CPU Performance: Web and Legacy Tests

While more the focus of low-end and small form factor systems, web-based benchmarks are notoriously difficult to standardize. Modern web browsers are frequently updated, with no recourse to disable those updates, and as such there is difficulty in keeping a common platform. The fast paced nature of browser development means that version numbers (and performance) can change from week to week. Despite this, web tests are often a good measure of user experience: a lot of what most office work is today revolves around web applications, particularly email and office apps, but also interfaces and development environments. Our web tests include some of the industry standard tests, as well as a few popular but older tests.

We have also included our legacy benchmarks in this section, representing a stack of older code for popular benchmarks.

All of our benchmark results can also be found in our benchmark engine, Bench.

WebXPRT 3: Modern Real-World Web Tasks, including AI

The company behind the XPRT test suites, Principled Technologies, has recently released the latest web-test, and rather than attach a year to the name have just called it ‘3’. This latest test (as we started the suite) has built upon and developed the ethos of previous tests: user interaction, office compute, graph generation, list sorting, HTML5, image manipulation, and even goes as far as some AI testing.

For our benchmark, we run the standard test which goes through the benchmark list seven times and provides a final result. We run this standard test four times, and take an average.

Users can access the WebXPRT test at http://principledtechnologies.com/benchmarkxprt/webxprt/

WebXPRT 3 (2018)

WebXPRT 2015: HTML5 and Javascript Web UX Testing

The older version of WebXPRT is the 2015 edition, which focuses on a slightly different set of web technologies and frameworks that are in use today. This is still a relevant test, especially for users interacting with not-the-latest web applications in the market, of which there are a lot. Web framework development is often very quick but with high turnover, meaning that frameworks are quickly developed, built-upon, used, and then developers move on to the next, and adjusting an application to a new framework is a difficult arduous task, especially with rapid development cycles. This leaves a lot of applications as ‘fixed-in-time’, and relevant to user experience for many years.

Similar to WebXPRT3, the main benchmark is a sectional run repeated seven times, with a final score. We repeat the whole thing four times, and average those final scores.

WebXPRT15

Speedometer 2: JavaScript Frameworks

Our newest web test is Speedometer 2, which is a accrued test over a series of javascript frameworks to do three simple things: built a list, enable each item in the list, and remove the list. All the frameworks implement the same visual cues, but obviously apply them from different coding angles.

Our test goes through the list of frameworks, and produces a final score indicative of ‘rpm’, one of the benchmarks internal metrics. We report this final score.

Speedometer 2

Google Octane 2.0: Core Web Compute

A popular web test for several years, but now no longer being updated, is Octane, developed by Google. Version 2.0 of the test performs the best part of two-dozen compute related tasks, such as regular expressions, cryptography, ray tracing, emulation, and Navier-Stokes physics calculations.

The test gives each sub-test a score and produces a geometric mean of the set as a final result. We run the full benchmark four times, and average the final results.

Google Octane 2.0

Mozilla Kraken 1.1: Core Web Compute

Even older than Octane is Kraken, this time developed by Mozilla. This is an older test that does similar computational mechanics, such as audio processing or image filtering. Kraken seems to produce a highly variable result depending on the browser version, as it is a test that is keenly optimized for.

The main benchmark runs through each of the sub-tests ten times and produces an average time to completion for each loop, given in milliseconds. We run the full benchmark four times and take an average of the time taken.

Mozilla Kraken 1.1

3DPM v1: Naïve Code Variant of 3DPM v2.1

The first legacy test in the suite is the first version of our 3DPM benchmark. This is the ultimate naïve version of the code, as if it was written by scientist with no knowledge of how computer hardware, compilers, or optimization works (which in fact, it was at the start). This represents a large body of scientific simulation out in the wild, where getting the answer is more important than it being fast (getting a result in 4 days is acceptable if it’s correct, rather than sending someone away for a year to learn to code and getting the result in 5 minutes).

In this version, the only real optimization was in the compiler flags (-O2, -fp:fast), compiling it in release mode, and enabling OpenMP in the main compute loops. The loops were not configured for function size, and one of the key slowdowns is false sharing in the cache. It also has long dependency chains based on the random number generation, which leads to relatively poor performance on specific compute microarchitectures.

3DPM v1 can be downloaded with our 3DPM v2 code here: 3DPMv2.1.rar (13.0 MB)

3DPM v1 Single Threaded3DPM v1 Multi-Threaded

x264 HD 3.0: Older Transcode Test

This transcoding test is super old, and was used by Anand back in the day of Pentium 4 and Athlon II processors. Here a standardized 720p video is transcoded with a two-pass conversion, with the benchmark showing the frames-per-second of each pass. This benchmark is single-threaded, and between some micro-architectures we seem to actually hit an instructions-per-clock wall.

x264 HD 3.0 Pass 1x264 HD 3.0 Pass 2

CPU Performance: Encoding Tests Gaming: World of Tanks enCore
POST A COMMENT

277 Comments

View All Comments

  • evernessince - Saturday, October 20, 2018 - link

    I'm sure for him money is a fixed resource, he is just really bad at managing it. You'd have to be crazy to blow money on the 9900K when the 8700K is $200 cheaper and the 2700X is half the price. Reply
  • Dug - Monday, October 22, 2018 - link

    Relative to how much you make or have. $200 isn't some life threatening amount that makes them crazy because they spent it on a product that they will enjoy. We spend more than that going out for a weekend (and usually don't have anything to show for it). If an extra 200 is threatening to your lively hood, you shouldn't be shopping for new cpu's anyway. Reply
  • close - Saturday, October 20, 2018 - link

    @ekidhardt: "I think far too much emphasis has been placed on 'value'. I simply want the fastest, most powerful CPU that isn't priced absurdly high."

