Twice the Cache - 17% Higher Latency

Both the Pentium 4 6xx and the new Extreme Edition share the same core, meaning they also have the same L2 cache.  When Intel first launched Prescott we noticed that in the move to the new architecture that cache latencies went up tremendously.  The increase in cache latencies was to be expected, as one tradeoff of a larger cache is that it takings longer to find and access data.  So when we heard that Intel was moving to a 2MB L2 cache with the 6xx series, we wondered how much slower the cache would get.

First we wanted to confirm that L1 cache latencies stayed the same, and they did at 4 cycles for the new Prescott 2M based core:

   Cachemem L1 Latency  ScienceMark L1 Latency
AMD Athlon 64 3 cycles 3 cycles
Intel Pentium 4 (Northwood) 1 cycle 2 cycles
Intel Pentium 4 (Prescott) 4 cycles 4 cycles
Intel Pentium 4 (Prescott 2M) 4 cycles 4 cycles
Intel Pentium M 3 cycles 3 cycles

Next up, was L2 cache latency.  In our review of the Pentium M processor on the desktop we discovered that its 10 cycle L2 cache was responsible for its solid performance in non "media rich" applications (e.g. office applications, OS performance).  The original Prescott had a 23 cycle L2 cache, and with a 2MB cache the latency has gone up to 27 cycles:

   Cachemem L2 Latency  ScienceMark L2 Latency
AMD Athlon 64 17 cycles 18 cycles
Intel Pentium 4 (Northwood) 16 cycles 16 cycles
Intel Pentium 4 (Prescott) 23 cycles 23 cycles
Intel Pentium 4 (Prescott 2M) 27 cycles 27 cycles
Intel Pentium M 10 cycles 10 cycles

While we're talking about "only" 4 cycles, at 3.6GHz that's 17% longer to access data from L2 cache.  Given Prescott's extremely lengthy pipeline, a 17% increase in L2 cache latency is not going to help minimize the downsides of such a long pipeline.  Also keep in mind that the only architectural change here is a larger L2 cache, so none of the normal tricks to help hide memory latencies are expanded upon in the new Pentium 4. 

What Intel is counting on is that the increase in hit rate provided by a 100% larger cache will outshine the 17% longer access to L2 cache.  Did Intel make the right bet?  In order to find out we took the new Pentium 4 660 (3.6GHz - 2MB L2) and compared it to the old Pentium 4 560 (3.6GHz - 1MB L2), with all other variables the same, let's see how much of an impact the extra megabyte of cache has in the real world.

In the business category, we see the added cache paying off a little. SYSMark shows good improvement in the document creation portion of its tests, while the Business Winstone makes some very good gains. Worldbench shows web browsing with Mozilla to have improved a good bit while our compression test and the ACDSee test show a loss in performance. These losses generally indicate areas where the test is more dependant on latency than cache hit rate. On the content creation side, adding Windows Media Encoder to the Mozilla test improves performance more than the individual Mozilla test. This is likely due to the fact that the large cache keeps Mozilla's data from being kicked out while Windows Media Encoder is working.

On the gaming front, Doom 3 is the only test we saw with any performance improvement. And the only other application to show a significant performance gain is Maya with more than a 43% gain. The huge gain in performance under Maya is likely a result of 1MB of cache being too small to fit models in while 2MB is enough. This seems to be a case where the test is very bandwidth sensitive rather than latency sensitive. Dropping most (if not all) of the data being worked on into the L2 cache offers a program a very large boost in apparent bandwidth.

As we can see, the unfortunate truth for performance on the 600 series is that most consumer data sets can fit into a 1MB cache just fine. The added cache does seem to help with multitasking from our limited investigation of the subject. The more threads that hit memory aggressively, the better chance we have of seeing a benefit from the 2MB cache. This is because less data from each thread will be kicked out of the cache, resulting in fewer pipeline stalls.

Unfortunately, most usage models that are a good fit for the 600 series are server and workstation workloads. Streaming data (using or encoding media), games, and most other consumer applications don't have the lots of big data requirement that can really separate the performance of the 1MB and 2MB parts.

As we've provided this chart and gone through the general impact of the benchmarks on Intel's new 600 line, we won't include analysis on the pages with our benchmark data. For those who are interested in a deeper look at the numbers and performance of all 5 new parts, graphs of each benchmark are included later in this article.

