• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

In Wake of Controversy, Intel-Paid Principled Technologies Retesting AMD Ryzen Processors

Raevenlord

News Editor
Joined
Aug 12, 2016
Messages
3,755 (1.18/day)
Location
Portugal
System Name The Ryzening
Processor AMD Ryzen 9 5900X
Motherboard MSI X570 MAG TOMAHAWK
Cooling Lian Li Galahad 360mm AIO
Memory 32 GB G.Skill Trident Z F4-3733 (4x 8 GB)
Video Card(s) Gigabyte RTX 3070 Ti
Storage Boot: Transcend MTE220S 2TB, Kintson A2000 1TB, Seagate Firewolf Pro 14 TB
Display(s) Acer Nitro VG270UP (1440p 144 Hz IPS)
Case Lian Li O11DX Dynamic White
Audio Device(s) iFi Audio Zen DAC
Power Supply Seasonic Focus+ 750 W
Mouse Cooler Master Masterkeys Lite L
Keyboard Cooler Master Masterkeys Lite L
Software Windows 10 x64
Well, that proverbial storm of dirty undies did serve to rile up some people over at Intel and their paid-for stint with Principled Technologies, whose name begs for a change for honesty's sake. In wake of the controversy regarding its... flawed... testing of AMD's Ryzen 2700X performance in gaming workloads, Principled technologies has now come forward to say it's retesting AMD's processors in less... biased circumstances.

Let's start with the glass half-full part of this retesting: initial reports of memory timings on AMD's system being set in an almost "whatever" kind of way apparently weren't fair, since Principled Technologies have said they used D.O.C.P. settings for AMD's XMP-equivalent memory settings (not properly disclosed in the initial report, so, it's their own fault this happened). The good stuff ends there, though; numerous other flaws in the methodology, such as the usage of AMD's stock cooling solutions against a Noctua cooler for the Intel system (which they'll now fix on retesting), and the usage of AMD's Game Mode on their consumer Ryzen processors, which meant the usually 8-core processor was working in a 4-core mode (really, now?)... The company will now retest both CPUs in a more even-footed way. How's that for a change?



View at TechPowerUp Main Site
 
Preorder disappeared from amazon. I am disappoint.

Glad they are retesting though. I wanna see old vs new side by side comparison.
 
Question: How can/will you ever trusted a site with so many flaws (and get paid by...) in regarding to their testing methodology?
 
I dont. I usually wait for reviews here. And even then, I care very little beacuse if I want something, I just get it.
 
This is called repairing the damage after the ship has sunk.

Like... stop wasting time. We know you're unreliable....
 
While they are @ it, why not invite Steve from Gamers Nexus and Steve from Hardware Unboxed to be present and supervise the benchmarks being run? That way, we can be sure them benchmarks are properly done!
 
Last edited:
I dont. I usually wait for reviews here. And even then, I care very little beacuse if I want something, I just get it.

LOL, I appreciate that perspective very much. :D

9900K will be so awesome if you have not updated your PC in a few generations. Those like me, with a sweet 8700K at 5.3Ghz and an 8086K at 5.4Ghz, the 9700K and 9900K will most likely just trade blows and mostly just overlap in Single Core performance. Multi-core of course the 8-cores will have an advantage over the 6-cores.

Yet, all my work apps are single and slightly threaded, so my sweet spot might just REMAIN at 6cores 12threads all the way until 10nanometer and 2020.

But yea, we really need to wait for independent reviews, still the Cine ST I don't believe there will be much difference with the 8700K 8086K 9900K or 9700K, plus or minus a very small amount, in either direction at least at stock vs stock.

How high the 9700K and 9900K will clock will tell "the other" story. LOL :)

I've got my 8086K 5.5Ghz Cine ST stable with a score of 236. The 9700K might beat that, maybe - maybe not, but not significantly enough for a real NEED to purchase a new CPU.


But like you said, if you REALLY want it for whatever reason, buy it and be happy. LOL :p


For getting work completed quickly, enthusiasts should really find out what their most used apps will benefit from most of the time. High Frequency or High Core Count (HCC).


This screenshot below is my most used daily work application, and as can be easily seen, it's a heavy single-threaded/slightly-threaded workload. More cores are NOT going to help me at all, only higher frequencies. More cores would actually slow my work down.

Yet from a hobby aspect, of course we should all buy whatever we want and be happy. Wooooooooo:toast:
 

Attachments

  • 5.3Ghz while typing.jpg
    5.3Ghz while typing.jpg
    42.8 KB · Views: 351
  • Cinebench Single Thread 5.5Ghz stable 8086K.png
    Cinebench Single Thread 5.5Ghz stable 8086K.png
    862.1 KB · Views: 428
Last edited:
"AMD's Game Mode on their consumer Ryzen processors, which meant the usually 8-core processor was working in a 4-core mode"

AMD's software does that? So it's not smart enough to know that it's Ryzen and not just cut half the cores?
 
