We are reviewing today VF2000 LED from Zalman, a hybrid cooler that fits video cards and CPUs, with a 92-mm fan and four heatpipes. Will it perform well on both tasks? Check it out!
VF2000 LED box is small, in cardboard paper. A transparent window allows you to see the cooler.
Inside the box we found the cooler, installation manual and hardware for CPUs and video cards, thermal compound and a FanMate 2 fan controller.
In Figure 3 you can have a general view of VF2000 LED.
In next pages we will see this cooler in detail.[nextpage title=”Zalman VF2000 LED”]
In Figure 4 you have a side view of the cooler. It has a very low profile, which is important for a VGA cooler.
In Figure 5, you can see it from the front side. Note that the four heatpipes are very close to each other at the base, but well spread at the heatsink.
In Figure 6 you can check the cooler back side. The heatsink fins are made of aluminum and the heatpipes are made of nickel-plated copper.
[nextpage title=”Zalman VF2000 LED (Cont’d)”]
In Figure 7 you have a top view of VF2000 LED. Its 92 mm fan is transparent and has blue LEDs, as the cooler name suggests. This fan has a three-pin connector and thus no automatic PWM speed control.
In Figure 8 you can check the base of the cooler, made of nickel-plated copper. It has a perfect mirrored look.
In Figure 9, you can see FanMate 2 fan controller (which also comes with VF900-Cu LED), its cable and three tubes of thermal compound that come with the cooler. Thanks, Zalman, for not saving on thermal compound quantity.
[nextpage title=”Installation on CPUs”]
VF2000 LED installation as a CPU cooler is relatively simple. First you must screw the clip set according to the CPU you have, Intel socket LGA775 (it does not support 1156 and 1366 sockets) or AMD sockets AM3, AM2+ and AM2, to the base of the cooler. Then install four screws to the clips and put the rubber washers on the screws, as you see in Figure 10.
After that, apply thermal compound, put the cooler on the CPU and screw four thumbnuts (with springs) on the solder side of the motherboard.
In Figure 12, you can see the cooler installed on our motherboard. It has a very low profile and will surely fit any case, even slim and SFF models.
In Figure 13, you can see the cooler installed in our case.
[nextpage title=”Installation on Video Cards”]
Installing VF2000 LED on a video card is as simple as on a CPU. You just need to attach four screws (not the same ones used for CPU installation) directly on the base of the cooler, put the cooler in place and attach it with the same thumbnuts.
We instaled the cooler on our PNY Verto GeForce 9600 GT video card. VF2000 LED comes with heatsinks for memory chips, which we did not install. The product is harmonious- and cool-looking, but has the problem of not blowing hot air outside the case. In Figures 14, 15, 16 and 17, you can see the cooler installed on our video card.
We try to install it on our GeForce GTS 250, but the position of the screws on VF200 LED was not compatible with the holes from this video card. So, we tested it only on our GeForce 9600 GT.
[nextpage title=”How We Tested”]
We are adopting the following methodology for our CPU cooler reviews.
First, we chose the CPU with the highest TDP (Thermal Design Power) we had available, a Core 2 Extreme QX6850, which has a 130 W TDP. The choice for a CPU with a high TDP is obvious. To measure the efficiency of the tested cooler, we need a processor that gets very hot. This CPU works by default at 3.0 GHz, but we overclocked it to 3.33 GHz, in order to heat it as much as possible.
We took noise and temperature measurements with the CPU idle and under full load. In order to achieve 100% CPU load on the four processing cores we ran Prime95 with the "In-place Large FFTs" option, and three instances of the StressCPU program, all at the same time.
We also compared the reviewed cooler to the Intel stock cooler (with copper base), which comes with the processor we used, and also with some other coolers we have tested using the same methodology.
Temperature measurements were taken with a digital thermometer, with the sensor touching the base of the cooler, and also with the core temperature reading (given by the CPU thermal sensor) from the from the SpeedFan program, using an arithmetic average of the four core temperature readings.
