Monday, October 21st 2019

AMD Ryzen 9 3950X Beats Intel Core i9-10980XE by 24% in 3DMark Physics

AMD's upcoming Ryzen 9 3950X socket AM4 processor beats Intel's flagship 18-core processor, the Core i9-10980XE, by a staggering 24 percent at 3DMark Physics, according to a PC Perspective report citing TUM_APISAK. The 3950X is a 16-core/32-thread processor that's drop-in compatible with any motherboard that can run the Ryzen 9 3900X. The i9-10980XE is an 18-core/36-thread HEDT chip that enjoys double the memory bus width as the AMD chip, and is based on Intel's "Cascade Lake-X" silicon. The AMD processor isn't at a tangible clock-speed advantage. The 3950X has a maximum boost frequency of 4.70 GHz, while the i9-10980XE isn't much behind, at 4.60 GHz, but things differ with all-core boost.

When paired with 16 GB of dual-channel DDR4-3200 memory, the Ryzen 9 3950X powered machine scores 32,082 points in the CPU-intensive physics tests of 3DMark. In comparison, the i9-10980XE, paired with 32 GB of quad-channel DDR4-2667 memory, scores just 25,838 points as mentioned by PC Perspective. Graphics card is irrelevant to this test. It's pertinent to note here that the 3DMark physics test scales across practically any number of CPU cores/threads, and the AMD processor could be benefiting from a higher all-core boost frequency than the Intel chip. Although AMD doesn't mention a number in its specifications, the 3950X is expected to have an all-core boost frequency that's north of 4.00 GHz, as its 12-core sibling, the 3900X, already offers 4.20 GHz all-core. In contrast, the i9-10980XE has an all-core boost frequency of 3.80 GHz. This difference in boost frequency, apparently, even negates the additional 2 cores and 4 threads that the Intel chip enjoys, in what is yet another example of AMD having caught up with Intel in the IPC game.
Sources: TUM_APISAK (Twitter), PC Perspective
Add your own comment

143 Comments on AMD Ryzen 9 3950X Beats Intel Core i9-10980XE by 24% in 3DMark Physics

#52
The Egg
xtreemchaos
this is rubbing salt into my cuts , daughters car has popped its engine today and ive just 10mins ago gave my 3950x money for a deposit for another one, woe is life your flying high and something kicks us in the gutts :banghead: . ive still got hope but my 2700x is going to last me a wee bit longer.
the 3950x is looking good.
Fate has certainly dealt you a cruel hand. Forced to survive, possibly for months, with only a mere 2700x. That's not a life anyone should have to live.
Posted on Reply
#53
xtreemchaos
The Egg
with only a mere 2700x
yep I have to be thankful for small mercys bro :)
Posted on Reply
#54
freeagent
Good for you, AMD. Way to get your shit together :clap:
Posted on Reply
#55
techguymaxc
The 10980xe physics score is nonsense.

My 7960x at 4GHz on all cores and with garbage RAM at only 2400 (and with horrible timings) scores 28615 in the physics test with a ton of open background applications and an instance of Plex Media Server running. Sorry, but a loss of 200MHz (with a gain of 2 cores) isn't going to drop the score 3000 points.

https://www.3dmark.com/3dm/40354402?

And before all the AMD zealots jump in, I'm cheering for AMD this round. I definitely want to build a 64-core Threadripper 3000 system.
Posted on Reply
#56
Xx Tek Tip xX
techguymaxc
My 7960x at 4GHz on all cores and with garbage RAM at only 2400 (and with horrible timings) scores 28615 in the physics test with a ton of open background applications and an instance of Plex Media Server running. Sorry, but a loss of 200MHz (with a gain of 2 cores) isn't going to drop the score 3000 points.
+1.
Posted on Reply
#57
EarthDog
techguymaxc
The 10980xe physics score is nonsense.

My 7960x at 4GHz on all cores and with garbage RAM at only 2400 (and with horrible timings) scores 28615 in the physics test with a ton of open background applications and an instance of Plex Media Server running. Sorry, but a loss of 200MHz (with a gain of 2 cores) isn't going to drop the score 3000 points.

https://www.3dmark.com/3dm/40354402?

And before all the AMD zealots jump in, I'm cheering for AMD this round. I definitely want to build a 64-core Threadripper 3000 system.
mmhmm...Mine at 4.4 GHz beat it...

https://www.techpowerup.com/forums/threads/amd-ryzen-9-3950x-beats-intel-core-i9-10980xe-by-24-in-3dmark-physics.260317/page-2#post-4136780
Posted on Reply
#58
ZoneDymo
EarthDog
You do realize that hot and power consumption are different things, right? For example, let's take fire... let's compare a lighter and a bonfire, both with yellow flames...which do you think would be more difficult to put out (has more energy)... even though the temps are the same?

