Frogboy Frogboy

DreamScene gets mixed reception

DreamScene gets mixed reception

CPU use, a tale of two video cards..

Windows DreamScene is Microsoft's animated wallpaper program that's currently in "preview" (beta) for users of Windows Vista Ultimate.  So far, reactions have been mixed.

While users seem to love the concept, the amount of CPU used to do it seems to vary wildly.  From 12% or so on my ThinkPad T60 to nearly 50% on systems that are much faster.

The difference? So far, it basically can be thrown to the feet of nVidia. Their drivers, which not only seem to crash PC games do double-duty in being very inefficient with DreamScene.

Below are the results on my laptop. Note the very low (as in not great) performance rating and yet its relatively low CPU use while running a DreamScene. It has an ATI card.

 

Lest someone thinks I have it out for nVidia, my primray machine has two nVidia 7950s in it. I just wish nVidia would take its drivers more seriously. They were crummy on Windows XP. Some may recall how you'd see reports about how WindowBlinds or WindowFX "slowed my machine to a crawl" and called our software bloated not realizing that no, our stuff runs great if the drivers are decent (like the ATI ones or the occasional nVidia one).

Now, Microsoft is getting a taste of what we've gone through. Maybe this time, nVidia will be motivated to worry about desktop drivers rather than trying to show they can get an extra frame per sec in a first person shooter.

Still, if you have a good machine, nVidia or ATI, the CPU use isn't a big deal in actual use, especially if the .DREAM file's content is in MPEG and not WMV (WMV is more CPU intensive).  And hopefully soon nVidia's drivers will get better. Because I want to get an Geforce 8800 but am not going to get one until they fix this stuff.

 

19,213 views 33 replies
Reply #26 Top
I'm glad that helped! The drivers that come with Vista are LAME.
Reply #27 Top
Cough...fraps...cough...cough...

1) after 10 minutes this idea put into practice gets annoying

2) its really a useless feature until they integrate some form of knowledge into the background desktop (i.e. stock market ticker, sports scores etc)

3) it will put useless wear on your hardware 8% system resources or 80%

4) stardock "created" desktopx widgets for the things mentioned in my point #2

5) will for once in its career, this company admit 2 things?
5a) Some stuff stardock does is 100% innovative, 50% creative, and done 100% top notch
5b) Some stuff is stupid, completely ineffective, and a complete failure

If ever there was a time to toot one's own horn about M$ asking Stardock to help them on a project, this was NOT it...after...StarDock frequently cites the numerous flaws found within or without other OSes and software...be more and realize you gotta know when to hold em...know when to fold em...know when to walk away, know when to run!
Reply #28 Top

5b) Some stuff is stupid, completely ineffective, and a complete failure

More accurately,

5b) Some stuff is not to my liking and I have no predilection towards its use.

Then, with the gambling analogy.....horses for courses....

Reply #29 Top
JunkInternetMail,

Those of us who have found great wallpaper videos that we enjoy and play constantly believe your opinion is simply that...YOUR opinion.
Different strokes for different folks, mate.

Thanks for sharing.
Reply #30 Top
Look all I'm asking for in this world is better software the is fun and maybe even some that helps productivity...stardock clearly has created numerous examples of both. This just isn't good for productivity (yet) or very much fun. It's also highly impractical when frogboy says stardocks goal was to reach the average user because vista ultimate is NOT for the average user nor are the system requirements to run it, thus making his statement in a previous introductory article point out the glaring failure of this project. The one app I thought was truly innovative and written excellently, Stardock gave up on developing (Bootskin) Then they scrapped an app like nevertrust which if developed correctly could have dramatically changed the way we browse the internet...

Just once, I'd love to see where a stardock fanboy or even employee says, "We tried, failed, and we are sorry, we will work harder in the future." In my opinion the company is growing to fast and the evidence lies within products such as the .dream format.

If Stardock ever becomes publicly traded I'd love to see their goals and aspirations as a company because I'm confused as a customer...are you a software company or a content provider? You have excellently demonstrated your abilites to write SUPERB software such as CursorXP, Object Bar, DesktopX, and the THinkDesk Suite. Focus on what you're great at and leave the other stuff alone until the dust settles a bit.
Reply #31 Top

The one app I thought was truly innovative and written excellently, Stardock gave up on developing (Bootskin)

When confronted with the issue of proprietary BIOS-use Bootskin became fundamentally unviable as an on-going product.  As it stands it will fail to function on several computer configurations so thus remains discontinued.

I am neither employee nor 'fanboy' [I still contend the latter is a derogatory term as it implies blind, unthinking devotion], but can safely say that Stardock has NOT failed with regards to Bootskin [for one], but were beaten by system disparity.

Please, one and all, find a better collective descriptor than 'fanboy'. It rankles fast.

SUPPORTERS of the SD software, etc. are NOT reliant on gestalt to function but each one of them can form intelligent and independent thought/opinion.

Please respect them as fellow community members/site users....

Reply #32 Top
Yeah, because having a different screen upon (getting rarer every day) reboot is "good for productivity" and "fun". Puh-lease.
Reply #33 Top
The simple explanation is that an animated desktop is quite annoying, some of us actually do serious work on our computers you know.

Now an interesting screen-saver makes sense because you are clearly not using your computer at the time but when I'm concentrating hard on object life-cycles/access in a multi-threaded C++ program I need to be concentrating on what I'm doing!

Microsoft should be concentrating on improving the value of some of their core tools.