    That, my good man, is the very definition of value. It happens automatically when you decide to take price into consideration the price. I also don't care about value, I just want a CPU with a good performance to price ratio. See what I did there? :)
    Reply
  • evernessince - Saturday, October 20, 2018 - link

    A little bit extra? It's $200 more then the 8700K, that's not a little. Reply
  • mapesdhs - Sunday, October 21, 2018 - link


    The key point being, for gaming, use the difference to buy a better GPU, whether one gets an 8700K or 2700X (or indeed any one of a plethora of options really, right back to an old 4930K). It's only at 1080p and high refresh rates where strong CPU performance stands out, something DX12 should help more with as time goes by (the obsession with high refresh rates is amusing given NVIDIA's focus shift back to sub-60Hz being touted once more as ok). For gaming at 1440p or higher, one can get a faster system by choosing a cheaper CPU and better GPU.

    There are two exceptions: those for whom money is literally no object, and certain production workloads that still favour frequency/IPC and are not yet well optimised for more than 6 cores (Premiere is probably the best example). Someone mentioned pro tasks being irrelevant because ECC is not supported, but many solo pros can't afford XEON class hw (I mean the proper dual socket setups) even if the initial higher outlay would eventually pay for itself.

    What we're going to see with the 9900K for gaming is a small minority of people taking Intel's mantra of "the best" and running with it. Technically, they're correct, but most normal people have budgets and other expenses to consider, including wives/gfs with their own cost tolerance limits. :D

    If someone can genuinely afford it then who cares, in the end it's their money, but as a choice for gaming it really only makes sense via the same rationale if they've also then bought a 2080 Ti to go with it, though even there one could retort that two used 1080 TIs would be cheaper & faster (at least for those titles where SLI is functional).

    If anything good has come from this and the RTX launch, it's the move away from the supposed social benefit of having "the best"; the street cred is gone, now it just makes one look like a fool who was easily parted from his money.
    Reply
  • Spunjji - Monday, October 22, 2018 - link

    Word. Reply
  • Total Meltdowner - Sunday, October 21, 2018 - link

    This comment reads like shilling so hard. So hard. Please try harder to not be so obvious. Reply
  • Spunjji - Monday, October 22, 2018 - link

    I think they placed just the right amount of emphasis on "value". Your post basically explains why it's not relevant for you in terms of you being an Intel fanboy with cash to burn. I'll elaborate.

    The MSRP is in the realm of irrational spending for a huge number of people. "Rational" here meaning "do I get out anything like what I put in", wherein the answer in all metrics is an obvious no.

    Following that, there are a HUGE number of reasons not to pre-order a high-end CPU, especially before proper results are out. Pre-ordering *anything* computer related is a dubious prospect, doubly so when the company selling it paid good money to paint a deceptive picture of their product's performance.

    Your assertion that Intel have never launched a bad CPU is false and either ignorance or wilful bias on your part. They have launched a whole bunch of terrible CPUs, from the P3 1.2Ghz that never worked, through the P4 Emergency Edition and the early "dual-core" P4 processors, all the way through to this i9 9900K which is the thirstiest "95W" CPU I've ever seen. Their notebook CPUs are now segregated in such a way that you have to read a review to find out how they will perform, because so much is left on the table in terms of achievable turbo boost limits.

    Sorry, I know I replied just to disagree which may seem argumentative, but you posted a bunch of nonsense and half-turths passed off as common-sense and/or logic. It's just bias; none of it does any harm but you could at least be up-front that you prefer Intel. That in itself (I like Intel and am happy to spend top dollar) is a perfectly legitimate reason for everything you did. Just be open and don't actively mislead people who know less than you do.
    Reply
  • chris.london - Friday, October 19, 2018 - link

    Hey Ryan. Thanks for the review.

    Would it be possible to check power consumption in a test in which the 2700x and 9900k perform similarly (maybe in a game)? POV-Ray seems like a good way to test for maximum power draw but it makes the 9900k look extremely inefficient (especially compared to the 9600k). It would be lovely to have another reference point.
    Reply
  • 0ldman79 - Friday, October 19, 2018 - link

    I'm legitimately surprised.

    The 9900k is starving for bandwidth, needs more cache or something. I never expected it to *not* win the CPU benchmarks vs the 9700k. I honestly expected the 9700k to be the odd one out, more expensive than the i5 and slower than the 9900k. This isn't the case. Apparently SMT isn't enabling 100% usage of the CPU's resources, it is allowing a bottleneck due to fighting over resources. I'd love to see the 9900K run against it's brethren with HT disabled.
    Reply

Log in

Don't have an account? Sign up now