Friday, October 23rd 2020

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

AMD has been improving its Zen core design, and with the latest Zen 3 IP found in Ryzen 5000 series CPUs, it seems like the company struck gold. Thanks to the reporting of VideoCardz, we come to know that AMD's upcoming Ryzen 5 5600X CPU has been benchmarked and compared to other competing offerings. In the CPU benchmark called PassMark, which rates all of the CPUs by multi-threaded and single-threaded performance, AMD's Ryzen 5 5600X CPU has taken the crown of the fastest CPU in the single-threaded results chart. Scoring an amazing 3495 points, it is now the fastest CPU for 1T workloads. That puts the CPU above Intel's current best—Core i9-10900K—which scores 3177 points. This puts the Zen 3 core about 10% ahead of the competition.

As a reminder, the AMD Ryzen 5 5600X CPU is a six-core, twelve threaded design that has a base clock of 3.7 GHz and boosts the frequency of the cores to 4.6 GHz, all within the TDP of 65 Watts. The CPU has 32 MB of level-3 (L3) cache and 3 MB of L2 cache.
Source: VideoCardz
Add your own comment

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

#76
Zach_01
birdie
I'm ignoring your posts from now on. You've failed to address the fact that Intel doesn't allow itself to raise prices when they release faster better products. You're trying to compare the 5600X to the 3600X which wasn't the entry level CPU, it was the 3600 which cost $200, so the difference is not $50 but $100, i.e. whopping 1.5 times. Good luck with AMD a-licking and vindicating their monopolistic behavior (because it is what is is).
Yeah, ignore the fact that entry level of 3000series is 3100 and 3300... You can close your eyes all you want. Still exist. Still there.
Are you playing blind?
A little help? Are you ignoring other things also except me?
Zach_01
And the 1200/1300/1400/1500/3100/3300 what exactly are? Sub-entry level or non existent CPUs?
You can cry all you want. 5600X is replacing 3600X and has price bump of 50$. That is a +20% on MSRP with at least the same performance uplift and most probably performs faster than any 6core.
You can try all you want but 5600X is replacing 3600X, you like it or not, understand it or not. I highly doubt the you dont, but just trying to distort reality for whatever reason. Couldnt care less why...

EDIT: typo
Posted on Reply
#77
birdie
Valantar
Relevance? One is a Skylake part, launched long before any Ryzen, and the other is Haswell, launched several years before that again.
No effing relevance, AMD releases a new generation of the same product (More cores? No! Some new features, maybe the AVX512 instruction set, or DL instructions? No! Maybe integrated graphics for a change? No!) at a whopping 1.5 times higher price.
Valantar
I mean, the 3600 was a fantastic value CPU, but the 5600X is promising to be noticeably faster - it's clocked higher and has much higher IPC, after all. It's still not as great value as the 3600 was, but it's not bad - and again, there will in all likelihood be a 5600 non-X.
Has Intel ever increased their prices 1.5 fold when their new CPUs offered fantastic increases in IPC and frequency? Should I remind you of Sandy Bridge which featured both these great advances vs. Clarkdale before it?
Valantar
What double standard? You're comparing different lineups and saying they are the same. There was no X SKU in the 3600 series. There was in the 1600 and 2600 series. There will in all likelihood be in the 5600 series, but it has as of yet not been launched. I would be shocked if there wasn't a ~$230 5600 non-X launched in a few months - and if that turned out to be true, I would indeed be rather pissed. You are allowing your apparent bias against AMD to make you preemptively angry long before the full 5000 series lineup is launched; you are creating a reason for yourself to be mad. Please stop.
AMD has deliberately chosen not to release the Ryzen 5 5600 to increase their margins - exactly the thing which Intel and NVIDIA are hated for. I don't care if they release the 5600 at a later time, I'm talking about their initial lineup. You and no one else know if 5600 will ever get released. Period. Intel and NVIDIA get ripped apart when they release new generations of products at a higher price. AMD now does exactly the same and all I see is tons of lame excuses to make it look OK'ish. God, I'm tired of this stupid conversation. F it and I'm done with it.
Valantar
First off: acting like a monopolist does not make you one.
Wow. I'm out of words. Never seen logic being skewed and twisted like this time around. Another person whose messages I will no longer see. For the 20th, I don't understand why this behavior is condoned while Intel and NVIDIA are destroyed for it.
Posted on Reply
#78
kapone32
If AMD had launched these X Skus at the price of the already existing 3000 weries skus they would have to have prematurely drop the price of those parts. As much as AMD really gained in mind share and revenue they have only been financially solvent for about 1 year. As much as people would like to complain about their tactics being the same as Intel if you already own a B450 to X570 board you just need a BIOS update to make it work. So much anger for such a small issue. If you don't want to pay $299 US for the 5600X just wait until a chip comes out that you want. This is the last iteration of AM4 (as far as we know) so expect that it there will be lot's of choices for all of us including the upcoming 4000 series APUs.
Posted on Reply
#79
Valantar
birdie
No effing relevance, AMD releases a new generation of the same product (More cores? No! Some new features, maybe the AVX512 instruction set, or DL instructions? No! Maybe integrated graphics for a change? No!) at a whopping 1.5 times higher price.

