Original Link: https://www.anandtech.com/show/1308
AMD Athlon 64 2800+: A Cheaper Newcastle
by Derek Wilson on April 27, 2004 7:00 AM EST- Posted in
- CPUs
Welcome to the Newcastle based Athlon 64 solution that runs at a clock speed of 1.8GHz (200MHz slower than the Athlon 64 3000+): the AMD Athlon 64 2800+.
The Athlon 64 2800+ has somewhat made its way stealthily into the market place. As is usually the case, the higher performance (more expensive) parts are the ones that companies push the hardest and enthusiasts are most interested in. It is always exciting to read about how fast something can get done, or new possibilities with emerging technology, but it isn't practical to expect everyone to run out and buy the highest performing chunk of silicon available (no matter how much we all may want to do so).
Often, price is much more important to a purchasing decision than pure performance, especially in the business world where even small price differences can add up very quickly with volume purchases. The trick has always been to find the best value for the money, which is much easier said than done. For inexpensive performance, the currently available option is the Athlon XP line of processors.
Until the cheaper Athlon 64 based Athlon XP solutions come around, we will have to hope that lower performance, lower priced Athlon 64 processors will be able to deliver the performance that we expect from the current generation of hardware at prices that will play nicely with others.
Is It Cheap Enough?
The prices that we have been seeing around the internet put the AMD Athlon 64 2800+ at near USD$185 shipped. This will put the chip squarely in a comfortable upper mid-range (or lower high end?) tier at just about the same price as the 800MHz FSB Intel Pentium 4 2.8GHz parts. This price might not make the new 2800+ cheap enough for everyone's budget, but the price does fit the market segment, and now, people who were looking in this price range will have another option.The next cheapest Athlon 64 system is the 3000+ (which was also the first 512kB L2 based Athlon 64), whose street price is somewhere around $240, while the Barton 3000+ Athlon XP counterpart is available for something like $130. When the math is done, what we end up seeing is that the price of the Athlon 64 2800+ falls just about between the two flavors of 3000+ offered by AMD.
Today, we will be taking the opportunity to see if the performance of the AMD Athlon 64 2800+ will maintain a performance level worthy of its price. The unique price layout of these processors means that the value of the chip (considering price and performance) will be readily apparent from the individual benchmarks; the performance of the 2800+ should fall somewhere near the average score of the two 3000+ models. Of course, this will be different on different benchmarks, as each chip has its advantages and disadvantages.
It is also important to note that prices do not usually scale linearly. When a brand new high end chip comes out, it can often be priced much higher than its performance gain over the previous leader would warrant. At the same time, near the bottom of the spectrum, a small change in price can lead to a larger percent increase in performance as a CPU's perceived "value" approaches what it cost to make the chip. Prices also fluctuate greatly over time, so our comparison of averages will be more of an interesting indicator than a hard and fast rule.
Performance Test Configuration
Performance Test Configuration | |
Processor(s): | AMD Athlon 64 2800+ AMD Athlon XP 3000+ AMD Athlon 64 3000+ AMD Athlon 64 3200+ AMD Athlon 64 3400+ AMD Athlon 64 FX51 AMD Athlon 64 FX53 Intel Pentium 4 3.2GHz EE Intel Pentium 4 3.4GHz EE Intel Pentium 4 3.2GHz Intel Pentium 4 3.0GHz Intel Pentium 4 2.8GHz Intel Pentium 4 3.2EGHz Intel Pentium 4 3.0EGHz Intel Pentium 4 2.8EGHz |
RAM: | 2 x 512Mb OCZ 3500 Platinum Ltd 2 x 512Mb OCZ 3200 EL ECC Registered 2:3:3 2 x 512Mb Mushkin ECC Registered High Performance 2:3:2 |
Hard Drives | Seagate 120GB 7200 RPM (8MB Buffer) |
Video AGP & IDE Bus Master Drivers | VIA Hyperion 4.51 (12/02/03) Intel Chipset Drivers |
Video Card(s): | Sapphire ATI Radeon 9800 PRO 128MB (AGP 8X) |
Video Drivers: | ATI Catalyst 4.1 |
Operating System(s): | Windows XP Professional SP1 |
Motherboards: | Intel D875PBZ (Intel 875P Chipset) FIC K8-800T (VIA K8T800 Chipset) ASUS SK8V (VIA K8T800 Chipset) |
The Athlon XP 3000+ is a Barton based processor running on a 333MHz FSB. The "E" appended to the speed of Intel's processors indicate a Prescott based core. The Athlon 64 3000+ is a 512kB L2 Clawhammer (half the cache is disabled) that should have similar performance to similarly clocked Newcastle based processors (of course, when we have Newcastle based 3000+ processors available, we'll test this).
General Usage & Content Creation Performance
Business Winstone 2004
For our first look at a benchmark, the performance of the 2800+ falls short of our average by 0.6. Not a very big number in terms of Winstone performance (and certainly any margin of error is amplified when comparing one number to two others), but it is 37.5% of the total range between the 3000+ models off the mark.
Content Creation Winstone 2004
Here, the average score of the 3000+ processors is 30.7 while the Athlon 64 2800+ scored a 30.6, which puts it right in line with our average performance for an average price theory.
These benchmarks possibly indicate that the 2800+ is a better value when it comes to content creation style computing rather than business computing (or perhaps that the XP line of processors has a better relative integer performance).
SYSmark 2004
The entire SYSmark run, which includes all the individual tests that we will be reporting, took anywhere between two and three hours to run. The two main suites (Internet Content Creation and Office Productivity) each took up half of the time (in contrast to Winstone, where the business benchmark completes much faster than content creation).SYSmark 2004 Overall
The overall score, which takes a little of everything into account, shows that the Athlon 64 2800+ falls just one "sysmark" shy of the average of the two 3000+ processors, which is exactly where we would expect it to fall based on its pricing.
