Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [Various] AMD’s Zen Eight Core CPU Delivers Double the Performance of the FX 8350 (Update 2)
New Posts  All Forums:Forum Nav:

[Various] AMD’s Zen Eight Core CPU Delivers Double the Performance of the FX 8350 (Update 2) - Page 54

post #531 of 564
Quote:
Originally Posted by Travieso View Post

he made a speculation by himself and then was disappointed with his own speculation.


"AMD gonna launch 4 core Zen with the same performance of i5 6600K at the same price point, that would be really disappointing... blah... blah..."

i mean... ok anyone can determine performance of 4 core Zen on 40% ipc increase and 8 core performance, that's fair point.

but how did he know about the price point ? AMD haven't stated anything about Zen's price yet and there's absolutely no rumour about it.

he's disappointed with Zen's price that was also made up by himself. that's totally nonsense.

There's no 4C/8T Zen in the FX range anyway, it's going to be a 6C/12T and an 8C/12T.

The 4C/8T is going to be an APU and is rumoured to have the graphics power of the current consoles.

All the CPU's look to be pretty good in their own regard, hoping for aggressive pricing.
post #532 of 564
Quote:
Originally Posted by Buris View Post

Even a stacked 16 core/32 thread AMD server CPU would obviously not even have a chance against intel if intel's already offering stacked 24 cores (that means they can fit 12 cores on a non-stacked chip for lga 2011), which means a 6-core mainstream skylake could be easily produced. potentially even an 8-core.
Intel's dies are all monolithic with the exception of the chips that have EDRAM. This means that intel doesnt stack or MCM a pair of 12c dies, they literally make a single massive 24C die for their biggest chips. Its how they got 18c into Haswell-EX, and 15c into IB-EX.

There's also the rumor floating around from the CERN leaks that AMD intends on making a 32c/64t dual-die chip using a pair of 16c dies in a manner similar to what they did for pretty much all socket G34 processors.
Leviathan
(17 items)
 
Charred
(10 items)
 
 
CPUMotherboardGraphicsGraphics
Xeon E5-2690 Biostar TPower X79 PNY GTX 660 2GB HP GT 440 OEM 
RAMHard DriveHard DriveHard Drive
Gskill Ripjaws 4x2GB 1600mhz Seagate Barracuda 500GB Seagate Barracuda 1.5TB Western Digital Caviar Blue 640GB 
Hard DriveCoolingOSMonitor
Patriot Pyro 60GB Xigmatek Gaia Windows 7 Ultimate Acer S230HL 
MonitorKeyboardPowerCase
Princeton 1280x1024 19" Logitech K120 Seasonic G550 Xclio Nighthawk 
Mouse
Logitech MX310 
  hide details  
Reply
Leviathan
(17 items)
 
Charred
(10 items)
 
 
CPUMotherboardGraphicsGraphics
Xeon E5-2690 Biostar TPower X79 PNY GTX 660 2GB HP GT 440 OEM 
RAMHard DriveHard DriveHard Drive
Gskill Ripjaws 4x2GB 1600mhz Seagate Barracuda 500GB Seagate Barracuda 1.5TB Western Digital Caviar Blue 640GB 
Hard DriveCoolingOSMonitor
Patriot Pyro 60GB Xigmatek Gaia Windows 7 Ultimate Acer S230HL 
MonitorKeyboardPowerCase
Princeton 1280x1024 19" Logitech K120 Seasonic G550 Xclio Nighthawk 
Mouse
Logitech MX310 
  hide details  
Reply
post #533 of 564
Quote:
Originally Posted by Travieso View Post

he made a speculation by himself and then was disappointed with his own speculation.


"AMD gonna launch 4 core Zen with the same performance of i5 6600K at the same price point, that would be really disappointing... blah... blah..."

i mean... ok anyone can determine performance of 4 core Zen on 40% ipc increase and 8 core performance, that's fair point.

but how did he know about the price point ? AMD haven't stated anything about Zen's price yet and there's absolutely no rumour about it.

he's disappointed with Zen's price that was also made up by himself. that's totally nonsense.

Okay. Well you might have a point there. Nonetheless, lets all be nice. cheers.gif
     
Reply
     
Reply
post #534 of 564
cinebench has always been favourable to amd - its the rest of the benchmarks that fail - And i am one that wants amd to succeed thumb.gif
post #535 of 564
Quote:
Originally Posted by ladcrooks View Post

cinebench has always been favourable to amd - its the rest of the benchmarks that fail - And i am one that wants amd to succeed thumb.gif

favourable to amd

cinebench

lachen.gif
post #536 of 564
Quote:
Originally Posted by ladcrooks View Post

cinebench has always been favourable to amd - its the rest of the benchmarks that fail - And i am one that wants amd to succeed thumb.gif

https://www.ftc.gov/sites/default/files/documents/cases/091216intelcmpt.pdf
PC
(19 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 7 1700 Crosshair VI Hero GTX970 @1492MHz F4-3200C14-8GFX 
Hard DriveHard DriveHard DriveOptical Drive
Crucial MX200 Samsung 850 Evo a few HDD's LG BluRay BH16NS40 
CoolingOSOSMonitor
Noctua NH-D15 Windows 10 Pro Arch Linux LG W2442PA 
MonitorKeyboardPowerCase
Viewsonic XG2401 Cherry MX-Board 3.0 Corsair RM850i Fractal Define R5 
MouseMouse PadAudio
Logitech G403 Roccat Taito OnBoard + Focusrite Scarlett 18i8 
  hide details  
Reply
PC
(19 items)
 
  
CPUMotherboardGraphicsRAM
Ryzen 7 1700 Crosshair VI Hero GTX970 @1492MHz F4-3200C14-8GFX 
Hard DriveHard DriveHard DriveOptical Drive
Crucial MX200 Samsung 850 Evo a few HDD's LG BluRay BH16NS40 
CoolingOSOSMonitor
Noctua NH-D15 Windows 10 Pro Arch Linux LG W2442PA 
MonitorKeyboardPowerCase
Viewsonic XG2401 Cherry MX-Board 3.0 Corsair RM850i Fractal Define R5 
MouseMouse PadAudio
Logitech G403 Roccat Taito OnBoard + Focusrite Scarlett 18i8 
  hide details  
Reply
post #537 of 564
Quote:
Originally Posted by ladcrooks View Post

cinebench has always been favourable to amd - its the rest of the benchmarks that fail - And i am one that wants amd to succeed thumb.gif

It was favorable to AMD in the Phenom II era, but not since. The construction cores perform very poorly in Cinebench relative to their performance in other benchmarks.

However, I expect Zen to change that. It will probably excel in highly SIMD loads as well as regaining the mixed-compute (integer+floating+memory) dynamic that gave the Phenom II the "multi-media" strength it had.

Zen can, in theory - when peaked, concurrently execute 4 integer ops, two memory ops, and four floating point ops every cycle (or for as long as it can keep the instructions coming).

This is well ahead of Intel's capabilities for concurrent executions and lends itself well to SIMD and SMT scaling. Zen+ need only feed the execution assets more efficiently while helping the instructions retire quickly to gain a 15% boost. The core design is simply that capable (from what we know).
post #538 of 564
Quote:
Originally Posted by EightDee8D View Post

favourable to amd

cinebench

lachen.gif

Nowdays it actually is, since R15 uses just SSE3 instructions. If (when) Cinebench would use modern instructions such as AVX2, it would be a complete bloodbath compared to current situation.
post #539 of 564
Quote:
Originally Posted by The Stilt View Post

Nowdays it actually is, since R15 uses just SSE3 instructions. If (when) Cinebench would use modern instructions such as AVX2, it would be a complete bloodbath compared to current situation.

But AMD jaguar is good even in AVX...

Jaguar core would simple destroy vishera and zambezi also steamroller.


Edited by Themisseble - 6/9/16 at 8:50am
post #540 of 564
Quote:
Originally Posted by EightDee8D View Post

favourable to amd

cinebench

lachen.gif

Glad i made you chuckle - lets say i meant in, general, video editing - cos they do, in amd terms and with winrar thats their best results tongue.gif

But i do have luuuuuuuuuve for amd wubsmiley.gif
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Rumors and Unconfirmed Articles
Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [Various] AMD’s Zen Eight Core CPU Delivers Double the Performance of the FX 8350 (Update 2)