Ryzen 1800X @ 3.6GHz With Turbo Disabled – Outperforms Intel’s 8-Cores In 6/8 Tests CPU Benchmarks

pencea

Weaksauce
Joined
Jun 21, 2016
Messages
74
Ryzen 1800X @ 3.6GHz With Turbo Disabled – Outperforms Intel’s 8-Cores In 6/8 Tests CPU Benchmarks

The tests include integer math, floating point performance, prime numbers, encryption, compression, sorting, SSE performance and physics. The AMD Ryzen 7 1800X outperformed every other CPU in 6 out of the 8 tests. Including Intel’s fastest 8 core chip, the $1099 Broadwell-E i7 6900K.

While in Passmark’s single-threaded performance test clock for clock the 1800X here manages to deliver 4% more performance than Intel’s Broadwell-E.

http://wccftech.com/amd-ryzen-7-1800x-8-core-benchmarks/
 
Can't be true. No way. No how....

More crazy speculative BS fake benchmarks. The leak is real... the tech press is fake.
 
Can't be true. No way. No how....

Passmark baseline N774164.

Not removed yet, you can check for yourself.

And yes, it does match 6900k.

Choo Choo Motherfuckers.

EDIT: On another hand, Memory results are just as bad as they were.
 
Can't be true. No way. No how....

More crazy speculative BS fake benchmarks. The leak is real... the tech press is fake.

What's fake about the tech press? I mean aside from wccf posting every rumor under the sun, but most people know that now.
 
Ryzen 1800X @ 3.6GHz With Turbo Disabled – Outperforms Intel’s 8-Cores In 6/8 Tests CPU Benchmarks

The tests include integer math, floating point performance, prime numbers, encryption, compression, sorting, SSE performance and physics. The AMD Ryzen 7 1800X outperformed every other CPU in 6 out of the 8 tests. Including Intel’s fastest 8 core chip, the $1099 Broadwell-E i7 6900K.

While in Passmark’s single-threaded performance test clock for clock the 1800X here manages to deliver 4% more performance than Intel’s Broadwell-E.

http://wccftech.com/amd-ryzen-7-1800x-8-core-benchmarks/

Keep in mind that the older 6950X can usually outclock the newer 6900K. I can hit 4.5GHz all day on any 6950X I've ever laid my hands on. The 6900K? Not so much. 4.3GHz is the best I've been able to do on those. The IPC difference isn't enough to close the 200MHz gap between the two CPUs making the older 5960X the faster of those two CPUs for enthusiasts.
 
I don't play passmark......

With that said I'll buy a new AMD rig if it outperforms my skylake in games.

I won't, mostly because I don't need it, but I'll certainly recommend AMD if their chipsets, boards and drivers pass muster too. Of course, not only do we need to test games on shipping silicon, we also need to test OC vs. OC, and I'm thinking that this is where AMD is going fall short, if only a little, given how low most of their parts are clocked.
 
I won't, mostly because I don't need it, but I'll certainly recommend AMD if their chipsets, boards and drivers pass muster too. Of course, not only do we need to test games on shipping silicon, we also need to test OC vs. OC, and I'm thinking that this is where AMD is going fall short, if only a little, given how low most of their parts are clocked.

Agreed. And I don't think skylake or kabylake owners have to worry about Ryzen. Pretty sure they will either still be better or roughly par with the 1700/1800X
 
Now the question becomes which RyZen would be the best for gaming with max overclocks? The 6 core if they clock faster, the 4 core or the 8core? A lot of stuff to sort out. Can I take an eight core and run it with just 4 cores at super clocked speeds of 4.5ghz+ or does it not make a difference with the number of cores when OC? Really I could careless about benchmarks that are low res, low settings etc. That is totally meaningless for saying one is better then the other in gaming and is not real testing in a nutshell. At gaming resolution and settings it may just not matter at all and that should be what is said if that is the case. If the experience is the same at max playable settings and that is that.
 
