Overclock.net › Forums › Industry News › Hardware News › [The Tech Report] Radeon HD 7950 vs. GeForce GTX 660 Ti revisited
New Posts  All Forums:Forum Nav:

[The Tech Report] Radeon HD 7950 vs. GeForce GTX 660 Ti revisited - Page 43  

post #421 of 745
Quote:
Originally Posted by badrapper View Post

What card hangs with the 7970 or 680?
what cards hangs with the low power, cool, silent cards?
What card hangs with the big boys when resolution is cranked up?
What card has the overclock potential to stomp the 680 easily or 7970?
What card is more future proof for games?
What card is it that costs less but is a high performance card?
What card has more VRAM & bigger Bus Width?
What card is Game & Compute capable?
What card can play multi screens easily at higher res?
Why Its the HD7950 (The 660TI could never fill that role apart from niche roles)
TechReport achieved its goal, which was site hits since its rapidly going the way of the dinosaur like the rest of the biased sites. Nobody likes a biased site that is nVidia focused only mad.gif nVidia can afford to pay those sites whilst AMD cannot

Quote:
Originally Posted by badrapper View Post

[quote name="Forceman" url="/t/1334760/snip.
Quote:
Ok so why was the 680/670 not included... we know why bcoz it would have made them look bad also. That 660ti overclocked card comes very close to the 680 and 670.. why did nvidia ask for them cards to be left out or the 7970?
Most would overclock or have that option when the times comes than not to be able to

I don't even know......

What is this?!?
    
CPUMotherboardGraphicsRAM
Intel i7 5820K AsRock Extreme6 X99 Gigabyte GTX 980 Ti Windforce OC 16 GB Corsair Vengeance LPX 
Hard DriveHard DriveCoolingOS
Samsung 840 EVO 250GB - HDD Speed Edtition Samsung SM951 512 GB - I still hate Samsung!  Noctua NHD14 Windows 10 
MonitorMonitorMonitorKeyboard
Achieva Shimian QH270-Lite Overlord Computer Tempest X27OC  Acer Predator XB270HU Filco Majestouch 2 Ninja 
PowerCaseMouseMouse Pad
Seasonic X-1250 Fractal Design R5 Razer Naga Razer Goliathus Alpha 
AudioAudio
AKG K702 65th Anniversary Edition Creative Sound Blaster Zx 
  hide details  
    
CPUMotherboardGraphicsRAM
Intel i7 5820K AsRock Extreme6 X99 Gigabyte GTX 980 Ti Windforce OC 16 GB Corsair Vengeance LPX 
Hard DriveHard DriveCoolingOS
Samsung 840 EVO 250GB - HDD Speed Edtition Samsung SM951 512 GB - I still hate Samsung!  Noctua NHD14 Windows 10 
MonitorMonitorMonitorKeyboard
Achieva Shimian QH270-Lite Overlord Computer Tempest X27OC  Acer Predator XB270HU Filco Majestouch 2 Ninja 
PowerCaseMouseMouse Pad
Seasonic X-1250 Fractal Design R5 Razer Naga Razer Goliathus Alpha 
AudioAudio
AKG K702 65th Anniversary Edition Creative Sound Blaster Zx 
  hide details  
post #422 of 745
Quote:
Originally Posted by badrapper View Post

[quote name="Forceman" url="/t/1334760/snip.
Ok so why was the 680/670 not included... we know why bcoz it would have made them look bad also. That 660ti overclocked card comes very close to the 680 and 670.. why did nvidia ask for them cards to be left out?
Most would overclock or have that option when the times comes than not to

