Monday, December 26th 2016

AMD Ryzen Performance Review Leaked: Promising

French tech print magazine "Canard PC" is ready with early benchmarks of an AMD Ryzen 8-core processor. The scan of a page from its Ryzen performance review article got leaked to the web, revealing three key performance takeaways. In the first selection of tests, Canard PC put Ryzen through synthetic CPU-intensive tests that take advantage of as many CPU cores/threads as you can throw at them. These include the likes of H.264 and H.265 video encoding, WPrime, Blender, 3DSMax 2015, and Corona. Ryzen was found to be faster than the quad-core Core i7-6700K, and the six-core i7-6800K, but somewhere between the i7-6800K and the eight-core i7-6900K.

The next selection of tests focused on PC gaming, with a list of contemporary AAA titles, including "Far Cry 4," "Battlefield 4," "The Witcher 3: Wild Hunt," "Anno 2070," "GRID: Autosport," and "ARMA III." Here, the Ryzen sample was found to be underwhelming - it was slower than the Core i5-6600 quad-core chip clocked at 3.30-3.90 GHz; but faster than the i5-6500, clocked at 3.20-3.60 GHz. The fastest chip in the table is the i7-6700K (4.00-4.20 GHz). The reviewer still notes that Ryzen has a decent IPC gain unseen from the AMD stable in a while.
In the final segment, the reviewers tested the power-consumption of the processor. AMD rates the TDP of the Ryzen 8-core chip at 95W, which was desperately needed from a chip built on the 14 nm node. Here it was noted that Ryzen made a tremendous performance/Watt leap over the 32 nm FX-8370 "Vishera." It consumes 93W, just under the 96W consumed by the Core i7-6900K eight-core chip, and slightly more than the 85W consumed by the 22 nm Core i7-4790K "Devil's Canyon" quad-core chip. The 14 nm i7-6800K six-core chip draws 83W, and the quad-core 14 nm i7-6700K draws 62W.

In all, the reviewer concludes that Ryzen could give the DIY performance CPU market the stir it badly needed, and could give Intel a shake-down, but it boils down the pricing.
Source: Reddit
Add your own comment

118 Comments on AMD Ryzen Performance Review Leaked: Promising

#76
mastrdrver
james888An 8 core chip doesn't have the thermal headroom to have higher clocks. Might AMD do a 4 core/4 thread cpu with higher clocks that still sits in the 95w TDP range?
Seeing as how the 8 core 16 thread has a TDP of ~95w, I'd expect a 4 core 8 thread to have a TDP of like ~60W.
Posted on Reply
#77
R0H1T
mastrdrverSeeing as how the 8 core 16 thread has a TDP of ~95w, I'd expect a 4 core 8 thread to have a TDP of like ~60W.
It could be higher than that, really depends on the clock speed of the 4c version. The deca core 6950x has a TDP of 140W whilst quad core 7700k sits at 91w, similarly the regular 7700 is rated for 65W but 7700T is just 35W in terms of TDP.

Usually the best bins are rated for low(er) power i.e. TDP & generally go into laptops as well.
Posted on Reply
#78
Melvis
phanbueythe leak is the link - look at gaming.
All I see is a lower clocked CPU getting beat by a higher clocked CPU? by 1-2% nothing surprising there! I dont see anything getting this so called "spanked" you mention
Posted on Reply
#79
Joker2988
and how you know which is which if is this leak even real prof of performance
Posted on Reply
#80
efikkan
R0H1TThanks for correcting me but this just reinforces my point even further i.e. people don't need to upgrade say from a Skylake to 97xx on the basis of more IPC, which posters continually harp about, because they think ST performance is all that matters.
Like I always say; if you have a Sandy-Bridge or newer and don't need more cores, then there is no point upgrading a working chip.
While single threaded performance remains important to use cases such as gaming; any power user should buy at least 6 cores at this point, since people don't run single programs any more, even while gaming.
Posted on Reply
#81
Joker2988
always think for future when your buying a new pc want to change stuff after a year or after 10 i for one got phenom II 945, 12gb ddr2, gtx 1070 can run almost any game at ultra but i need to upgrade so i will get ryzen 8 core and 32gb ram 3200mhz. I believe with this build should for 5 years not have problems with games but i hope for at least 10. :p


most games these days are gpu intensive than cpu. bf1 ultra settings with vsync on, 1080p i have constant 60 fps it dont uses more than 50% of my gpu usage.

punctuation its overrated :)
Posted on Reply
#82
EarthDog
Punctuation...use it. :)

