Overclock.net › Forums › Industry News › Hardware News › [PCPOP]Haswell Core i7 4770K retail version of the first test
New Posts  All Forums:Forum Nav:

[PCPOP]Haswell Core i7 4770K retail version of the first test - Page 32

post #311 of 352
Quote:
Originally Posted by ohhgourami View Post

I'm just hoping Microcenter will have their i5 deal on launch week. I've had to use my old C2D laptop for 2 months - so sluggish and noisy.

lachen.gif

not going to happen unfortunately for atleast a month. They should be list price come sunday
Core I7 5960X
(10 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core I7 5960X ASRock X99 OC Formula MSI GTX 970 Gaming MSI GTX 970 Gaming 
RAMHard DriveCoolingOS
G. Skill 16GB DDR4 2600 Sandisk Extreme II 240GB SSD Custom 480 Water Setup Windows 10 
PowerCase
AX1500I Corsair 900D 
  hide details  
Reply
Core I7 5960X
(10 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core I7 5960X ASRock X99 OC Formula MSI GTX 970 Gaming MSI GTX 970 Gaming 
RAMHard DriveCoolingOS
G. Skill 16GB DDR4 2600 Sandisk Extreme II 240GB SSD Custom 480 Water Setup Windows 10 
PowerCase
AX1500I Corsair 900D 
  hide details  
Reply
post #312 of 352
Looks like its just another 3770k with better IGP, i.e still playing a lottery and hoping for a good chip. 4.5Ghz at 1.26v is easy!
Current rig
(11 items)
 
  
Reply
Current rig
(11 items)
 
  
Reply
post #313 of 352
Quote:
Originally Posted by svenge View Post

You've got it backwards. "Ticks" are die-shrinks (e.g. Ivy Bridge's size reduction from Sandy Bridge), "tocks" are new micro-architectures at the same size as the previous tick (e.g. Haswell, which is the same size as Ivy Bridge).

Broadwell will presumably be a 14nm die-shrink "tick" of Haswell (which is currently 22nm). Then Skylake would be the new micro-architecture "tock" at 14nm.
Ahh, Thanks for informing me of that I've got it all mixed up.
post #314 of 352
Quote:
Originally Posted by svenge View Post

You've got it backwards. "Ticks" are die-shrinks (e.g. Ivy Bridge's size reduction from Sandy Bridge), "tocks" are new micro-architectures at the same size as the previous tick (e.g. Haswell, which is the same size as Ivy Bridge).

Broadwell will presumably be a 14nm die-shrink "tick" of Haswell (which is currently 22nm). Then Skylake would be the new micro-architecture "tock" at 14nm.
Ahh, Thanks for informing me of that I've got it all mixed up.
post #315 of 352
Quote:
Originally Posted by Stay Puft View Post

Lets say 225w for a 4.6-4.8Ghz 3770K plus 125W-130w for each 660. I think 460-480w for an ivy bridge setup compared to my overclocked 660's/bloomfield rig using 550w. Yeah totally worth it for power consumption alone lachen.gif

Ivy has 85 W, and overclocked has about 130 W, if someone didn't screwed up voltage. I wonder how you arrived at 225W for Ivy. It's the same 3D transistors as HW anyway. (Just different voltage regulators, but as long as user had great MB, and was willing to properly set stuff manually, it's not that bad difference.)

Actually I seen proper graphs for Ivy with voltage, temperature and power usage in W as 3D graphs. High voltage is bad for Ivy, and would be bad for HW. Low voltage is quite fine.
Quote:
Originally Posted by Tomalak View Post

Not really a problem, since disabling those power states fixes it?
Company that developed my PSU said they developed zero power PSU 2 years ago already, because they started to work on that problem before it even arose. Then they released a list of PSUs that are NOT compatible. (Basically as long as you didn't buy value version, or had a recent PSU you were fine.)