Now the question becomes which RyZen would be the best for gaming with max overclocks? The 6 core if they clock faster, the 4 core or the 8core? A lot of stuff to sort out. Can I take an eight core and run it with just 4 cores at super clocked speeds of 4.5ghz+ or does it not make a difference with the number of cores when OC? Really I could careless about benchmarks that are low res, low settings etc. That is totally meaningless for saying one is better then the other in gaming and is not real testing in a nutshell. At gaming resolution and settings it may just not matter at all and that should be what is said if that is the case. If the experience is the same at max playable settings and that is that.

I believe the purpose with the low settings is to remove the Vid Card as a factor.
 
I believe the purpose with the low settings is to remove the Vid Card as a factor.
That does not just automatically make the cpu the limiting component - it could be memory limitations or latency, chipset latency, API or software how it is written that would never really tax fully the cpu. Games that are written for few cpu cores - cpu is running at 30% load and people call it cpu limited -> software limited and not using the cpu fully. IPC may help but better software using all of the cpu if possible would be better yet.

In other words running games at ridicules settings for let say a high end system shows nothing about the real gaming performance yet it is done over and over again by many sites with conclusions that are not accurate. If at gaming resolutions and settings and there is a difference then yes call it like it is.
 
Games that are written for few cpu cores - cpu is running at 30% load and people call it cpu limited -> software limited and not using the cpu fully.
Software limited is not a thing. Unless you have a formal proof there is some performance left on the road beyond "But it does not use my 16 Atom cores!111"
 
How do you play Passmark, may i ask?

You do know this is a hardware and overclocking enthusiast site right? Not everyone uses their system for gaming, or only gaming. To MANY people, more cores is a damn good thing (provided the IPC is good). Your options right now, are use an older Xeon setup and sacrifice modern amenities, or give Intel $1k+.
 
You do know this is a hardware and overclocking enthusiast site right?
Of course, but that does not answer the question: how the hell do you play Passmark? It's not like it's workloads are actually useful in real life with their miniscule sizes.
To MANY people, more cores is a damn good thing (provided the IPC is good).
I have already said it: i do plan on purchasing Ryzen rig in the summer, once i am done with my spring semester. For sort of stuff i do having 16 threads even at just "good" per core performance is useful.
Your options right now, are use an older Xeon setup and sacrifice modern amenities, or give Intel $1k+.
Ebay 2 2666v3 qual samples, what is the issue :p? Intel gets nothing, you get 20 cores at decent frequency.
 
It is looking more and more like I will be getting the 1700X and the Asus X370 Pro or Crosshair VI Hero. (Probably the hero just to have more of everything but it depends on the cost.) These benches look right in line with what we were more or less expecting at those clocks. I already have my 16GB of DDR 4 ram and I have my Noctua NH-D15 AM4 bracket kit on the way. Looking forward to a new build but, I will have to see if I am going to just plug and play or redo my computer from scratch.
 
Of course, but that does not answer the question: how the hell do you play Passmark? It's not like it's workloads are actually useful in real life with their miniscule sizes.

I have already said it: i do plan on purchasing Ryzen rig in the summer, once i am done with my spring semester. For sort of stuff i do having 16 threads even at just "good" per core performance is useful.

Ebay 2 2666v3 qual samples, what is the issue :p? Intel gets nothing, you get 20 cores at decent frequency.

The passmark benchmarks directly indicate what we were pretty much told to expect anyways. Games will run fine, that is a fact.
 
Software limited is not a thing. Unless you have a formal proof there is some performance left on the road beyond "But it does not use my 16 Atom cores!111"

Software limitations are a real thing regardless of what hardware is being used. Mocking him does not change that fact and he already provided the proof. Does not matter though, I will be the new owner of a Ryzen System in less than 2 weeks now. :)
 
Except you know exactly what I am talking about and Ryzen is not any of those, nice try though. :p:confused::rolleyes:

I do, and you left it wide-open ;)

Still, we need to see game benchmarks, if what we're talking about is gaming, and we need to see OC vs. OC as well as stock vs. stock. These compute benchmarks are great for isolating performance that pertains to very specific workloads, but they don't really show how the whole system works and clear the platform as a whole of potential deficiencies.
 
