Principled Technologies Issues Updated Test Results

Toggle sidebar Toggle sidebar
K

Krazie_Ivan

Honorable
Aug 22, 2012
102
0
10,680
it's nice that PT was immediately receptive to the criticism & retested. it would have been better if they just applied all the changes/fixes, & maybe invited an experienced 3rd party (Tech Jesus comes to mind) to be there for some oversight & prevent further critique. results are still skewed, even if just slightly, but it's beating a dead horse now.

i'd say this fiasco will hopefully teach certain people to wait for multiple independent reviews before ordering, but lets be realistic... most people will never know any of this happened.

(not so) eagerly awaiting all the NDA benchmark results from cherry-picked review samples next week...
 
nitrium

nitrium

Distinguished
Jul 27, 2009
1,552
99
20,590
It's just confirming what everyone already knew, even based off of the old tests: Intel's chip is faster than AMD's equivalent but is an absolutely terrible bang for the buck and thus only attractive to people either obsessed with having the very fastest chip regardless of value or irrational Intel fanbois.
 
D

da.hilloulin

Oct 13, 2018
2
0
10
Yeah, now in many games the 2700x is at 10% or so of the 9900k. The cooling could improve maybe by 5%. And then you are at spitting distance...
The price difference is clearly not in the favour of the 9900k.
I wonder who would buy the 9900k instead of buying the 2700x and putting the extra cash in the GPU instead, or the RAM, or SSD, or anything really.
But once again it will be the "processor of choice" because everyone wants the bigger Amazon commission. (Even though probably it would make more money to recommend to upgrade the GPU)
 
R

rantoc

Distinguished
Dec 17, 2009
1,859
1
19,780
@1080p, who beside perhaps a few who plays in the extreme fps competition plays in that resolution with a rig in the price class those chips will land in? @1440p+ the diff is perhaps a handful frames in the majority of games, sure the intel is better but at double the price... worth it?
 
barryv88

barryv88

Distinguished
May 11, 2010
121
33
18,720
The biggest gripe that I have for a while now, is not even based on the numbers from benchmark results by these tech companies (sorry that includes you Tom's), but the over usage of the word "faster".
In many ways, a beefy chip such as the 2700X is no slouch at gaming whatsoever. And in sheer performance, it falls slightly short from comparative Intel chips on bench results. But does it really mean that these Intel chips are "faster"? If both chips score in the region of say 130-150fps in our favorite games, will every average Joe gamer out there see a difference if you game on both systems and have to tell them apart?
No. No human on this planet possesses such unique abilities to tell that difference. Your FPS is FAR over the 60fps mark. You don't run faster, you don't shoot faster. It's all smooth on both systems. Stop using the "faster" term. Its BS.
 
A

anywhere.craig

Prominent
Sep 13, 2018
3
0
510
Based on an average Australian hardware retailer, the 9900K + Noctua NH-U14S costs $964 while the 2700X with included cooler costs $469. Double the price does not equal double the performance and at higher resolutions the results should be closer. Even factoring the total system cost into consideration the performance per dollar benefit would be marginal.
 
A

AnimeMania

Distinguished
Dec 8, 2014
334
18
18,815


You mean they found exactly what the people who commissioned the study asked them to find.
 
F

fiber4now

Oct 13, 2018
2
0
10
Simply amazing how the AMD fans on this site do not realize the difference between Creators and Game modes are pretty insignificant thus the difference in gaming with 4 or 8 Ryzen cores are also insignificant . AMD realized this thus the two modes in the first place.

One other thing it strikes me as odd that most AMD fanboys claim that PT did not do all the work arounds that a guy from Gamers Nexus would know about but the 9900K and the 2700X will mostly be sold in OEM systems that the user would never try unless AMD itself patched it into their drivers
 
K

Krazie_Ivan

Honorable
Aug 22, 2012
102
0
10,680


Game Mode is for Threadripper, not Ryzen - as noted in the software bundle when you DL it, and by every reputable tech reviewer covering this story. Depending on the game, these modes do impact perf, sometimes by quite a bit. PT's own retest even shows this.