Considering a lot of PSU are converting to 12V, then they are converting from 12V DC to 3.3 and 5V directly, there is always some load on PSU, as long as PSU didn't shut down.
Edited by Raghar - 5/29/13 at 6:04pm
post #316 of 352
Quote:
Originally Posted by Raghar View Post

Ivy has 85 W, and overclocked has about 130 W, if someone didn't screwed up voltage. I wonder how you arrived at 225W for Ivy. It's the same 3D transistors as HW anyway. (Just different voltage regulators, but as long as user had great MB, and was willing to properly set stuff manually, it's not that bad difference.)

Actually I seen proper graphs for Ivy with voltage, temperature and power usage in W as 3D graphs. High voltage is bad for Ivy, and would be bad for HW. Low voltage is quite fine.
Company that developed my PSU said they developed zero power PSU 2 years ago already, because they started to work on that problem before it even arose. Then they released a list of PSUs that are NOT compatible. (Basically as long as you didn't buy value version, or had a recent PSU you were fine.)

Considering a lot of PSU are converting to 12V, then they are converting from 12V DC to 3.3 and 5V directly, there is always some load on PSU, as long as PSU didn't shut down.

You think Ivy uses 130W at 4.6Ghz? lachen.gif Its not even close



Under cinebench at 4.4Ghz i'm pulling 444w according to my Killawatt.
Edited by Stay Puft - 5/29/13 at 6:14pm
Core I7 5960X
(10 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core I7 5960X ASRock X99 OC Formula MSI GTX 970 Gaming MSI GTX 970 Gaming 
RAMHard DriveCoolingOS
G. Skill 16GB DDR4 2600 Sandisk Extreme II 240GB SSD Custom 480 Water Setup Windows 10 
PowerCase
AX1500I Corsair 900D 
  hide details  
Reply
Core I7 5960X
(10 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core I7 5960X ASRock X99 OC Formula MSI GTX 970 Gaming MSI GTX 970 Gaming 
RAMHard DriveCoolingOS
G. Skill 16GB DDR4 2600 Sandisk Extreme II 240GB SSD Custom 480 Water Setup Windows 10 
PowerCase
AX1500I Corsair 900D 
  hide details  
Reply
post #317 of 352
Quote:
Originally Posted by Stay Puft View Post

Lets say 225w for a 4.6-4.8Ghz 3770K plus 125W-130w for each 660. I think 460-480w for an ivy bridge setup compared to my overclocked 660's/bloomfield rig using 550w. Yeah totally worth it for power consumption alone lachen.gif

Quote:
Originally Posted by Stay Puft View Post

You think Ivy uses 130W at 4.6Ghz? lachen.gif Its not even close



Under cinebench at 4.4Ghz i'm pulling 444w according to my Killawatt.

You just proved a 0.8 ghz higher clocked Ivy Bridge uses ~160 watts less than an i7 920... rolleyes.gif

Totally not a big difference at all.
Millenium Falcon
(24 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 4930k MSI Big Bang Xpower II EVGA GTX 690 Patriot Viper II Sector 7 
Hard DriveHard DriveHard DriveHard Drive
OCZ Deneva 2 Corsair Force 3 Maxtor Western Digital Green 
Optical DriveCoolingCoolingCooling
Samsung BD/DVD-RW Swiftech MCP655 x2 Black Ice GTX 480 Black Ice GTX 280 
CoolingCoolingCoolingCooling
Alphacool Repack Dual D5 Watercool Heatkiller 3.0 Alphacool GTX 690 fullcover Bitspower Big Bang Xpower II fullcover 
OSMonitorKeyboardPower
Windows 8.1 64-bit Professional 3x Dell S2340 Max Keyboard Durandal CoolerMaster V1000 
CaseMouseMouse PadAudio
Azza Genesis 9000B Logitech G700 Roccat Alumic Onkyo HT-S9100THX 
  hide details  
Reply
Millenium Falcon
(24 items)
 
  
CPUMotherboardGraphicsRAM
Core i7 4930k MSI Big Bang Xpower II EVGA GTX 690 Patriot Viper II Sector 7 
Hard DriveHard DriveHard DriveHard Drive
OCZ Deneva 2 Corsair Force 3 Maxtor Western Digital Green 
Optical DriveCoolingCoolingCooling
Samsung BD/DVD-RW Swiftech MCP655 x2 Black Ice GTX 480 Black Ice GTX 280 
CoolingCoolingCoolingCooling
Alphacool Repack Dual D5 Watercool Heatkiller 3.0 Alphacool GTX 690 fullcover Bitspower Big Bang Xpower II fullcover 
OSMonitorKeyboardPower
Windows 8.1 64-bit Professional 3x Dell S2340 Max Keyboard Durandal CoolerMaster V1000 
CaseMouseMouse PadAudio
Azza Genesis 9000B Logitech G700 Roccat Alumic Onkyo HT-S9100THX 
  hide details  
Reply
post #318 of 352
Quote:
Originally Posted by Tsumi View Post


You just proved a 0.8 ghz higher clocked Ivy Bridge uses ~160 watts less than an i7 920... rolleyes.gif

According to that graph it does but my numbers suggest otherwise. Even if those numbers were correct 160W's is not a reason to solely upgrade to haswell.

Especially when you look at this graph

350x700px-LL-ed73cbcf_GTX680-games-1.jpeg
Core I7 5960X
(10 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core I7 5960X ASRock X99 OC Formula MSI GTX 970 Gaming MSI GTX 970 Gaming 
RAMHard DriveCoolingOS
G. Skill 16GB DDR4 2600 Sandisk Extreme II 240GB SSD Custom 480 Water Setup Windows 10 
PowerCase
AX1500I Corsair 900D 
  hide details  
Reply
Core I7 5960X
(10 items)
 
  
CPUMotherboardGraphicsGraphics
Intel Core I7 5960X ASRock X99 OC Formula MSI GTX 970 Gaming MSI GTX 970 Gaming 
RAMHard DriveCoolingOS
G. Skill 16GB DDR4 2600 Sandisk Extreme II 240GB SSD Custom 480 Water Setup Windows 10 
PowerCase
AX1500I Corsair 900D 
  hide details  
Reply
post #319 of 352
Quote:
Originally Posted by ComputerRestore View Post

From what I've seen of current games that use more than 4 Cores, it still wont make a huge difference. AMD will still need to improve on their single core performance to be competitive Apples to Apples.

i.e. - 25% CPU usage over 8 Cores vs 50% CPU usage over 4 Cores - Intel still has stronger per core performance so as long as it has CPU headroom it will be better.

CPU performance doesn't make much of a difference in most games anyway, I went from an FX-4170 to an i5 3570k and most games still play equally as well.
Quote:
Originally Posted by AlphaC View Post

Please read this.
http://www.overclockers.ru/lab/53063_3/Vybor_optimalnogo_processora_stoimostju_170-350_dlya_igrovogo_PK.html#25

for things like Skyrim AMD is absolute garbage , a FX-8350 clocked at 4.7GHz is slower than a Core i5-2320 @ 3.6GHz


and here's Dragon Age Origins


These are very poorly threaded.

I have zero idea why they're getting such a big difference..I didn't see that on my FX vs my i5 at equal clocks.
There was a difference of like 5fps, but it was just as playable and fluid on both platforms.
Quote:
Originally Posted by Kane2207 View Post

Yeah, a bit of FUD there really isn't it?

4 out of 37 benchmarks were 'won' by the 8350 rolleyes.gif

http://www.anandtech.com/bench/Product/697?vs=551

10-13 of those benchmarks are completely useless in the real world depending on if you game at 1024x768 or not...And are mostly won by Intel.