While they are @ it, why not invite Steve from Gamers Nexus and Steve from Hardware Unboxed to be present and supervise the benchmarks being run? That way, we can be sure them benchmarks are properly done!
Um Steve, from GN, lives virtually next door however PT ain't paying the other Steve, from down under, for his trip that's for sure.
 
AMD's software does that?

It doesn't, you need to explicitly set that option. You can also disable an arbitrary amount of cores 2/4/6/8, among many other things. It's a simple configuration tool that has most of the functionality you get from the BIOS. And it's a tool that only the most hardcore of the bunch would care to fiddle with so it doesn't need to be smart, this is not meant to be used by the average user. That being said you don't do testing like this, you test things out of the box if you really want to pertain that you are 100% fair.
 
Last edited:
LOL, I appreciate that perspective very much. :D
Youre welcome :laugh:

9900K will be so awesome if you have not updated your PC in a few generations. Those like me, with a sweet 8700K at 5.3Ghz and an 8086K at 5.4Ghz, the 9700K and 9900K will most likely just trade blows and mostly just overlap in Single Core performance. Multi-core of course the 8-cores will have an advantage over the 6-cores.
Im mainly interested in it so I can crunch.
 
It doesn't, you need to explicitly set that option. You can also disable an arbitrary amount of cores 2/4/6/8, among many other things. It's a simple configuration tool that has most of the functionality you get from the BIOS. And it's a tool that only the most hardcore of the bunch would care to fiddle with so it doesn't need to be smart, this is not meant to be used by the average user. That being said you don't do testing like this, you test things out of the box if you really want to pertain that you are 100% fair.

That's such a shame people cannot find an extra 30minutes to tune a processor that will benefit them for years and years to come, both AMD and Intel via motherboard makers provide performance boosts that are FREE and only a few clicks away.

This per-core boost I set up (below) can also be used by the average user, it's just some easy maneuvering in bios that will gain anyone much more performance - an increased Single Core Turbo Boost, much faster than Intel provides out of the box.

Your processor will still return to idle with SpeedStep enabled.

I'm running this schemata on Noctua single-stack Air, no need for water cooling or anything fancy.

It's a shame enthusiasts do not spend a few minutes taking advantage of both AMD's and Intel's extra FREE features.

Every 8700K/8086K will not be able to create a 5.6GHz single core Turbo Boost, but every 8700K/8086K can most likely create a sweet 5.3Ghz single core boost set up like 5.3-5.2-5.1-5.0-5.0 5.0.

You do not need to own a 5.3Ghz all cores all threads stable 8700K or 8086K to create a 5.3Ghz single core 5.2Ghz dual core 5.1Ghz triple core etc. super boost setup.

Only need to spend some quality time with your processor and find out what it's capable of. :)

Set it up, and SAVE to overclock profile number one. :toast:

Same personal custom Turbo Boosts can be applied with the upcoming 9700K and 9900K, easy and so simple. :peace:
 

Attachments

  • Intel Turbo Boost Max Technology 4.0 LOL.jpg
    Intel Turbo Boost Max Technology 4.0 LOL.jpg
    150.3 KB · Views: 463
  • IMG_4112.JPG
    IMG_4112.JPG
    1.5 MB · Views: 378
Last edited:
While they are @ it, why not invite Steve from Gamers Nexus and Steve from Hardware Unboxed to be present and supervise the benchmarks being run? That way, we can be sure them benchmarks are properly done!

Yeah they're the white knights we all need!
 
Pre-ordered the 9900K.
 
Pre-ordered the 9900K.

I would have waited until Siliconlottery.com bins a few on October 26th.

But still.

WOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOT! :peace:
 
I don't have the patience for that nor do I wanna pay more of a premium over what I already am.

I also wanted to utilize the Amazon rewards that I had
 
Last edited:
I don't have the patience for that nor do I wanna pay more of a premium over what I already am.

I also wanted to utilize the Amazon rewards that I had

Enjoy your 9900K 8core 16thread beauty to the maximum bro. :p
 
Last edited:
PT seems somewhat out of touch with what a current gamer uses for a rig in 2018. I can’t think of a single person I know of, that has 64GB of RAM in their gaming rig. And I can count the ones with 32GB on one hand. 16GB is the norm.
 
Its good to see that they are going to retest and then should see different results.

The sad part is intel originally agreed with there results by saying that is what we get in our labs as well, and they told PT to use Ryzenmaster on the 2700X.......

So intel you going to retest as well then huh? :shadedshu:
 
Back
Top