Overclock.net › Forums › Industry News › Hardware News › [Eteknix] Pascal GTX 1080 Async Compute Explored
New Posts  All Forums:Forum Nav:

[Eteknix] Pascal GTX 1080 Async Compute Explored - Page 19  

post #181 of 200
Quote:
That's like judging a car by "This Volkswagen took me from point A to point B, this Tesla took me from point A to point B just the same, they're near identical"
you didnt actually dispute anything with this sentence:p


like I said I read that in reviews on a bunch of sites that I trust

on the other hand I dont know who this Linus guy is
The Green Beast
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 3770K @ 4500 Mhz ASRock Z77 Pro3 Gigabyte GTX 1080 G1 Gaming Crucial Ballistix 2x8GB DDR3-1600 
Hard DriveHard DriveHard DriveCooling
SSD Crucial M550 500GB SSD Samsung 850 Evo 1TB HDD Seagate 7200rpm 3TB Cooler Master Hyper 212 EVO; Xilence X5 
OSMonitorKeyboardPower
Windows 10 Pro 64-bit Acer Predator XB271HU 27" IPS Gsync 1440p 165Hz CM Storm QuickFire XT Cherry Red 800W modular 
CaseMouseMouse PadAudio
Fractal Design Define R4 Black Logitech G900 Chaos Spectrum SteelSeries QcK+ 4mm SK Gaming Realtek On-board 
  hide details  
The Green Beast
(16 items)
 
  
CPUMotherboardGraphicsRAM
Intel i7 3770K @ 4500 Mhz ASRock Z77 Pro3 Gigabyte GTX 1080 G1 Gaming Crucial Ballistix 2x8GB DDR3-1600 
Hard DriveHard DriveHard DriveCooling
SSD Crucial M550 500GB SSD Samsung 850 Evo 1TB HDD Seagate 7200rpm 3TB Cooler Master Hyper 212 EVO; Xilence X5 
OSMonitorKeyboardPower
Windows 10 Pro 64-bit Acer Predator XB271HU 27" IPS Gsync 1440p 165Hz CM Storm QuickFire XT Cherry Red 800W modular 
CaseMouseMouse PadAudio
Fractal Design Define R4 Black Logitech G900 Chaos Spectrum SteelSeries QcK+ 4mm SK Gaming Realtek On-board 
  hide details  
post #182 of 200
Quote:
Originally Posted by ChevChelios View Post

you didnt actually dispute anything with this sentence:p


like I said I read that in reviews on a bunch of sites that I trust

on the other hand I dont know who this Linus guy is

I would be OK with that, if those sites you trust would at least test the technologies they're reviewing beyond using them like normal users do, and judging by eye or experience.

I like Guru3D for example, I trust them but not entirely, same with TechSpot and Linustechtips, I've seen them all fail at some point, I only trust repeatable comprehensive data, even this test doesn't gain 100% of my trust but its all we've got, and science doesn't work by choosing what to believe, it's more along the lines of what we can prove.

I mean the review sites you mention don't even go into details beyond the specs of each Technology and how it affects gaming, that's it, there are no input lag or latency benchmarks, you can't make anything out of that.

There was a time when G-Sync was absolutely superior to FreeSync, a time when FreeSync disabled the Overdrive function of all FreeSync monitors, one where FreeSync wouldn't work below 55Hz, a time where FreeSync was garbage, but many people forget that G-Sync also had Ghosting, Blur, Overdrive issues and Tearing issues when it first Launched, Bias memory at its best.

Both technology have matured now, G-Sync before FreeSync, but at this point Freesync have matured enough to be at its full potential, being a software feature with no Module Overhead it does have the potential to be faster than G-Sync as these tests suggest, again more Research is needed, less of "I read a site saying this is better" and more research with comprehensive data.

Also LinusTechTips is way bigger than any Tech Review site IMO, they're the only ones who have tried this, dismissing this because you don't know or don't like the guy is foolish.
post #183 of 200
Quote:
Originally Posted by Dargonplay View Post

I would be OK with that, if those sites you trust would at least test the technologies they're reviewing beyond using them like normal users do, and judging by eye or experience.

