Gigabyte GA-EP45-UD3P P45 Motherboard Review

[MAC]

Benchmarking Methodology

Benchmarking Methodology

For this review, our benchmarking section will focus solely on the Gigabyte EP45-UD3P, in stock and overclocked configuration. The reasoning behind this approach is that there is often a miniscule clock-for-clock performance difference between two motherboards that share an identical chipset and declaring one product a winner based on a 1% performance advantages seems futile to us.

We have outlined the two setups in the sample graph to the right. The blue results will indicate the performance at the BIOS defaults. The only changes that will be made are to the memory in order to get it running at its rated frequency and timings. Every other setting will be at the default that the BIOS sets. The red results are going to display the overall best overclock we found for this particular setup.

These overclock results are again, the best overall overclock we could manage with this motherboard and the hardware used in the review. As was mentioned in the overclocking section, we used Everest bandwidth benchmarks and a few other performance tests to determine which settings were ultimately faster, not just higher. For all of the benchmarks, appropriate lengths are taken to ensure an equal comparison through methodical setup, installation, and testing. The following outlines our testing methodology:

A) Windows is installed using a full format.

B) Intel Chipset drivers and accessory hardware drivers (audio, network, GPU) are installed followed by a defragment and a reboot.

C) Programs and games are then installed followed by another defragment.

D) Windows updates are then completed installing all available updates followed by a defragment.

E) Benchmarks are each ran three times after a clean reboot for every iteration of the benchmark unless otherwise stated, the results are then averaged.

We have listed the benchmark versions on each graph as results can vary between updates. That should about cover everything so let's see what kind of numbers this motherboard puts up in the suite of benchmarks we have chosen.

 

[MAC]

Memory Benchmarks

Memory Benchmarks

Everest Ultimate v4.50

Everest Ultimate is the most useful tool for any and all benchmarkers or overclockers. With the ability to pick up most voltage, temperature, and fan sensors on almost every motherboard available, Everest provides the ability to customize the outputs in a number of forms on your desktop. In addition to this, the memory benchmarking utility provides a useful tool of measuring the changes to your memory sub-system.

As you can see, our 50% memory overclock (800Mhz to 1200Mhz) contributed to an impressive increase in overall memory bandwidth. Having said that, the copy bandwidth is actually approximately 100MB/s slower than we would expect with these settings, but that is a mere 2% difference and we are confident that a future bios release will rectify this minor performance downfall.

Here we see a 40% drop in latency, which is downright excellent. This overclocked result is a mere 0.3ns worse than we achieved with the more mature EP45-DQ6, which proves that the EP45-UD3P is already a very strong performer.


ScienceMark v2.0

Although last updated almost 3 years ago, ScienceMark v2.0 remains a favorite for accurately calculating bandwidth on even the newest chipsets.

As in Everest, ScienceMark reveals a monumental 50% bandwidth increase going from the stock to overclocked settings. With this level of scaling, it is clear that the EP45-UD3P is efficient at turning the increased frequencies into raw megabytes per second.

 

[MAC]

System Benchmarks

System Benchmarks

SuperPi Mod v1.5

When running the SuperPI 32MB benchmark, we are calculating Pi to 32 million digits and timing the process. Obviously more CPU power helps in this intense calculation, but the memory sub-system also plays an important role, as does the operating system.

The increased CPU, front side bus, and memory clocks really have an opportunity to shine in SuperPI, reducing the overall time by almost 5 minutes. That is huge in a benchmark in which many spend countless hours trying to reduce their score by mere tenths of a second.


PCMark Vantage

The latest iteration of the popular system benchmark is PCMark Vantage from the Futuremark crew. The PCMark series has always been a great way to either test specific areas of a system or to get a general over view of how your system is performing. For our results, we simply run the basic benchmark suite which involves a wide range of tests on all of the sub-systems of the computer.

Although the difference in scores may seem minimal when compared to the massive gains that we have previously shown, PCMark Vantage tends to lean more towards real world performance and seeing a 20% gain is very respectable.


Cinebench R10

Another benchmarking community favorite, Cinebench renders an intense 2D scene relying on pure processing power. This is arguably one of the the most efficient programs tested today at utilizing all available cores. We will be running both the single-threaded and multi-threaded benchmarks.

Cinebench is an effective way of testing pure processing power, and the results demonstrate almost perfect scaling with the increase in CPU clocks.