"work arounds" is the wrong choice of terminology, and i suspect you know it (given your use of "fanboys" to disparage). It's not the same as "configuring", or even "optimizations" ...both of which are needed equally on Intel & AMD systems, but were not equally applied by PT - XMP doesn't auto-apply itself on Intel. Correctly configuring a PC is what separates a reputable builder or OEM from un-recommendable poor quality fly-by-night junk. Or in this case, validity of benchmark results.
 
K

kinggremlin

Distinguished
Jul 14, 2009
574
41
19,010


Someone who already has all those things. A lot of those people exist. There are also very specific use cases where the 9900k will dominate the value charts vs anything AMD makes or Intel's HEDT platform like Adobe Premiere when it uses the IGP for acceleration.
 
S

siman0

Distinguished
Jan 16, 2011
89
0
18,630
Tests are still invalid. They are using an 80 dollar cpu cooler on an already more expensive cpu. This is effectively compairing a 600 dollar cpu vs a 300 dollar cpu, and thats not double the preformance. The 9900k looks like **** IMHO...
 
K

kinggremlin

Distinguished
Jul 14, 2009
574
41
19,010


Price doesn't "invalidate" performance tests. PT could have chosen the $4600 AMD Epyc 7601 for the comparison, and it would have gotten destroyed by the 9900k. Still a perfectly valid test, though useless from the standpoint that no one would be cross shopping those 2 CPU's. The fastest gaming chips were chosen from each side to test and a lot of buyers WILL be cross shopping a 2700x and 9900k. It's not the responsibility of performance testers to determine value, that is solely up to the individual buyer.
 
S

steve15180

Distinguished
Dec 31, 2007
34
18
18,535
Actually, it's still not quite right. Aside from not fixing the coolers, they still insisted
on using 64gb dual rank memory without any changes from their original review. So,
they got a big improvement in AMD performance while still leaving quite a bit on the
table. Intel will still be faster, but by even less than this at 1080p. Go up from there,
and it begs the question "What are you paying for?"
 
S

steve15180

Distinguished
Dec 31, 2007
34
18
18,535


Actually, if you have all of those things, you'd be nuts to buy if for any video editing. As with SLI, the programs can't use different GPUs if their not the same. Therefore, you're arguing that the 9900k's IGP would be better than, oh, say a 1080 for acceleration? From what I understand, a 1060 would blow away IGP in Adobe Premiere. Why would I buy a 9900k and disable a discrete graphics care?
 
B

baddog02

Distinguished
Jun 27, 2011
1
0
18,510
Let's not forget that a 2800x is still out there somewhere waiting for this exact moment. This moment where, after additional testing, it only turns out that Intel is around 10% better in most games at twice the price, and only in the 100+ fps range...where most of us are doing 1440 or 4k and are definitely not hitting 100+fps with any card out there, lol. So this test was useless all around, and then it was redone with the stock cooler still in place on the 2700x even AFTER Prancipled's(around the facts) integrity was BEYOND in question. Ha. Just shut down. You're done here. All for a mediocre $600 chip. I hope the payoff was worth it.

Sincerely,
Mr. Waiting here patiently for a $350 2800x that runs 20% slower speeds but still matches a 9900
 
T

t99

Honorable
Jul 16, 2014
756
1
11,215
Who test in 2018 at only 1080p lol. 9900k such a waste of money for gaming this only proved to either buy a i5 8600k, 2700x or 2600/x. 2600 cost 3 times as less and at 1440p with a 1080ti it's going to be so so close in fps, I bet less than 15% on average.

9900k only seems ideal if you are perhaps a paid streamer for high fps games. If you are spending money you work for it's really hard to justify paying double for such minimal gains.

How many people who spend 500$ on a cpu even have a 1080p display?
 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom