News:

Willkommen im Notebookcheck.com Forum! Hier können sie über alle unsere Artikel und allgemein über Notebook relevante Dinge disuktieren. Viel Spass!

Main Menu

Post reply

The message has the following error or errors that must be corrected before continuing:
Warning: this topic has not been posted in for at least 120 days.
Unless you're sure you want to reply, please consider starting a new topic.
Other options
Verification:
Please leave this box empty:

Shortcuts: ALT+S post or ALT+P preview

Topic summary

Posted by Anon
 - March 16, 2020, 23:38:31
Quote from: Curious on March 16, 2020, 03:54:34
The intend of the author is very suspicious.   A managing editor will put the comparison of a low power mobile cpu to desktop chip in the subtitle?

To catch more views or simply showing off his IQ?

The author seems to bias more to Intel. Check out his other article, where he compared the upcoming AMD's desktop 4000 series APU (4 core?) with the mobile 4000 series APU (which has much lower TDP, even though it has more cores).
Posted by Danny 90210
 - March 16, 2020, 15:54:46
Intel has very long history of cheating the benchmark...

And this result in 15 years damaging in customer money... crating monopoly and stay the last 5 years in 14nm

If you supporting this... you supporting monopoly and anti innovation
Posted by daniel creanga
 - March 16, 2020, 08:40:35
This is just a stupid article...
Posted by Curious
 - March 16, 2020, 03:54:34
The intend of the author is very suspicious.   A managing editor will put the comparison of a low power mobile cpu to desktop chip in the subtitle?

To catch more views or simply showing off his IQ?

Posted by James4591
 - March 16, 2020, 01:29:18
Single thread performance isn't even important for workloads any longer. UNIX and UNIX-like operating systems and environments have been using multithread workloads for years, and only Windows has finally decided to do the same.

Applications are taking advantage of multithread workload features of COUs to spread the workload and boost processing capabilities for better performance.

Single thread is only useful for testing purposes. It had no valid functional usage in applications other than a staunch developer refusing to get with the times.

Passmark isn't even relevant anymore. Geekbench and other tools have far better results and can really show what your CPU is able to do.
Posted by Dick Licker
 - March 16, 2020, 00:36:51
This pathetic excuse for an article should be taken down and thrown in the trash.

Lrn2test you Intel fanboys.

Nuff said...
Posted by abag
 - March 15, 2020, 21:59:33
are you people literally retarded? like literally dropped on your head when little stupid?

you are comparing a desktop processor against a mobile U (extra efficient processor) you incredible goshddarn moerons!

just because passmark didn't test the 4000 version desktop processor doesn't mean intel is better.
Posted by Samuel F Baniul
 - March 15, 2020, 20:36:35
Total, honest, we are not tied.

This site blows.

No.

This site.

No. Not us.
news site.

This item blows.

Not news.

No, we are like Trump.

We can say anything, and well , generate cash like a banshee.

  Intell customer service service announcement.

Spend around $450 more and we can give you a 6%  increase.

Well worth the cash.

Intel fan boys , well suck it.

Here is a , supposed , reliable, news source.

AND Published.

on an archive.

For everyone, too look back on.

This is only saved in an archive.

Sad times my friends.


Intell:


We are the the better.

This site just showed HOW much they need advertising . Do Not Trust these Journalists/ News site anymore.

They EVEN published the findings.

They did this !.!.!

And keep changing the posts to suit their needs.

Hey,. !

How can I get these dollars so my daughter doesn't die from cancer?

Fucking liars.. 


Block these so called journals. Money, hungry, pussys. From your life.

And let them come back too reply.


You are my only lifeline to this world. Please help me?


Posted by eeee
 - March 15, 2020, 20:32:35
Intel: synthetic benchmarks are worthless.
Passmark: releases new version.
Intel: synthetic benchmarks are all that matter.
Posted by james delorsa
 - March 15, 2020, 20:10:17
Quote from: Shepard on March 15, 2020, 19:49:55
To the lost herd: AVX 512 has nothing to do with that result, as 9900ks does not support it. Although Intel does have the single thread performance lead, we are not talking bulldozer level gaps. Actually we do have even AMD leading ST performance for select workloads, but why have passmark selected the least favorable approach to AMD?

Intel does have the single threaded lead, but its solely limited to clock speed and Turbo values. (You can also include potential OC headroom as intel's chip do much better in terms of max frequency all around.)

Ryzen 3000 has higher IPC clock for clock over Skylake, but Skylake on 14++++(do i need more plus'?) clocks much higher and thus has the advantage typically.

The results are obviously skewed in a way that makes intel look better than they actual are.  Like.. I'm not disagreeing that a CPU like the 9900KS shouldn't be in the number one spot, but the testing and overall results make no sense.
Posted by james delorsa
 - March 15, 2020, 19:56:09
Uhh... this isn't very credible or their testing things that are clearly skewed towards intel via clock speed settings.

I was using a 7700k since launch and my new 3900x completely decimates the thing.. Even in single threaded work loads, CPU-Zs variation of single threaded performance testing is over 50 points higher than my 7700k.. which is reflective of reports that the 3000 series has slightly higher IPC over  Skylake's multiple revisions..  it's a 4.3 vs 4.4 with "turbo" without "manually overclocking".  My applications and REAL world use also reflect this result.

Just look at the clock speeds tested.. This isn't a very credible and they're purposely placing CPU's higher and lower than they should be.  7700k has a 4.2 base speed sure, but the thing literally runs at 4.5Ghz in a single core work load.. 4.4 multi core turbo without touching anything in the bios assuming you're using at least a half way decent $20-30 cooler and not throttling. The 9000 series chips run at anywhere from 4.6-5.0 in the same situation. 

I just find it hard to believe that a 3.8ghz base clock 3900x is significantly behind a 4.0ghz base clock 9900k? Like what the hell are you even testing in your proprietary software?
Posted by Shepard
 - March 15, 2020, 19:49:55
To the lost herd: AVX 512 has nothing to do with that result, as 9900ks does not support it. Although Intel does have the single thread performance lead, we are not talking bulldozer level gaps. Actually we do have even AMD leading ST performance for select workloads, but why have passmark selected the least favorable approach to AMD?
Posted by Lars Lund
 - March 15, 2020, 19:42:11
Intel aee cheatinbg.
Anyone ever running one of many programs, called cheatmachine u can speed up the cpu speed,
So lets take a eksempel thay easy to understand, u are running a commodore 64 emulator. The emulator lets u run the game x10 faster
But hard to play 10 times faster then it suppose to go.
So u running a old demo but 10 times faster
Now u start benchmark program in cheat machine, 3dmark. Running cpu speed 1,3x enouge to beat amd. But it is cheating.
And cheat machine are detected by 3dmark anti cheat system, but northing prevent to write there own "cheay machine"
Posted by Weylenn
 - March 15, 2020, 18:30:58
Of course intel leads single core, but when the 3700x exceeds a 7700 in single core and it is nowhere to be found.. That's a little odd. And the gap isn't as big as it used to be. Another problem is that AVX512 is limited to HEDT as an instruction set, so... I don't see the point in including a desktop processor.
Posted by PAVEL BONEV
 - March 15, 2020, 17:58:53
Except you are not buying single thread CPU, check this out - AMD Threadripper is unbeatable so far. The new top of the line
Intel Core i9-10980XE - 32,561 just a position after
AMD Ryzen 9 3900X - 32,829.
AMD Ryzen 9 3950X - 39,417.
Check PassMark high end CPU chart.