That is, unless real world consists of running tools that have perfectly working multi-threaded code in a single-thread and Sysmark, take that into account and the FX-8350 didn't too bad for a CPU that is usually $75 cheaper or so. Hell, if you're running Linux stuff the i7 3770k and FX-8350 are even closer due to ICC not being the main compilation option on Linux.
Quote:
Originally Posted by kingduqc View Post

Intel do just fine in any type of game and AMD only in multi threaded games.

AMD does just as well as Intel in most games, short of SC2 you likely won't notice a difference.
And it's funny. a 4.5Ghz i5 3570k cannot max out Sins of a Solar Empire...from 2008. It literally still lags when you've got big battles going.
Quote:
Originally Posted by vampirr View Post

I thought Crysis 3 can use more than 4 cores ;(

It can, 1.3 fixed HT so now the i7 3770k can use its extra threads...Of course it's ahead of the FX there, the FX competes with the i5 in price and performance while coming up to the i7 sometimes.
Quote:
Originally Posted by Kane2207 View Post

Steamroller beating Haswell is a very bold claim indeed

It all depends on clocks, if Haswell doesn't clock too much better than Ivy (Doubtful from what we've heard in some reviews, likely according to others) and SR OCs as well or better than PD, it might come up close to the Haswell i5 in some situations (ie 7zip, x264 encoding, etc) especially considering that the lack of a shared decoder should be a decent performance increase. (1c per module performance on an FX-8150 bumped up performance 9%-40% depending on the benchmark used, but that's the cache too)
Quote:
Originally Posted by KloudZero View Post

In these same games I often see threads on their respective forums about both reduced performance or outright incompatibilities from both AMD's processor and GPUs products.

That is completely and utterly false, I've never seen something (recently, anyway) that had a bug or problem on AMD but not Intel...and as for GPUs, nVidia's drivers really aren't any better outside of SLI latency.
Quote:
Originally Posted by Maiky View Post

Bahh, countering with a game where the AMD 8 core is clocked at 4Ghz while the Intel's are at 3.4 and 3.5Ghz at the most is just nonsense. Put all CPU's at 4.5Ghz and we will see who comes out ahead and by what margin.

The 2500K alone which is a 2.5 year old i5 will destroy an 8350 in any game when having both CPU's at the same exact clock. Under-clock that 8350 to 3.3Ghz and watch how it drops from 3rd to 10th place on that list. Lets compare apples to apples.

AMD needs a rabit.

Did you completely ignore that FX-6300 in there? 3.5Ghz and just under the 3.4Ghz i7 3770k.
Quote:
Originally Posted by A Bad Day View Post

Why tinker with clock rates? Shouldn't it be performance per watt/price?

AMD said that they implemented Clock Gate Mesh technology, which allows them to increase the clock rate without increasing power consumption.


IPC and clock rates don't matter. Instructions Per Seconds is what we should be using instead of IPC.
This.
My system
(22 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i5 3770k @ 4.6Ghz ASRock Z77 Pro3 Powercolor Radeon HD7950 3GB @ 1150/1370 4x4GB G.Skill Ares 2200Mhz CL11 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 840 250GB Western Digital Black 1TB WD1002FAEX Samsung Spinpoint EcoGreen 2TB Pioneer DVR-220LBKS 
CoolingCoolingCoolingOS
Noctua NH-D14 Scythe Gentle Typhoon 1850rpm Corsair AF140 Quiet Edition Arch Linux x86-64, amdgpu 
OSMonitorMonitorKeyboard
Windows 10 Bloatfree Edition BenQ G2220HD BenQ G2020HD Ducky Shine III Year of the Snake, Cherry Blue 
PowerCaseMouseMouse Pad
Silverstone Strider Plus 600w Lian Li Lancool PC-K60 SteelSeries Sensei Professional Artisan Hien Mid Japan Black Large 
AudioOther
ASUS Xonar DX NZXT Sentry Mesh 30w Fan Controller 
  hide details  
Reply
My system
(22 items)
 
