AMD Ryzen 5 5600X Takes the Crown of the Fastest CPU in Passmark Single-Thread Results

[TheUn4seen]

Nice.... if true. It's not like AMD ever used untrue claims and "fanboy fantasy" marketing to sell products which turned out to be a huge disappointment, right?.
But again, great if true.

 

[thesmokingman]

I see AMD has finally done it. They've gotten the deniers all in a defensive frenzy. They didn't freak out when it was the 5950x, but no, a 5600x. Cannot have that now! lmao

 

[Valantar]

Nice.... if true. It's not like AMD ever used untrue claims and "fanboy fantasy" marketing to sell products which turned out to be a huge disappointment, right?.
But again, great if true.

Well, this is a Passmark result, not AMD marketing. The result may well be submitted by someone inside AMD with the purpose of having it "leak", but unless they are either LN2 overclocking the chip or outright faking the result, there isn't much they could do to make it look better than it is. Traditional overclocking on Ryzen has so far meant losing ST performance in favor of MT, after all.

 

[8BitZ80]

I see AMD has finally done it. They've gotten the deniers all in a defensive frenzy. They didn't freak out when it was the 5950x, but no, a 5600x. Cannot have that now! lmao

AMD released more budget-oriented processors at the Zen 2 release (3600, 3600X, 3700X, 3800X, 3900X), but this time around they've focussed more on releasing the higher-end processors with Zen 3 (5600X, 5800X, 5900X, 5950X).

The reality is that AMD hasn't even released any bang-for-buck processor SKUs yet, so the accusations of 50% price increases and price gouging tactics are highly exaggerated.

 

[Jism]

When using the default intel spec for power limits its pretty shitty only slightly faster than a 2700X in MT workloads....... Gaming isn't Affected though. Not sure about 21% though either but most Reviews reviewed the chip with all limits removed.

And is'nt having the chip locked to a 65W TDP spec vs a AMD at 65W spec single core, a good sign? At the end of the day it's how the chip performs. And if this is true your looking at a overall winner coming (finally) from team AMD.

 

[TheUn4seen]

Well, this is a Passmark result, not AMD marketing. The result may well be submitted by someone inside AMD with the purpose of having it "leak", but unless they are either LN2 overclocking the chip or outright faking the result, there isn't much they could do to make it look better than it is. Traditional overclocking on Ryzen has so far meant losing ST performance in favor of MT, after all.

The thing is that I wouldn't put faking the result above what this particular company is willing to do in an attempt to generate hype. They were willing to outright lie to their investors and clients not that long ago, so... meh, I'll wait for the actual third party reviews of the actual end-user product before even considering a purchase.

 

[thesmokingman]

The thing is that I wouldn't put faking the result above what this particular company is willing to do in an attempt to generate hype. They were willing to outright lie to their investors and clients not that long ago, so... meh, I'll wait for the actual third party reviews of the actual end-user product before even considering a purchase.

You might want to look in the own mirror. It's ironic that you suggest that diarrhea given that Intel is being sued by their investors for fraud.

 

[TheUn4seen]

You might want to look in the own mirror. It's ironic that you suggest that diarrhea given that Intel is being sued by their investors for fraud.

To be honest, I have nothing but contempt for all corporations and, by extension, their fanboys, irrespective of the market segment they represent. I buy a product for what it is, after it is available on the market, gets reviewed by several sources and has proven itself. AMD products are routinely overhyped before release with outlandish claims which almost every time turn out to be worth their weight in fecal material. I'm not saying Intel is any better, no corporation is, but AMD is the corporation which often used social media and hype to aggressively spread outright lies about their upcoming products.

 

[thesmokingman]

To be honest, I have nothing but contempt for all corporations and, by extension, their fanboys, irrespective of the market segment they represent. I buy a product for what it is, after it is available on the market, gets reviewed by several sources and has proven itself. AMD products are routinely overhyped before release with outlandish claims which almost every time turn out to be worth their weight in fecal material. I'm not saying Intel is any better, no corporation is, but AMD is the corporation which often used social media and hype to aggressively spread outright lies about their upcoming products.

It seems to me that you are focused only on AMD when Intel keeps lying to everyone and is actually being sued by their investors right now for that. Under Su, AMD have delivered on these so-called outlandish claims. But yet you don't bat for any team... ok.