I do, and you left it wide-open ;)

Still, we need to see game benchmarks, if what we're talking about is gaming, and we need to see OC vs. OC as well as stock vs. stock. These compute benchmarks are great for isolating performance that pertains to very specific workloads, but they don't really show how the whole system works and clear the platform as a whole of potential deficiencies.

I would still say that the premise of Ryzen is that you can buy cheaper 8C16T and get your price performance kicks from there.

In gaming Intel would still be faster for a few reasons one of them is clock speed on 4 cores games that do not scale well on 4+ would still "easily" outperform Zen counterparts.
For me this is not important.
 
I don't play passmark......

With that said I'll buy a new AMD rig if it outperforms my skylake in games.

Is that because you are a one-dimensional computer user and love those Neanderthal limited one thread games???? You should really try well engineered games like those from Stardock. They all have multithreading, because Wardell was a developer for os/2 in the old days. Multl-threading is burned into his brain. The Neanderthal developers are going to be left behind. The trend is for game development to take advantage of more cores and more threads. Intel has been holding on to the 4 core processor as their main stay and financially penalizing those who want more cores and more threads. They have actually retarded software development with this greedy, monopolistic behavior. I am extreemely happy that AMD has developed Ryzen and aggressively priced it so that 6 and 8 core chips with high performance are affordable for so many more people.
 
I want to see some Insurgency benchmarks. That game eats CPUs like it's its job.

Actually the games that supposedly eat cpu are actually not eating it all. They are thread-bare games that require you to burn a whole lot of watts to play on a single thread. They stink period. Only one -dimensional tunnel-vison gamesters like those games and encourage pathetic game developers to keep creating this trash. Then you reward monopolist Intel foe subsidizing those game developers who produce this trash. Win, win for Intel. Lose, lose for consumers.
 
Actually the games that supposedly eat cpu are actually not eating it all. They are thread-bare games that require you to burn a whole lot of watts to play on a single thread. They stink period. Only one -dimensional tunnel-vison gamesters like those games and encourage pathetic game developers to keep creating this trash. Then you reward monopolist Intel foe subsidizing those game developers who produce this trash. Win, win for Intel. Lose, lose for consumers.

Yeah it's really too bad because it's such a good game. I know it's based on the Source engine, could that have anything to do with it?
 
Is that because you are a one-dimensional computer user and love those Neanderthal limited one thread games???? You should really try well engineered games like those from Stardock. They all have multithreading, because Wardell was a developer for os/2 in the old days. Multl-threading is burned into his brain. The Neanderthal developers are going to be left behind. The trend is for game development to take advantage of more cores and more threads. Intel has been holding on to the 4 core processor as their main stay and financially penalizing those who want more cores and more threads. They have actually retarded software development with this greedy, monopolistic behavior. I am extreemely happy that AMD has developed Ryzen and aggressively priced it so that 6 and 8 core chips with high performance are affordable for so many more people.

err nope lol.

Software development only moves forward if the hardware is available and the API's can take advantage of that. And then hardware sales are based on software (adoption rate) can take advantage of the hardware.

There is no such thing as Neanderthal developers, you might have some lazy devs. But every meet a lazy engine developer? I haven't, those guys are usually top of their game, because they have to be, its a pretty competitive field in game development, probably the highest paid too. Once a company has a good engine developer, they don't want to loose them, because it makes a lot of headaches later on without them, essential the engine they are making or using, becomes laboriously hard to make modifications unless a new guy can sit down and understand the code and thought process of the original programmer, which can take up to months of time.

Most Dev teams want to do more than what is released, but can't because they are restricted by the publisher or time, or money , or all over these three things. People that make games, love making games, that is why they are doing it. These people are not 9 to 5 desk jockeys. Making games is hard work because tech moves so fast in gaming, new things happen every six months that need to be stayed on top of. Everything from asset development to programming. This is what makes PC development so much harder than consoles. Consoles have a self life of 5 years or so, maybe down to 3 years now, but during that time everything stays the same. PC's just doesn't work that way.

