeed slower in renders. I have been unable to produce an equal or slower result in Bryce5 when using the nasty 4 (reflection, refraction, bump, volumetrics). If you can render a good image without these, perhaps you should do a tute? lol anyhow, even tho I have not successfully been able to bring the time down to equal or below...I can bring it down to within 3% which is entirely acceptable. However, the method that this requires is silly, but until corel fixes this then here is what to do. In my trial, I used an image that contains all of the nasty 4. This image averaged close to 1 hour for a render. You can see the time differences in the attached image. You can see that a b4 file directly rendered in b5 is a bit slower than if you directly convert it to b5 then reopen it. You can also see that if you ctrl-alt-delete and close everything but the systray, explorer, and bryce5 it is very near the same time as b4. What this is all pointing to is that Bryce thinks it is running in the background all of the time. It does not appear to ever be taking the "primary" window unless YOU tell windows to take it. I have noticed that when you click "render" that the button on the start bar actually pops out. This button is one way that windows determines the program with focus and designates the primary cpu usage. If it is out..then duh..it can be considered "background". Be sure to click it back in. btw, Im not making excuses for the slower renders. Im just trying to help you find a way to bring the times down until Corel gets this problem fixed. Additionally, it should be stated that the Bryce5 download trial is lots slower than the actual version of bryce5. Corel is aware of the problem and is working on it. We can only hope that they fix the slow renders for everyone. oh..and for those of you that wanted to see how much faster b4 rendered when it was the only thing running, that is included in the pict too. In windows, you should ALWAYS be running a memory manager. But for this test, I turned mine off. With it on, it made little difference with only the 3 items running in windows. Windows memory management sucks. Simply installing a freebee mem manager will fix not only Bryce's hog for RAM, but also other program. Get rambooster if you dont already have it and set it to free about 5% of your ram every 30 seconds. Do not accept its default setup (its defaut is every second). Leaving it at that setting will make windows spend more time fixing the ram that doing your renders. hope this helps some...and at least for now you have a way to work around the slow times... BT

That's interesting. Under Win2k you can observe how much CPU time a app is taking by bringing up the Task Manager (I'm sure you know this, but I'm just saying it for the record.) There is no difference in CPU usage % when Bryce has focus or when it is minimized. And from watching the number of Page Faults, the Memory Delta, the current Virtual Memory, Peak Memory, blah blah blah, my plan is to go in and tweak some memory related settings (page memory size, etc) and see what happens. That said, given how Windows works, I can completely see where terminating all other running processes would free up CPU time for Bryce. :-) TomD

I have Memory Turbo fo rmy memory manager and it works great, Rambooster is pretty good too for a free app. Explorer and Systray are all you need to have Windows running. A couple of other things I would mention to speed up Window 98SE are>> Right click on my computer and go to properties: go to performance, and then click the file system button, this should be set to Network Server, it uses the cache better. Set your Virtual Memory to 2-2.5 times the amount of ram you have. Don't have a bunch of quicklaunch icons on your taskbar, this eats up cpu usage. No wallpaper. No screensaver, just go to black after 20 minutes or so. Disable startup items, by going to Run, and typing msconfig and going to the startup tab, you can stop things from loading at startup there. Thanks for the tests Brycetech, everything helps, and it looks like you might be onto something. kaom

thanx guys Im not sure of the difference between win2k and winME. My understanding is that 2k is a lot different (more like NT). If that be the case, then the steps to make it the primary are different and I also understand that those will allow you to set a priority (?). I know win95 used to do that. I still think its a priority thing, but shrug...I dont have the code in front of me to debate it accurately. That is just a guess, but the render times are actual observations. Anyhow, this is all just to help get the times down. Its not really worth the effort if the render is only an hour or so. But, from the start B5 time...a 6 hour render on B4 would be a 7 hour render on B5...but simple changes, you can make it a 20 hour render in B4 a 21 hour render in B5. later BT

Bryce has used Idle-time processing since Bryce 4. That actually made Bryce 4 slightly slower than Bryce 3 depending on what you were doing while rendering, but also made it smoother and it multitasks better. And Bryce 4 got more idle time in Win98 original than in later Win98's. Windows does more on its own with nothing running in SE and ME. Under the original Win98 Bryce 4 was getting 97-98% of the processor time, but after applying Windows updates to the equivalent of SE, it rarely got more than 92%. Bummed me out. Why all this concern with Bryce 4 projects in Bryce 5? I think Bryce 5 projects render no slower than Bryce 4 if the project is completely Bryce 5. At least that was my experience when I duplicated projects from scratch. Bryce 4 projects do render slightly slower in B5 and I'm not sure why. Spit

BTW...just a comment. No matter what people say, I don't believe in running those ram booster programs. People set their disk cache size so high and then wonder why they don't have any free memory left and then use an extra program to dump it. It takes processor cycles too. Doesn't make sense to me. I actually had more ram free by limiting my disk cache to about 32 megs than running a ram cleaner. And I do NOT recommend setting your machine as a server either. It just keeps more and more stuff in ram that you'd rather it flushed. Servers need that, we don't. Spit