We are a participant in the Amazon Services LLC Associates Program, an affiliate advertising program designed to provide a means for us to earn fees by linking to Amazon.com and affiliated sites.
[nextpage title=”Introduction”]
We finally got to review processors designed to the low-end market, i.e., the most inexpensive processors on the market. Nowadays Intel has the advantage on the high-end market, but who gets to be the best at the low cost PC market? Read our review and find out.
Everyone who goes shopping for a PC dreams of taking home one with a high-end processor. The thing is that most of the times they can only afford the simplest model, and then the options AMD Sempron and its Intel competitor, Celeron D.
Reviewing inexpensive processors is not as easy as it seems. We have to use motherboards that are compatible with the processor purpose, i.e., we cannot use a USD 200-board to test a USD 50-processor.
That’s why we have chosen two inexpensive motherboards that are extremely popular on the market: a PCChips P21G v3.1 for the Celeron D and a Gigabyte GA-K8VM800M Ver. 2.1 for the Sempron.
Both have similar characteristics. They use VIA chipsets with the same graphics engine (P4M800 + VT8237R on the PCChips board, and K8M800 + VT8237R on the Gigabyte board) and are used by several PC manufacturers and assemblers.
Figure 1: Platform used for reviewing the AMD Sempron processor.
Figure 2: Platform used for reviewing the Intel Celeron D processor.
As you can see, both motherboards have AGP 8x slot and two Serial ATA 150 MB/s ports, besides 6-channel sound and Fast Ethernet (10/100 Mbps) LAN.
On the next page we will talk a more about the CPUs included in our review.
[nextpage title=”CPUs Included In Our Review”]
We have already written several tutorials on Sempron and Celeron D CPUs: “Sempron vs. Celeron D“, “All Celeron Models“, and “All Sempron Models“, but let’s take a more careful look at the processors included in our review.
The Sempron model chosen was the 3000+, that’s the most popular Sempron among the models sold.
The main characteristics of the Sempron 3000+ tested are:
- Socket: 754
- Internal clock: 1.8 GHz
- Memory clock (external clock): 400 MHz (200 MHz DDR)
- HyperTransport bus clock: 1,600 MHz (800 MHz DDR)
- L1 memory cache: 128 KB
- L2 memory cache: 128 KB
- Voltage: 1.40 V
- Maximum temperature: 69º C
- Thermal power: 62 W
- Manufacturing process: 90 nm
- Core: Palermo
- 64-bit extensions enabled
- SSE3 instruction set and NX Disable
- Memory controller: Single channel
- More information: www.amd.com
- Average price in the USA*: USD 57.00
Figure 3: AMD Sempron 3000+ CPU.
Figure 4: Technical specs of the Sempron 3000+ tested.
The Celeron D chosen was the 331, for being the first Socket LGA775 Celeron to support the EM64T technology, that allows the processor to access more RAM memory, and the XD (eXecute Disable, also known as NX bit) technology, that prevents some kinds of viruses from attacking the PC.
The main characteristics of the Celeron D 331 tested are:
- Socket: 775
- Internal clock: 2.66 GHz
- Memory clock (external clock): 533 MHz (133 MHz QDR)
- L2 memory cache: 256 KB
- Voltage: 1.25 – 1.4 V
- Thermal power: 84 W
- Manufacturing process: 90 nm
- Core: Prescott
- 64-bit extensions enabled
- SSE3 instruction set and eXecute Disable
- More information: www.intel.com
- Average price in the USA*: USD 54.00
Figure 5: Intel Celeron D 331 CPU.
Figure 6: Technical specs of the Celeron D 331 tested.
* Researched at Shopping.com on the day we published this review.[nextpage title=”How We Tested”]
During our benchmarking sessions, we used the configuration listed below. Between our benchmarking sessions the only variable was the CPU being tested and the motherboard, since each CPU benchmarked required a different type of motherboard.
Hardware Configuration
- Socket 754 motherboard: Gigabyte GA-K8VM800M Rev. 2.1 (BIOS FF, December 15, 2005) for the AMD Sempron 3000+ CPU.
- Socket LGA775 motherboard: PCChips P21G V3.1 (BIOS 060308S, March 8, 2006) for the Celeron D331 CPU.
- Memory: One OCZ EL DDR PC-4000 Gold GX XTC 1 GB module installed at DDR single channel (3-4-3-8 timings).
