AMD Further Unveils Zen Processor Details

The video is in the link in the OP of this thread.



And I think you are right KazeoHin , it's more of a GPU bottleneck but I guess still a good sign as I bet most people here want to play games on this thing.
 
cybereality said:
The video is in the link in the OP of this thread.



And I think you are right KazeoHin , it's more of a GPU bottleneck but I guess still a good sign as I bet most people here want to play games on this thing.
Click to expand...


Yes, BUT playing at 4K with a single GPU is the best way to make a slow processor seem fast. I run a slow-as-a-dog server CPU in my main rig (great for rendering and multitasking) and I imagine that my in-game experience at 4K is the same as an i7 6700. The GPUs are the bottleneck. Bring that down to the more popular high-refresh 1080p or 1440p/ ultrawide and the CPU starts having to keep up with faster frames.
 
cybereality said:
The video is in the link in the OP of this thread.



And I think you are right KazeoHin , it's more of a GPU bottleneck but I guess still a good sign as I bet most people here want to play games on this thing.
Click to expand...


They may as well have demoed minesweeper because you'll be hard pressed to find a CPU intensive game at 4K
 
Ieldra said:
They may as well have demoed minesweeper because you'll be hard pressed to find a CPU intensive game at 4K
Click to expand...

If they showed it running a the game on Acer's new 200Hz 1080p monitor, that would be a better demonstration...
 
KazeoHin said:
If they showed it running a the game on Acer's new 200Hz 1080p monitor, that would be a better demonstration...
Click to expand...


Was it an RX480 running Deus Ex on Zen at 4K? That would be doubly concerning

CPU demos at 1080p, please. Always 1080p.
 
Ieldra said:
Was it an RX480 running Deus Ex on Zen at 4K? That would be doubly concerning

CPU demos at 1080p, please. Always 1080p.
Click to expand...

The system was running a furryX, and it looked like it was not at max refresh, but often cameras make monitors seem stuttery when they aren't
 
lolfail9001 said:
Well, the most popular game on Steam is CPU intensive at 4k.
Click to expand...

DotA 2?

Speaking of which, who are all these people who play DotA 2?

It's the largest game on steam.by a wide margin, but I don't think I've ever spoken to anyone who has played it.

South Koreans? It seems like the type of game South Koreans would be into.
 
Last edited:
Ieldra said:
I'm at 1440p and I can't hold 300fps all the time, although I am running completely maxed out with 8xmsaa
Click to expand...

Uhh. Most people consider maxing out to being able to vsync to their monitor refresh.

For me that is 60hz. I consider anything above that placebo.
 
Zarathustra[H] said:
Uhh. Most people consideracing out launching to theironitor refresh.

For me that is 60hz. I consider anything above that placebo.
Click to expand...

Nah, I have my refresh at 100hz with fast sync enabled, maybe it is placebo but I swear csgo feels more responsive with uncapped fps

I think he meant csgo not Dota 2

Lots of people play Dota 2 :p many people in developing countries because it's free, but many westerners too. It runs on anything and looks pretty good, I don't like classic mobas though, third person ones are way more fun Imo
 
Ieldra said:
Nah, I have my refresh at 100hz with fast sync enabled, maybe it is placebo but I swear csgo feels more responsive with uncapped fps

I think he meant csgo not Dota 2
Click to expand...

Ugh. Some editing is in order. I hate typing on phones.

My desktop isn't up and running yet...
 
So I must have posted this 3-4 times now, but...

Over a year ago....

There was a leak stating that ZEN was going to max out in top form at 3Ghz base, 3.5Ghz turbo due to the process it was produced on. There was another note that besides an additional 10% increase in IPC ZEN + was hoping to add 200-400Mhz additional headroom on the same process.

It was also mentioned that ZEN can turbo higher but the thermals go nuts if you do. So the 4c version...with near extreme cooling might be able to be overclocked to 4Ghz.
 
DukenukemX said:
You do have a point, as AMD would have clocked it higher if they could, and they didn't. We know from Polaris that AMD's 14nm process doesn't promote clock speeds very much. The BroadWell-E CPUs don't clock well either, and turbo as high as 3.5Ghz. Otherwise 3.0Ghz is the rated base speed for a 5960X.
Click to expand...


If they weren't so damn expensive :ROFLMAO:
Screenshot_20160819-061815_zpsclludykk.png


mmm
Guaranteed overclock of 4.4 Ghz on all 10 cores
 
Zarathustra[H] said:
Ugh. Some editing is in order. I hate typing on phones.

My desktop isn't up and running yet...
Click to expand...

I have trouble with this forum on mobile, copy paste is non-functioning half the time, text selection/highlighting is the problem. Drives me nuts, and it's only this forum... God knows why, doesn't he... Kyle...

Mackintire said:
So I must have posted this 3-4 times now, but...

Over a year ago....

There was a leak stating that ZEN was going to max out in top form at 3Ghz base, 3.5Ghz turbo due to the process it was produced on. There was another note that besides an additional 10% increase in IPC ZEN + was hoping to add 200-400Mhz additional headroom on the same process.

It was also mentioned that ZEN can turbo higher but the thermals go nuts if you do. So the 4c version...with near extreme cooling might be able to be overclocked to 4Ghz.
Click to expand...

Speaking of temps, why are FX CPUs rated for such low temps? I was shocked, friend of mine with an OC 8350 once saw my load temps in p95 and thought I was on a suicide run (70c).

Apparently something like 64C is his limit, what
 
Ieldra said:
I have trouble with this forum on mobile, copy paste is non-functioning half the time, text selection/highlighting is the problem. Drives me nuts, and it's only this forum... God knows why, doesn't he... Kyle...



Speaking of temps, why are FX CPUs rated for such low temps? I was shocked, friend of mine with an OC 8350 once saw my load temps in p95 and thought I was on a suicide run (70c).

Apparently something like 64C is his limit, what
Click to expand...

I think it is how AMD measures the temperatures. My old Phenom II had a limit of 60 degrees. It's in where the T-Junction is measured and how far from the central cores it is.
 
KazeoHin said:
I think it is how AMD measures the temperatures. My old Phenom II had a limit of 60 degrees. It's in where the T-Junction is measured and how far from the central cores it is.
Click to expand...

Yeah but it's like 40c lower than Intel's threshold (for core temps), I feel like there's more to it than just the temp sensor being far from the cores
 
Steve said:
I think everyone is really hoping for the Zen to be stellar so that AMD once again gives Intel a run for their money and, in turn, spurs healthy competition and innovation.
Click to expand...

This is ultimately all I want. The Athlon Xp/AMD64 era were exciting times on that front. If AMD puts out something great, I'd like to see what Intel can respond with instead of giving us minimal IPC improvements every gen as they have for the last few years.
 
rat said:
This is ultimately all I want. The Athlon Xp/AMD64 era were exciting times on that front. If AMD puts out something great, I'd like to see what Intel can respond with instead of giving us minimal IPC improvements every gen as they have for the last few years.
Click to expand...

Intel haven't been slacking off, it's just hard to eke out more single thread performance, you start getting diminishing returns with respect to complexity of scheduling/control logic.

The next big jump should come from change in materials with gallium based process increasing viable clock speeds significantly
 
64c is not the limit of the FX cpus

They thermal throttle at 70c, and have a hard shut down at 90c. They will also throttle if the cpu socket temperature goes above 75c.

Problem is most of the AMD cpu's start getting unstable above 60c. Not really a stock clock problem, but definitely causes an overclocked system to go unstable. Before I was watercooled on my FX I could only do 4.8ghz and hitting 60c. I could push the voltage more, but it would just go unstable. With the water cooling cpu doesn't go above 45c, and I'm at 5ghz with more voltage.
 
Last edited:
Shintai said:
We also had benchmarks with Bulldozer beating Intel with 40% and Barcelona beating Intel with 50%. So I wouldn't put too much in it yet. Specially not in a benchmark that can easily be tampered and a custom image run on unknown hardware and settings. All you know is its a Zen CPU, its a 6900K and it runs some sort of Blender.

The AOTS benchmark leak had much more details to compare.
Click to expand...

To you only shit that matters is what leads to zen being a fail. You come out so one sided that it isn't even funny. you really have stock invested in intel? Common man give it a rest and wait until final numbers. They should clock for clock comparison what else you want, you are believing more a game benchmark to describe actual performance in all possible scenarios? Common get real and enough with the amd bashing. Wait until the chip is out.
 
Ieldra said:
Broadwell level single thread performance would be a huge victory, I just won't believe it until I see hard numbers across a range of benchmarks.
Click to expand...

True thanks for saying that. Lol. I think if this even gets in with 10% that's a win. Amd doesn't have to kill it, they just have to become competitive.
 
ManofGod said:
Really? Sorry but, I did not see any such thing so, perhaps you just remember incorrectly?
Click to expand...
He hates amd bro! You can put a magnifying glass in front of him to see better he will probably still find a way to bash amd! Lol
 
ManofGod said:
No, it will not be 800 to 1000$ in my opinion. However, I could see $300 to $450 as long as the 4c/8t one does ok.
Click to expand...
Didn't AMD say there will be no 4c/8t part? Minimum 6c/12t.
Shintai said:
The AOTS benchmark leak had much more details to compare.
Click to expand...
In all probability the AOTS leak was not intended, which is why the results were removed.
 
Hypergreatthing said:
Clock to clock speed doesn't matter when the standard clock speed of a 6900K is 3.2 (turboed to 3.7 under load) compared to a zen at 3.0.