I like Guru3D for example, I trust them but not entirely, same with TechSpot and Linustechtips, I've seen them all fail at some point, I only trust repeatable comprehensive data, even this test doesn't gain 100% of my trust but its all we've got, and science doesn't work by choosing what to believe, it's more along the lines of what we can prove.

I mean the review sites you mention don't even go into details beyond the specs of each Technology and how it affects gaming, that's it, there are no input lag or latency benchmarks, you can't make anything out of that.

There was a time when G-Sync was absolutely superior to FreeSync, a time when FreeSync disabled the Overdrive function of all FreeSync monitors, one where FreeSync wouldn't work below 55Hz, a time where FreeSync was garbage, but many people forget that G-Sync also had Ghosting, Blur, Overdrive issues and Tearing issues when it first Launched, Bias memory at its best.

Both technology have matured now, G-Sync before FreeSync, but at this point Freesync have matured enough to be at its full potential, being a software feature with no Module Overhead it does have the potential to be faster than G-Sync as these tests suggest, again more Research is needed, less of "I read a site saying this is better" and more research with comprehensive data.

Also LinusTechTips is way bigger than any Tech Review site IMO, they're the only ones who have tried this, dismissing this because you don't know or don't like the guy is foolish.
Eh, there's not really any overhead for VRR, maybe a handful of clock cycles for calculating the overdrive offset, but that work would be done on streaming data(pixel by pixel basis), not waiting for the whole frame to buffer before working on it. Oh, and AMD has to do that too, but it's using GPU resources instead of a FPGA. So maybe a few microseconds, certainly not 10 milliseconds..
1
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 930 @ 3.6 Gigabyte x58a ud3r Gigabyte GV-R6870C-1GD Mushkin Redline 
Hard DriveOptical DriveMonitorKeyboard
2x 1TB Spinpoint samsung dvd burner Samsung p2370 + Mitsubishi Diamond Pro 930SB WASD keyboards v1 semi custom w/ cherry browns 
PowerCaseMouse
Antec CP-850 Antec P183 CM Storm Spawn 
  hide details  
1
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 930 @ 3.6 Gigabyte x58a ud3r Gigabyte GV-R6870C-1GD Mushkin Redline 
Hard DriveOptical DriveMonitorKeyboard
2x 1TB Spinpoint samsung dvd burner Samsung p2370 + Mitsubishi Diamond Pro 930SB WASD keyboards v1 semi custom w/ cherry browns 
PowerCaseMouse
Antec CP-850 Antec P183 CM Storm Spawn 
  hide details  
post #184 of 200
Quote:
Originally Posted by TranquilTempest View Post

Eh, there's not really any overhead for VRR, maybe a handful of clock cycles for calculating the overdrive offset, but that work would be done on streaming data(pixel by pixel basis), not waiting for the whole frame to buffer before working on it. Oh, and AMD has to do that too, but it's using GPU resources instead of a FPGA. So maybe a few microseconds, certainly not 10 milliseconds..

I'm genuinely curious to hear your opinion as to why Crysis 3 capped at 200 FPS by the default engine cap and uncapped with Vsync OFF but Gsync/FreeSync on yielded 20ms on FreeSync and 30ms on Gsync in 5 different consistent tests.

The graphics pipeline having to go through one extra step talking to the G-Sync Hardware Module would be my bet, but again, more research is needed.
post #185 of 200
Quote:
Originally Posted by Glottis View Post

and play DOOM at 23fps? there's more to graphics technologies than just asyc.


WTH how can nvidia be over 2.5x better?
post #186 of 200
Quote:
Originally Posted by atomicmew View Post

WTH how can nvidia be over 2.5x better?

Because the first AMD card in that list is a 380x
post #187 of 200
Quote:
Originally Posted by atomicmew View Post

WTH how can nvidia be over 2.5x better?
it's not, that's an old benchmark.
post #188 of 200
Quote:
Originally Posted by Dargonplay View Post

I'm genuinely curious to hear your opinion as to why Crysis 3 capped at 200 FPS by the default engine cap and uncapped with Vsync OFF but Gsync/FreeSync on yielded 20ms on FreeSync and 30ms on Gsync in 5 different consistent tests.