- Hard disk: Samsung SpinPoint SP0411N (7,200 rpm, 40 GB, ATA-133).
- Video Card: NVIDIA Geforce 6600GT 128 MB AGP.
- Video resolution: 1024x768x32 75Hz.
- Power Supply: Seventeam ST420BKV-03F.
Software Configuration
- Windows XP Professional installed using NTFS.
- Service Pack 2.
- Direct X 9.0c.
Driver Versions
- NVIDIA video driver version: 91.31 WHQL.
- VIA chipset driver version: Hyperion Pro 5.09A.
- Realtek sound driver version: A3.90.
Used Software
- SYSmark2004 – Patch 2
- PCMark05 Professional 1.1.0
- 3DMark2001 SE Build 3.3.0
- 3DMark06 Professional 1.0.2
- Quake III Arena 1.32
- Doom 3
- Sandra Lite 2007.SP1 (2007.6.10.99)
- Super Pi Mod 1.5
- GamingHeaven Photoshop Benchmark V2
- DVD Shrink 3.2.0.15
We adopted a 3% error margin; thus, differences below 3% cannot be considered relevant. In other words, products with a performance difference below 3% should be considered as having similar performance.
[nextpage title=”SYSmark2004″]
We benchmarked the CPUs’ overall performance by using the SYSmark2004, a program that simulates the use of real-world applications. Thus, we consider this the best software to measure, in practical terms, the system performance.
The benchmarks are divided into two groups:
- Internet Content Creation: Simulates the authoring of a website containing text, images, videos and animations. The following programs are used: Adobe After Effects 5.5, Adobe Photoshop 7.01, Adobe Premiere 6.5, Discreet 3ds Max 5.1, Macromedia Dreamweaver MX, Macromedia Flash MX, Microsoft Windows Media Encoder 9, McAfee VirusScan 7.0 and Winzip 8.1.
- Office Productivity: Simulates the use of an office suite, i.e., simulates sending e-mails, word processing, spreadsheets, presentations, etc. The following programs are used: Adobe Acrobat 5.05, Microsoft Office XP SP2, Internet Explorer 6.0 SP1, NaturallySpeaking 6, McAfee VirusScan 7.0 and Winzip 8.1.
This software delivers several results, all of them using a specific SYSmark2004 unit. First we have a SYSmark2004 overall score. Then we have a group result for each batch listed above. And for each batch, we have specific results: 3D Creation, 2D Creation and Web Publication for Internet Content Creation; and Communication, Document Creation and Data Analysis for Office Productivity.
You can see the results in the graphic below:
| Sempron 3000+(1.8 GHz) | Celeron D 331(2.66 GHz) | Difference | |
| SYSmark2004 – Score | 138 | 110 | +25.45% |
| Internet Content Creation – Score | 149 | 120 | +24.17% |
| 3D Creation | 144 | 110 | +30.91% |
| 2D Creation | 169 | 132 | +28.03% |
| WEB Publication | 136 | 118 | +15.25% |
| Office Productivity – Score | 128 | 101 | +26.73% |
| Communication | 122 | 94 | +29.79% |
| Document Creation | 139 | 102 | +36.27% |
| Data Analysis | 123 | 106 | +16.04% |
Sempron 3000+ (1.8 GHz) achieved performance, in average, 26% superior to the Celeron D 331 (2.66 GHz) on SYSmark2004.
Although the Intel CPU has 860 MHz more than the AMD model tested, it was beaten in all the tests performed by SYSmark. Really impressive.
[nextpage title=”PCMark05 Professional”]
PCMark05 Professional measures the system performance by running several tests. We selected two batches for our comparisons, System and CPU.
The System batch performs the following tests: HDD XP Startup, Physics and 3D, 2D Transparent Window, 3D Pixel Shader, Web Page Rendering, File Decryption, 2D Graphics Memory – 64 lines, HDD General Usage and three multithreading tests.
The CPU batch performs the following tests: File Compression, File Decompression, File Encryption, File Decryption, Image Decompression, Audio Compression and two multithreading tests.
The results are given in a PCMark05 specific unit.
| Sempron 3000+(1.8 GHz) | Celeron D 331(2.66 GHz) | Difference | |
| PCMark05 – System | 2718 | 2632 | 3.27% |
| PCMark05 – CPU | 2537 | 2993 | 17.97% |
On system batch, the Sempron 3000+ (1.8 GHz) was 3.27% faster than the Celeron D 331 (2.66 GHz), that is, they both achieved almost the same performance.