Stardock's dev's I would not put them up there near the capabilities of the UE dev's, ID's dev's, Crytek's Dev's, big, big difference between the capabilities of Stardock's engine vs the later 3.

Intel had nothing to do with the API's either. Intel, yeah forced everyone to stick to 4 cores and lower, but hell, no AMD to push em to do anything different so milking the market was their best option. If you were in Intel's shoes what would you do? Push the market forward, or milk the market, and then when you need to push it forward, you can do it at that time? I would do the later, its business, making money and keeping costs lower or same is beneficial in the long term.
 
Last edited:
To everyone saying not to focus on games, wasn't it posted in the news recently that gaming drives the high end computer market? Therefore, doesn't it make sense to have gaming as one of the biggest factors to judge Ryzen on?
 
Yes it does make sense. Ryzen is going to be adopted by enthusiasts and gamers because of the lack of the APU, those are the only people that will be interested in them. (outside of work, like 3d artists, video editors, etc that want high core counts at reasonable prices)

Why would anyone purchase a lets say 1400 Ryzen for 200 bucks (motherboard and CPU) then stick a 200 dollar GPU on top of that, unless they want to play games.....
 
To everyone saying not to focus on games, wasn't it posted in the news recently that gaming drives the high end computer market? Therefore, doesn't it make sense to have gaming as one of the biggest factors to judge Ryzen on?

In terms of desktop PC sales, this may have some truth to it, but not when it comes to CPU design. This is why Intel CPU's have been a wash for the last four generations: they've been focused on stuff other than games, particularly stuff that will improve the mobile computing experience and stuff that will lower enterprise/datacenter TCO. Because that's what actually sold.
 
In terms of desktop PC sales, this may have some truth to it, but not when it comes to CPU design. This is why Intel CPU's have been a wash for the last four generations: they've been focused on stuff other than games, particularly stuff that will improve the mobile computing experience and stuff that will lower enterprise/datacenter TCO. Because that's what actually sold.


Added to the fact there was no one pushing them for more too.
 
In terms of desktop PC sales, this may have some truth to it, but not when it comes to CPU design. This is why Intel CPU's have been a wash for the last four generations: they've been focused on stuff other than games, particularly stuff that will improve the mobile computing experience and stuff that will lower enterprise/datacenter TCO. Because that's what actually sold.

I doubt that actually is the case, I'm more of the belief that it has become significantly harder to improve IPC. The easy way is with new instruction sets, but software developers are slow to adopt new instruction sets. Improving IPC lowers frequency needed to obtain the same performance, which lowers overall power usage, so why wouldn't Intel want to increase IPC with the goal of selling to mobile and enterprise? The only place I see a possible argument is the manufacturing process being optimized towards low power rather than high frequencies.
 
I doubt that actually is the case, I'm more of the belief that it has become significantly harder to improve IPC. The easy way is with new instruction sets, but software developers are slow to adopt new instruction sets. Improving IPC lowers frequency needed to obtain the same performance, which lowers overall power usage, so why wouldn't Intel want to increase IPC with the goal of selling to mobile and enterprise? The only place I see a possible argument is the manufacturing process being optimized towards low power rather than high frequencies.

Intel has been actively doing this for the past 3 or so generations of CPU's. When was the last time we had so many Y, S, T, U and others sku's come out first, and then the mainstream and enthusiast chips come out?

Looking at Kaby Lake, it is well optimized for lower end sku's where we can see the IPC change, but for their higher end chips, those IPC games are held back by some kind of bottleneck.
 
Added to the fact there was no one pushing them for more too.

In the desktop space, agreed. AMD's new products, just by AMD's sheer need to survive by grabbing marketshare to generate and sustain revenue, will force Intel to adjust somewhat.

On the other hand, I can't really complain; outside of gaming, on the desktop, there's been very little need for faster parts, and Intel hasn't let their pricing go wild either.
 
Back
Top