CPU Tests: Synthetic

Most of the people in our industry have a love/hate relationship when it comes to synthetic tests. On the one hand, they’re often good for quick summaries of performance and are easy to use, but most of the time the tests aren’t related to any real software. Synthetic tests are often very good at burrowing down to a specific set of instructions and maximizing the performance out of those. Due to requests from a number of our readers, we have the following synthetic tests.

Linux OpenSSL Speed: SHA256

One of our readers reached out in early 2020 and stated that he was interested in looking at OpenSSL hashing rates in Linux. Luckily OpenSSL in Linux has a function called ‘speed’ that allows the user to determine how fast the system is for any given hashing algorithm, as well as signing and verifying messages.

OpenSSL offers a lot of algorithms to choose from, and based on a quick Twitter poll, we narrowed it down to the following:

  1. rsa2048 sign and rsa2048 verify
  2. sha256 at 8K block size
  3. md5 at 8K block size

For each of these tests, we run them in single thread and multithreaded mode. All the graphs are in our benchmark database, Bench, and we use the sha256 results in published reviews.

(8-3c) Linux OpenSSL Speed sha256 8K Block (1T)(8-4c) Linux OpenSSL Speed sha256 8K Block (nT)

GeekBench 5: Link

As a common tool for cross-platform testing between mobile, PC, and Mac, GeekBench is an ultimate exercise in synthetic testing across a range of algorithms looking for peak throughput. Tests include encryption, compression, fast Fourier transform, memory operations, n-body physics, matrix operations, histogram manipulation, and HTML parsing.

I’m including this test due to popular demand, although the results do come across as overly synthetic, and a lot of users often put a lot of weight behind the test due to the fact that it is compiled across different platforms (although with different compilers).

We have both GB5 and GB4 results in our benchmark database. GB5 was introduced to our test suite after already having tested ~25 CPUs, and so the results are a little sporadic by comparison. These spots will be filled in when we retest any of the CPUs.

(8-1d) Geekbench 5 Multi-Thread(8-1c) Geekbench 5 Single Thread

CPU Tests: Legacy and Web CPU Tests: SPEC
Comments Locked

126 Comments

View All Comments

  • YB1064 - Monday, January 4, 2021 - link

    Thermals look horrendous! 102 C even when throttled is bad. Even with water, I don't think you will see much improvement. Better have a chiller on hand!
  • GeoffreyA - Monday, January 4, 2021 - link

    The sun should buy a few million of these fellows. Will help a lot with temperature when sunspots get too cool.
  • at_clucks - Monday, January 4, 2021 - link

    Nooo, as the article puts it, it's "not as bad as it sounds". People just needlessly freak out when they processor FREAKING BOILS WATER on "an open test bed with a chunky copper cooler" and needing 270W at full load. But in reality it's just "more about thermal gradients inside the processor and how easily the thermals can move" so it's all good. I can feel the temps dropping just by reading these comforting words.

    I guess the only thing that can make a current gen Intel CPU look good is another current gen Intel CPU. Here's hoping that they get their act together for the next gen(s) or we'll keep having to read reviews where Intel competes against Intel as the only way to get some praise.
  • at_clucks - Monday, January 4, 2021 - link

    Read the comment above as if there was an edit button to fix all the grammatical... inconsistencies.
  • 1_rick - Monday, January 4, 2021 - link

    It's no worse than the article itself.
  • Smell This - Monday, January 4, 2021 - link

    as the article puts it, it's "not as bad as it sounds"
    _______________________________________

    Sadly, the presentation makes it much worst. Bad enough smoked with the Ryzen 5900x ... there is little to no downward price pressure on an 18-month old AMD 3900x at $499.

    Even worst --- 18 months ago the HEDT Intel 12/24 i9-79xxX was $1,190 and was slobber-knocked by the 3900x . . .
  • GeoffreyA - Wednesday, January 6, 2021 - link

    "Here's hoping that they get their act together"

    Rocket Lake should cover the ground performance-wise, but it's going to be disastrous when it comes to power and heat. Would be nice to see Sunny Cove running on 7 or 5 nm, but I suspect even there, Zen 3 would use less power, owing to its more economical design.
  • powerarmour - Monday, January 4, 2021 - link

    It's a genuinely horrible CPU compared to the competition, Intel have regressed so much on performance-per-watt that even Apple are stealing their lunch now.
  • shelbystripes - Monday, January 4, 2021 - link

    It’s embarrassing (for Intel) how easily Apple justified the Intel-to-ARM switch with the M1. Stagnation is a bitch.
  • Operandi - Monday, January 4, 2021 - link

    Part stagnation part major technological fumbles. Intel definitely played it safe on the architectural level but you could argue that was the right call given the state of the market. What is really killing them is the dropped ball on the fabrication front.

    Regardless though Apple's move to their own designs was going to happen even if Intel was stumbling all over the place, that move has been planned for over a decade.

Log in

Don't have an account? Sign up now