Clock speed is fairly irrelevant when a zen will have trouble going higher than 3.0 ghz.
Click to expand...
Do you even engineering sample brah
 
Ocellaris said:
As usual, AMD's plan is to give people a glimmer of hope with no third party verification, then tell them to wait six months for something important to happen.
Click to expand...
ai people who don't bother reading, hot hardware, PC perspective etc all were there even the biggest Intel fanboy Ryan Shrout was mega impressed he is singing praises, if he is exited so am I.
 
Shintai said:
When samples are out to partners, mobo makers etc. Then there are no secrets anymore. Specially not when you are fabless. Its not prototypes anymore.
Click to expand...
Polaris ES was 800MHz at this point when it was out to partners, etc. Engineering samples are about testing interface and core stability, not performance numbers.
 
More than happy with my 4770k, but when it comes to upgrade time, I hope that there is once again a decent choice to make.

Remember the Athlon XP days fondly!
 
Deathroned said:
ai people who don't bother reading, hot hardware, PC perspective etc all were there even the biggest Intel fanboy Ryan Shrout was mega impressed he is singing praises, if he is exited so am I.
Click to expand...

Yeah but this is all based on AMD saying 'we outperform broadwell-e'.

They literally ran blender CPU bench on a project and you can see both tests finish at more or less the same time, but no hard data is given.

I'm not familiar with blender as a CPU benchmark so I checked their results DB and it's a huge squash near the top with all kind of weird results.

Other than that they demoed a game at 4K.

Basically they didn't provide any performance numbers at all, everyone is impressed because AMD said they're better than broadwell - which would indeed be impressive if true.
 
Ieldra said:
Yeah but this is all based on AMD saying 'we outperform broadwell-e'.
They literally ran blender CPU bench on a project and you can see both tests finish at more or less the same time, but no hard data is given.
I'm not familiar with blender as a CPU benchmark so I checked their results DB and it's a huge squash near the top with all kind of weird results.
Other than that they demoed a game at 4K. Basically they didn't provide any performance numbers at all, everyone is impressed because AMD said they're better than broadwell - which would indeed be impressive if true.
Click to expand...

The hard data is that they ran both at 3 GHZ.
You are funny
Why do you lack knowledge of what Blender does then still comment on it. Because your lack of knowledge you claim no performance numbers , the performance is that it beats the Intel cpu at equal cores threads and gigahertz in the blender demo.
AMD did not say there faster they showed it , as in everyone could see it performs better in the blender demo.
 
Pieter3dnow said:
The hard data is that they ran both at 3 GHZ.
You are funny
Why do you lack knowledge of what Blender does then still comment on it. Because your lack of knowledge you claim no performance numbers , the performance is that it beats the Intel cpu at equal cores threads and gigahertz in the blender demo.
AMD did not say there faster they showed it , as in everyone could see it performs better in the blender demo.
Click to expand...

They ran both at 3GHz... That's not performance data, they showed blender finishing first on Zen, but no numbers.

Even then, go look at the blender results DB, it's not like you need some intricate knowledge of blender to interpret results....

Among the CPUs that outperform a 6950X there's six year old Gulftown Xeons

Perhaps instead of saying genuinely dumb things like 'you lack knowledge of blender' you could apply some of that idle gray matter you have eh?
 
Ieldra said:
They ran both at 3GHz... That's not performance data, they showed blender finishing first on Zen, but no numbers.
Even then, go look at the blender results DB, it's not like you need some intricate knowledge of blender to interpret results....
Among the CPUs that outperform a 6950X there's six year old Gulftown Xeons. Perhaps instead of saying genuinely dumb things like 'you lack knowledge of blender' you could apply some of that idle gray matter you have eh?
Click to expand...
You have no clue whatsoever I highlighted your conversation points then you claim I am saying that you lack knowledge , you are saying you lack knowledge
And if you want performance numbers you are free to buy a Zen for yourself or wait for the reviews ...
Ieldra said:
Yeah but this is all based on AMD saying 'we outperform broadwell-e'.
They literally ran blender CPU bench on a project and you can see both tests finish at more or less the same time, but no hard data is given.
I'm not familiar with blender as a CPU benchmark so I checked their results DB and it's a huge squash near the top with all kind of weird results.
Other than that they demoed a game at 4K.
Basically they didn't provide any performance numbers at all, everyone is impressed because AMD said they're better than broadwell - which would indeed be impressive if true.
Click to expand...
 
Pieter3dnow said:
You have no clue whatsoever I highlighted your conversation points then you claim I am saying that you lack knowledge , you are saying you lack knowledge
And if you want performance numbers you are free to buy a Zen for yourself or wait for the reviews ...
Click to expand...

I said I'm not familiar with blender CPU test, meaning I don't know if it scales well with core count, clock, FP performance...

So I went and looked the results database? Did you do that? Did you do anything at all? No. Instead you come here and claim AMD presented hard data when they did not, they did not even specify whether quad channel memory was used on the Intel system.

Anyway here is deus ex at 4k

dem_3840_12.png
 
You must log in or register to reply here.
Back
Top