Because they didn't test the 670 or 680? You can just as easily ask why the 7870 or 7970 wasn't included. They picked two cards to test. That's it. You can ascribe all the conspiracy theory motives you want, but not every review tests every card in every possible combination.
post #423 of 745
That high clocked 660 TI is a really good option for folks who don't like overclocking, it's at a good price point.
Gaming Rig(2012)
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 3770K ASRock Z77 Extreme 4 Asus GTX 660 TI DirectCU II OC 8GB Ripjaws 1600 
Hard DriveHard DriveCoolingOS
Samsung 840 250GB  Samsung Spinpoint F3 Corsair H80 Windows 7 64 
MonitorPowerCaseAudio
Asus VW246H Seasonic X650 Gold Lancool K-58W(modded) Xonar DX > Klipsch Promedia 2.1/JVC HA-RX700 
  hide details  
Gaming Rig(2012)
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 3770K ASRock Z77 Extreme 4 Asus GTX 660 TI DirectCU II OC 8GB Ripjaws 1600 
Hard DriveHard DriveCoolingOS
Samsung 840 250GB  Samsung Spinpoint F3 Corsair H80 Windows 7 64 
MonitorPowerCaseAudio
Asus VW246H Seasonic X650 Gold Lancool K-58W(modded) Xonar DX > Klipsch Promedia 2.1/JVC HA-RX700 
  hide details  
post #424 of 745
Quote:
Originally Posted by yesitsmario View Post

That high clocked 660 TI is a really good option for folks who don't like overclocking, it's at a good price point.

True that but has nothing to do with comparing the card here in OCN. HD 7950 has a lot of OC potential. I mean stock is 800Mhz. Getting 1.2Ghz is 50% OC. Considering how well GCN scales you are getting close to 40% increased performance if you OC the memory too. GTX660 Ti cant even reach 40% with number let alone performance.
Ishimura
(21 items)
 
Silent Knight
(13 items)
 
 
CPUMotherboardGraphicsRAM
Intel Core i7 3770K @ 4.6GHz ASRock Z77E-ITX eVGA GTX 1080 Ti Hybrid AMD Radeon R9 16GB DDR3-2400MHz  
Hard DriveHard DriveCoolingCooling
SanDisk Ultra II 960GB Toshiba X300 5TB Corsair H100i GTX eVGA Hybrid Water Cooler  
CoolingOSMonitorKeyboard
4x GentleTyphoon AP-15 Windows 10 Pro 64-Bit Philips Brilliance BDM4065UC 4K Razer BlackWidow Chroma  
PowerCaseMouseMouse Pad
eVGA SuperNOVA 750 G3 Define Nano S Logitech G502 Proteus Core PECHAM Gaming Mouse Pad XX-Large 
AudioAudioAudioAudio
Audioengine D1 DAC Mackie CR Series CR3 Audio-Technica ATH-M50 Sennheiser HD 598 
Audio
Sony XB950BT 
CPUMotherboardGraphicsRAM
AMD Phenom II X4 955 @ 4.2GHz ASUS M4A79XTD EVO AMD Radeon HD 7970 3GB @ 1200/1500 2x 4GB G.SKILL Ripjaws X DDR3-1600 
Hard DriveHard DriveHard DriveCooling
OCZ Agility 3 60GB WD Caviar Green 1.5TB 2 x Seagate Barracuda 2TB XSPC Raystorm 
CoolingCoolingOSPower
EK-FC7970 XSPC RS360 Windows 10 Pro 64-Bit Corsair TX750 
Case
NZXT Switch 810  
  hide details  
Ishimura
(21 items)
 
Silent Knight
(13 items)
 
