On 10 Aug 2006, at 8:33 PM, Tyler Turner wrote:
Radeon 9000. Both are mobility versions.
Huh? I thought you had desktop computers. Are you saying you have
mobility (i.e. laptop) GPU's installed in your desktop machines?
Do they affect the visual quality of
the display in
any way?
From what I've read, they do.
I'm asking you what you *see* on your screen. Let me put it this way:
if you took a screenshot of a Finale document with the slider all the
way up, and another screenshot with the slider all the way down, is
there any difference between the two?
You're really missing a big point here. The controls
in this dialog are tailor made by the video card
manufacturer for the specific card.
Are you sure? For Macs with retail video cards, the ATI control
panels are the same across the entire family of cards. Isn't it
possible that the slider for the Radeon 9000 works the same way
across that family of Radeons?
It's a graphic card setting, and all it does is
provide a quick way to turn off sets of specific
features for the video card.
But what features are those? You still haven't said. What if turning
the slider all the way up enables some 3D-only feature like, say,
anisotropic filtering, that could cause a significant performance hit
without actually improving 2D image quality? Or -- and this seems
more likely with the situation you describe -- what if it enables
excessive screen caching, filling up the memory on your video card
and forcing the screen caches to be spooled out to RAM, or even
virtual memory? This is something that sometimes happens on Macs with
video cards with insufficient video RAM. (How much VRAM does your
Radeon 9000 have, anyway?)
If a video card doesn't support one of these features,
the feature wouldn't be in the dialog in the first
place.
It might if, for instance, nVidia gave you a generic slider for all
7XXX-series cards, or even all 7300-series cards, which (as we've
mentioned) have very different features and performance depending on
what flavor of 7300 you're using. nVidia and ATI usually release one
set of drivers for each family of cards, not each individual card.
But even if the slider only enables graphics features supported by
the specific card in your machine, that doesn't mean that the cards
-- especially the older cards you have, like the Radeon X600 (based
on the 9600 chipset, released 2003) and the 9000 (released 2002) --
can execute those features *well*. (Especially if you have, as you
say, the much less powerful Mobility versions of those cards!) My
concern is, if turning the slider up causes a performance hit without
actually affecting 2D visual quality, it may be that this slider is
inappropriately invoking a demanding 3D feature in a 2D environment.
You still haven't said what visual difference this
slider makes in
Finale (if any).
Actually I did, in one of my very first posts. I
stated that normal scrolling up and down was slower
with the slider disabled, but dragging the screen
around via right-click drag (which results in a much
more consistent set of redraws of the screen image) is
much slower with it turned up.
One more time: I'm not asking about PERFORMANCE. I'm asking about
pure image quality. When you're just looking at the Finale document
-- not scrolling or doing anything else, but just looking at the
screen -- is there any visual difference when the slider is up vs.
down?
Also -- I'm assuming you've installed the latest drivers for both
these cards, correct?
But anyway, it's absolutely unimportant for me to show
that I need to have that slider turned up in Finale. I
have to have that slider turned up for OTHER
applications that I run, including 3D applications, at
times concurrently with Finale.
I still don't understand why turning this slider up degrades Finale
performance (when dragging), but improves performance in other apps.
I hope you're not suggesting
that a person should have to change that slider
setting for working with various applications.
The fact that you need to turn this slider up for maximum performance
in some apps and turn it down for maximum performance in others is
not something I've ever heard anyone complain about before. It sure
sounds to me like something weird is going on. Again, it would help
if you could describe specifically what features are enabled or
disabled at various slider positions on your system. Surely this is
covered in the graphics card documentation somewhere?
If you can't produce some evidence that adjusting that
slider does something more than enable/disable video
card features that are implemented by the video
hardware, I think we're at a stand still. Are modern
video cards significantly more powerful than the ones
I'm running? Yes. But are they enough more powerful so
that even the cheaper ones can convert the 2 frames
per second I'm seeing at 1024X768 32-bit up to 40 or
more frames per second at 2048X1536? I'll believe it
when I see it.
Fair enough. But this all started when you were complaining the Mac
Pro's stock video card was not "reasonable," and that overall
performance (even in non-3D apps) would be better with just one (dual-
core) processor, but a better video card than the 7300 GT.
So, if you'll allow me to put aside all of our other disagreements
for the moment, I'd like to make a simple, narrow claim: I doubt a
more expensive video card would significantly improve the speed of
non-3D tasks in that specific computer.
In partial support of that view, I pointed out that the 7300 GT is
not an underpowered card. I linked to all sorts of reviews that
explain why the 7300 GT is the best budget card currently on the
market, and that it's based on the same chipset as the midrange 7600
GT. It has a decent amount of video RAM (256 MB). In the Mac Pro
benchmarks, the 7300 GT gets over 90 fps in UT2004 tests (versus 65
fps for a Quad G5 with GeForce 6600 -- which was a midrange card
when it came out.) One review mentioned that the 7300 GT was capable
of running Exposé smoothly even on the 30" Cinema display
(2560x1600). And given that the 7300 GT is capable of running very
demanding 3D games like Splinter Cell: Chaos Theory at a resolution
2048x1536 with an average fps of 18, I don't think there's any reason
to expect that it will choke on 2D Finale redraws.
[Source: http://www.ixbt.com/video2/g73-3-d.shtml#p17]
But, we shall see. Hopefully, as we begin to see more Mac Pro
benchmarks emerge, some additional light will be shed on this matter.
Cheers,
- Darcy
-----
[EMAIL PROTECTED]
http://secretsociety.typepad.com
Brooklyn, NY
_______________________________________________
Finale mailing list
[email protected]
http://lists.shsu.edu/mailman/listinfo/finale