 

[efikkan]

And is'nt having the chip locked to a 65W TDP spec vs a AMD at 65W spec single core, a good sign? At the end of the day it's how the chip performs. And if this is true your looking at a overall winner coming (finally) from team AMD.

I don't think we can compare it that way. Intel's TDP means the CPU will throttle to that power usage after a few seconds (typical 28 or 56 seconds), while AMD's TDP isn't a hard limit at all, it's more of a cooling guide. Real benchmarks will show how well it performs, but it will probably be good.

 

[Zach_01]

3600nonX 65W TDP

R15/20 SMT = 86~88W PPT (Power Package Tracking) = Max power draw limit (sustained if temp/current within limits)
R15/20 ST = 50~55W PPT

AMD’s TDP rate is for required cooling capacity on specific CPU/case/ambient temperatures.

 

[SL2]

That looks like it's from a NotebookCheck review. Care to link it? Of course absolute performance in mobile depends a lot on the implementation (chassis, cooling, VRM, fan curves, etc.), but at least on paper TGL should boost higher than ICL in the short term and sustain slightly higher base clocks in the long term.

Yup. https://www.techpowerup.com/forums/...ark-single-thread-results.273696/post-4375982

Yeah, it's all about implementation when it comes to laptops, like when the low end IL i5 beats a "faster" TL i7 in R20 MT.

But please be aware that Tiger Lake is not slower than the 4800U, it only has Fewer Cores™.
(Not directed at you Valantar)

It's not like AMD ever used untrue claims and "fanboy fantasy" marketing to sell products which turned out to be a huge disappointment, right?.

Maybe, but it doesn't matter here. AMD doesn't own Passmark, and they didn't post this, Videocardzzzzzzz did. It's a screenshot from Passmark.

Why do people still think that AMD is behind this?

Is it because the OP mixed AMD slides with the Passmark screenshot?

 

[Valantar]

To be honest, I have nothing but contempt for all corporations and, by extension, their fanboys, irrespective of the market segment they represent. I buy a product for what it is, after it is available on the market, gets reviewed by several sources and has proven itself. AMD products are routinely overhyped before release with outlandish claims which almost every time turn out to be worth their weight in fecal material. I'm not saying Intel is any better, no corporation is, but AMD is the corporation which often used social media and hype to aggressively spread outright lies about their upcoming products.

You're definitely right that AMD has a history of really shitty marketing, but I have two notes on that:
- The truly shitty marketing has been limited to their graphics cards, and almost entirely to RTG under Raja Koduri. Not saying it was his doing, but we haven't seen anything like the Vega launch and "Poor Volta" since.
- Lisa Su does seem to be on a "sell products on their own merits" rampage through the various PR departments at AMD. She clearly didn't have much say with RTG under Koduri (his desire for near total independence is widely documented), but my impression is that things have been changing since he left. PR wise the RDNA launch was a pretty straightforward affair, after all.

 

[RandallFlagg]

You might want to look in the own mirror. It's ironic that you suggest that diarrhea given that Intel is being sued by their investors for fraud.

You might want to go look at what AMD said in the past, and how that compared to reality.

But since you're not gonna do that, I'll show you. AMD doesn't exactly lie on these, but they're not going to show you everything either.

AMD announcement, 3600X gaming performance - pretty much a tie right? :

Reality (note the 9600k is over 10% faster in their geometric mean). This is where you'll see significantly more stutter less consistent performance from a 3600X than a 9600K. Note that at the time, there was only a $13 price difference here :

And even in averages, the 9600K was better. Even my lowly and less expensive 10400 whips the new and more expensive 3600XT by almost 5% in average FPS. << - That's reality. And the 10600K? We're talking 7.4% faster average FPS than a 3600XT.

 

[Zach_01]

You might want to go look at what AMD said in the past, and how that compared to reality.

But since you're not gonna do that, I'll show you. AMD doesn't exactly lie on these, but they're not going to show you everything either.

AMD announcement, 3600X gaming performance - pretty much a tie right? :

View attachment 173401



Reality (note the 9600k is over 10% faster in their geometric mean). This is where you'll see significantly more stutter less consistent performance from a 3600X than a 9600K. Note that at the time, there was only a $13 price difference here :

View attachment 173402

And even in averages, the 9600K was better. Even my lowly and less expensive 10400 whips the new and more expensive 3600XT by almost 5% in average FPS. << - That's reality. And the 10600K? We're talking 7.4% faster average FPS than a 3600XT.

View attachment 173405

Can you help me with my math please? Where exactly are you seeing the 10% difference? Because I only see a 2%... (TPU)
And in Geomean the 3600X is infront...

 

[Valantar]

You might want to go look at what AMD said in the past, and how that compared to reality.

But since you're not gonna do that, I'll show you. AMD doesn't exactly lie on these, but they're not going to show you everything either.

AMD announcement, 3600X gaming performance - pretty much a tie right? :

View attachment 173401



Reality (note the 9600k is over 10% faster in their geometric mean). This is where you'll see significantly more stutter less consistent performance from a 3600X than a 9600K. Note that at the time, there was only a $13 price difference here :

View attachment 173402

And even in averages, the 9600K was better. Even my lowly and less expensive 10400 whips the new and more expensive 3600XT by almost 5% in average FPS. << - That's reality. And the 10600K? We're talking 7.4% faster average FPS than a 3600XT.

View attachment 173405

Uhm, did you miss the fact that the top CPU in that chart from Tom's is OC'd to 5GHz, and that the stock 9600K is essentially tied with, but slightly lower than the 3600X?

 

[RandallFlagg]

Can you help me with my math please? Where exactly are you seeing the 10% difference? Because I only see a 2%... (TPU)
And in Geomean the 3600X is infront...

It's actually 2.7% - closer to 3% than 2% - in the averages. The 9600K OCs to 5.0 pretty much 100% of the time, most are getting 5.1. That's a 13% difference actually. For $13. Which would explain why the 3600X never actually commanded that price.

 

[Zach_01]

Ohh oh OC...
Yeah... because this is how any corporation markets a CPU. By telling users that they can push power draws to 200W levels and gain a hefty performance of 10%.

Good to know!

And the difference is 2.3%...

 

[RandallFlagg]

Ohh oh OC...
Yeah... because this is how any corporation markets a CPU. By telling users that they can push power draws to 200W levels and gain a hefty performance of 10%.

Good to know!

And the difference is 2.3%...

The 3600X is 99.6 and the 9600K is 102.3

102.3/99.6 X 100% = 102.71%

You are looking at the 3600XT which is running about the same price as the 9700K right now. Edit: Assuming you can find a 3600XT.

 

[Zach_01]

The 3600X is 99.6 and the 9600K is 102.3

102.3/99.6 X 100% = 102.71%

You are looking at the 3600XT which is running about the same price as the 9700K right now. Edit: Assuming you can find a 3600XT.

You're right I did see the XT.
I stand corrected!

Have your +0.4% of performance along with that +10% of 200W.

 

[Valantar]

It's actually 2.7% - closer to 3% than 2% - in the averages. The 9600K OCs to 5.0 pretty much 100% of the time, most are getting 5.1. That's a 13% difference actually. For $13. Which would explain why the 3600X never actually commanded that price.

Yet the vast majority of users have no idea how to overclock, no interest in learning to do so, and run their parts at stock. Heck, most non-enthusiast computer builders don't even know they have to activate XMP manually. What is possible and what is reality for most people are not the same thing.

 

[RandallFlagg]

You're right I did see the XT.
I stand corrected!

Have your +0.4% of performance along with that +10% of 200W.

It's 179W on torture test. Note that is (far) less than a 2700X stock on the same test:

 

[dir_d]

How can you talk about the performance of a 3600 then turn around and give power numbers for a 2700x, come on man...

 

[Zach_01]

It's 179W on torture test. Note that is (far) less than a 2700X stock on the same test:

View attachment 173421

I never said anything about OC, you keep bringing it up.
Here we are talking about the 5000 series CPUs and you have gone back 2 gens for both Intel and AMD talking about OC... to proof what exactly? How superior Intel Is?

Well have it...

Raja Koduri to Present at Samsung Foundry Forum amid Intel's Outsourcing Efforts

Intel's chief architect and senior vice president of discrete graphics division, Mr. Raja Koduri, is said to be scheduled to present at Samsung Electronics Event day. With a presentation titled "1000X More Compute for AI by 2025", the event is called Samsung Foundry SAFE Forum. It is a global...

www.techpowerup.com

 

[Makaveli]

How can you talk about the performance of a 3600 then turn around and give power numbers for a 2700x, come on man...

Guy is reaching real hard with his posts.

Let me sum it up in one line.

"I prefer intel inside"