 
CPUMotherboardGraphicsRAM
Intel Core i7 3770K @ 4.6GHz ASRock Z77E-ITX eVGA GTX 1080 Ti Hybrid AMD Radeon R9 16GB DDR3-2400MHz  
Hard DriveHard DriveCoolingCooling
SanDisk Ultra II 960GB Toshiba X300 5TB Corsair H100i GTX eVGA Hybrid Water Cooler  
CoolingOSMonitorKeyboard
4x GentleTyphoon AP-15 Windows 10 Pro 64-Bit Philips Brilliance BDM4065UC 4K Razer BlackWidow Chroma  
PowerCaseMouseMouse Pad
eVGA SuperNOVA 750 G3 Define Nano S Logitech G502 Proteus Core PECHAM Gaming Mouse Pad XX-Large 
AudioAudioAudioAudio
Audioengine D1 DAC Mackie CR Series CR3 Audio-Technica ATH-M50 Sennheiser HD 598 
Audio
Sony XB950BT 
CPUMotherboardGraphicsRAM
AMD Phenom II X4 955 @ 4.2GHz ASUS M4A79XTD EVO AMD Radeon HD 7970 3GB @ 1200/1500 2x 4GB G.SKILL Ripjaws X DDR3-1600 
Hard DriveHard DriveHard DriveCooling
OCZ Agility 3 60GB WD Caviar Green 1.5TB 2 x Seagate Barracuda 2TB XSPC Raystorm 
CoolingCoolingOSPower
EK-FC7970 XSPC RS360 Windows 10 Pro 64-Bit Corsair TX750 
Case
NZXT Switch 810  
  hide details  
post #425 of 745
Quote:
Originally Posted by Mad Pistol View Post

There's a problem with that line of thinking, and more importantly, there is a fatal flaw in Tech Report's testing methodolgy that most people haven't realized up to this point.
First, there is a threshold of latency that Tech Report tests for, and any frame time that falls below that gets counted against the card in question. The one thing that most of you don't know is that as the average framerate falls, the time it takes to produce each frame grows exponentially.
For example: 60 fps = 16.67ms per frame
50 fps = 20ms per frame
40 fps = 25ms
30 fps = 33.3
25 fps = 40ms <- the number is growing more and more each time.
20 fps = 50ms
etc.
Once you fall below a certain threshold of "acceptable performance" you start getting a lot more extra time for producing frames. That can sometimes lead to skewed numbers if you're not careful.
Second, and this one is very important. Everything was done using FRAPS on gameplay segments. The tests were not pre-recorded and not able to be reproduced with 100% accuracy. When you're talking about frame latency, that alone is enough to invalidate their testing methodology. If you're going to measure frame latencies, YOU HAVE TO REPRODUCE THE TESTS EXACTLY BETWEEN EACH CARD!!! As soon as you throw any sort of variance into the equation, that's when the results become dicey to the point that they can even be invalidated given the right set of circumstances (see below).
Third, they did not test different resolutions or different settings of each card. They tested one resolution and one crop of settings. Some games were at 2560x1440, while others were at 1080P, but there wasn't a single game where multiple resolutions were tested. When has ANY review site settled with one set of settings and published their results as law? That's a very narrow-sighted approach to benchmarking.
Fourth, and I've said this many times before, the 660 Ti is overclocked to an extreme level straight from the factory, while the 7950 boost is basically a stock card. This is not ok. The cards are so close at stock that when you overclock either card, it will pull ahead. It is unfair to compare a heavily overclocked card with a stock card. Period. End of discussion.
Hopefully that sheds some light on some things that TR screwed up on with this test. Everyone (I mean everyone) should take all the results in this test with a grain of salt. The smell of bias is overwhelming on this one.

Oh my.

First of all: what you see as a fatal flaw in Tech Report's method is actually its greatest strength. The more severe the fps dips, the more noticeable and annoying the stutter. Human eye indeed detects the frame time growth exponentially below a certain fps threshold. That threshold is in fact individual, which is why it's good practice to publish the results with several thresholds. This is exactly what Tech Report does.

Also, while each test run can never be the same with actual gameplay, the variance can be neglected by statistics. Tech Report tested each sequence five times to start with. AFAIK, that's more control runs than any other hardware site. Then there is the fact that their sample size is huge: thousands of frames and as such thousands of frame time data points for each title. Occasional lag spikes get drowned in the sheer volume of samples. That is another very strong argument for their method, not against it.

On the settings, I agree. It would be great to see the results with various settings. That would give really good insight on the way each card is capable of handling different levels of eye candy. But since they need to do actual analysis on their test data and not just read a single data point like avg fps, I can understand why they haven't gone there.