The sound pressure level (SPL) was measured with a digital noise meter, with its sensor placed 4" (10 cm) from the fan. We turned off the video board cooler so it wouldn’t interfere with the results, but this measurement is only for comparative purposes, because a precise SPL measurement needs to be made inside an acoustically insulated room with no other noise sources, which is not the case here.
- Processor: Core 2 Extreme QX6850
- Motherboard: Gigabyte EP45-UD3L
- Memory: 4 GB G.Skill F2-6400CL5S-2GBNY (DDR2-800/PC2-6400 with 5-5-5-15 timings), configured at 800 MHz
- Hard drive: 1 TB Seagate Barracuda 7200.12 (ST31000528AS, SATA-300, 7200 rpm, 32 MB buffer)
- Video card: PNY Verto Geforce 9600 GT
- Video resolution: 1680×1050
- Video monitor: Samsung Syncmaster 2232BW Plus
- Power supply required: Seventeam ST-350BKV
- Case: 3RSystem K100
- Windows XP Professional SP3
We adopted a 2 °C error margin, i.e., temperature differences below 2 °C are considered irrelevant.
[nextpage title=”Our Tests”]
On the tables below you can see our results. We ran the same tests with the coolers shown on below tables. Each test ran with the CPU idle and then with the CPU fully loaded. On BigTyp 14Pro, TMG IA1, NH-U12P and ISGC-300 the tests were done with the fan at full speed and at minimum speed. The other coolers were connected directly to the motherboard and it controls the fan speed based on CPU load level and temperature on PWM models. ISGC-400, iCEAGE Prima Boss, Megahalems Rev. B, Thermaltake SpinQ VT, Zalman CNPS10X Flex, Tuniq Tower 120 Extreme, Tuniq Propeller 120 and Zalman VF2000 LED were tested at minimum speed on idle test and at maximum speed on full load test.
|Cooler||Room Temp.||Noise||Fan Speed||Base Temp.||Core Temp.|
|Intel stock||14 °C||44 dBA||1000 rpm||31 °C||42 °C|
|BigTyp 14Pro (min)||17 °C||47 dBA||880 rpm||29 °C||36 °C|
|BigTyp 14Pro (max)||17 °C||59 dBA||1500 rpm||26 °C||34 °C|
|Akasa Nero||18 °C||41 dBA||500 rpm||26 °C||35 °C|
|Cooler Master V10||14 °C||44 dBA||1200 rpm||21 °C||26 °C|
|TMG IA1 (max)||16 °C||47 dBA||1500 rpm||22 °C||30 °C|
|TMG IA1 (min)||16 °C||57 dBA||2250 rpm||21 °C||30 °C|
|Zalman CNPS10X Extreme||16 °C||44 dBA||1200 rpm||21 °C||29 °C|
|Thermaltake ISGC-100||18 °C||44 dBA||1450 rpm||35 °C||49 °C|
|Noctua NH-U12P (low)||15 °C||42 dBA||1000 rpm||20 °C||30 °C|
|Noctua NH-U12P||15 °C||46 dBA||1400 rpm||20 °C||28 °C|
|Noctua NH-C12P||17 °C||46 dBA||1400 rpm||23 °C||28 °C|
|Thermaltake ISGC-200||21 °C||43 dBA||1100 rpm||31 °C||35 °C|
|Schythe Kabuto||22 °C||42 dBA||800 rpm||29 °C||34 °C|
|Arctic Cooling Alpine 11 Pro||20 °C||43 dBA||1500 rpm||32 °C||39 °C|
|ISGC-300 (min)||18 °C||42 dBA||800 rpm||26 °C||30 °C|
|ISGC-300 (max)||18 °C||46 dBA||1400 rpm||24 °C||26 °C|
|SilverStone NT06-E||21 °C||66 dBA||2600 rpm||30 °C||41 °C|
|Zalman CNPS9700 NT||22 °C||48 dBA||1700 rpm||28 °C||35 °C|
|Scythe Mugen-2||17 °C||41 dBA||700 rpm||25 °C||30 °C|
|ISGC-400 (min)||17 °C||44 dBA||850 rpm||24 °C||30 °C|
|Cooler Master Vortex 752||20 °C||48 dBA||1700 rpm||32 °C||44 °C|
|iCEAGE Prima Boss (min)||22 °C||42 dBA||1000 rpm||29 °C||36 °C|
|Evercool Buffalo||17 °C||51 dBA||1850 rpm||22 °C||29 °C|
|Scythe Big Shuriken||20 °C||42 dBA||900 rpm||31 °C||39 °C|
|Cooler Master Hyper TX3||21 °C||44 dBA||1700 rpm||30 °C||39 °C|
|Titan Skalli||20 °C||43 dBA||1200 rpm||27 °C||34 °C|
|Prolimatech Megahalems Rev. B||21 °C||40 dBA||800 rpm||28 °C||32 °C|
|Zalman CNPS9900 NT||23 °C||45 dBA||900 rpm||30 °C||34 °C|
|Cooler Master Hyper N620||21 °C||44 dBA||1200 rpm||28 °C||34 °C|
|Nexus LOW-7000 R2||23 °C||46 dBA||1400 rpm||33 °C||42 °C|
|Evercool HPK-10025EA||20 °C||54 dBA||1900 rpm||27 °C||34 °C|
|Evercool HPH-9525EA||23 °C||50 dBA||1900 rpm||38 °C||49 °C|
|iCEAGE Prima Boss II||23 °C||42 dBA||1000 rpm||29 °C||35 °C|
|Thermaltake SpinQ VT||24 °C||45 dBA||950 rpm||32 °C||39 °C|
|Titan Fenrir||21 °C||42 dBA||950 rpm||29 °C||35 °C|
|Zalman CNPS 10 Flex||23 °C||40 dBA||800 rpm||32 °C||39 °C|
|Tuniq Tower 120 Extreme||24 °C||43 dBA||1100 rpm||30 °C||37 °C|
|Gelid Tranquillo||22 °C||41 dBA||850 rpm||29 °C||36 °C|
|Cooler Master Hyper 212 Plus||20 °C||45 dBA||1200 rpm||27 °C||35 °C|
|Spire TherMax Eclipse||20 °C||58 dBA||2300 rpm||25 °C||34 °C|
|Tuniq Propeller 120||20 °C||43 dBA||1050 rpm||24 °C||33 °C|
|Nexus VCT-9000||20 °C||44 dBA||600 rpm||28 °C||37 °C|
|Coolink Corator DS||19 °C||45 dBA||1050 rpm||25 °C||32 °C|
|CoolIT ECO||17 °C||43 dBA||900 rpm||–||32 °C|
|Zalman VF2000 LED||17 °C||43 dBA||1300 rpm||28 °C||36 °C|
CPU Fully Loaded
|Cooler||Room Temp.||Noise||Fan Speed||Base Temp.||Core Temp.|
|Intel stock||14 °C||48 dBA||1740 rpm||42 °C||100 °C|
|BigTyp 14Pro (min)||17 °C||47 dBA||880 rpm||43 °C||77 °C|
|BigTyp 14Pro (max)||17 °C||59 dBA||1500 rpm||35 °C||70 °C|
|Akasa Nero||18 °C||48 dBA||1500 rpm||34 °C||68 °C|
|Cooler Master V10||14 °C||54 dBA||1900 rpm||24 °C||52 °C|
|TMG IA1 (max)||16 °C||47 dBA||1500 rpm||27 °C||63 °C|
|TMG IA1 (min)||16 °C||57 dBA||2250 rpm||25 °C||60 °C|