Has Intel ever increased their prices 1.5 fold when their new CPUs offered fantastic increases in IPC and frequency? Should I remind you of Sandy Bridge which featured both these great advances vs. Clarkdale before it?
Okay, here's some basic math:
Ryzen 3600X: $249
Ryzen 5600X: $299.
299/249 = 1.2008
1.2008 ≠ 1.5

For what seems to be the hundredth time: the 5600X is not in the same tier as the 3600.
birdie
AMD has deliberately chosen not to release the Ryzen 5 5600 to increase their margins - exactly the thing which Intel and NVIDIA are hated for. I don't care if they release the 5600 at a later time, I'm talking about their initial lineup. You and no one else know if 5600 will ever get released. Period. Intel and NVIDIA get ripped apart when they release new generations of products at a higher price. AMD now does exactly the same and all I see is tons of lame excuses to make it look OK'ish. God, I'm tired of this stupid conversation. F it and I'm done with it.
Sure, part of why there isn't yet a 5600 is that while we wait for it, some people will get impatient and buy the X instead, increasing AMD's margins a tad and helping them amortize the R&D cost for the new architecture faster with higher ASP parts. For the rest of us, we can wait a bit, not get stressed out over stupid stuff, and get a non-X when it launches.

As for this double standard you're speaking of: I don't believe I've ever criticized either Intel or Nvidia for launching high end parts first and filling out the rest of the lineup later. As I said, it's standard industry practice, and not inherently problematic (partly due to the amortization of R&D costs I mentioned above).

And as for whether a 5600 non-X will be released: as I've gone into at length above, there have always been both X and non-X SKUs in the Ryzen 5 xx60 series. There is no reasonable reason for there not to be this time around, but you are for some reason expecting there not to be. You are the one expecting a change from the norm here, the one making a new assumption, and as such the burden of proof is on you. As for my reaction if there, against all odds, is no non-X coming, I've already gone into that too:
Valantar
I would be shocked if there wasn't a ~$230 5600 non-X launched in a few months - and if that turned out to be true, I would indeed be rather pissed.
If you actually read the posts of the people you were arguing against we would be having a much more productive discussion here.
birdie
Wow. I'm out of words. Never seen logic being skewed and twisted like this time around. Another person whose messages I will no longer see. For the 20th, I don't understand why this behavior is condoned while Intel and NVIDIA are destroyed for it.
To be a monopolist, you need to be in a position where you can have a monopoly. Intel still outsells AMD by at least 3-4X. That is hardly a monopoly, right? As for you entering into a discussion, making a bunch of sensationalist claims with no backing, and then lashing out at and ignoring the people arguing against you, well ... that's on you.
Posted on Reply
#80
kapone32
Valantar
Okay, here's some basic math:
Ryzen 3600X: $249
Ryzen 5600X: $299.
299/249 = 1.2008
1.2008 /= 1.5

For what seems to be the hundredth time: the 5600X is not in the same tier as the 3600.


