Overclock.net › Forums › Benchmarks › Benchmarking Software and Discussion › [OFFICIAL]--- Top 30 --- Unigine 'Valley' Benchmark 1.0
New Posts  All Forums:Forum Nav:

[OFFICIAL]--- Top 30 --- Unigine 'Valley' Benchmark 1.0 - Page 1144

post #11431 of 11436
Quote:
Originally Posted by Mydog View Post

helpinghand.gif well well do I sense jealousy here, I always buy the high end GPU's I got 780 Ti's and 290X's here as well as some old 7970's. I've not looked into why the 290's doesn't perform in Vally but what I do know is that I had more trouble with them in BF4 than the 780 Ti's.

Why would I be jealous of a clearly broken benchmark. I find it ridiculous that people hold any value to Valley results. There's a reason why this thread's activity resembles your local radio shack.
Storm
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-4770K Asus Maximus Vi Hero XFX 7970 GHZ CFX Corsair Vengeance 4x8GB 
Hard DriveCoolingOSMonitor
128GB Evo Corsair H100 Windows 8.1 VE278Q 
PowerCaseMouseMouse Pad
SuperNOVA 1300 G2 Corsair 800D Razor Deathadder Desk 
Audio
ATH 700 
  hide details  
Reply
Storm
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7-4770K Asus Maximus Vi Hero XFX 7970 GHZ CFX Corsair Vengeance 4x8GB 
Hard DriveCoolingOSMonitor
128GB Evo Corsair H100 Windows 8.1 VE278Q 
PowerCaseMouseMouse Pad
SuperNOVA 1300 G2 Corsair 800D Razor Deathadder Desk 
Audio
ATH 700 
  hide details  
Reply
post #11432 of 11436
Quote:
Originally Posted by Mydog View Post

helpinghand.gif well well do I sense jealousy here, I always buy the high end GPU's I got 780 Ti's and 290X's here as well as some old 7970's. I've not looked into why the 290's doesn't perform in Vally but what I do know is that I had more trouble with them in BF4 than the 780 Ti's.
Quote:
Originally Posted by LurkerToPoster View Post

Why would I be jealous of a clearly broken benchmark. I find it ridiculous that people hold any value to Valley results. There's a reason why this thread's activity resembles your local radio shack.

...don't mean to get in between your discussion, but calling Valley 'broken' doesn't' make an sense to me...it is just different., and MyDog has a point to some extent. I have run Valley w/ anything from a 3770K w/1x to 4x 670X on 3770K to a 5960X with a single 290X - and everything in-between - re Intel on CPUs and AMD / NVidia on GPUs...1x 290X lightning, 2x7990s / 4x GPUs and 1x to 4x 780 TI Classifieds...

....Valley addresses system memory and PCIe bus differently than most any other 'apps', and multi-GPU drivers ' play a big role...why else can I score 165.5 w/ 2x 780 TI Classifieds, but drop down well below that w/ 3x or 4x 780 TIs at the same settings for everything else ?

...but all that said, I don't think it's a simple case of 'jealousy'...within AMD for example, 7970 vs 290x is just a world of difference, to stick within one manufacturers'' Valley portfolio, never mind comparing AMD to NVidia...
Edited by Joa3d43 - 9/28/14 at 4:32pm
post #11433 of 11436
Quote:
Originally Posted by Joa3d43 View Post


...don't mean to get in between your discussion, but calling Valley 'broken' doesn't' make an sense to me...it is just different., and MyDog has a point to some extent. I have run Valley w/ anything from a 3770K w/1x to 4x 670X on 3770K to a 5960X with a single 290X - and everything in-between - re Intel on CPUs and AMD / NVidia on GPUs...1x 290X lightning, 2x7990s / 4x GPUs and 1x to 4x 780 TI Classifieds...

....Valley addresses system memory and PCIe bus differently than most any other 'apps', and multi-GPU drivers ' play a big role...why else can I score 165.5 w/ 2x 780 TI Classifieds, but drop down well below that w/ 3x or 4x 780 TIs at the same settings for everything else ?

...but all that said, I don't think it's a simple case of 'jealousy'...within AMD for example, 7970 vs 290x is just a world of difference, to stick within one manufacturers'' Valley portfolio, never mind comparing AMD to NVidia...

No discussion, but well said and kind of what my experience is too. AMD GPU's seem to scale more like one expect with x3 and x4 compared to Nvidia where the scaling is negativ. You have proven that it's not only the GPU memory but also system memory that this benchmark likes fast from what I've seen.
post #11434 of 11436
He's kinda right though. If you blame memory bandwidth for low FPS in valley, then why are the 290X's and 290's not topping the chart?
     
CPUMotherboardGraphicsRAM
Intel Core i7-3610QM Processor Lenovo NVIDIA GeForce GTX660M 2GB  8.0GB PC3-12800 DDR3 SDRAM 1600 MHz 
Hard DriveHard DriveOptical DriveCooling
Mushkin Enhanced Atlas Series120GB SATA III MLC... 1TB 5400RPM (Storage) Blu-ray/DVD-RW Cm X-Slim 
OSMonitorKeyboardAudio
Windows 8 pro 15.6" FHD LED Glare Wedge 1920x1080 Backlit Keyboard JBL speakers 
  hide details  
Reply
     
CPUMotherboardGraphicsRAM
Intel Core i7-3610QM Processor Lenovo NVIDIA GeForce GTX660M 2GB  8.0GB PC3-12800 DDR3 SDRAM 1600 MHz 
Hard DriveHard DriveOptical DriveCooling
Mushkin Enhanced Atlas Series120GB SATA III MLC... 1TB 5400RPM (Storage) Blu-ray/DVD-RW Cm X-Slim 
OSMonitorKeyboardAudio
Windows 8 pro 15.6" FHD LED Glare Wedge 1920x1080 Backlit Keyboard JBL speakers 
  hide details  
Reply
post #11435 of 11436
Occamrazor had an explanation for why the 980 is weak in Unigine's benches yet really strong in the Futuremark benches. I didn't really understand it (I'm just a forum guy not a chip engineer) but he said something about how the GM204 is designed optimizes it for certain workloads but does nothing in other circumstances. PM him over in the Titan Club if you want a detailed explanation! thumb.gif
post #11436 of 11436
I dont mean to sound negative but wouldn't that mean they have to "support" different workloads as in, some games, benches dont get full on benefits of maxwell?
This sounds like it would be a bad idea in terms of selective performance. I guess I don't understand it.
New Posts  All Forums:Forum Nav:
  Return Home
Overclock.net › Forums › Benchmarks › Benchmarking Software and Discussion › [OFFICIAL]--- Top 30 --- Unigine 'Valley' Benchmark 1.0