Turn off the Ad Banner  

To print: Select File and then Print from your browser's menu.

    -----------------------------------------------
This story was printed from CdrInfo.com,
located at http://www.cdrinfo.com.
-----------------------------------------------


Appeared on: Thursday, August 25, 2005
Intel D 840 Dual Core CPU


1. Introduction

In our last review concerning Intel's CPUs, we had tested the flagship Single-core processor from Intel, the 3.73GHz Extreme Edition which was also extremely expensive. This time, we have in our hands the 840 which is a Dual-Core CPU, running at 3.20GHz with 1MB L2 cache. Dual-Core processors are the new generation of CPUs and both Intel and AMD have already released not one, but over three different models.

The one we will review here costs approximately half the money of the P4 3.73GHz Extreme Edition and it is rather interesting to compare their performance. When you are about to get a new processor, most times, money will define which one is the most appropriate for your needs, otherwise we would all purchase the most expensive, thinking that we have also got the best. Is this true however?

Architecture 90 nm technology
L2 Cache 2x1MB
Clock Speed 3.20 GHz
Front Side Bus 800 MHz
Chipset - Intel 955X Express chipset
- Intel 945P Express chipset
- Intel 945G Express chipset
- Intel E7230 Chipset
Socket LGA775

Retail Package

Processor Features & Definitions

Architecture

Basic design of a microprocessor. May include process technology and/or other architectural enhancements.

Cache (MB/KB)

A temporary storage area for frequently accessed or recently accessed data. Having certain data stored in a cache speeds up the operation of the computer. Cache size is measured in megabytes (MB) or kilobytes (KB).

Clock Speed (GHz/MHz)

Speed of the processor's internal clock, which dictates how fast the processor can process data. Clock speed is usually measured in GHz (gigahertz, or billions of pulses per second).

Front Side Bus (GHz/MHz)

The connecting path between the processor and other key components such as the memory controller hub. FSB speed is measured in GHz or MHz.

The stock cooler that Intel includes in its retail version for the D 840 resembles the one used for the P4 series of processors. If you look more carefully however, you'll notice that there are some distinctive differences. The copper base, which exists in all of Intel's stock coolers, is more voluminous. The fins are also of a different design. This indicates that the heat dissipation required is going to be much greater than in single-core processors. This is something we will check on later.

Click to Enlarge
Old cooler

Click to Enlarge
New Cooler

2. Specifications

In Device Manager, you'll notice that your system will list the D 840 as two processors, while from CPU-Z you can read the details for each one separately.

When the CPU runs in idle, the temperature according to Asus Probe II and the bios is:

When we ran PCMark04 and 05, due to running simultaneous multiple processes, the temperature increased to:

This is a very high temperature and the stock cooler has to work overtime in order to restore temperatures back to "normal".

Here's a rundown of our testbed used to test the Intel D 840 3.2GHz processor:

System Specifications:
CPU: Intel Pentium D 840
Cooler: stock
Case: Antec
Motherboard: Asus P5WD2E-Deluxe
Power supply: CoolerMaster 450W
Memory: 2X512MB Corsair CM2X512-8000UL (dual channel)
VGA: Asus 6800GT PCI-E
Hard Disk Drive: WD800JD 80GB 7200RPM
OS: Windows XP Pro SP2
Drivers: 7.1.8.4
DirectX: v9.0

Benchmarks & Applications used


3. SiSoft Sandra

SiSoftware Sandra is a 32-bit and 64-bit Windows system analyser that includes benchmarking, testing and listing modules. It also provides information about the CPU, chipset, video adapter, ports, printers, sound card, memory, network, Windows internals, AGP, ODBC Connections, USB2, Firewire etc.

The CPU Arithmetic benchmark tests how your CPU stacks up to other typical combinations:

Multi-Processor (SMP) support for up to 32/64 CPUs & SMT.
- Test takes about 20/30 seconds on P6-class PCs.
- Test is repeated 3 to 5 times to get a realistic result.
- Results outside the confidence limits (5%) are discarded.
- Command Line Switch: CPUAABench
- Requirements: None
- OS (Win32) Support: Windows 2000, XP, 2003
- OS (Win32 CE) Support: PocketPC 2000/2002/2003, SmartPhone 2002, 2003
- OS (Win64) Support: Windows XP/2003

Sandra's CPU arithmetic benchmark suite uses 2 famous benchmarks (Dhrystone and Whetstone) for stress testing the processor. Also, it includes a new version of the Whetstone test that makes use of a processor's SSE2 instructions to show the performance boost an application might gain should it be optimized for SSE2. Unfortunately, this does not always represent true real-life performance, but is useful for comparing the speed of various CPUs.