The best way to compare the 2 is the 3100 to 3300x. Both 4 cores but very different. One is readily available and the other is on back order world wide. As I keep saying if you have had the pleasure of owning a 3300x to compare to even the 2600 you would be very excited for the prospect of a 6 core single CCX solution. Let us all just hope that the price remains at $299.
Posted on Reply
#81
Zach_01
Valantar
As for you entering into a discussion, making a bunch of sensationalist claims with no backing, and then lashing out at and ignoring the people arguing against you, well ... that's on you.
He entered this with intention to do more than that, but I highly doubt he can...
birdie
I know AMD fans are super excited but I'd love to shit on your parade.
I guess this is productive... in a way...
Posted on Reply
#82
crimsontape
I come here for the comments. I wasn't disappointed. You are all awesome.

No one like price hikes, but I'm sure this isn't all money in the pocket for AMD. I wonder if they bought IP to move their architecture forward. That would explain a hike. Might be the same IP that's finding itself in Zen2 and new consoles... Possibly a deferred cost model? Simple supply and demand for their CPUs is another factor. They have the product and increased market share to show for it now, no?

This isn't monopoly. This is capitalization. And they're doing it very well. They're not fixing the market and choke-holding us to buy their products. Intel arguably has become that, but only thanks to the same approach: capitalize. Which, for them, I believe was mostly a question of volume of supply (which is only shaky in the last few years), and establishing ISAs which other vendors then pay to implement and use. More IP.

There's only guilt when you fix/manipulate markets, like price-fixing (as ATI and NVIDIA once did some 15 years ago?) and purposefully breaking an existing market solution to artificially encourage the adoption of a bad product, for which you're the sole provider (electric cars, intel MMX).
Posted on Reply
#83
Mats
birdie
Has Intel ever increased their prices 1.5 fold when their new CPUs offered fantastic increases in IPC and frequency?
Neither have AMD. 36 % isn't 50 %. :D

3600X    249        5600X    299    20 % higher
3700X 329 5800X 449 36 %
3800X 399 5800X 449 12.5 %
3900X 499 5900X 549 10 %
3950X 749 5950X 799 6.7 %

Yeah we all miss the successors of the 3600 and the 3700X, but if prices seem a bit high, just wait.
More models will most likely show up, prices will go down, and Rocket Lake will probably affect prices in some way.
Stop comparing the 3700K to the 5800X, as we will most likely see some product in between there at some point.
The successor to the 3800X is the 5800X, obviously.
birdie
Should I remind you of Sandy Bridge which featured both these great advances vs. Clarkdale before it?
You're the one who has to be reminded before you make yet another bitter post about not having the money for Ryzen 5000. Give it a rest already.

Clarkdale was a low end dual core.
Sandy Bridge was the successor to Lynnfield, the latter had lower clock speeds but still a very capable CPU. Both launched at over 300USD.

Now stop being bitter, we don't care.
Posted on Reply
#84
RandallFlagg
birdie
I know AMD fans are super excited but I'd love to shit on your parade. You know why? AMD has finally outperformed an Intel uArch from ... 2015. This might sound like a great achievement but honestly it's just because Intel has completely f*ed up their 10nm transition. Yeah, their latest 10nm++ node (first - Cannon Lake, second - Ice Lake and now Tiger Lake) allows to boost to 4.8GHz at the expense of insane power consumption and they've made changes to the Willow Cove Core architecture which sometimes translate to a lower performance than Ice Lake:



In short Intel has turned from an indisputable x86 performance leader to something else entirely and AMD has quickly seized the opportunity to significantly increase their prices. An entry level Ryzen 5000 CPU, Ryzen 5 5600X, is now 50% (!) more expensive than its Ryzen 3000 counterpart, Ryzen 5 3600. There's nothing to be excited about. One struggling monopoly has been replaced by another.
What is funny here is how hard you had to look to cherry pick a benchmark that showed AMD over Tiger Lake. Right below that same image you posted, is the one below, where a 15W limited Tiger Lake is nearly twice as fast as the 4750G Pro (using AVX) - and who in their right mind would disable this if they are doing this type of thing (which, most won't be doing this) :




And then there are all the benchmarks of real-world things people do with their laptops that you ignored, and where Tiger Lake completely rips apart AMD.

