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 Braggingrights
 - April 16, 2020, 20:55:19
So Intel win gaming again
Posted by hsgu
 - April 16, 2020, 01:31:08
I agree this seem to be misleading.
Posted by Alex544
 - April 14, 2020, 02:55:44
Quote from: gruffi on April 14, 2020, 01:11:38
Quote from: Alex544 on April 13, 2020, 11:50:57
a quick search on geekbench results shows the i9-9900 scoring nearly 40,000  multicore and 6000+ single core
would the i9-10900F with 2 extra cores and 4 extra threads somehow score less than that, on average? I think not! Let me downclock my Ryzen 5 2600 to 2ghz, run Geekbench, and make a title called "Zen+ Ryzen 5 2600 gets clobbered by 6 year old i7-3770!!!"
Geekbench scores can vary a lot. Especially from OEM to OEM. Search for HP scores of the i9-9900. You will find something like ~5700 single core and a little more than 30k multi core. So, the results of the i9-10900F look legit.

And don't get confused. Nowadays Intel is heavily limited by their 14nm process and thermal throttling. Especially with lower TDPs. Under full load it doesn't matter that much if 8 cores run at 3.5 GHz or 10 cores run at 3 GHz. Scores will be quite similar. A 10-core design also means more uncore logic. Which could limit the power budget for all cores.
It's "legit" but it's very misleading to say Intel's upcoming desktop flagship (locked) CPU is being put to "shame" by AMD's mobile CPU when it's scoring less than even Intel's very own i9-9900 with 2 less cores.
And no, 2 extra cores won't drop the frequency by 500mhz lol. 4.6ghz all-core for the i9-10900 has been confirmed, which makes its all-core boost right in line with the i9-9900, despite having 2 more cores. Desktop CPUs aren't as thermally (and consequently power) limited as mobile CPUs. Unless you're using the crappy stock cooler, you can expect the i9-10900 to score higher than its predecessor.
Posted by gruffi
 - April 14, 2020, 01:11:38
Quote from: Alex544 on April 13, 2020, 11:50:57
a quick search on geekbench results shows the i9-9900 scoring nearly 40,000  multicore and 6000+ single core
would the i9-10900F with 2 extra cores and 4 extra threads somehow score less than that, on average? I think not! Let me downclock my Ryzen 5 2600 to 2ghz, run Geekbench, and make a title called "Zen+ Ryzen 5 2600 gets clobbered by 6 year old i7-3770!!!"
Geekbench scores can vary a lot. Especially from OEM to OEM. Search for HP scores of the i9-9900. You will find something like ~5700 single core and a little more than 30k multi core. So, the results of the i9-10900F look legit.

And don't get confused. Nowadays Intel is heavily limited by their 14nm process and thermal throttling. Especially with lower TDPs. Under full load it doesn't matter that much if 8 cores run at 3.5 GHz or 10 cores run at 3 GHz. Scores will be quite similar. A 10-core design also means more uncore logic. Which could limit the power budget for all cores.
Posted by Augirre
 - April 13, 2020, 21:13:20
What we are seeing with Intel is thermal throttling. Finally, even robust cooling options cannot keep up with a 14nm 10 core chip. A 9900k gets a higher score because it's not drawing insane power to get their single core up to unattainable clocks. If you limited the chip to 4.8ghz, it would probably fare a ton better, and maybe even operate in the advertised TDP.

The AMD chip happily follows TDP pretty closely, and dissipating that small amount of heat is relatively easy, even in a laptop. The Intel 65w part is likely drawing 200w, meaning the OEM probably didn't oversize the cooling system enough to account for Intel's ignoring of power envelope.

There are limits to how much heat you can realistically dissipate out of a TIM as small as an i9. And Intel can't abandon it's cooler compatibility (Tim size) without pissing even more people off.
Posted by polievka
 - April 13, 2020, 20:55:22
Quote from: Padmakara on April 13, 2020, 12:49:39
Looks like Apple is moving to Amd, on geekbench 5 there are a lot of imacs with 3700x-3900x ryzen.
These are very good news.
iMacPro1,1 Ryzen 3900x
1335
Single-Core Score
12279
Multi-Core Score

Those are Hackintoshes, jesus.
Posted by Padmakara
 - April 13, 2020, 12:49:39
Quote from: Alex544 on April 13, 2020, 11:50:57
a quick search on geekbench results shows the i9-9900 scoring nearly 40,000  multicore and 6000+ single core
would the i9-10900F with 2 extra cores and 4 extra threads somehow score less than that, on average? I think not! Let me downclock my Ryzen 5 2600 to 2ghz, run Geekbench, and make a title called "Zen+ Ryzen 5 2600 gets clobbered by 6 year old i7-3770!!!"
Looks like Apple is moving to Amd, on geekbench 5 there are a lot of imacs with 3700x-3900x ryzen.
These are very good news.
iMacPro1,1 Ryzen 3900x
1335
Single-Core Score
12279
Multi-Core Score
Posted by Alex544
 - April 13, 2020, 11:50:57
a quick search on geekbench results shows the i9-9900 scoring nearly 40,000  multicore and 6000+ single core
would the i9-10900F with 2 extra cores and 4 extra threads somehow score less than that, on average? I think not! Let me downclock my Ryzen 5 2600 to 2ghz, run Geekbench, and make a title called "Zen+ Ryzen 5 2600 gets clobbered by 6 year old i7-3770!!!"
Posted by Padmakara
 - April 13, 2020, 11:50:09
There is a lot of lobby not to use amd in high end devices.
But look at the bright side, If you consumers continue to ask OEMs for Amd in high end laptops, and mid range with better screens and not going for Intel systems, they will be Forced to build the laptops with Amd that consumers want, maybe from next year.
But people should ask for that...
Otherwise nothing will change and Intel has a lot of money to give.
Also amd gpus 5600m and 5700m will come soon, so people will have the equivalent of 2060 and 2070. Unfortunately they won't compete with nvidia's super...
So only the cosumers can change the current situation asking for better screens and high end devices with AMD
Posted by toven
 - April 13, 2020, 11:37:27
Quote from: Dharan on April 13, 2020, 11:08:50
I never imagined that there will be a time in the future where we will compare an AMD mobile processor with an Intel Desktop processor for performance parity...  :o :o  It is a hard lesson for anyone who sits lazy on their success.  8) You better run now rather than being forced in the future to do catch up.  8) The later is more painful...  ::)
They are being forced to do catch up now. Increasing cores and clock to their maximum resulting in much higher TDP and BOM.
Posted by fdsofldmos
 - April 13, 2020, 11:23:40
It is all thanks to TSMC.
Posted by Dharan
 - April 13, 2020, 11:08:50
I never imagined that there will be a time in the future where we will compare an AMD mobile processor with an Intel Desktop processor for performance parity...  :o :o  It is a hard lesson for anyone who sits lazy on their success.  8) You better run now rather than being forced in the future to do catch up.  8) The later is more painful...  ::)
Posted by Redaktion
 - April 13, 2020, 10:47:50
A 10-core Intel Core i9-10900F desktop processor has been spotted as part of an HP system on Geekbench, where it racked up strong single-core and multi-core scores. Unfortunately for the upcoming Comet Lake chip, its multi-core score lagged behind the result we recorded for an AMD Ryzen 9 4900HS in an Asus Zephyrus G14 laptop.

https://www.notebookcheck.net/10-core-Intel-Core-i9-10900F-desktop-CPU-does-well-in-Geekbench-multi-core-test-but-then-gets-put-to-shame-by-8-core-AMD-Ryzen-9-4900HS-mobile-APU.461162.0.html