DivX Converter v6

Now that we have ran some of the more 'synthetic' benchmarks, it is time for a real-life VOB to DivX encoding task. We will take a 1.08GB VOB rip of the cult-classic movie Full Metal Jacket and convert it into DivX using the default multi-media setting of DivX converter v6. DivX fully utilizes both cores of the processor and will rely heavily on all aspects of the system for performance.

Even with the advent of multi-core processors, video conversion is still one of the most time consuming endeavour that PC users are face with. As a result, being able to shave almost 8 minutes off a 40 minute task is a very tangible improvement.

Lame Front End

Unlike the DivX Converter, Lame Front End is not multi-threaded and only utilizes a single core. While this will obviously limit performance, we should still recognize significant time savings going from the stock to the overclocked settings. We will be encoding a WAV rip of Santana’s Supernatural album and converting it to MP3 using the VBR 0 quality preset.

Our overclocked system is RIAA's worst nightmare, able to rip and convert an entire CD album to high-quality MP3s in under 2.5 minutes, which is almost 25% faster than the stock configuration.


Photoshop CS3

For the image editing portion of this review, we will use Photoshop CS3 in coordination with Driver Heaven’s Photoshop Benchmark V2, which is an excellent test of CPU power and memory bandwidth. This is a scripted benchmark that individually applies 12 different filters to a 60MB JPEG, and uses Photoshop’s built-in timing feature to provide a result at each test stage. Then it’s simply a matter of adding up the 12 results to reach a final figure.

A 24% -- 18 second improvement between stock and overclocked configurations is very respectable and should convince all image-editing enthusiasts to overclock their systems.


WinRAR

The last of our real-life tests will be with the highly popular WinRAR v3.71, which has a built-in benchmark that can measure both single-threaded and multi-threaded archive extraction performance.

Along with being the first application that we install on a fresh Windows installation, WinRAR is a tool that we use countless times per day, so any performance improvement can provide some very tangible time savings. With the overclocked configuration, archive extraction was an impressive 38% faster than at stock. This improvement is much greater than our 26% CPU overclock, and this is because WinRAR extraction is heavily dependent on the memory sub-system, so our 50% memory overclock plays a significant role.

 

[MAC]

Gaming Benchmarks

Gaming Benchmarks

Futuremark 3DMark 06

We will continue to include the venerable 3DMark 06 because it is less GPU bound than Vantage, and thus a better indicator of overall system performance.

Based on the static SM2.0 and SM3.0 scores that we observed, and the relatively minor 5% performance improvement, it is clear that our reference-clocked GeForce 8800 GT 512MB is becoming a bottleneck at high CPU speeds.


Futuremark 3DMark Vantage

You asked for it, so we have finally included 3DMark Vantage, Futuremark’s latest release in their renowned line of 3D benchmarking software. This latest DX10-only 3DMark comes with a variety of presets, but for our tests will be use the standard Performance preset which is suitable for a much greater range of system specifications than the other more demanding presets.

Since 3DMark Vantage is such a GPU-intensive application, there was only a slight 4% performance difference between the stock and overclocked configurations. A higher-end graphics card would be better able to take advantage of the increased system clocks.

Crysis - Sphere Benchmark

While Crysis is intensely GPU dependent, we added it to our gaming benchmarks to see how system changes can improve in-game performance on a mid-level system. We utilized the Sphere level demo for our benchmarks, and ran it in DX10 mode with a resolution of 1680x1050 with all detail levels set to medium.

Our mid-level GeForce 8800GT 512MB is simply overmatched in Crysis. As a result of this bottleneck, the increased system clocks do not translate into noticeably improved gaming performance. However, the critical minimum frame rate has inched closer towards the baseline 30FPS level that is required to maintain a smooth(-ish) gaming experience.

Company of Heroes: Opposing Fronts

This test consists of actual gameplay using a single mission (Authie: Boudica's Boys from the British Campaign) since it holds a bit of everything the game has to offer: vehicle battles, artillery barrages and house-to-house fighting. We recorded framerates using FRAPS up until the 15 minute mark of the mission and excluding any in-game briefings / cutscenes. The game was run in DirectX 9 mode at 1680x1050 with all detail levels set to high.

Similar to Crysis, COH: OP is quite graphically intensive, but the increased system clocks do provide a noteworthy 9% performance boost over the stock configuration.