 Impact of L2 Cache Size on Performance (1MB vs. 2MB - 3.60GHz)
   1MB L2  2MB L2  2MB Performance Advantage
Business/General Use Performance
Business Winstone 2004 21.4 24.2 13.0%
SYSMark 2004 - Communication 137 137 0.0%
SYSMark 2004 - Document Creation 201 218 8.4%
SYSMark 2004 - Data Analysis 184 186 1.0%
Microsoft Office XP with SP-2 522 520 0.3%
Mozilla 1.4 459 422 8.0%
ACD Systems ACDSee PowerPack 5.0 547 558 -2.0%
Ahead Software Nero Express 6.0.0.3 545 550 -0.9%
WinZip Computing WinZip 8.1 412 411 0.2%
WinRAR 479 469 -2.0%
Multitasking Content Creation Performance
Content Creation Winstone 2004 32.7 33.9 3.7%
SYSMark 2004 - 3D Creation 231 231 0.0%
SYSMark 2004 - 2D Creation 288 279 -3.1%
SYSMark 2004 - Web Publication 206 203 -1.0%
Mozilla and Windows Media Encoder 676 601 11.1%
Video/Photo Creation & Editing
Adobe Photoshop 7.0.1 342 342 0.0%
Adobe Premiere 6.5 461 468 -1.5%
Roxio VideoWave Movie Creator 1.5 287 276 3.8%
Audio/Video Encoding
MusicMatch Jukebox 7.10 484 470 2.9%
DivX Encoding 55.3 55.4 0.2%
XviD Encoding 33.9 33.4 -1.4%
Microsoft Windows Media Encoder 9.0 2.57 2.56 -0.3%
Gaming
Doom 3 84.6 88.6 4.7%
UT2004 59.3 60.4 1.9%
Wolfenstein: ET 97.2 95.5 -1.7%
3D Rendering
Discreet 3dsmax 5.1 (DX) 268 266 0.7%
Discreet 3dsmax 5.1 (OGL) 327 329 -0.6%
SPECapc 3dsmax 6 1.64 1.62 -1.1%
Professional 3D
SPECviewperf 8 - 3dsmax-03 17.04 17.11 0.4%
SPECviewperf 8 - catia-01 13.87 13.57 -2.2%
SPECviewperf 8 - light-07 14.3 13.83 -3.3%
SPECviewperf 8 - maya-01 13.12 18.85 43.7%
SPECviewperf 8 - proe-03 16.7 16.5 -1.2%
SPECviewperf 8 - sw-01 13.09 13.33 1.8%
SPECviewperf 8 - ugs-04 15.31 13.82 -9.7%


Index An Interesting Observation: Prescott 2M's Die
POST A COMMENT

71 Comments

View All Comments

  • KingofL337 - Tuesday, February 22, 2005 - link

    Why would anyone buy a P4 for EMT64? When AMD64 is
    a full implementations of 64 not just a poor incomplete copy?

    In Soviet Russia, Computer Reboot You!

    Reply
  • Zebo - Tuesday, February 22, 2005 - link

    "The original Prescott was a Sunday launch. "

    Well see what I mean..:)
    Reply
  • Hans Maulwurf - Tuesday, February 22, 2005 - link

    And remember hyperthreading - it uses only a small area on the die and increases power consumption significantly!

    Oh, and I would still like to know weather it ws 1T or 2T on the Athlon.
    Reply
  • Viditor - Tuesday, February 22, 2005 - link

    Reflex - First, you may very well be quite correct!
    Second, not all parts of the die are equal...for example, the ALU runs at twice the clockspeed of the core. The areas affected by 64bit modes MAY be disproportionally higher than the rest of the CPU (I really don't know, which is why I'm asking for a test...).
    Third, the design for 64bit on Intel is quite different that on AMD. AMD designed the chip to be hybrid from the ground up, Intel had to "retrofit" their Netburst architecture to accomodate it...while they both function very similarly, their incorporation into the chip is quite dissimilar (e.g. AMD has no double-pumped ALU)
    Reply
  • DerekWilson - Monday, February 21, 2005 - link

    Intel often does sunday launches --

    The original Prescott was a Sunday launch.

    And there was at least a couple others that I can't recall at the moment.

    I've seen other sites say something to the affect of this being a sneaky launch, and I think don't think that is accurate.

    I, for one, would prefer Intel not launch parts on a Sunday. But that's how its been and likely how it will be. :-/
    Reply
  • Zebo - Monday, February 21, 2005 - link

    What's up with Intel sneaking around in the dark for..Sunday night launches on a holiday weekend told me all I needed to know about this new chip release.

    In sure they still sell billions over AMD but the message is clear from enthusiasts prosective. If you want performance, quiet and cool you buy AMD A64's.(ageing I might add)
    Reply
  • johnsonx - Monday, February 21, 2005 - link

    In Soviet Russia, message clears YOU!


    damn, it's hard to stop....
    Reply
  • johnsonx - Monday, February 21, 2005 - link

    The message is clear: Soviet Russia has failed.

    Enough already... let's move on to the next catch phrase (if there must be a next one...).

    Reply
  • Reflex - Monday, February 21, 2005 - link

    Vidiator - 64bit uses very little die space, even once activated its not going to consume any really noticable amount of power. On an Athlon64 its estimated to be about 10% of the core. Considering how much larger a P4 core is it would account for even less percentage wise. I am not including cache in that measurement either which as you saw accounts for 50% or better.

    So technically it may draw a watt or two, but its not going to change the results significantly...
    Reply
  • Viditor - Monday, February 21, 2005 - link

    danidentity - "Power consumption is not going to change depending on whether you're running 64-bit apps or not"

    Is there a reason you expect this? My own rationale is that Intel (I'm assuming here) probably dials down a few things (ALU logic, BISTs, unneeded repeaters, etc...) unless the CPU is operating in 64bit only mode or compatiblity mode (as opposed to legacy mode).
    As an example, AMD64 doesn't use the extra registers unless it's in one of the 64bit modes...

    I don't know if there will be a swing in power consumption, but I am curious to see any empirical evidence one way or the other...

    Derek - Thanks for the heads up on the test (Powernow usage...)!
    Reply

Log in

Don't have an account? Sign up now