|Zalman CNPS10X Extreme||16 °C||51 dBA||1900 rpm||24 °C||50 °C|
|Thermaltake ISG-100||18 °C||50 dBA||1800 rpm||58 °C||93 °C|
|Noctua NH-U12P (low)||15 °C||42 dBA||1000 rpm||28 °C||59 °C|
|Noctua NH-U12P||15 °C||46 dBA||1400 rpm||25 °C||54 °C|
|Noctua NH-C12P||17 °C||46 dBA||1400 rpm||37 °C||76 °C|
|Thermaltake ISGC-200||21 °C||48 dBA||1900 rpm||42 °C||68 °C|
|Scythe Kabuto||22 °C||47 dBA||1200 rpm||38 °C||63 °C|
|Arctic Cooling Alpine 11 Pro||20 °C||51 dBA||2300 rpm||49 °C||85 °C|
|ISGC-300 (min)||18 °C||42 dBA||800 rpm||36 °C||64 °C|
|18 °C||46 dBA||1400 rpm||31 °C||56 °C|
|SilverStone NT06-E||21 °C||66 dBA||2600 rpm||39 °C||96 °C|
|Zalman CNPS9700 NT||22 °C||56 dBA||2600 rpm||34 °C||63 °C|
|Scythe Mugen-2||17 °C||46 dBA||1300 rpm||28 °C||54 °C|
|ISGC-400 (max)||17 °C||47 dBA||1400 rpm||36 °C||69 °C|
|Cooler Master Vortex 752||20 °C||55 dBA||2300 rpm||48 °C||92 °C|
|iCEAGE Prima Boss (max)||22 °C||53 dBA||2000 rpm||35 °C||59 °C|
|Evercool Buffalo||17 °C||51 dBA||1850 rpm||32 °C||67 °C|
|Scythe Big Shuriken||20 °C||50 dBA||1500 rpm||51 °C||85 °C|
|Cooler Master Hyper TX3||21 °C||53 dBA||2700 rpm||39 °C||66 °C|
|Titan Skalli||20 °C||47 dBA||1550 rpm||37 °C||69 °C|
|Prolimatech Megahalems Rev. B||21 °C||61 dBA||2600 rpm||30 °C||51 °C|
|Zalman CNPS9900 NT||23 °C||56 dBA||2000 rpm||34 °C||54 °C|
|Cooler Master Hyper N620||21 °C||50 dBA||1650 rpm||32 °C||56 °C|
|Nexus LOW-7000 R2||23 °C||53 dBA||1900 rpm||45 °C||74 °C|
|Evercool HPK-10025EA||20 °C||54 dBA||1900 rpm||39 °C||69 °C|
|Evercool HPH-9525EA||23 °C||50 dBA||1900 rpm||58 °C||100 °C|
|iCEAGE Prima Boss II||23 °C||56 dBA||2100 rpm||32 °C||56 °C|
|Thermaltake SpinQ VT||24 °C||52 dBA||1500 rpm||40 °C||68 °C|
|Titan Fenrir||21 °C||50 dBA||1600 rpm||33 °C||58 °C|
|Zalman CNPS 10 Flex||23 °C||61 dBA||2600 rpm||33 °C||59 °C|
|Tuniq Tower 120 Extreme||24 °C||56 dBA||1900 rpm||35 °C||60 °C|
|Gelid Tranquillo||22 °C||46 dBA||1450 rpm||31 °C||60 °C|
|Cooler Master Hyper 212 Plus||20 °C||52 dBA||1900 rpm||32 °C||64 °C|
|Spire TherMax Eclipse||20 °C||58 dBA||2300 rpm||29 °C||73 °C|
|Tuniq Propeller 120||20 °C||55 dBA||1900 rpm||36 °C||68 °C|
|Nexus VCT-9000||20 °C||50 dBA||850 rpm||43 °C||88 °C|
|Coolink Corator DS||19 °C||56 dBA||1800 rpm||32 °C||62 °C|
|CoolIT ECO||17 °C||54 dBA||1850 rpm||–||62 °C|
|Zalman NF2000 LED||17 °C||51 dBA||2200 rpm||43 °C||97 °C|
The next graph shows how many degrees Celsius the CPU core was hotter than room temperature during our idle tests.
