Sapphire HD 4670 Ultimate 512MB Passive Video Card Review

[SKYMTL]

Dead Space

Dead Space

To benchmark Dead Space, we used a timed 10 minute gameplay session around in Chapter 2: Intensive Care on Hard Difficulty in which we battle a few necromorphs and enter a zero-G environment. Seriously, what good is a benchmark without action sequences? “Walkthroughs” are unacceptable around here… Once again FRAPS was used.

1440 X 900

​

1680 X 1050

​

 

[SKYMTL]

Left 4 Dead

Left 4 Dead

For benching Left 4 Dead, we used a pre-recorded 5 minute timedemo taken on the Death Toll campaign during The Church mission. Framerates were captured with FRAPS.

1440 X 900

1680 X 1050

 

[SKYMTL]

Need for Speed Undercover

Need for Speed Undercover

Benchmarking NFS: Undercover proved to be an interesting process but we finally decided on the Ocean and Wilson Circuit race for a worst case scenario benchmark run. The benchmark run involves running 4 laps of the circuit at the hardest difficulty setting with 7 other racers and a high traffic volume. The total amount of time usually ranged between 3:30 and 3:40 of gameplay time recorded in FRAPS.

1440 X 900

1680 X 1050

 

[SKYMTL]

Far Cry 2 (DX9)

Far Cry 2: DX9

Even though Far Cry 2 has its own built-in benchmarking tool with some flythroughs and “action scenes”, we decided to record our own timedemo consisting of about 7 minutes of game time. It involves everything from run-and-gun fights to fire effects.

1440 X 900

1680 X 1050

 

[SKYMTL]

Far Cry 2 (DX10)

Far Cry 2: DX10

1440 X 900

1680 X 1050

 

[SKYMTL]

Fallout 3

Fallout 3

For these Fallout 3 benchmarks we decided to use one of the more graphically intensive portions of the game: a gameplay session which starts at the exit from Vault 101, through Springvale, over a small hill towards Washington and then back towards Megaton. Within this session two ants were fought near Springvale. All in all, the gameplay time is about 6 minutes.

1440 X 900

1680 X 1050

 

[SKYMTL]

X3: Terran Conflict

X3: Terran Conflict

After playing through the game trying to find a suitable zone for benchmarking purposes, we finally gave up and began using the stand-alone benchmarking tool from Egosoft. Believe it or not, after over 20 hours playing X3: Terran Conflict, it seems the benchmarking tool combines the worst and best case scenarios of the game quite accurately. The minimum framerates you see here are the averages for the 4 sections of the benchmark.

1440 X 900

1680 X 1050

 

[SKYMTL]

Crysis: Warhead (DX9)

Crysis: Warhead DX9

To benchmark Warhead, we recorded a 5 minute timedemo on the Ice level which included ranged and hand to hand combat.

1440 X 900

1680 X 1050

 

[SKYMTL]

Crysis: Warhead DX10

Crysis: Warhead DX10

1440 X 900

1680 X 1050

 

[SKYMTL]

Heat & Acoustics / Power Consumption

Heat & Acoustics

We all know that the passive cooling offered by this card would all be for naught if temperatures climbed to extremes but somewhat surprisingly the Ultimate’s aluminum-fin setup proved to be an efficient heat killer. This was done on an open tech benching without much air movement other than an Antec Spotcool set to Medium placed about 9” away. These results actually impressed me considering I thought the small heatsink assembly would be simply overwhelmed by the heat from the core. I am happy to say that I was wrong and gladly give the Sapphire HD 4670 Ultimate my blessings as a perfectly adequate passively cooled card. These are truly excellent results.

In a HTPC environment, silence is king and in this regard the Sapphire HD 4670 Ultimate reigns supreme since you don’t need a high speed fan a-buzzin’ for it to stay cool. Altogether, Sapphire has a lot to be proud of in this regard.

Power Consumption

For this test we hooked up our power supply to a UPM power meter that will log the power consumption of the whole system twice every second. In order to stress the GPU as much as possible we once again use the Batch Render test in 3DMark06 and let it run for 30 minutes to determine the peak power consumption while letting the card sit at a stable Windows desktop for 30 minutes to determine the peak idle power consumption. We have also included several other tests as well.

Power consumption is well within the margin of error when compared to the stock card’s results. That being said, the HD 4670 is an extremely efficient card overall which is a definite benefit to any of those who are running this in a lower-powered HTPC.