The overclocking point of view could have been mentioned, I agree. But anyone calling bias because "they compared a stock 7950 to a high OC 660Ti" is clueless. That is one of the fastest 7950's compared to one of the fastest 660Ti's, both around the same price. That is as fair as a stock graphics card comparison gets. OC isn't their focus, their focus is their unorthodox testing method, which gives new information to those who are ready to absorb it. I give them credit for it, that is what sets them apart from the dozens of no-news hardware sites.
Edited by specopsFI - 12/8/12 at 4:32pm
The Air Tunnel
(10 items)
 
  
CPUMotherboardGraphicsRAM
i7-4820K (4500MHz@1.28V) P9X79 Sapphire R9 290 Tri-X New Edition Team Group Vulcan 4x4GB 2133MHz 
Hard DriveCoolingOSPower
2x240GB SSD @RAID 0 Noctua NH-D14 SE2011 Windows 10 Pro Corsair AX750 
CaseAudio
Rosewill Armor EVO hiFace+AudioGD+DT770Pro 
  hide details  
The Air Tunnel
(10 items)
 
  
CPUMotherboardGraphicsRAM
i7-4820K (4500MHz@1.28V) P9X79 Sapphire R9 290 Tri-X New Edition Team Group Vulcan 4x4GB 2133MHz 
Hard DriveCoolingOSPower
2x240GB SSD @RAID 0 Noctua NH-D14 SE2011 Windows 10 Pro Corsair AX750 
CaseAudio
Rosewill Armor EVO hiFace+AudioGD+DT770Pro 
  hide details  
post #426 of 745
As expected, you guys stand behind TR's testing methodology regardless of it's inherent flaws. The test was rigged. I do not understand how anyone cannot see this.
Everest - Intel
(19 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 4790k Gigabyte Z97X Gaming 7 MSI Geforce GTX 1080 Ti Gaming X 16GB (2x8) Patriot Viper 1866Mhz  
Hard DriveHard DriveOptical DriveCooling
Seagate 3TB, WD 500GB HDD, WD 640GB HD Samsung 850 EVO 512GB Samsung DVD-Burner Corsair H110 w/ Dual Aerocool DS 140mm fans 
OSMonitorMonitorKeyboard
Windows 10 Pro Dell S2716DG (1440p, 144hz Gsync) AOC U3477 PQU (3440x1440 IPS) Logitech G810 Orion Spectrum 
PowerCaseMouseMouse Pad
Evga SuperNOVA 750 G2 NZXT Phantom 530 Black Logitech G502 Proteus Core Corsair MM400 
AudioAudioAudio
Creative Sound Blaster E5 DAC/AMP Sennheiser HD 598 Headphones HyperX Cloud Headset 
  hide details  
Everest - Intel
(19 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 4790k Gigabyte Z97X Gaming 7 MSI Geforce GTX 1080 Ti Gaming X 16GB (2x8) Patriot Viper 1866Mhz  
Hard DriveHard DriveOptical DriveCooling
Seagate 3TB, WD 500GB HDD, WD 640GB HD Samsung 850 EVO 512GB Samsung DVD-Burner Corsair H110 w/ Dual Aerocool DS 140mm fans 
OSMonitorMonitorKeyboard
Windows 10 Pro Dell S2716DG (1440p, 144hz Gsync) AOC U3477 PQU (3440x1440 IPS) Logitech G810 Orion Spectrum 
PowerCaseMouseMouse Pad
Evga SuperNOVA 750 G2 NZXT Phantom 530 Black Logitech G502 Proteus Core Corsair MM400 
AudioAudioAudio
Creative Sound Blaster E5 DAC/AMP Sennheiser HD 598 Headphones HyperX Cloud Headset 
  hide details  
post #427 of 745
mad.gifmad.gifmad.gifmad.gif

As it turns out the Zotac 670 Twincooler, more or less the 670 version of the 660Ti in the article, is sold only in China....

That crushes me

I wanted a shorty card with some muscle....
    