The next graph gives you an idea on how many degrees Celsius the CPU core was hotter than room temperature during our full load tests.
[nextpage title=”Our Tests (Cont’d)”]
We ran some simple tests to check the performance of Zalman VF2000 LED, measuring the GPU core temperature with the aid of the SpeedFan software, and sound pressure level (SPL) with a digital noise meter set 4" (10 cm) from the video card, with the GPU at full load running folding@Home GPU client. For this measurement, we turned off the CPU cooler, so its noise wouldn’t interfere. Please keep in mind that sound pressure level measurement is just for comparative purposes, because a precise measurement would have to be done in an acoustically insulated enviroment, which we don’t have.
We compared the results from VF2000 LED with the VGA stock cooler, with an Arctic Cooling NV Silencer 5 VGA cooler, a Zalman VF900-Cu LED cooler at minimum and maximum speeds and with an Evercool Turbo2 cooler, all installed on our GeForce 9600 GT. You can check the results in the table below.
|Cooler||Noise||Room Temp.||Core Temp.||Temp. Diff.||Speed|
|Stock Cooler||56 dBA||21 °C||53 °C||
|NV Silencer 5||52 dBA||21 °C||48 °C||27 °C||–|
|VF900-Cu LED (min.)||43 dBA||21 °C||46 °C||25 °C||1300 rpm|
|VF900-Cu LED (max.)||49 dBA||21 °C||40 °C||19 °C||2400 rpm|
|Evercool Turbo 2||41 dBA||18 °C||50 °C||32 °C||–|
|VF2000 LED (min.)||43 dBA||17 °C||35 °C||18 °C||1300 rpm|
|VF2000 LED (max.)||51 dBA||17 °C||35 °C||18 °C||2400 rpm|
On the graph below you can compare the temperature difference between the GPU core and the room. Remember that, the lower the value, the better performance is.
[nextpage title=”Main Specifications”]
Zalman VF2000 LED main features are:
- Application: Socket LGA775, AM3, AM2+ and AM2 processors, compatible video cards.
- Fins: Aluminum.
- Base: Copper.
- Heat-pipes: Four copper heat-pipes.
- Fan: 92 mm.
- Nominal fan speed: 2,350 rpm.
- Fan air flow: Not informed.
- Maximum power consumption: Not informed.
- Nominal noise level: 29 dBA.
- Weight: 10.23 oz (290 g).
- More information: https://www.zalman.com
- Average price in the US*: USD 46.00
* Researched at www.quietpcusa.com on the day we published this review.
Zalman really innovated with a hybrid cooler model, which fits video cards and CPUs. The problem was that as a CPU cooler it achieved a very poor performance.
Because of its great looks and excellent construction quality, at first we thought it would perform better that other low-profile CPU coolers we’ve tested, but unfortunately this proved to not be tr
ue. It performed worst than all other CPU coolers we tested to date at full load (except for the Intel stock cooler). At least it has a low noise level.
As a VGA cooler the situation is a little different. It proved to be better than VF900-Cu LED, and kept the GPU at the same temperature independently if we set the fan at low or at high speed, showing that the product was not working at its limit. The product, however, is limited in terms of installation options: unlike VF900-Cu LED, which has a lot of installation options, VF2000 LED has only one possible screws position, which means it will not fit video cards with holes set on a different position. Because of this, we could not install it on our GeForce GTS 250. Other issue is that it takes two slots (plus the video card slot), which can be a problem if you use these two slots with expansion cards or you intend to build SLI or CrossFireX arrays.
Considering Zalman VF2000 LED is not inexpensive, has a poor CPU cooling performance and has a limited installation configuration for video cards, there is no way we can recommend it.