Navigate:

Advice
Beginners
BIOS Guide
CPUs
Links
Mainboards
Memory
Network
Storage
Video/Sound Cards
Contact
Forum
SiteMap
Sponsors
WebNews
Home
 . .

Prices:

Mainboards

ABIT
ASUS
Chaintech
Shuttle
Soyo
Tyan

CPU
Intel
P4 2.4C-800
P4 2.6C-800
P4 2.8C-800
P4 3.0-800
P4 3.2-800

AMD
AthlonXP
XP 1700+
XP 2000+
XP 2400+
XP 2500+
XP 2700+
XP 3000+
XP 3200+

Athlon64
Athlon64 3200+
Athlon64 FX-51

Opteron
Opteron 240
Opteron 242
Opteron 244
Opteron 246

Memory

Corsair
Crucial
Kingston
Mushkin
OCZ

What are you
shopping for?

































































































































































LOSTCIRCUITS
SHORTCUTS:
The Market
Technical Hi-Lights
Test Configuration and Benchmark Overview
3dsmax 8.0
Maya 7.0 (Phoenix)
Springmark
SPEC ViewPerf
Final Words
Update: Gaming Performance

Give us some feedback

 nVidia Quadro FX1400
The Midrange Solution for Almost Anything
(Review by MS, June 13, 2006)

Springmark

Springmark renders a spring with 1 million vertices / loop. While it is visually not too exciting, the variable model size (integer numbers of loops) allows to scale the model to either fit into the LFB or else pushing it out into the system memory, or, for those that are really going for the extreme - beyond the system memory to force swapfile access. Each loop occupies 36 MB of memory, that is, two loops will tie up 72 MB, three loops fill 108 MB and so on. In most cases, we see a natural decline in frames per second that is inverse to the model size up to the point where the model no longer fits into the on-board graphics memory. If we plot processed vertices/sec, the results are in most cases a constant value up to the point where the model gets larger than the locl memory space.

.

.

One of the features of essentially all modern graphics cards is the capability of dynamically partition the on-board memory to either geometry or else to texture storage. As a result of this flexible allocation space, almost the entire memory space can be filled up with either set of data, which is in contrast to older (consumer) architectures using a fixed (50/50) partitioning of the local frame buffer. The graph for the FX3000 clearly shows this effect, up to 180 MB model size, no degradation in performance occurs (256 MB LFB). In the case of the FX1400 with its 128 MB LFB, we would have expected a similar linear relation up to a model size of 108 MB, however, as it is quite obvious especially from the lower graph, even a 72 MB model seems to exceed the geometry allocation of the onboard memory. We discussed this issue with nVidia and were told that the flexible allocation is handled now on the level of the drivers rather than on the BIOS or even hardware level. However, we did not get any explanation for the behavior of the FX1400 as we found it, other than that "Springmark is pathologic for artifacts of all sorts". Whatever it is, appears to be isolated to the FX1400, though.

nVidia Quadro FX 1400

Next Page:    => SPEC ViewPERF 8.1 =>

If you enjoyed reading this article and found it useful, please consider making a small donation to LostCircuits.
Thank you!
General disclaimer: This page only reflects the author's personal opinion and assumes no responsibility whatsoever regarding any of the contents or any damages that may occur explicitly or implicitly from reading the contents of this site. All names and trademarks mentioned in this review are the exclusive property of the respective parent companies.
All contents of this site are protected by international copyright laws. Reproduction of the contents even in parts is not allowed except after written permission by the author and referral to this site.
Copyright 1998 - 2008 LostCircuits