We included these numbers for the sake of completeness, but we still aren't sure about the value and meaning of these numbers. We do know that this time around, AMD did have equal input into the creation of the benchmark, and while the FX-53 does put in a good showing for AMD, the benchmark seems to favor Intel's architectures.
SYSmark 2004 Internet Content Creation
SYSmark 2004 3D Content Creation
SYSmark 2004 2D Content Creation
SYSmark 2004 Web Publication
SYSmark 2004 Office Productivity
SYSmark 2004 Communications
SYSmark 2004 Document Creation
SYSmark 2004 Data Analysis
DirectX 9 Performance
As we have seen before, processor speed has much less impact on DirectX 9 games than does video card power. Most of the differences in these numbers are negligible, but Aquamark shows some interesting numbers when comparing the performance of Athlon 64 based processors and the Athlon XP.Aquamark
As we can see, most of the frames per second scores are too close together to really mean anything, except where the Athlon XP 3000+ is concerned. In this benchmark, the 3000+ lags its nearest neighbor (the Athlon 64 2800+) by about 8.3%. The 2800+ is much closer to the 512kB L2 based 3000+ in this benchmark.
Aquamark tries to isolate the impact of the CPU and GPU on overall performance by giving different scores for both. We can't really be sure what this score is based on, but it is interesting to look at when trying to understand the effect of different CPUs on the benchmark. We can see once again that the Barton 3000+ lags its 64 bit brother by quite a good margin.
Halo
GunMetal
DirectX 8 Performance
Unreal Tournament 2003 Performance
Unreal Tournament 2004 is coming down the line very shortly, and we will be adding that to our benchmarking suite as soon as possible. For now, we will stick with the venerable UT2K3 benchmark in the usual flyby and botmatch flavors.Here, we see that the trend set in the DX9 games continues into DX8 with the Athlon 64 processors much closer in performance than the Athlon XP processor.
The trend continues when looking at botmatch performance (which is more heavily CPU dependant than the flyby).
WarCraft III: The Frozen Throne
And still more of the same with the Newcastle 2800+ outperforming the average of the two 3000+ processors by about 27% of their range.
OpenGL Performance
Quake 3: Arena
Once again, the Q3A engine is ubiquitous enough among games that it warrants being included in the benchmarks. On a side note, this benchmark is interesting as there is a change in concavity in the graph of our benchmark data. This gives us an example of a benchmark where the fast get faster and the slow get slower.With our average performance number here being 387.6, the Athlon 64 2800+ has beaten the price/performance curve once again (this time by only 4% of the range).
Jedi Knight: Jedi Academy
Wolfenstein: Enemy Territory
In all the OpenGL games, price/performance of the 2800+ is marginally better than linear in the price range and performance of the two 3000+ models.
DivX 5.1.1 Encoding
Once again, we have used Gordian Knot and the documentation available at Doom9 (http://www.doom9.org).The performance of the 2800+ falls dead center between the 3000+ models in this benchmark. We are still going to see performance on the low end of this benchmark as AMD processors don't perform as well as Intel processors when faced with the task of encoding data.
3D Rendering Performance
3D Studio Max R5
Both the XP 3000+ and the 64 2800+ perform poorly in this test, but oddly, the 3000+ beats out the 2800+ in this test. This is the only test where the Athlon XP outperforms any of the Athlon 64 based processors.
Lightwave 7.5
But we can see that under Lightwave, the tables are turned. The Barton Athlon XP 3000+ performs horribly in comparison to the other processors in this test. The 2800+ isn't a screamer either, but it is much better suited to this benchmark than the 3000+. It is likely that Lightwave is sensitive to memory speed, and memory speed differences could have had a part to play in this dichotomy in 3D rendering performance.
Update: It has been pointed out that Lightwave makes heavy use of SSE2 when available, which explains the numbers we see in this benchmark.
Development Workstation Performance
Here, we have a tie between the Barton 3000+ and the Newcastle 2800+, but it should be noted that this benchmark is not very granular. All of these scores are based on the third of three consecutive builds in order to make sure all necessary files are cached.
Final Words
Our makeshift measure of relative price/performance has shown that the 2800+ is a solid value for the money. The processor performs similarly to the slightly faster 512kB L2 Clawhammer based 3000+, and it offers consumers another entry point into the current generation of Athlon 64 processors.As new processors continue to emerge, we will start seeing prices falling on the older generation of hardware, which will likely change the dynamics of the price/performance issue. Athlon 64 based solutions will still have the value of a relative longevity over the Athlon XP platform, especially considering the fact that current entry level Athlon 64 based processors are equal only by the top of the line Athlon XP processors.
Down the line, the Athlon XP for socket 754 will help push along the transition toward current generation technology, but those who are really interested in longevity and upgradeability will wait for the 939 socket coming out later this year. The past couple of years have seen a few too many socket and platform changes to really leave any meaningful upgrade path, but the industry has yet to build a system that is truly upgradeable as time goes on. One of the big problems is finding a way to build a platform that leads current CPU designs enough to support the needs of future technology. Maybe someday.
The cost of the 2800+ may still be a little too much for everyone, and the average cost of a K8 motherboard is also higher than a nice nForce 2 board. As new sockets and platforms begin to emerge, it will be important for AMD and Intel to realize that affordable entry level systems will be key in maintaining high market penetration for their products.
But for now, if you're in the market for a sub $200 processor that doesn't sacrifice on performance, the Newcastle Athlon 64 2800+ will deliver some very nice bang for your buck.