CPUMotherboardGraphicsRAM
Intel i7 5820K AsRock Extreme6 X99 Gigabyte GTX 980 Ti Windforce OC 16 GB Corsair Vengeance LPX 
Hard DriveHard DriveCoolingOS
Samsung 840 EVO 250GB - HDD Speed Edtition Samsung SM951 512 GB - I still hate Samsung!  Noctua NHD14 Windows 10 
MonitorMonitorMonitorKeyboard
Achieva Shimian QH270-Lite Overlord Computer Tempest X27OC  Acer Predator XB270HU Filco Majestouch 2 Ninja 
PowerCaseMouseMouse Pad
Seasonic X-1250 Fractal Design R5 Razer Naga Razer Goliathus Alpha 
AudioAudio
AKG K702 65th Anniversary Edition Creative Sound Blaster Zx 
  hide details  
    
CPUMotherboardGraphicsRAM
Intel i7 5820K AsRock Extreme6 X99 Gigabyte GTX 980 Ti Windforce OC 16 GB Corsair Vengeance LPX 
Hard DriveHard DriveCoolingOS
Samsung 840 EVO 250GB - HDD Speed Edtition Samsung SM951 512 GB - I still hate Samsung!  Noctua NHD14 Windows 10 
MonitorMonitorMonitorKeyboard
Achieva Shimian QH270-Lite Overlord Computer Tempest X27OC  Acer Predator XB270HU Filco Majestouch 2 Ninja 
PowerCaseMouseMouse Pad
Seasonic X-1250 Fractal Design R5 Razer Naga Razer Goliathus Alpha 
AudioAudio
AKG K702 65th Anniversary Edition Creative Sound Blaster Zx 
  hide details  
post #428 of 745
Quote:
Originally Posted by Forceman View Post

Because they didn't test the 670 or 680? You can just as easily ask why the 7870 or 7970 wasn't included. They picked two cards to test. That's it. You can ascribe all the conspiracy theory motives you want, but not every review tests every card in every possible combination.


Ok let me rephrase... Why was the 670 not included since the 7950 competes with it or the 7870 with the 660ti
post #429 of 745
Quote:
Originally Posted by badrapper View Post

Ok let me rephrase... Why was the 670 not included since the 7950 competes with it or the 7870 with the 660ti

Because that isn't what the article is about...

It is about how close the 660Ti is to the 7950. We already know that the GTX 670 is faster than the GTX 660 Ti, so simple logic dictates that if the 660Ti is equal to the 7950, then the GTX 670 would be as fast or faster.....

How is that hard to understand?
    
CPUMotherboardGraphicsRAM
Intel i7 5820K AsRock Extreme6 X99 Gigabyte GTX 980 Ti Windforce OC 16 GB Corsair Vengeance LPX 
Hard DriveHard DriveCoolingOS
Samsung 840 EVO 250GB - HDD Speed Edtition Samsung SM951 512 GB - I still hate Samsung!  Noctua NHD14 Windows 10 
MonitorMonitorMonitorKeyboard
Achieva Shimian QH270-Lite Overlord Computer Tempest X27OC  Acer Predator XB270HU Filco Majestouch 2 Ninja 
PowerCaseMouseMouse Pad
Seasonic X-1250 Fractal Design R5 Razer Naga Razer Goliathus Alpha 
AudioAudio
AKG K702 65th Anniversary Edition Creative Sound Blaster Zx 
  hide details  
    
CPUMotherboardGraphicsRAM
Intel i7 5820K AsRock Extreme6 X99 Gigabyte GTX 980 Ti Windforce OC 16 GB Corsair Vengeance LPX 
Hard DriveHard DriveCoolingOS
Samsung 840 EVO 250GB - HDD Speed Edtition Samsung SM951 512 GB - I still hate Samsung!  Noctua NHD14 Windows 10 
MonitorMonitorMonitorKeyboard
Achieva Shimian QH270-Lite Overlord Computer Tempest X27OC  Acer Predator XB270HU Filco Majestouch 2 Ninja 
PowerCaseMouseMouse Pad
Seasonic X-1250 Fractal Design R5 Razer Naga Razer Goliathus Alpha 
AudioAudio
AKG K702 65th Anniversary Edition Creative Sound Blaster Zx 
  hide details  
post #430 of 745
Quote:
Originally Posted by PostalTwinkie View Post