In fact, Tiger Lake as a laptop chip is as fast and quite often faster than the 10900K in many instances. For example, the Kraken 1.1 scores of 630/631 are better than the previous best score for any system - 730ms scored by a 10900K. Tiger Lake is hitting 630/631ms. That is a huge difference.

If any significant part of this gets in to Rocket Lake, Zen 3 won't be on top for very long.

Web Browsing :









Rendering :



Play a MMO :



Play an FPS :

Posted on Reply
#85
Mats
Tiger Lake is very capable. Overall, it has a faster GPU but a slower CPU compared to Renoir, whit the latter being no surprise given it only has 4 cores.
Even a 6 core TL would possibly come close to Renoirs CPU performance.

www.notebookcheck.net/Asus-ZenBook-14-UX425E-Review-11th-Gen-Core-i7-Tiger-Lake-Debut.495286.0.html

Now I don't know why Intel still can't make 10 nm and 8 cores. Low yields with Cannon Lake and Ice Lake, I get that, but even to this day? Intel's 10 nm is a train wreck that keeps on pushin'. :D The architecture seems decent tho.
Posted on Reply
#86
RandallFlagg
Mats
Tiger Lake is very capable. Overall, it has a faster GPU but a slower CPU compared to Renoir, whit the latter being no surprise given it only has 4 cores.
Even a 6 core TL would possibly come close to Renoirs CPU performance.

www.notebookcheck.net/Asus-ZenBook-14-UX425E-Review-11th-Gen-Core-i7-Tiger-Lake-Debut.495286.0.html

Now I don't know why Intel still can't make 10 nm and 8 cores. Low yields with Cannon Lake and Ice Lake, I get that, but even to this day? Intel's 10 nm is a train wreck that keeps on pushin'. :D The architecture seems decent tho.
Tiger Lake does not have a slower CPU, it has less cores. Look at the benchmarks.

This is the R15 ST Tiger Lake vs 4800U - Tiger Lake is 27.1% faster here. :





Keep in mind Tiger is an Intel laptop chip, I'm now comparing to full on desktop chips -

This is vs the R5 3600 desktop chip, Tiger is 18.5% faster :



This is vs the R7 3800XT - the fastest AMD desktop chip here, Tiger is 8% faster:



And here, the fastest desktop chip you can buy for this ST benchmark the 10900K, is beaten by a Tiger Lake laptop chip:

Posted on Reply
#87
zlobby
Dredi
Not really relevant to the news article, but I wonder how long it’ll take for userbenchmark to change its point system (again) after the launch? Will it become just a memory latency test? XD
Hey, as long as intel win, it's OK.

We saw a long time ago that money overpower dignity.
Posted on Reply
#88
Mats
RandallFlagg
Tiger Lake does not have a slower CPU, it has less cores. Look at the benchmarks.
Well, that's pretty much what I said. :rolleyes:
Overall, it has a faster GPU but a slower CPU compared to Renoir, with the latter being no surprise given it only has 4 cores.
Even a 6 core TL would possibly come close to Renoirs CPU performance.
Please read before you reply.
Also mobile vs desktop doesn't make much difference when talking ST, since power limit becomes less of a limiting factor.

www.cpu-monkey.com/en/compare_cpu-amd_ryzen_7_pro_4750g-1560-vs-amd_ryzen_7_4800u-1142

Posted on Reply
#89
RandallFlagg
Mats
Well, that's pretty much what I said. :rolleyes:



Please read before you reply.
Also mobile vs desktop doesn't make much difference when talking ST, since power limit becomes less of a limiting factor.
I did read it, but saying the "CPU is slower" is false, fewer cores is not the same thing as having a slower CPU. That depends entirely on what you're doing, and most things are still single thread limited. Just open up task manager and watch as you do things, you'll go single core limited long before you go multi-core limited the vast majority of the time. Then there is the spectre of 6 and 8 core Tiger Lake in Q1. Then if you get Rocket Lake multi-core high freq and high power limits, the picture becomes quite a bit clearer.