That poor 1070 is so limited by that processor, yikes.
Posted on Reply
#83
phanbuey
MelvisAll I see is a lower clocked CPU getting beat by a higher clocked CPU? by 1-2% nothing surprising there! I dont see anything getting this so called "spanked" you mention
You must be missing that i5 moniker in front of it...
Posted on Reply
#84
Melvis
phanbueyYou must be missing that i5 moniker in front of it...
You must be missing the half a GHz clock speed difference?
Posted on Reply
#85
Fx
entropicconsider that it was an engineering sample that never went above 3.4 in the tests, that the mobo probably wasnt very high quality and these benchmarks make for a good baseline, it can only go up from here, especially since lisa su said on new horizon that 3.4 will be the minimum base clock for all ryzen chips.
Yep, I'll be starting off with that and upgrade once they release something higher-end. This is all very exciting news.
DeathtoGnomesdidnt we just close a thread on this very "french review"?

Its not even the same chip that was in the announcement stream. Its not even a released chip. Its also been said that the clocks are lowered during the pre-release. Far as I'm concerned, its a bogus review of a stolen prototype, so not a review worth its salt.
While this is true, it still offers a bit of perspective on what its potential can achieve.
Posted on Reply
#86
phanbuey
MelvisYou must be missing the half a GHz clock speed difference?
I did - because it doesn't matter. The comment I was responding to was referring to game performance.

Here is my process: Look at chart, see score. See Intel midrange chip from last gen ahead of AMD high-end chip releasing in 2 months; see Intel high end desktop chip 20% ahead. Feel sad, point out reality of data.

Here is the alternative process: Look at chart, make excuses as to why score is so low. Discount data.
Posted on Reply
#87
GhostRyder
Seems alright, its definitely an improvement but since this is an early bench with an engineering sample we really have not gained much info on the chip yet. At least we get enough info to know we are getting a decently performing chip out of Ryzen, but its going to come down to the final product and price.
Posted on Reply
#88
mastrdrver
Someone is saying that the final picture in that magazine has a binary string that says: ZenOC@air=5ghz
phanbueyI did - because it doesn't matter. The comment I was responding to was referring to game performance.

Here is my process: Look at chart, see score. See Intel midrange chip from last gen ahead of AMD high-end chip releasing in 2 months; see Intel high end desktop chip 20% ahead. Feel sad, point out reality of data.

Here is the alternative process: Look at chart, make excuses as to why score is so low. Discount data.
So are you telling me that an old intel 6900k is better then the 6600k because it's higher on the chart and because of that Intel's newer processor is crap?
Posted on Reply
#89
phanbuey
mastrdrverSomeone is saying that the final picture in that magazine has a binary string that says: ZenOC@air=5ghz
So are you telling me that an old intel 6900k is better then the 6600k because it's higher on the chart and because of that Intel's newer processor is crap?
What?
Posted on Reply
#90
kruk
phanbueyWhat?
He is saying that you don't have a clue how game benchmarks work. If you look at the results, it's clear that results correlate with clock speed. The proper way to analyze this graph would be normalization by dividing the score with speed in MHz. You might be surprised with the result ...
Posted on Reply
#91
Nordic
mastrdrverSeeing as how the 8 core 16 thread has a TDP of ~95w, I'd expect a 4 core 8 thread to have a TDP of like ~60W.
Exactly. Would AMD make a 4core part with high clockspeeds? The only problem is I could see it competing with their 8 core part in some ways.
Posted on Reply
#92
phanbuey
krukHe is saying that you don't have a clue how game benchmarks work. If you look at the results, it's clear that results correlate with clock speed. The proper way to analyze this graph would be normalization by dividing the score with speed in MHz. You might be surprised with the result ...
OH MY GOD? IT SCALES with clockspeed!?!? well thanks for pointing that out, that changes EVERYTHING.

Except wait, nope... zen still has a lower clockspeed and won't really make sense for games.
james888Exactly. Would AMD make a 4core part with high clockspeeds? The only problem is I could see it competing with their 8 core part in some ways.
I think they will have to - the high end product looks geared for content creation.
Posted on Reply
#93
kruk
phanbueyOH MY GOD? IT SCALES with clockspeed!?!? well thanks for pointing that out, that changes EVERYTHING.

Except wait, nope... zen still has a lower clockspeed and won't really make sense for games.