The following results with various CPUs are more for reference reasons since there is no actual direct comparison for the D 840. However, we thought it would be interesting to include them in our graphs.

Although the actual speed of the D 840 is slower than other CPUs, its performance was extremely high in this test.

The CPU Multi-Media benchmark tests how your CPU stacks up against other typical combinations:

- Multi-Processor (SMP) support for up to 32/64 CPUs & SMT.
- Test takes about 20 seconds on P6-class PCs.
- Test is repeated 3 to 5 times to get a realistic result.
- Results outside the confidence limits (5%) are discarded.
- Command Line Switch: CPUMMBench
- Requirements: MMX (Enhanced), 3DNow! (Enhanced) or SSE(2/3) recommended
- OS (Win32) Support: Windows 2000, XP, 2003
- OS (Win64) Support: Windows XP/2003

The test has two parts:

Integer test - using integer data:
- MP SSE3 is used for better performance if supported.
- MP SSE2 is used for better performance if supported.
- MP SSE (Streaming SIMD) is used for better performance if supported.
- MP Enhanced MMX is used for better performance if supported.
- MP MMX is used for better performance if supported.
- MP ALU is used otherwise.
Floating-point test - using floating point data:
- MP SSE3 is used for better performance if supported.
- MP SSE2 is used for better performance if supported.
- MP SSE (Streaming SIMD) is used for better performance if supported.
- MP 3DNow! Enhanced is used for better performance if supported.
- MP 3DNow! is used for better performance if supported.
- MP FPU is used is used otherwise.

This test involves the generation of Mandelbrot Set fractals that are used to realistically describe and generate natural objects such as mountains or clouds. By using various multi-media extensions (MMX), 3DNow! and SSE(2/3) better performance is achieved.

The D 840 costs half the money of some of the above listed processors but its performance is much better.


4. PCMark04

PCMark04 is an application-based benchmark and a premium tool for measuring overall PC performance. It uses portions of real applications instead of including very large applications or using specifically created code. This allows PCMark04 to be a smaller installation as well as to report very accurate results. As far as possible, PCMark04 uses public domain applications whose source code can be freely examined by any user.

PCMark04 includes 4 categorized suites for benchmarking your computer. These include CPU, Graphics, Memory and a Hard Disk Drive benchmark. In our case we select only the CPU test suite.

The Dual-Core processor produced a very high score in the PCMark04 benchmark. This test includes multitasking, something we knew from the start was going to be a strong point for the Dual-Core CPU.

More analytically:

  Intel P4 3.73Ghz Intel P4 660 Intel P4 D 840
File Compression (MB/sec) 5.4 5.3 3.8
File Encryption (MB/sec) 70.5 65.6 57.4
File Decompression (MB/sec) 45.2 42.9 64.4
Image Processing (MPixels/sec) 18.5 16.9 27.5
Grammar Check (KB/sec) 3.0 2.4 4.7
File Decryption (MB/sec) 107.5 102.3 91.9
Audio Conversion (KB/sec) 2921.7 2779.1 2507.3
WMV Video Compression (FPS) 64.1 59.7 74.4
DivX Video Compression (FPS) 88.4 77.6 87.1

With some tests, the D 840 is slower than the 3.73GHz ED or even the P4 660 3.6GHz. This is logical since the D 840 runs at 3.2GHz, so for single process applications the D 840 is no different to a single core CPU. When there are multiple processes, then the D 840 is faster.


5. 3DMark01

3DMark2001 SE is a diagnostics tool for measuring the 3D game performance of PCs. It is entertaining and easy to use, which makes it "must have" software for all home PC users interested in 3D games. Even a beginner PC user can get a game performance measurement with 3DMark2001 SE.

Although the 3DMark series is dedicated to 3D graphics performance, it is rather interesting to see how different CPU architectures influence frame rates.

From this test it seems that the higher processor and Front Side Bus speeds of the 3.73GHz ED processor do make a difference. Does the same hold true when downloading from the internet or encoding a movie? This is something we will find out shortly.


6. 3DMark03

3DMark is a widely used and accepted benchmark that stresses the DirectX performance of a VGA card. For testing the performance of each card we use the 4 game benchmark 3DMark offers. The first is a DirectX 7 game, the second and the third use DirectX 8 and the last one stresses graphics in DirectX 9. A very strong point of 3DMark is that its VGA card measuring does not require a lot of CPU power. So the resulting fps are a good reference of a VGA card's rendering performance.

Same as with 3DMark01. The Gaming environment benefits from higher speed processors.


7. 3DMark05

3DMark05 is concentrated on measuring the performance of the DirectX 9 generation of 3D hardware using the technology of next generation games. Earlier 3DMark versions have usually been launched with the introduction of a new Microsoft DirectX version and with the introduction of a new hardware generation. This has somewhat limited how much 3DMark has utilized the new hardware features. At the time of this writing, DirectX 9 was introduced soon two years ago, and there is now DX9 generation hardware available all the way from the very high end to value and mobile parts. 3DMark05 can thereby utilize the features of DirectX 9 to a 100%. In fact, DirectX 9 hardware is required for every game test of 3DMark05, resulting in the ultimate tool for DirectX 9 hardware performance comparison.

3DMark05 is the newest representative from Futuremark and simulates very closely, the latest game environments which are more CPU intensive than older generation games. All you need is a very fast graphics adapter and a CPU that isn't going to hold it back. And in this case, the D 840 managed to outperform the rival, single core processors.


8. PassMark Benchmark Suite

CPU Benchmarks

The suite of PassMark Performance Tests contains multi-process CPU tests such as:

- Integer (32-bit addition, subtraction, multiplication and division)
- Floating Point (32-bit addition, subtraction, multiplication and division)
- WIN32: MMX (matrix multiplication) or WIN64: 64bit Integer Matrix Multiplication
- WIN32: SSE (128-bit SSE operations such as addition, subtraction and multiplication) or WIN64: 64bit Floating Point Matrix operations
- WIN32: 3Dnow! (3D transformation of an array of 32-bit vertices) or WIN64: 64bit Floating Point Matrix operations
- Compression
- Encryption
- Image Rotation (Rotate image co-ordinates in memory)
- Random String Sorting

Integers are whole numbers such as 23, 459532, -26. Floating point numbers contain a fractional part, e.g. 1.003, 98394.2. These two types of numbers are treated quite differently inside a computer, which is why PerformanceTest differentiates between the two.

For each of the integer tests, a large array of random 32 bit integers is processed using the particular mathematical operations listed below. For the floating point tests, single precision floating point numbers (32 bit) are used.

The "PassMark rating" is a weighted average of all the other test results and gives a single overall indication of the computers performance. The bigger the number, the faster the computer. The "PassMark rating" can only be calculated if the results from all other tests are available. The value is calculated as follows.

Test Suite Weighting
Disk 20%
CD / DVD 9%
Memory 18%
3D Graphics 12%
2D Graphics 14%
CPU 27%
Total 100%

In our case however we only ran the CPU suite.

We chose to do "All" the tests for the CPU, as described above. Below you can see the difference between the D 840, the 3.73Ghz ED and the P4 660.

Test D 840 3.73Ghz ED P4 660

As we can see, once again the 3.6GHz and 3.73GHz processors are faster than the 3.2GHz D 840 for single processes. But for multitasking? Check below...

Multitasking Benchmarks

The Advanced Multi Process Test from the PassMark benchmark is designed to run a series of standard benchmark tests in parallel, to examine performance under these conditions. Multi-processor machines should excel under this kind of environment, as the load is split between the CPUs.

The tests include a number of the standard PassMark benchmarks along with some popular benchmarking algorithms and tests which emulate some day-to-day operations your CPU may be required to carry out and are particularly processor intensive.