That 2nd comment about power doesn't make sense. The 3800XT is 26% faster than the 4800U in single thread - the difference is entirely due to power limits since these are the same architecture chips. Diminishing returns, sure, but 26% is nothing to sneeze at. Yet that 105W desktop TDP part still loses to a 28W Tiger Lake part by 8%.

The per core performance difference between laptop Tiger Lake and the laptop Ryzen Renoir and desktop Ryzen Zen 2 parts is massive. For that matter the difference between Tiger Lake and Comet Lake is massive as well.
Posted on Reply
#90
Mats
RandallFlagg
fewer cores is not the same thing as having a slower CPU.
Nobody said so. TL is slower in many benchmarks, mostly because it only has 4 cores.
I did edit my last post so maybe you didn't get the comparison of the 15 W 4800U and the 65 W 4750G, both Renoir. The 15 W part barely runs slower in ST than the 65 W.
Judging by your logic, the 4800U would crush the 4750G if it could run at 65 W, which is not the case.
Posted on Reply
#91
theoneandonlymrk
Should be a new version out shortly to correct this great wrong, userbench clearly will need a new version too soon.
Posted on Reply
#92
RandallFlagg
Mats
Nobody said so. TL is slower in many benchmarks, mostly because it only has 4 cores.
I did edit my last post so maybe you didn't get the comparison of the 15 W 4800U and the 65 W 4750G, both Renoir. The 15 W part barely runs slower in ST than the 65 W.
Judging by your logic, the 4800U would crush the 4750G if it could run at 65 W, which is not the case.
You're making a false comparison and putting words in my mouth.

The 4800U is in fact crushed by the 3800XT by 26%. I just said that. If desktop variants of the tiger lake architecture gets a similar differential, it would crush a 10900K and a 3800XT by around 30% in single core.
Posted on Reply
#93
Mats
RandallFlagg
You're making a false comparison and putting words in my mouth.
You're the one who said that Tiger Lake is held beck in ST by being mobile, I said no, and showed you a direct comparison with desktop and mobile Renoir, an indication that power limit has nothing or very little to do with it.

What you say is holding back mobile TL must hold back mobile Renoir as well? That's your logic, and don't feel hurt that I carried it over to a different comparison, no offence was intended.
RandallFlagg
The 4800U is in fact crushed by the 3800XT by 26%.
So why not compare the 4800U with its desktop counterpart? By doing so we see exactly what happens when we go from mobile to desktop, ruling out a few differences.
Posted on Reply
#94
Valantar
RandallFlagg
That 2nd comment about power doesn't make sense. The 3800XT is 26% faster than the 4800U in single thread - the difference is entirely due to power limits since these are the same architecture chips. Diminishing returns, sure, but 26% is nothing to sneeze at. Yet that 105W desktop TDP part still loses to a 28W Tiger Lake part by 8%.
There are significant differences between how Intel and AMD treat power limits and turbo though, both on desktop and mobile. Mobile Intel parts tend to boost to higher power levels than mobile AMD parts, and IIRC single core boost power for TGL is significantly higher than for Renoir, so that at least partially explains the delta. TGL is still very fast obviously, and faster than Zen 2 in both IPC and boost clocks (including mobile TGL vs desktop Zen2). Zen 3 should surpass it though:
SKL (and derivatives): 100% IPC
Zen 2: ~107% IPC (though that doesn't include latency-sensitive applications like games, number is based on AT's SPEC numbers)
ICL (and TGL, as there shouldn't be any IPC change) : ~118% IPC
Zen 3: ~107*1.19= 127.33% IPC. So even with TGL clocking slightly higher Zen 3 should have the upper hand, at worst they'll tie.
Posted on Reply
#95
Mats
About Tiger Lake 1165G7 being fast, I dunno anymore. I guess I didn't read up enough in my own link. Shame on me.
Posted on Reply
#96
Makaveli
birdie
I know AMD fans are super excited but I'd love to shit on your parade. You know why? AMD has finally outperformed an Intel uArch from ... 2015. This might sound like a great achievement but honestly it's just because Intel has completely f*ed up their 10nm transition. Yeah, their latest 10nm++ node (first - Cannon Lake, second - Ice Lake and now Tiger Lake) allows to boost to 4.8GHz at the expense of insane power consumption and they've made changes to the Willow Cove Core architecture which sometimes translate to a lower performance than Ice Lake:



In short Intel has turned from an indisputable x86 performance leader to something else entirely and AMD has quickly seized the opportunity to significantly increase their prices. An entry level Ryzen 5000 CPU, Ryzen 5 5600X, is now 50% (!) more expensive than its Ryzen 3000 counterpart, Ryzen 5 3600. There's nothing to be excited about. One struggling monopoly has been replaced by another.
lol you came in here to Intel Fan boy troll and then got destroyed by logic.

Is the intel defense Fund not paying you enough?

i've seen post like this on numbers forums from fan boys you really can't take heat AMD is putting out, so lets go and try to rain on peoples parade to make myself feel better... Get a life!
Posted on Reply
#97
Th3pwn3r
Metroid
I dont know what lisa is doing but she is doing it right, new management, new plan, new thinking, new products, new records, new highs, new performance. When lisa got in, amd was so far behind intel and nvidia that only a miracle would turn the tides. Good work lisa.

I hope msi releases a bios for my b450 before january 2021 so i can purchase that 5600x or the 5900x.
It's more along the lines of what they're not doing. They're not milking ancient tech and expecting consumers to be happy like Intel has been. In my opinion AMD has advanced MOSTLY because they're pushing further in the NM process all while Intel sat on their ass. It was the classic story of the Tortoise And The Hare. The hare slept way too long and is now in a panic.
Posted on Reply
#98
RandallFlagg
Th3pwn3r
It's more along the lines of what they're not doing. They're not milking ancient tech and expecting consumers to be happy like Intel has been. In my opinion AMD has advanced MOSTLY because they're pushing further in the NM process all while Intel sat on their ass. It was the classic story of the Tortoise And The Hare. The hare slept way too long and is now in a panic.
AMD doesn’t push ‘nm’, that would be TSMC. AMD is just reaping the rewards of TSMCs leadership in that area.
Posted on Reply
#99
Valantar
Mats
About Tiger Lake 1165G7 being fast, I dunno anymore. I guess I didn't read up enough in my own link. Shame on me.

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.
Th3pwn3r
It's more along the lines of what they're not doing. They're not milking ancient tech and expecting consumers to be happy like Intel has been. In my opinion AMD has advanced MOSTLY because they're pushing further in the NM process all while Intel sat on their ass. It was the classic story of the Tortoise And The Hare. The hare slept way too long and is now in a panic.
Don't underestimate the value of the architectural development they've done. Remember, Zen (1) represented a >50% IPC improvement over ... ugh, Excavator? I could never tell all that heavy machinery apart. Never mind, Zen was a massive improvement in both performance/clock and performance/W even before accounting for the node improvement. That was just icing on the cake. And they've kept iterating on it rapidly, with yearly improvements (though admittedly Zen+ wasn't much to write home about). All the while Intel has just last year gotten around to actually improving their architecture's performance characteristics. Though it's also well worth mentioning that before things stagnated for Intel, we were seeing 5-10% IPC improvements per generation from them. ICL and TGL are ~18% up from SKL, so that's much better, but it took five years. In the mean time, AMD has delivered +~3%, +~15% and +~19% and on a much tighter schedule.
Posted on Reply
#100
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.
Posted on Reply
Add your own comment