On CPU batch, Celeron D 331 (2.66 GHz) was 17.97% faster than the AMD processor.
[nextpage title=”3DMark2001 SE”]
3DMark2001 SE simulates older games based on DirectX 8.1 and it’s sensitive to any changes on the hardware configuration.
You can see the results below.
| Processor | 3DMark2001 SE | Difference |
| Sempron 3000+ (1.8 GHz) | 15150 | |
| Celeron D 331(2.66 GHz) | 10684 | 41.80% |
On this benchmark, Intel Celeron was severely beaten by the AMD processor.
[nextpage title=”3DMark06 Professional”]
3DMark06 is the latest version of 3DMark franchise, measuring Shader 3.0 (i.e., DirectX 9.0c) performance. We run this software on its default configuration (1280×1024 resolution with no image quality settings enabled), checking the CPU batch results for comparison.
| Sempron 3000+(1.8 GHz) | Celeron D 331(2.66 GHz) | Difference | |
| 3DMark06 – Overall | 1398 | 1637 | 17.10% |
| 3DMark06 – CPU | 676 | 608 | 11.18% |
On graphics batch, Celeron D 331 CPU was better, being 17.10% faster than the AMD processor.
On CPU batch, Sempron 3000+ got even and was 11.18% faster.
[nextpage title=”Quake III Arena 1.32″]
Even though this is an old game, its importance comes from the fact its engine is used by several other games, like Jedi Knight II and Medal of Honor, and also because this game is very sensitive to any changes on the hardware configuration.
We used the demo four available on version 1.32 of Quake III and wrote down the obtained number of frames per second to make our benchmarking with this game. We ran this demo three times at 1024x768x32 resolution and all image quality settings on their default configuration and we picked the middle value for our comparisons, i.e., we discarded the highest and the lowest values.
Check the results below.
| Processor | Frames Per Second | Difference |
| Sempron 3000+ (1.8 GHz) | 244.9 | |
| Celeron D 331 (2.66 GHz) | 162.7 | 50.52% |
On Quake III, Sempron 3000+ was 50.52% faster than Celeron D 331, showing that the AMD CPU is a better choice if you like gaming.
[nextpage title=”Doom 3″]
Using the resolution 1024x768x32 High Quality, we ran demo1 four times and wrote down the obtained number of frames per second. The first result we discarded at once, since it is far inferior to the other results. This happens because at the first time we run the demo the game must load all textures to video memory, fact that doesn’t happen from the second time we run the demo on. From the three results left, we consider as our official result the middle result, i.e., we discard the highest and the lowest values. Curiously almost all times the values obtained at the second round on were the same.
A very important detail that we must mention is that Doom 3 has an internal FPS lock: it is only capable of generating 60 frames per second, even if your board is able to produce more frames per second than that. This is done in order to make the game to have the same “playability” sensation independently from the video card installed on the PC. This lock, however, is disabled in the game benchmarking mode.
Check out the results below.
| Processor | Frames Per Second | Difference |
| Sempron 3000+ (1.8 GHz) | 62.1 | |
| Celeron D 331 (2.66 GHz) | 35.7 | 73.95% |
Once again Sempron proves to be the best option for those who enjoy gaming. It was 73.95% faster than Celeron D 331 on Doom 3.
[nextpage title=”Sandra Lite 2007.SP1 (2007.6.10.99)”]
We used the Memory Bandwidth Benchmark module from Sandra to measure the maximum transfer rate that the memory was able to achieve. The result published is an arithmetic average from the Buffered Integer Stream and Buffered Float Stream results.
You can check the results on the graph below.
| Processor | Transfer Rate | Difference |
| Sempron 3000+ (1.8 GHz) | 2.693 MB/s | |
| Celeron D 331 (2.66 GHz) | 2.479 MB/s | 8.63% |
Both processors benchmarked access memory in single channel. On Celeron D it’s possible to use the dual channel feature since the motherboard chipset supports that feature. The motherboard used in this review didn’t have that feature.
At this benchmark, AMD processor’s memory transfer rate was 8.63% superior to the transfer rate achieved by Intel Celeron D.
[nextpage title=”Super Pi Mod 1.5″]
Super Pi is a benchmarking software created at Tokyo University that measures the time spent to calculate Pi up to 32 million decimal places. In our tests we calculated Pi with 1 million decimal places.
Check the results below, shown in seconds (the lower the result, the better).
| Processor | Seconds | Difference |
| Sempron 3000+ (1.8 GHz) | 53.109 | |
| Celeron D 331 (2.66 GHz) | 74.438 | 28.65% |
AMD Sempron 3000+ was 28.65% faster than the Intel CPU. It completed the test with an advantage of more than 21 seconds.
[nextpage title=”DriverHeaven Photoshop Benchmark V2″]
The best way to measure performance is by using real programs. The problem, though, is creating a methodology using real software that provides accurate results. For Photoshop CS2 there is a methodology created by the folks at GamingHeaven that is very accurate. Their script applies a series of 12 filters to a sample image and we wrote down the time taken for each filter to run. At the end, we have the results for each individual filter and we simply added them up to give the total time taken to run the 12 filters from GamingHeaven batch.
Check out the results below.
The results below are given in seconds, so the lower the number the better.
The AMD CPU accomplished all the tasks in 276.55 seconds, while it took the Intel CPU 288.45 seconds to perform the same tests, i.e., it took he AMD processor 11.9 seconds less to complete the test.
[nextpage title=”DVD Shrink 3.2.0.15″]
Similar to the test with Photoshop, this is a benchmarking that uses real applications in order to show the performance of the processors we reviewed.
DVD Shrink is a program to create backups of DVDs. It can remove the DVDs copy protection, compact and reduce its size in order to let you Record a double layer DVD9 movie into a single layer DVD5.
Our benchmarking consisted in measuring the amount of time the system took to create an ISO image of the film “Syriana” to the hard disk. For this test, we used all the program’s default options.
Check out the results below.
Once more, the lower the number the better, since the value indicates the number of minutes. It took Sempron 3000+ 1 minute and 69 seconds less to complete the DVD image creation to the hard disk, finishing the task in 22 minutes and 43 seconds, while it took the Intel CPU 24 minutes and 12 seconds.
[nextpage title=”Overclocking”]
Both motherboards used on the tests were simple boards, so they had few overclocking options. Even so it was still possible to have an Idea of the CPUs overclocking potential.
All of them have the clock multiplier locked.
With our Sempron 3000+, which runs internally at 1.8 GHz multiplying its 200 MHz base clock by 9, we managed to increase its base clock to 250 MHz, which made it run internally at 2.25 GHz, an increase of 25% over standard clock.
By using this overclocking, AMD processor accomplished the Super Pi benchmarking in 42.765 seconds, time 24.18% lower than that of the processor operating at its original frequency. A good result.
Figure 7: Sempron 3000+ (1.800 MHz) running at 2.250 MHz (250 MHz x 9).
With our Celeron D 331, which runs internally at 2.66 GHz multiplying its 133 MHz external clock by 20, we managed to increase its external clock to 165 MHz, which made it run internally at 3.30 GHz, an increase of 23.78% over standard clock.
Running at 3.3 GHz, Celeron D accomplished the Super Pi benchmarking in 64.8 seconds, time 14.84% lower than that of the processor operating at its original frequency.
Figure 8: Celeron D 331 (2.660 MHz) running at 3.300 MHz (165 MHz x 20).
AMD CPU benefited more from the frequency increasing, despite the fact that the Intel CPU achieved higher increase in megahertz than the AMD CPU (634 MHz vs. 450 MHz).
[nextpage title=”Conclusions”]
If Intel rules in the high-end market, AMD has proven to be a very respectful competitor in the low-end market.
Even with lower operating frequency, Sempron 3000+ performed well and was much superior to the Intel processor in almost all the tests.
In the benchmarking with Office apps the AMD CPU was in average 26% faster than the Intel CPU. Therefore it’s the best choice if you’re going to use the PC for Internet, texts, spreadsheets and office work.
If you use Photoshop, the AMD processor is also the best choice. It accomplished the 12 benchmarking tasks in 11.9 seconds less than the Intel processor.
In games AMD advantage was even bigger. Sempron 3000+ was up to 73% faster on Doom 3, using the same video card.
Besides being faster, the AMD processor consumes less power, heats less, makes less noise (since its fan is less demanded) and doesn’t demand special cases or power supplies.
So, for the moment, the best solution for low cost PCs is AMD, until Intel launches a low cost processor based on its new Core architecture.