The graphics pipeline having to go through one extra step talking to the G-Sync Hardware Module would be my bet, but again, more research is needed.
Could be AMD was CPU limited in that scenario due to driver overhead, even if the framerates were similar.
1
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 930 @ 3.6 Gigabyte x58a ud3r Gigabyte GV-R6870C-1GD Mushkin Redline 
Hard DriveOptical DriveMonitorKeyboard
2x 1TB Spinpoint samsung dvd burner Samsung p2370 + Mitsubishi Diamond Pro 930SB WASD keyboards v1 semi custom w/ cherry browns 
PowerCaseMouse
Antec CP-850 Antec P183 CM Storm Spawn 
  hide details  
1
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 930 @ 3.6 Gigabyte x58a ud3r Gigabyte GV-R6870C-1GD Mushkin Redline 
Hard DriveOptical DriveMonitorKeyboard
2x 1TB Spinpoint samsung dvd burner Samsung p2370 + Mitsubishi Diamond Pro 930SB WASD keyboards v1 semi custom w/ cherry browns 
PowerCaseMouse
Antec CP-850 Antec P183 CM Storm Spawn 
  hide details  
post #189 of 200
Quote:
Originally Posted by TranquilTempest View Post

Could be AMD was CPU limited in that scenario due to driver overhead, even if the framerates were similar.

The FreeSync setup being CPU limited would actually increase input lag results not reduce it... Wouldn't it? am I missing something here? FreeSync had 20ms while Nvidia G-Sync had 50% more input lag at 30ms, I don't see how having driver overhead or a CPU bottleneck on the FreeSync setup would in any way shape or form increase performance leading to reduced input lag compared to Nvidia's setup.
Edited by Dargonplay - 6/15/16 at 7:29am
post #190 of 200
Quote:
Originally Posted by Dargonplay View Post

The FreeSync setup being CPU limited would actually increase input lag results not reduce it... Wouldn't it? am I missing something here? FreeSync had 20ms while Nvidia G-Sync had 50% more input lag at 30ms, I don't see how having driver overhead or a CPU bottleneck on the FreeSync setup would in any way shape or form increase performance leading to reduced input lag compared to Nvidia's setup, like I don't even...

You know what flip queue/max prerendered frames is? That buffer fills up if you're GPU or display limited(this includes framerate caps outside of the game engine), and it still adds latency if you set it to its minimum value. Being CPU limited or using an in game framerate cap will keep that buffer completely empty, as the bottleneck is now in front of that buffer instead of after it. There are other points of buffering, but that's just one many people are familiar with.

Take a length of highway that passes X cars per second. If the bottleneck restricting it to X cars per second is at the end of that length of highway, the highway fills up and each individual car drives slow, waiting on the car in front of it. If the bottleneck is at the start of the length of highway, each individual car can drive fast, because everything after the bottleneck is less restrictive(nobody to wait for).
Edited by TranquilTempest - 5/20/16 at 8:45pm
1
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 930 @ 3.6 Gigabyte x58a ud3r Gigabyte GV-R6870C-1GD Mushkin Redline 
Hard DriveOptical DriveMonitorKeyboard
2x 1TB Spinpoint samsung dvd burner Samsung p2370 + Mitsubishi Diamond Pro 930SB WASD keyboards v1 semi custom w/ cherry browns 
PowerCaseMouse
Antec CP-850 Antec P183 CM Storm Spawn 
  hide details  
1
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 930 @ 3.6 Gigabyte x58a ud3r Gigabyte GV-R6870C-1GD Mushkin Redline 
Hard DriveOptical DriveMonitorKeyboard
2x 1TB Spinpoint samsung dvd burner Samsung p2370 + Mitsubishi Diamond Pro 930SB WASD keyboards v1 semi custom w/ cherry browns 
PowerCaseMouse
Antec CP-850 Antec P183 CM Storm Spawn 
  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 › [Eteknix] Pascal GTX 1080 Async Compute Explored