.
Now I see it. You are right, this is the final product. They will produce ONLY ONE cpu model: It will be named Ryzen, it will have 8C/16T, 3.15 GHz base and 3.3 GHz boost. It's all clear now. How could I be so blind? /s

STOP TROLLING!!
Posted on Reply
#95
phanbuey
krukNow I see it. You are right, this is the final product. They will produce ONLY ONE cpu model: It will be named Ryzen, it will have 8C/16T, 3.15 GHz base and 3.3 GHz boost. It's all clear now. How could I be so blind? /s

STOP TROLLING!!
Look I'm sorry if i am hurting your feelings by pointing out that, at least the first model, that is shipping in late feb /march is the 8ct/16 and its not ideal for games (not factoring in overclocking) as it is matched/outperformed by a 58W $250 part. Feel free to divide the numbers however you would like and make any assumptions regarding future models that will make you feel better.

I am not, nor was I ever talking about the IPC of ryzen or how it is inferior in games; but about the actual chip that is rumored to come out; pre any overclocks is not going to be the ideal product for gamers.

I will be waiting for a higher clocked quad or hex.
Posted on Reply
#96
kruk
phanbueyLook I'm sorry if i am hurting your feelings by pointing out that, at least the first model, that is shipping in late feb /march is the 8ct/16 and its not ideal for games (not factoring in overclocking) as it is matched/outperformed by a 58W $250 part. Feel free to divide the numbers however you would like and make any assumptions regarding future models that will make you feel better.

I am not, nor was I ever talking about the IPC of ryzen or how it is inferior in games; but about the actual chip that is rumored to come out; pre any overclocks is not going to be the ideal product for gamers.

I will be waiting for a higher clocked quad or hex.
Nobody except AMD (and mabye some lucky reviewers) knows what the final base/turbo clock of the first released model will be, but feel free to compare a prototype to a already released CPU in whichever way you would like - if that will make you feel better. I rest my case ...
Posted on Reply
#97
phanbuey
krukNobody except AMD (and mabye some lucky reviewers) knows what the final base/turbo clock of the first released model will be, but feel free to compare a prototype to a already released CPU in whichever way you would like - if that will make you feel better. I rest my case ...
umm.. thats the point of this thread lol, its pure speculation with what very little data we have to go on. And it gives us a very rough glimpse of what the performance will be like.
Posted on Reply
#98
efikkan
krukThe proper way to analyze this graph would be normalization by dividing the score with speed in MHz. You might be surprised with the result ...
Actually no, gaming performance is better with higher clocks, but it's not proportional to clock frequency.
There are a lot of other factors that matters, like i7-6900K is beating i5-6600K (if I read the chart correctly), even when i7-6900K is an "older" architecture and has "lower clocks". There are several reasons for this, first of all boosting works differently, especially on CPUs with many cores. Different amounts of L3 cache matters, and of course more cores can sometimes free up resources.

As always, real world performance is the only thing that matters. I wish more of these benchmarks were performed with background applications to get a more realistic workload.
Posted on Reply
#99
Melvis
phanbueyI did - because it doesn't matter. The comment I was responding to was referring to game performance.

Here is my process: Look at chart, see score. See Intel midrange chip from last gen ahead of AMD high-end chip releasing in 2 months; see Intel high end desktop chip 20% ahead. Feel sad, point out reality of data.

Here is the alternative process: Look at chart, make excuses as to why score is so low. Discount data.
What? soooo clock speed doesnt matter for gaming is that what your saying? are you really that dense? :slap:

:roll: 20% ahead? wow I think the problem is you have no idea how to read, you might need to go back to the classroom boy and learn the basic's first before commenting and making yourself look like a total dumb ass!

Maybe you work for intel? and your scared? and your job is to inflict as much BS into these kind of threads to mislead people? ether way your comments are so incorrect and have no real meaning that you have just told us all that read this thread how in-mature and full of trolling BS you are :)
Posted on Reply
#100
phanbuey
MelvisWhat? soooo clock speed doesnt matter for gaming is that what your saying? are you really that dense? :slap:

:roll: 20% ahead? wow I think the problem is you have no idea how to read, you might need to go back to the classroom boy and learn the basic's first before commenting and making yourself look like a total dumb ass!

Maybe you work for intel? and your scared? and your job is to inflict as much BS into these kind of threads to mislead people? ether way your comments are so incorrect and have no real meaning that you have just told us all that read this thread how in-mature and full of trolling BS you are :)
I like the rage. makes me happy, tells me you care. Also the engrish... in-mature lol.

The 6700K is 20% ahead of the Zen in those leaks under gaming :cool:
Posted on Reply
Add your own comment
Nov 21st, 2026 22:22 PST change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts