Where does 64-bit help?

Although the performance that will sell the Athlon 64 today has nothing to do with this, the 64-bit part of the equation will definitely play a role in the processor's future. With no final release of the 64-bit version of Windows XP, there is no popular OS support (we will touch on Linux support as well as Win64 support shortly) and no real application support at this time, but where will the 64-bitness of the Athlon 64 help?

There are three main categories that you can split up the performance benefits into: 32-bit applications running on a 32-bit OS, 32-bit applications on a 64-bit OS and 64-bit applications on a 64-bit OS; we will be analyzing each one of these scenarios individually.

Case 1: 32-bit apps under a 32-bit OS

At the launch of the Athlon 64, the predominant operating environment will be running 32-bit applications under a 32-bit OS. All performance benefits the K8 architecture will show here are courtesy of the on-die memory controller, improved branch predictor, higher clock speed and more robust TLBs - none of the performance improvements you'll see in this case will have anything to do with the 64-bit capabilities of the processor.

Case 2: 32-bit apps under a 64-bit OS

When Windows XP 64-bit Edition is officially released (a public beta is due out at the time of publication), many users will be running their 32-bit applications under the 64-bit OS.

Outside of the performance improvements that we just outlined in Case 1, there are a couple of additional benefits the Athlon 64 may offer users. Currently under Windows, although you have a physical memory limit of 4GB, any given process can only use up to 2GB of memory; the remaining 2GB is reserved for use by the OS. With the 32-bit applications under a 64-bit OS scenario, each 32-bit application could be given a full 4GB of memory to work with, instead of being limited to the 2GB Windows process size limitation. Unfortunately this benefit isn't really "plug 'n play" as the application would have to be aware that it can use the added memory, which in the vast majority of cases would require a new patch to be made available.

The second benefit the Athlon 64 could offer in this scenario comes from the availability of additional registers. Although the 32-bit application would still only be compiled to use the regular set of 8 general purpose registers and standard set of FP and SSE2 registers, the 64-bit OS would be able to reference and use all of the registers at its disposal. The performance benefits that you would see here exist in any sort of task handling that the OS would be doing (switching between applications) as well as just regular Windows performance. Granted that the performance improvements seen here should be negligible, considering the extra overhead that does exist when running 32-bit applications in a 64-bit environment (more on this in a bit).

Case 3: 64-bit applications under a 64-bit OS

The final scenario is the one that shows the most promise, yet has the least amount of application support today - running a 64-bit app under a 64-bit OS. Here, the benefits are numerous; not only do you get the performance improvements courtesy of the Athlon 64's architecture, but each application now has full access to the increased number of registers and each application can use much more than 4GB of memory.

Although the Athlon 64 can support 64-bit memory addressability, for demand reasons it only supports 40-bit of physically addressable memory - or ~137GB, not exactly a limiting factor at this point.

The performance improvements developers are expecting to see under this final scenario has been estimated to be in the 10 - 20% range in tasks that are not memory bound, meaning those areas where the application is using less than 2 - 4GB of memory in the first place will still see sizable performance gains courtesy of the availability of more registers. We will investigate a few of these scenarios to substantiate (or refute) these claims later on in the article.

Performance improvements where you are memory bound will be even more impressive; just think about how slow swapping to disk is and how much faster keeping everything in memory makes your computer.

An Early Christmas present from AMD: More Registers AMD's Gem: Athlon 64
Comments Locked

122 Comments

View All Comments

  • AgaBooga - Tuesday, September 23, 2003 - link

    Where is the P4EE in the memory tests?
  • Anonymous User - Tuesday, September 23, 2003 - link

    Personally this was rather anti-climatic for me. It's certainly not a Intel killer that all the hype proclaimed. AMD for business, Intel for content, and a throwup for gaming. Same as it has been for awhile.
  • Anonymous User - Tuesday, September 23, 2003 - link

    #27 & #28 (amd fanboy double post)

    It SHOULD be up there with the P4EE because the PRESCOTT will be coming right around the corner! Face it, AMD did not put out a killer and Intel is sitting pretty in 2004.
  • Anonymous User - Tuesday, September 23, 2003 - link

    #20 are you serious? Did you just comment in the forum without looking at the review or did you actually look at the review. AMD is not "lagging" behind Intel. They are right up there with them. Look at the benchmarks and you will see the CURRENTLY AVAILABLE Athlon64 easily matches a NOT CURRENTLY AVAILABLE P4EE.
  • Anonymous User - Tuesday, September 23, 2003 - link

    #20 are you serious? Did you just comment in the forum without looking at the review or did you actually look at the review. AMD is not "lagging" behind Intel. They are right up there with them. Look at the benchmarks and you will see the CURRENTLY AVAILABLE Athlon64 easily matches a NOT CURRENTLY AVAILABLE P4EE.
  • Anonymous User - Tuesday, September 23, 2003 - link

    AMD, Pamela Anderson called. She want's to know how she can get a bust as big as yours. I have two words for AMD- "Segway" and "Scooter."
  • Anonymous User - Tuesday, September 23, 2003 - link

    nForce3 performance bug

    Time to re-do the benchmarks, Anand.

    Your FX-51 benchmarks are inaccurate.

    http://www20.tomshardware.com/cpu/20030923/athlon_...

    Nvidia: NForce-3 Bug

    The extremely low AGP performance of the NForce3 can be clearly attributed to problems with the HyperTransport channel interface to the Northbridge. That is proven by the benchmark results and the performance differences of up to 33.2 percent. Details about this can be found in the benchmark section of this article.

    Originally, Nvidia had planned to also integrate a SATA RAID controller in the Southbridge. Although the controller is included in the current NForce 3, Nvidia deactivated this feature. The reason was that error-free operation was not possible. For this reason, we decided to use additional boards based on the VIA K8T800 chipset.

    Nvidia (Athlon 64 FX, or alternatively GeForce FX - related names) may be a more high-profile partner for AMD than VIA. However, we would point out that VIA, with the K8T800 chipset, currently offers a clearly better solution for the Athlon 64.

  • Anonymous User - Tuesday, September 23, 2003 - link

    What is that smell?

    AMD just let loose with a huge turd!
  • Anonymous User - Tuesday, September 23, 2003 - link

    #4 You may be right (I don't think so but let say you are), but then ask yourself - where is the Pentium 4 Extreme Edition? There is no mention of this CPU at Intel web site at all, there is no datasheet and no batch numbers. Today, it is only a prototype CPU, such as Prescott is. They managed to build few Gallatin B1 cores that are able to work at this frequency and then remarked them. This CPU is not reality, only OEMs can buy it in very limited quantities, but end users can't. I think a 3 GHz Athlon 64 FX on 90nm prototype would perform far the best in this review... and it would be the same policy as with this Pentium 4 Extreme Edition.
  • Anonymous User - Tuesday, September 23, 2003 - link

    #17 Answers:

    1. Athlon 64's memory controller is very fast as you can see from the benchmarks. Dual channel is only needed in some situations to give decent performance. HT operates at 800 MHz with DDR and 16 bits, thus giving 3.2 GB/s each way (6.4 GB/s). Not so bad for a I/O and AGP interface only bus

    3. S754 is a lower end platform while S940 is an Opteron platform. AMD will introduce S939 early next year and will continue to produce CPUs for all those sockets. S940 A64 FX will, however, disapper in the end of next year.

    6. HyperTransport "Tunnel" system allows for practically unlimited number of chipset combinations, thus a PCI Express will only require to add another Tunnel or integrate it into current chipsets.

Log in

Don't have an account? Sign up now