Does Lucid Virtu Work?
We detail how we tested the
efficacy of the Lucid Virtu software in our
Intel Z68 Test Setup page, but essentially we measured the performance of a purely Processor Graphics-based system and then uninstalled the PG driver and installed a GeForce GTX 570 1.3GB. After measuring the performance of this system, we re-installed the PG driver, installed the Lucid Virtu software and then ran the test suite a third time.
This convoluted procedure was necessary to gauge whether, when and to what extent the Virtu software was working.
Results Analysis
Our initial testing with a
GeForce GTX 590 3GB went disastrously. No matter how many naughty words we shouted at the system, or evil glares we gave it, the card just didn't work. Swapping the GTX 590 3GB for a GeForce GTX 570 1.3GB was a convoluted process, as we had to switch the display cable from the PG output to the card, then re-boot after the GeForce driver updated, then switch the cable back, then re-start once Virtu was happy and then disable the second monitor that had mysteriously enabled itself. Long winded doesn't do it justice.
It wasn't always a case of clear sailing after this, either. Despite Arma II being listed as a recognised game, Virtu still tried to run it on the PG. All we ever saw was a box made of three Arma-coloured lines. At least in 3DMark11 we saw no performance drop from using Virtu (there was a slight performance increase, but we can put this down to the vagaries of 3DMark 11 and solar flares). We saw no difference in performance from the slider in the Virtu interface either, whether we left it on default or slid it all the way to either Quality or Performance.
We were also disappointed to see that Virtu assigned the BadaBoom encode to the GTX 570 1.3GB, despite the PG being both faster and more power efficient. BadaBoom has a historical alignment with Nvidia, but we'd hoped that Virtu would choose more sensibly which graphics unit to use regardless of this. At least Virtu made the right choice when it came to video playback, opting for the more power-frugal PG rather than the wasteful GTX 570 1.3GB,
The most disappointing aspect of Virtu was that it actually produced a power draw increase when the PC was idle. With the GTX 570 1.3GB handling Windows (with Aero enabled), the PC only drew 95W from the wall, but the extra effort of firing up the PG increased the power draw by 7W.
We tried Virtu on two different motherboards from two different manufacturers and found the same results - it's clearly the software that's at fault.
-
With Nvidia GeForce GTX 570 1.3GB only
-
Lucid Virtu (PG + GTX 570 1.3GB)
0
1000
2000
3000
4000
5000
6000
3D Marks (P), higher is better
-
With Nvidia GeForce GTX 570 1.3GB only
-
Lucid Virtu (PG + GTX 570 1.3GB)
Frames per second (fps), higher is better
-
Intel Processor Graphics only
-
With Nvidia GeForce GTX 570 1.3GB only
-
Lucid Virtu (PG + GTX 570 1.3GB)
-
Intel Core i5-2600K only (Turbo Boost disabled)
0
25
50
75
100
125
150
175
Seconds, lower is better
-
Intel Processor Graphics only
-
With Nvidia GeForce GTX 570 1.3GB only
-
Lucid Virtu (PG + GTX 570 1.3GB)
Watts, lower is better
-
Intel Processor Graphics only
-
With Nvidia GeForce GTX 570 1.3GB only
-
Lucid Virtu (PG + GTX 570 1.3GB)
-
Intel Core i5-2500K only
Watts, lower is better
-
Intel Processor Graphics only
-
With Nvidia GeForce GTX 570 1.3GB only
-
Lucid Virtu (PG + GTX 570 1.3GB)
Watts, lower is better
Want to comment? Please log in.