As the tests are quite varied, as are their results, it unfortunately means that no graphing feature exists as is common amongst the other Advanced Tests.

Test Name Intel 3.73 Intel D 840 Primes found
Prime Number Search 51341 66161 Primes Found
Sorting Random Strings 153388 155960 Thousand strings sorted per second
Compression 264 338 KBytes per second
Encryption 1409.74 2011.54 KBytes per second
Rotation 22.78 35.89 Rotation per second
MMX Addition 49.87 71.37 Million Ops/Sec
MMX Multiplication 46.91 66.65 Million Ops/Sec
Integer addition 75.35 120.87 Million Ops/Sec
Whetstone 226.07 212.99 Million Ops/Sec
Dhrystone 254.73 330.22 Million Ops/Sec
Memory Read 289.39 374.48 MB/sec
Memory Write 289.33 374.72 MB/sec
Disk Access 0.15 0.14 MB/sec

There is no doubt that the Dual-Core is better equipped to handle multiple processes. This test confirms this.

Further details about each test are given below:

The Tests
- Prime Number Search:
This test uses a simple brute force prime number search algorithm. It begins at 3 and checks every integer to see if it is a member of the set of primes. (A prime number is one which is divisible only by 1 and itself). The test reports the amount of primes found and the largest prime. Note that for longer test periods, the rate of Primes Per Second found will drop as the distance between primes gets larger.

- Sorting Random Strings:
The test uses qsort to continually sort 1000 strings of 256 characters. Once the test duration expires, the test reports the speed of the sorting in thousands of strings per second.

- Compression:
The Compression test uses an Adaptive encoding algorithm based on source code from Ian H. Witten, Radford M. Neal, and John G. Cleary in an article called “Arithmetic Coding for Data Compression”. The system uses a model which maintains the probability of each symbol being the next encoded. It reports a compression rate of 363% for English text, which is slightly better than the Huffman method. This test reports its results in KBytes/Sec compressed.

- Encryption:
The encryption test uses the Blowfish enciphering algorithm. It is based on the C implementation by Paul Kocher. Data is enciphered using a 16byte key in blocks of 4 KB. The test reports in KBytes/Sec.

- Rotation:
The rotation algorithm is a PassMark developed algorithm using standard rotation mathematics. The test uses a 400 x 400 pseudo image with one byte of data per pixel. The image is rotated degree by degree, with each degree counting as one rotation. The test reports in rotations per second.

- MMX Addition:
- MMX Multiplication:

These are the standard PassMark benchmarks. See MMX Tests for more information. These tests are only included with the version of PT that is for the 32-bit version of Windows.

- Maths Integer addition:
This is a standard PassMark benchmark. See Maths Tests for more information.

- Whetstone:
The Whetstone benchmark is one commonly used to test floating point arithmetic. The algorithm is based on the original C source code written by Rich Painter. The benchmark is designed to test the speed of commonly used floating point CPU instructions. The test reports in MIPS (Million Instructions Per Second.)

- Dhrystone:
The Dhrystone benchmark is one commonly used to test a machines integer arithmetic. The algorithm is based on the C source code for Dhrystone version 2 by Reinhold P. Weicker. The test reports in MIPS.

- Memory Read:
- Memory Write:

These are the standard PassMark benchmarks. See Memory Tests for more information.

- Disk Access:
This test uses the same test engine as the PassMark Advanced Disk Test. The test uses a file size of the current available RAM, a block size of 4K, the Win32 uncached interface, a 50%/50% RW level, it is 100% Sequential and executed in Synchronous mode. The test reports its results in MB/Sec.


9. Prime95

Prime95 is basically a Mersenne prime number discovery program. It's a great example of Distributed Computing, but it's most loved by overclockers for its powerful system stress-testing and benchmarking abilities. It can torture-test your CPU to produce its maximum heat or spit out any errors due to excessive overclocking. Also, using the blend test, it can consume all your physical memory and run error-checking tests on your modules to ensure their quality. However, right now, we're mostly interested in it for its benchmarking capabilities.

Here are the times for the Intel D840, Pentium 4 3.73Ghz ED and 660, and AMD Athlon 64 FX-55.(less is better)

The 3.73GHz ED seems to offer the best performance.


10. Audio/Video Encoding, Compression

Benchmarks are good for reference and comparison reasons but what about real performance? For this reason we did some simple tests using popular utilities for tasks that most users perform every day.

dBPower Tests

In the following graph, you can see how the three different CPUs performed while encoding the same audio file into different formats. For this we used a whole music CD with a size of 750MB. We ripped the AudioCD using EAC and then with dBPower encoded from one format to another. The times needed for each task are given below (the lower, the better off course :-)

As we've mentioned previously, the D 840 does not differ from a single core processor when it comes to single thread processes.

The settings with dBPower were the same for all CPUs. Analytically:

DVD Shrink Tests

Compressing a movie from DVD9 to DVD5 is a very common task. For this reason, we ripped an original movie (which we own of course :-), using DVD Decrypter. Then with DVD Shrink, we compressed it in order to fit on a DVD5 disc. The size of the original movie was 6.85GB and we crunched it down to 4.463GB. Below you can see how many frames per second each CPU can process and the total time needed for the encoding.

It is amazing to see how many fps the D 840 can process. This is the highest framerate we have seen. The encoding time required is one and a half minutes less than with the 3.73GHz ED. So, for video processing with DVD Shrink, the D 840 is simply excellent.

DivX Tests

Another very common task is to convert a movie into the DivX format. For this we used a vob file sized 202MB which we converted to AVI with Flask MPEG. The settings are shown below:

Unfortunately, not as good a performance when encoding to DivX as we saw with DVDShrink and the D 840 was the slowest of the three processors.

WinRAR Tests

Using a folder with 101MB of a variety of files, we checked how long each of the CPUs needed to compress and extract them.

When compressing, the P4 3.73GHz was a good deal faster while with decompression, the D 840 was slightly faster.

The performance of the D 840 is very much dependent on the application in question and more specifically, how that application has been coded, i.e. whether it has been designed to take advantage of multi-threading or not. Multiple threads means two or more tasks from the same process running simultaneously, something that is well suited to the D 840.


11. Half Life 2

Half life 2 is no doubt the most anticipated pc game of all times.

Physics - From pebbles to water to 2-ton trucks respond as expected, as they obey the laws of mass, friction, gravity, and buoyancy.

Graphics - Source's shader-based renderer, like the one used at Pixar to create movies such as Toy Story® and Monster's, Inc.®, creates the most beautiful and realistic environments ever seen in a video game.

AI - Neither friends nor enemies charge blindly into the fray. They can assess threats, navigate tricky terrain, and fashion weapons from whatever is at hand.

We have mentioned this in the past and we will repeat it again, a very good and fast card like the 6800 series from NVidia, the X800 or X850 from ATI, require a lot of CPU processing in order to perform at their maximum. Although the D 840 was not the fastest at the lower resolutions, at 1600X1200 it proved more capable.


12. Overclocking

In order to overclock the CPU, we manually started to increase, step by step, the FSB, beginning from the default 200. When we set the FSB to 247 or 250, we were able to access Windows and work but not for long. As soon as we ran PCMark05 or SiSoft Sandra 2005, the system crashed and the blue screen appeared. So we decreased the FSB frequency to below 247.

Finally, with the FSB frequency set to 240, everything seemed to work ok and all the benchmarks finished successfully without any errors. The actual speed at this point was 3.85GHz instead of the default 3.2GHz.

It is highly possible with a better cooling system, the CPU could run higher than 3.85GHz. While we ran our benchmarks, the temperature reached 77°C.

- The Tests

The increased FSB frequency raised the performance. The difference in performance was very noticeable compared to the default setting.

SiSoft Sandra 2005 Arithmetic Benchmark
3.2Ghz
3.85Ghz

SiSoft Sandra 2005 Multimedia Benchmark
3.2Ghz
3.85Ghz

Amazing difference in SiSoft Sandra's benchmark's.

In games, the overclocked CPU dramatically increased the framerates.


13. Running Multiple Processes

At this point, it would be rather interesting to see how the D 840 performs some tasks while some other processes are running in the background. For example, if you are an extreme gamer but you also want to download from the internet, will this affect the actual frame rate and by how much? How about encoding a movie while listening to music ?