  
CPUMotherboardGraphicsRAM
Intel Core i5 3770k @ 4.6Ghz ASRock Z77 Pro3 Powercolor Radeon HD7950 3GB @ 1150/1370 4x4GB G.Skill Ares 2200Mhz CL11 
Hard DriveHard DriveHard DriveOptical Drive
Samsung 840 250GB Western Digital Black 1TB WD1002FAEX Samsung Spinpoint EcoGreen 2TB Pioneer DVR-220LBKS 
CoolingCoolingCoolingOS
Noctua NH-D14 Scythe Gentle Typhoon 1850rpm Corsair AF140 Quiet Edition Arch Linux x86-64, amdgpu 
OSMonitorMonitorKeyboard
Windows 10 Bloatfree Edition BenQ G2220HD BenQ G2020HD Ducky Shine III Year of the Snake, Cherry Blue 
PowerCaseMouseMouse Pad
Silverstone Strider Plus 600w Lian Li Lancool PC-K60 SteelSeries Sensei Professional Artisan Hien Mid Japan Black Large 
AudioOther
ASUS Xonar DX NZXT Sentry Mesh 30w Fan Controller 
  hide details  
Reply
post #320 of 352
Quote:
Originally Posted by Stay Puft View Post

You think Ivy uses 130W at 4.6Ghz? lachen.gif Its not even close



Under cinebench at 4.4Ghz i'm pulling 444w according to my Killawatt.
still... ivy looks like a pretty effecient cpu compared to the rest...
    
CPUMotherboardGraphicsGraphics
3770k (3233C Batch)  G1.Sniper 3 sapphire 7950 reference sapphire 7950 reference 
RAMHard DriveHard DriveHard Drive
ares 32GB Samsung 840 Pro 256GB 2.5" MZ-7PD256BW Crucial M4 CT128M4SSD2 128GB Western Digital WD1002FAEX  
Optical DriveCoolingOSMonitor
usb iomega superslim dvd Zalman CNPS12X 7 ultimate x64 SP1 ZR30w 
MonitorMonitorMonitorMonitor
u2412mb u2412mb u2412m u2412m 
MonitorMonitorMonitorKeyboard
V7 D24W33 LA2405wg E157FP dell sk-8115 
PowerCaseMouseMouse Pad
SeaSonic Platinum-860 860W ATX12V RSV-R4000 Rackmount Server Chassis TeamScorpion X-Luca xtrac ripper xxl 
AudioOtherOtherOther
Klipsch ProMedia 2.1 logitech c920 webcam logitech C920 Webcam serato sl1 
Other
usb 2.0 hub 
  hide details  
Reply
    
CPUMotherboardGraphicsGraphics
3770k (3233C Batch)  G1.Sniper 3 sapphire 7950 reference sapphire 7950 reference 
RAMHard DriveHard DriveHard Drive
ares 32GB Samsung 840 Pro 256GB 2.5" MZ-7PD256BW Crucial M4 CT128M4SSD2 128GB Western Digital WD1002FAEX  
Optical DriveCoolingOSMonitor
usb iomega superslim dvd Zalman CNPS12X 7 ultimate x64 SP1 ZR30w 
MonitorMonitorMonitorMonitor
u2412mb u2412mb u2412m u2412m 
MonitorMonitorMonitorKeyboard
V7 D24W33 LA2405wg E157FP dell sk-8115 
PowerCaseMouseMouse Pad
SeaSonic Platinum-860 860W ATX12V RSV-R4000 Rackmount Server Chassis TeamScorpion X-Luca xtrac ripper xxl 
AudioOtherOtherOther
Klipsch ProMedia 2.1 logitech c920 webcam logitech C920 Webcam serato sl1 
Other
usb 2.0 hub 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
Overclock.net › Forums › Industry News › Hardware News › [PCPOP]Haswell Core i7 4770K retail version of the first test