I've had a 90C 5W IC and my 200W CPU runs cooler....while associated, don't let it fool you... the processor will still run where it is supposed to. ;)
Not sure what you are on about tbh, power consumption = heat in the world of processors, TDP means nothing.
Posted on Reply
#59
Solaris17
Dainty Moderator
Hm, I take this with a grain of salt. I'm not saying its wrong, and the differences between the new and old arcs I PERSONALLY find negligible to justify upgrade. But something about this seems off.

My 7980xe only boosts to 4.2 (In the test it determined max freq was only a little over 4ghz mind you). And while im not saying the 10980xe could beat ryzen, and while I know its not an architectural wormhole of difference I Find it hard to believe that chip only managed only a few thousand point faster than my CPU at bone stock (non over clocked) with not only a 400mhz boost and ark advantage but prime testing conditions.

I had a multitude of things open and like 5 different messaging platforms.

https://www.3dmark.com/3dm/40354603?
Posted on Reply
#60
ZoneDymo
Solaris17
Hm, I take this with a grain of salt. I'm not saying its wrong, and the differences between the new and old arcs I PERSONALLY find negligible to justify upgrade. But something about this seems off.

My 7980xe only boosts to 4.2 (In the test it determined max freq was only a little over 4ghz mind you). And while im not saying the 10980xe could beat ryzen, and while I know its not an architectural wormhole of difference I Find it hard to believe that chip only managed only a few thousand point faster than my CPU at bone stock (non over clocked) with not only a 400mhz boost and ark advantage but prime testing conditions.

I had a multitude of things open and like 5 different messaging platforms.

https://www.3dmark.com/3dm/40354603?
Where did we even get the results from the 10980xe from?
Posted on Reply
#61
Vayra86
fynxer
Sorry for saying this but OMG, you really don't realize that if Intel does not sell they will HAVE TO lower prices to match performance per dollar.

You really think they can run a cpu product line up and say, hey, it does not matter if it is not selling, let's just put the cpu's in a warehouse and let them collect dust.

Intel DOES NOT HAVE THE UPPER HAND ANYMORE, so they do not have the luxury to be more expensive.

Just for a real world check, Intel has already LOWERED THEIR PRICE TWICE in main stream. WHY would they lower price twice if they have no problem with being more expensive.

I would be so bold to say that Intel are already planing for a third price cut in the main stream segment.


AMD top main stream cpu with PCIe 4.0 (PLEASE observe i am stating MAIN STREAM) is more powerful than Intel's (so called HEDT) top cpu. So you are telling me you would PAY MORE for Intel HEDT 18 core with PCIe 3.0 when you can get an AMD 16 core with PCIe 4.0 that performs better for far less.

Get real man.
Wait and see. The other explanation is that Intel has a LOT of wiggle room, not just in pricing but also in their long term contracts, and overall market penetration. They really don't have to go mental on the consumer side of things.
Posted on Reply
#62
Solaris17
Dainty Moderator
ZoneDymo
Where did we even get the results from the 10980xe from?
Forgive me for being naive, but I'm not sure if this is a play on how bad PC perspective is, or if you actually missed it in the first post.
Posted on Reply
#63
ZoneDymo
Solaris17
Forgive me for being naive, but I'm not sure if this is a play on how bad PC perspective is, or if you actually missed it in the first post.
no no, I got that, but I find it odd we have a screenshot of the 3950x run but not for the 10980XE, where is it?
Posted on Reply
#64
Solaris17
Dainty Moderator
ZoneDymo
no no, I got that, but I find it odd we have a screenshot of the 3950x run but not for the 10980XE, where is it?
idk, but that might even play into my point. I like get the Intel hate, and while I personally have nothing against AMD (having 2 generations of ryzen in my house) I find it a bit hard to believe, not that it couldn't beat the new chip, but that it is only so marginally better than my score on a loaded system.