We are pretty sure that if we ran enough tasks simultaneously, the D 840 would eventually reach its limits. But our purpose is not to overload the processor but to see how its performance deteriorates as more tasks are loaded. For this reason, we selected the following combinations of tasks.

Our first attempt involved measuring the framerate in HL2 and Doom3 while in the background we ran DVD Shrink encoding, in low priority mode. DVD Shrink was reading from the Primary HD and writing to an external USB drive. The games were installed on the Primary HD.

As you can see, the framerate is not greatly affected and you would still be able to play your favorite game while at the same converting your personal movies. As was reasonable to expect, DVD Shrink encoding time increased from 6:17min to 10:10min but this is not too important since you won't notice it while you're playing.

How about now, we encode in DVD Shrink and at the same time convert 16 wav files to MP3? Each task separately needed 6:17min for the encoding in DVD Shrink and 4:10min for converting 16 wav files from a full AudioCD of 700MB to MP3. When we ran these two processes at the same time, the times changed to 11:46min for DVD Shrink and 4:36min for dBpower. Since we knew that DVD Shrink would require longer, we ran dBpower a second time. As we can see from the times, although dBpower wasn't greatly affected, DVD Shrink took almost twice as long to complete.

One final note, if every task uses the same HD, then it is highly possible to end up with a bottleneck as your HD tries to service all the i/o requests. For example, if DVD Shrink reads from one partition of the HD and writes to another on the same physical drive while dBpower does something similar and then you run your games from the same HD, then this is going to be too much for the HD.


14. Conclusion

With the new Dual-Core series from both Intel and AMD, a new era in processors has started. They are already available at varying speeds as well as prices. It is rather interesting to put them to the test and compare them against the single core generation of processors to see what performance is like and whether it is worth moving up this new generation of CPU. We all knew that the Extreme Edition Pentium at 3.73Ghz was the fastest CPU from Intel and for this reason we chose to compare it with the D 840, also from Intel.

A head to head of the D 840 with an AMD Dual-Core would also be very interesting and it is something we are planning to do in the near future.

The overall performance of the D 840 is very good. Although the CPU is supposed to be better in multiprocessing, and of course it is, we also liked its performance with single processes. Our CPU benchmarks, especially SiSoft Sandra 2005, PCMark and PassMark, reported extremely high scores for the D 840 which indicates how fast the specific CPU is. In game environments, the D 840 might not be as fast as the 3.73GHz ED but believe us, it won't let you down. Instead, you will be able to play your favorite games while at the same time encode your movies in the background without any significant performance drop.

Something which also impressed us was the framerate that the CPU produced during DVD Shrink's Deep Analysis and off course the very fast times for the full encoding. One thing that is noticeable though, is that the performance of the the D 840 differs from application to application and while it may be ideal for multi-threaded applications, we got the feeling that most applications don't currently take advantage of this feature. Perhaps when dual core processors are more common place, applications will follow suit.

The temperature levels were rather high, as they were with the ED series, but this time they were even higher. At idle state, the CPU reported 54°C but during our tests, with the default FSB frequency, when running our benchmarks the temperature reached 71°C. Perhaps a better cooler would make things better. For testing the CPU with higher FSB frequencies, we had to set the CPU temperature level in our monitoring software higher, in order to avoid the continuous temperature alert dialog box.

If you like overclocking, the D 840 is for you. The CPU can run without problems and with the stock cooling system in place, up to 3.85GHz instead of the default 3.2GHz. We also managed to get the CPU up to 4.00GHz and although Windows started normally, when we ran the PCMark tests, we saw the blue screen. It maybe that the 77°C core temperature was responsible for this and a better cooling system would allow for even higher speeds.

The D 840 costs around US$550 on the e-market. The respective prices for AMD's 4200+ Dual-Core are US$490 while the 4600+ close to US$700. We believe that the price is reasonable given the overall performance but we would like to wait before giving a final judgement until we have reviewed the rival CPU from AMD.



Home | News | All News | Reviews | Articles | Guides | Download | Expert Area | Forum | Site Info
Site best viewed at 1024x768+ - CDRINFO.COM 1998-2014 - All rights reserved -
Privacy policy - Contact Us .