Because that isn't what the article is about...
It is about how close the 660Ti is to the 7950. We already know that the GTX 670 is faster than the GTX 660 Ti, so simple logic dictates that if the 660Ti is equal to the 7950, then the GTX 670 would be as fast or faster.....
How is that hard to understand?

well your logic is flawed sence they didnt test a 670 other sites have shown a 7950 trading blows with the 670. For all we know if they put a 670 in the 660Ti could of beat it also hard to say sence they didnt do it all these benchmarks are a butch of bull anyway if you happy with your card and what you paid for it you we should be happy.
SABERWOLF
(18 items)
 
XEON-WOLF
(18 items)
 
 
CPUMotherboardGraphicsRAM
Intel Core i7-6800K MSI X99A GAMING PRO CARBON EVGA GTX 1080 FTW Hydro Copper G.SKILL TridentZ Series 
Hard DriveCoolingCoolingCooling
Samsung EVO 850 500 EK-SBAY Dual DDC 3.2 PWM Serial Alphacool NexXxoS ST30 Full Copper Radiator 360 Hardware Labs Black Ice SR2 Multiport Black Car... 
CoolingCoolingOSMonitor
EK-CoolStream RAD XTC 420 koolance 380i Windows 10 Pro ASUS VW246H Black 24" 2ms 
PowerCaseMouseAudio
SeaSonic M12D SS-850 850W  Thermaltake core x71 CM Storm Sentinel Advance II Corsair Void 
CPUMotherboardGraphicsRAM
X5675@4400 1.344 ASUS P6X58D-E XfX RX480 RS F3-12800CL7T-6GBPI 6-7-6-21 1T 
RAMHard DriveHard DriveHard Drive
F3-12800CL9S-4GBRL WDC WD1001FALS-00J7B1 ST31000340AS WDC WD7500AAKS-00RBA0 
Hard DriveOptical DriveCoolingOS
Hitachi HDS723020BLA642 ihas 624-A SilverStone Heligon HE01 windows 10 64 Pro 
MonitorPowerCase
Vizio 42inch lcd tv CORSAIR CMPSU-950TX Modded Rocketfish 
  hide details  
SABERWOLF
(18 items)
 
XEON-WOLF
(18 items)
 
 
CPUMotherboardGraphicsRAM
Intel Core i7-6800K MSI X99A GAMING PRO CARBON EVGA GTX 1080 FTW Hydro Copper G.SKILL TridentZ Series 
Hard DriveCoolingCoolingCooling
Samsung EVO 850 500 EK-SBAY Dual DDC 3.2 PWM Serial Alphacool NexXxoS ST30 Full Copper Radiator 360 Hardware Labs Black Ice SR2 Multiport Black Car... 
CoolingCoolingOSMonitor
EK-CoolStream RAD XTC 420 koolance 380i Windows 10 Pro ASUS VW246H Black 24" 2ms 
PowerCaseMouseAudio
SeaSonic M12D SS-850 850W  Thermaltake core x71 CM Storm Sentinel Advance II Corsair Void 
CPUMotherboardGraphicsRAM
X5675@4400 1.344 ASUS P6X58D-E XfX RX480 RS F3-12800CL7T-6GBPI 6-7-6-21 1T 
RAMHard DriveHard DriveHard Drive
F3-12800CL9S-4GBRL WDC WD1001FALS-00J7B1 ST31000340AS WDC WD7500AAKS-00RBA0 
Hard DriveOptical DriveCoolingOS
Hitachi HDS723020BLA642 ihas 624-A SilverStone Heligon HE01 windows 10 64 Pro 
MonitorPowerCase
Vizio 42inch lcd tv CORSAIR CMPSU-950TX Modded Rocketfish 
  hide details  
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Hardware News
This thread is locked  
Overclock.net › Forums › Industry News › Hardware News › [The Tech Report] Radeon HD 7950 vs. GeForce GTX 660 Ti revisited