Seems odd. Id look at it the same way if AMD lost the bench, it just doesn't make a ton of sense. With that kind of gap, I could beat the physics score doing a clean boot and overlocking a few hundred mhz and probably still keep my clocks under the 10980xe at that!
Posted on Reply
#65
Super XP
ratirt
Apparently he doesn't understand that. He is one of those people that will say: "Intel was always expensive and "competitive", why change that? Intel will always be more expensive because that's the way it always has been. It doesn't matter the new lineup sucks. It will still cost twice that much because it is Intel. "
That's because those that are not in the KNOW, which is the majority of people, see Intel CPUs as better quality CPUs over AMD. People even go back to the days of old when comparing both Intel and AMD, and claim AMD has various issues with its processors, then gives an example of something that happened back in the year 2000 LMAO. The majority of people are CLUELESS.

Though with the age of the internet, more and more people are doing there own research. Today more people recognize AMD as a high quality CPU & GPU designer, but still that past false stigma of lesser quality over Intel still rides on people's minds.

That said, just picture Intel as a stubborn child that refuses to eat its food for example, well that is Intel on there stance to even have competition.
I've read somewhere several months ago off (Seeking Alpha? I think) where many that hold Intel stock were complaining that AMD should never have been allowed to release such a processor such as ZEN, as to compete with Intel, because that prevented the Intel stock price from reaching a high potential. Anyhow my point is the arrogance of the actual company INTEL and its Stockholders is ludicrous., they WHINE like a bunch of freaking Babies.
Posted on Reply
#66
xtreemchaos
for me its what ever gives you the best performance for the best price on the day, ive never been into the red or blue camp or green or red but it a bit academic for me now as ive spent me money :( but who knows what ill be getting for Christmas if I drop the hints to her in doors ;)
Posted on Reply
#67
EarthDog
ZoneDymo
Not sure what you are on about tbh, power consumption = heat in the world of processors, TDP means nothing.
Maybe I'm explaining it wrong...

I don't think I mentioned anything about TDP. But 100W power used on Intel and AMD processors doesn't equal the same temperature (for many reasons). Look back in the day... FX chips, at using more watts than Intel and Intel yet the Intel's ran with a higher temperature. As I said, I had a 5W IC on a mining ASIC board run 90C... it's more than power used that equal the hot temps.

The analogy was to show the different amounts of energy for the same temperature. :)

... but that isn't really what this thread is about so... I digress (PM box is open. ;)).
Posted on Reply
#68
The Egg
After watching/reading GN's multiple recent pieces on TDP just within the past week, it's very clear that manufacturer TDP numbers mean jack squat.
Posted on Reply
#69
techguymaxc
ZoneDymo
no no, I got that, but I find it odd we have a screenshot of the 3950x run but not for the 10980XE, where is it?
This is why several Intel HEDT high-core-count users have posted their own results in this thread. The claimed 10980xe results are too low to be believable.
Posted on Reply
#70
EarthDog
techguymaxc
The claimed 10980xe results are too low to be believable.
Or just wondering exactly how low its all core boost is to be that slow. :)

Maybe 3.6 GHz or something? I can see 4.2-4.2GHz with two less c/t and better SMT efficiency catch up...
Posted on Reply
#71
techguymaxc
EarthDog
Or just wondering exactly how low its all core boost is to be that slow. :)

Maybe 3.6 GHz or something? I can see 4.2-4.2GHz with two less c/t and better SMT efficiency catch up...
Specs for the 10980xe have already leaked. 3.8GHz all-core boost, 4.8GHz maximum Turbo Boost 3.0 clocks. That's not slow for an HEDT chip. Not by a long shot. Previous generation HEDT chip base clocks are often in the high 2GHz - low 3GHz range. This is actually quite a beastly chip.
Posted on Reply
#72
EarthDog
I've only seen the base at 3.6.. I would've hoped for a bit more than 4.8 ghz all core... but yes, improved over last gen, no doubt.
Posted on Reply
#74
techguymaxc
EarthDog
I've only seen the base at 3.6.. I would've hoped for a bit more than 4.8 ghz all core... but yes, improved over last gen, no doubt.
Apparently all the rumors were wrong and the base clocks are actually only 3.0GHz, same as preceding generations!
https://ark.intel.com/content/www/us/en/ark/products/198017/intel-core-i9-10980xe-extreme-edition-processor-24-75m-cache-3-00-ghz.html

Which means this all makes a lot more sense now. Oh well, almost no one buys an HEDT chip and doesn't overclock, so I don't really care about stock performance.
Posted on Reply
#75
ShrimpBrime
Vya Domus
Who cares about the scores. What's impressive is that both CPUs perform in the same ballpark but the 3950X uses probably like what, half the power ?
Its being compares to a 3950x?
Why not compare the AMD 3970x instead? Maybe only 2/3s the power then?
Posted on Reply
Add your own comment