Team Fortress 2

As our last gaming benchmark, we will use the addictive and CPU-intensive Team Fortress 2. For this test, we made an action-packed 30-minute timedemo on the “2_Fort” map with a constant 20-24 player load. This test represents a worst-case scenario because it is a small map with a high number of players on the screen at all times, placing a significant load on the CPU. The resolution was set to 1680x1050 with all settings on high.

Team Fortress 2 is based on Valve's Source engine, which is well-known for having excellent CPU scalability, and this fact is quite evident in our results. Our overclocked configuration posted a 33% FPS gain, which suggests that the game not only benefits from the additional processing power, but also from the increase in front side bus and memory clocks. The fast-paced FPS experience is much more fluid due to the significant higher average frame rate and reduction in frame rate dips.

 

[MAC]

Voltage Regulation / Temperature Testing

Voltage Regulation

Our voltage regulation testing will focus on the various voltages and the differences encountered between what is selected in the BIOS, what is reported by EasyTune6 (when available), and what is reported by a digital multi-meter (DMM). We have found five voltage read points on the motherboard so the vCORE, PLL, CPU Termination/VTT, vNB, and vDIMM will be recorded with our DMM. Both vCORE and vNB will be read from the underside of the motherboard at the output side of the corresponding inductor for each voltage supply. PLL and VTT voltages will be read from points on the top of the motherboard. vDIMM will be read directly from a DIMM slot's VDD pin. The ground point used for all readings will be a screw hole. Here are a few photos showing the various read points used.

​

Now that we have established where the read points are, let’s have a look at the results. These measurements were taken at 9x400Mhz, the highest officially supported FSB, and the setting that will most likely be used by those seeking mild overclocks. Everything else in the BIOS is set to auto. Just to clarify, the vCore (LLC) section is the vCore results with Load-Line Calibration enabled. Here are our extensive findings:

As a whole, the EP45-UD3P has terrific voltage output. There is very little variance between what you select in the BIOS and what the board actually outputs. The DIMM, NB, PLL, and VTT voltages are very accurate and stable, showing no variances between idle and load states. There is some line droop (vDroop) on the vCORE when going from idle to load, but it is not substantial enough to negatively affect stability nor overclocking. Also, once Load-Line Calibration is enabled, the vDroop is totally eliminated. Since it is such a crucial setting, let's take a closer look at the vCORE's characteristics with a two hour OCCT stress test.

The vCore dips when the stress test actually kicks-in, but otherwise the vCore line is very solid and there were only miniscule spikes, none of which were worrisome. Evidently, GIGABYTE's engineers have done an excellent job with this model's six-phase power design.

With Loadline Calibration enabled the vCore line is straight as an arrow displaying zero variance even during load changes. This is very, very impressive and we are extremely pleased that GIGABYTE have finally implemented this worthwhile feature.

Temperature Testing

Now it's time to see if the EP45-UD3P's new cooling system is as good as it looks. Since this is a mainstream model motherboard it does not come with temperature sensors for the northbridge and southbridge. Therefore, we relied on our trusty digital thermometer for all temperature measurements. We set the system to its overclocked configuration, and then OCCT was run for two hours. The temperatures were recorded at twenty minute intervals throughout the two hour test and the results averaged out. The ambient temperature was 21°C/71.6°F.

As you can see, the northbridge cooler reached 43°C/109.4°F and the southbridge cooler measured 37°C/98.6°F. Both of these temperatures are quite good considering the 500Mhz FSB. The MOSFET coolers hovered around the 33°C/91.4°F mark, which is surprisingly cool. Could this be a result of Ultra Durable 3's claimed cooling advantages? It is quite possible. Remember, none of the above components benefited from any direct cooling, so all these figures essentially represent a worst-case scenario. In a regular case with one or two 120MM fans, temperatures should be even lower. As a result, we can definitely state that the EP45-UD3P's cooling system is excellent and should prove to be adequate for all but the most extreme overclocking efforts.

 

[MAC]

Installation

Installation

For those of you who are interested in seeing how exactly the components fit on the motherboard, we have included a new Installation Section which should give you a general idea of the clearance levels on the EP45-UD3P.

As you can see, the northbridge and MOSFET coolers do not interfere with the Thermalright Ultra-120 heatsink and should not have any compatibility issues with most popular CPU cooler. Having said that, we have seen some third-party pictures indicating that the mounting bracket on Tuniq Tower 120 comes into contact with the northbridge cooler, so keep that in mind.​

The memory slots are spaced far enough away from the CPU socket so that even very wide heatsinks should not hang over the DIMM slots nor come into contact with tall heatspreaders. Likewise, the memory slot clips do not come into contact with the back of the graphics card.​

These two images demonstrate how GIGABYTE's engineers have finally managed to ensure that graphics cards will not interfere or block SATA ports...by putting the ports BEHIND the graphics cards, pure genius! All SATA ports are accessible no matter what graphics card configuration you choose.​

We only encountered to minor issues during our installation. First, as mentioned previously, a single-slot graphics card installed in the second PCI-E x16 slot can potentially block the bottom PCI slot, while a dual-slot model will definitely block the slot. Secondly, depending on your choice of CPU cooler and the direction that you install it in, there can be a clearance issue between the 8-pin CPU power connector and the CPU fan. The simple solution is to plug in the power connector before installating the fan and perhaps bend the cable a bit.​

​

Lastly, those who use a CPU cooler that requires a back plate will be glad to know that we did not encounter any clearance issues when installating our Thermalright Ultra-120.​

 

[MAC]

Conclusion

Conclusion

In the ultra-competitive mainstream motherboard market, Gigabyte's challenge was to create a new P45 model that would stand above the rest. Have they succeed? Well let's compare the EP45-UD3P to its closest competitor, the ASUS P5Q PRO. Both have two mechanical PCI-E x16 2.0 slots, both have 8 SATA II ports, both have similar RAID functionality, both have eSATA and FireWire ports, both have 8-channel HD audio, both have advanced energy saving technologies...and we could go on and on. However, while they may initially appear quite evenly matched, Gigabyte's newest model distances itself with its distinctly superior layout, better BIOS, greater overclocking capabilities, a better cooling system, two PCI-E Gigabit LAN controllers, integrated TPM data encryption chip, and EasyTune6. As a result, we believe that it is fair to state that Gigabyte really have created a motherboard that belongs at the top of Mainstream Mountain.

The layout in itself is a tremendous achievement. Almost everything has been placed exactly where it should be to maximize usability and minimize compatibility issues, namely on the edge of the board. In particular, we are glad to see that no SATA ports need to be sacrificed when using a CrossFire configuration, which is an issue that needlessly affects far too many motherboards.

Admittedly, we were a little disappointed with our fully stable overclocking results. However, we did have unusually high expectations for this model, and a 100% stable 515Mhz is still a very good result. In fact, it is more than 99% of users will need for achieving a high CPU overclock for 24/7 use. More importantly though, these results were achieved with an immature bios, and based one our experiences with the EP45-DQ6, we know that Gigabyte will shortly release a BIOS that further improves overclocking. This hidden potential can be seen in our <I>maximum benchable</i> results, which were quite simply jaw-dropping.

Related to this (potential) overclocking prowess is the EP45-UD3P's excellent voltage regulation. Whatever voltage you set in the BIOS is nearly always what the motherboard outputs, and even under load the voltages remain steady. The only (very) minor exception to this is the vCore, but thanks to the introduction of Load-Line Calibration, vDroop can be effectively eliminated.

What can we say about Ultra Durable 3 itself? Well the higher quality solid capacitors, MOSFETs, and chokes are obviously a welcome addition, as they potentially increase the durability and longevity of the motherboard. The 2 oz copper PCB layers may have played a role in the low MOSFET temperature we recorded, as well in our extreme overclocking achievements, but there is obviously no way for us to verify this directly.

The EP45-UD3P does have one Achilles' Heel at the moment: price...at least here in Canada. Due to a number of economic factors, this model's price has been steadily increasing over the last few weeks and most retailers are selling it for between $180-$205 CAD, which is a $20-30 more than we would expect. On the other hand, our fortunate American neighbours can find this model for 135$ USD (and under), and at that price it is the best Bang for the Buck P45 motherboard on the market. Period.

By the way, those of you who would like a DDR3 version of this motherboard will be glad to know that the EP45T-UD3P should be available in the retail channel in early November.


Pros

- Excellent layout.
- Great cooling system.
- Impressive overclocking potential.
- CrossFire support.
- Fantastic BIOS options.
- Load-Line Calibration now available.
- EasyTune6.
- Price (in the United States).


Cons

- Price (in Canada).
- Immature BIOS prevents maximum FSB overclock.

Our thanks to Gigabyte for making this review possible!​

Gigabyte GA-EP45-UD3P Motherboard Review Comment Thread​