3Digest MOBILITY: January 2007
|
[an error occurred while processing this directive]
Monthly drivers report and popular 3D
accelerators comparison
Monthy 3Digest is the project of iXBT.com and Digit-Life.com dedicated to informing you about the performance of a large number of graphics cards under Windows XP in the past month.
It's our 80th issue of 3Digest.
January 2007
Starting this year we begin benchmarking notebooks with rather powerful Middle-End or even Hi-End graphics.
We clearly understand the impossibility of benchmarking a notebook graphics card separately from a notebook. We are bound to the given notebook configuration, which may differ radically in terms of RAM and CPU. And even the core logic plays its role. Nevertheless, more and more users are willing to know the difference between various notebook graphics cards, market share of which grows each year.
So, how do we compare devices with very different system resources and no clear GPU/memory clock rate regulations (since each vendor clocks its notebooks graphics as he wishes.)
Thus we had no choice, but to go the "hard way". We will be gathering statistics for different notebooks from the angle of their graphics subsystems. Therefore in our digest we'll be benchmarking resolutions from 1024x768 to 1600x1200. Of course, these are rather relative, since many notebooks have wide displays, but there's huge variety of matrices and sizes, many having their peak resolutions. We believe that even the highest-end mobile machine with the most powerful graphics can fail even at 1600x1200. Do not forget that while having the same markings as desktop solutions, notebooks graphics can have pared-down clock rates, pipelines, etc.
As our statistics database grows, we'll be publishing newer issues of 3Digest MOBILITY with updated charts and ratings.
We are open to your opinions, creative criticism and suggestions. We would like this material to be interesting and wanted. We must also say it straight: questions about the absence of certain notebooks in our ratings will be rejected. We usually can't select models to test, so we get what we're offered.
CONTENTS
- Testbed configurations and test software
- Tested videocards list
- What's new in 3Digest
- Test summary diagrams and digest bottom line
- 3D accelerator rating calculations for the end of the month
- Screenshot gallery
- Game Quality Section
- Game Quality Control Section
1. Testbed configuration
Of course, each notebook is a testbed of a kind, so there's no unified testbed configuration.
All notebooks have Windows XP Professional Service Pack 2, DirectX 9.0c installed. 22" Mitsubishi monitor was used. (22").
The list of benchmarks is here
2. Notebook list
Below are links to pages for each of the benchmarked notebooks.
Each page contains brief information about the notebook and, most
of all, its graphics subsystem.
Speaking of quality values, they are identical to those of desktop
graphic cards. So, you can see the quality of 3D produced
by tested cards in our monthly updated Screenshot
gallery.
Card list:
- ASUS
W2PC (RADEON X1700 Mobility)
- Fujitsu-Siemens
AMILO Xi1546 (RADEON X1800 Mobility)
- RoverBook
Voyager V700 WP (GeForce 7900 Go)
- Roverbook
Nautilus W790 (GeForce 7900 Go SLI)
Archive
(video cards, information on which is not updated anymore)
3. January 2007 news and current matters
- 3.1. The following drivers were used:
- NVIDIA ForceWare for all NVIDIA-based cards:
- ATI Catalyst for all ATI-based cards:
4. Summary charts of notebook graphics performance for January 2007
All willing to see the complete results in Excel 2003 format can
download this RAR
3.0 or this ZIP
archive.
Before you start browsing charts, we would like you to revise the
differences between desktop and mobile graphics with the same markings.
You should remember that while mobile GPUs might be named the same
as their discrete desktop modifications, they are ALWAYS pared-down
in something. For example, a usual Radeon X1800 has 16 rendering pipelinesm
while its mobile variant has only 12. And even if you find desktop
and mobile graphics identical in architecture and pipelines, the latter
would always be underclocked.
So if you see a notebook featuring, say, ATI Mobility X1800, you
shouldn't expect an X1800 XT. At best, it may be an underlocked X1800
XL. As a rule, mobile graphics performance ranges between similar
desktop solutions and desktop cards of a lower level. E.g., X1800
Mobility performance will be between those of X1800 XT and X1600 XT.
Sometimes, it may just be a mere marketing trick. You won't find X1700
Mobility in desktop series, though it's architecture is identical
to that of X1600. The performance can be even lower though due to
lower clock rates.
And another thing: while all desktop cards (with certain exceptions)
have a clear clock rate regulations, the clock rates of mobile GPUs
and memory vary according to vendor needs. Some may prefer lower energy
consumption and heat emission and reduce clock rates. And some may
wish superb performance at the expense of battery life and cooling
system noise.
So, if you see GPU test results for a certain notebook, do not consider
them correct for all notebooks with similar graphics. This is another
difficulty of comparing mobile graphics. Even the same name, like
GeForce 7900 Go, won't guarantee same performance in every one notebook.
- 4.1. Far Cry, Research (No HDR)
- 4.2. Far Cry, Research (HDR)
- 4.3. Splinter Cell Chaos Theory (No HDR)
- 4.4. Splinter Cell Chaos Theory (HDR)
- 4.5. Call Of Juarez
- 4.6. Prey
- 4.7. 3DMark05 MARKS
- 4.8. Company Of Heroes
- 4.9. F.E.A.R.
- 4.10. Serious Sam II (No HDR)
- 4.11. Serious Sam II (HDR)
- 4.12. 3DMark 06, Shader Model 2.0 MARKS
- 4.13. 3DMark 06, Shader Model 3.0 MARKS
- 4.9. Archive
(information from the previous 3Digests absent in this issue)
- 4.15 Notes on graphics subsystems:
- If a particular card is not in the summary, i.e. its results are either 1, or 100, it means the given benchmark hasn't work or has worked incorrectly with this card.
- Obviously, the notebook based on GeForce 7900 Go SLI showed the highest performance, which was natural, judging by its features. It's persistently followed by the single GeForce 7900 Go. But it's hard to make any ambiguos conclusions about the remaining two machines, since they perform nearly on par. While X1800 should be faster than X1700, different clock rates, processors and memory capacity nearly negated the advantage of X1800 Mobility.
5. Notebook 3D Graphics Rating
The rating was calculated using the gathered gaming benchmark results.
All willing to conduct their own rating calculations with own values are invited to download this Excel XP table in RAR 3.0 or ZIP format.
The method of rating calculation is described here.
Calculations were conducted considering the following assumptions:
- Percentage of performance and quality demand ratio:
- performance priority - 60%
- quality priority - 40%
- Only the HIGHEST resolution available in games was used to calculate prospect ratings.
- The usability rating indicates performance and features of a card (numerator) related to its price (denominator).
WE UNDERSTAND THAT PERFORMANCE IS HEAVILY AFFECTED BY SYSTEM RESOURCES, OTHER THAN GRAPHICS CARD. THEREFORE THESE RATINGS ARE RATHER RELATIVE.
"Usability" Rating:
GENERAL usability rating
| 01. |
RoverBook Nautilus W790 |
577
|
1177
|
66100
|
| 02. |
Fujitsu-Siemens AMILO Xi1546 |
518
|
923
|
47700
|
| 03. |
RoverBook Voyager V700 WP |
418
|
917
|
56700
|
| 04. |
ASUS W2Pc |
320
|
912
|
75000
|
"Prospect" Rating:
GENERAL prospect rating
| 01. |
RoverBook Nautilus W790 |
577
|
1177
|
66100
|
| 02. |
Fujitsu-Siemens AMILO Xi1546 |
518
|
923
|
47700
|
| 03. |
RoverBook Voyager V700 WP |
418
|
917
|
56700
|
| 04. |
ASUS W2Pc |
320
|
912
|
75000
|
Ingame quality
6. Games screenshot gallery
All test screenshots are compiled into a gallery where you can
qualify the work of a chosen videocard. All shots were taken at
the same 1024x768 resolution at the maximum quality. Screenshot
gallery
7. Game Artefacts Gallery
Game
Quality Section
8. Game Quality Control Section
Game
Quality Control Section
We thank all the companies that provided us with graphics cards and other equipment for our 3Digest, including:
Foxconn Russia,
MSI Russia,
Gainward,
HIS (and personally Peter Yueng)
NVIDIA Russia,
ATI Technologies Russia,
Sapphire Technology Russia,
ASUStek Russia,
Gigabyte Russia,
AOpen Russia,
Bench'emAll! (and personally Alexander Kondratyuk AKA Render for helping us optimize bechmarking).
The PSU for the testbed was kindly provided by HIPER
|
|
The Dell 3007WFP monitor was kindly provided by NVIDIA
|
06.01.2007
|
|
 |
|
|
|
