• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Pixel Fillrate and Turing/Pascal

eraser666

New Member
Joined
Feb 7, 2019
Messages
16 (0.01/day)
Limited by GPC? the max Pixel Fillrate of TU106 is ROPs*frequency/4*3, the same as GP106(max Pixel Fillrate=ROPs*frequency/3*2).

You can use TU116(1660Ti,3GPC 48ROP) or TU106-200A(2060,3GPC 48ROP) comparing with TU104-400(2070,3GPC 64ROP) to verify this fact.

Remember! Use a same GPU and MEM frequency to verify this fact in Feature Test of 3DMark06 or 3DMark Vantage.

LOL!


QQ图片20190220143540.png
 
Last edited:
Joined
May 8, 2016
Messages
1,735 (0.60/day)
System Name BOX
Processor Core i7 6950X @ 4,26GHz (1,28V)
Motherboard X99 SOC Champion (BIOS F23c + bifurcation mod)
Cooling Thermalright Venomous-X + 2x Delta 38mm PWM (Push-Pull)
Memory Patriot Viper Steel 4000MHz CL16 4x8GB (@3240MHz CL12.12.12.24 CR2T @ 1,48V)
Video Card(s) Titan V (~1650MHz @ 0.77V, HBM2 1GHz, Forced P2 state [OFF])
Storage WD SN850X 2TB + Samsung EVO 2TB (SATA) + Seagate Exos X20 20TB (4Kn mode)
Display(s) LG 27GP950-B
Case Fractal Design Meshify 2 XL
Audio Device(s) Motu M4 (audio interface) + ATH-A900Z + Behringer C-1
Power Supply Seasonic X-760 (760W)
Mouse Logitech RX-250
Keyboard HP KB-9970
Software Windows 10 Pro x64

Pixel Fillrate = GPC count x 16 pixels x Core clock (close to max. Boost clock to be accurate).

Turing (overall) : If GPC x 16 > ROP, Pixel Fillrate = ROP x Core clock, else use GPC x 16.

Did I get it right ?
 

eraser666

New Member
Joined
Feb 7, 2019
Messages
16 (0.01/day)

Pixel Fillrate = GPC count x 16 pixels x Core clock (close to max. Boost clock to be accurate).

Turing (overall) : If GPC x 16 > ROP, Pixel Fillrate = ROP x Core clock, else use GPC x 16.

Did I get it right ?

Excellent!
 
Joined
May 8, 2016
Messages
1,735 (0.60/day)
System Name BOX
Processor Core i7 6950X @ 4,26GHz (1,28V)
Motherboard X99 SOC Champion (BIOS F23c + bifurcation mod)
Cooling Thermalright Venomous-X + 2x Delta 38mm PWM (Push-Pull)
Memory Patriot Viper Steel 4000MHz CL16 4x8GB (@3240MHz CL12.12.12.24 CR2T @ 1,48V)
Video Card(s) Titan V (~1650MHz @ 0.77V, HBM2 1GHz, Forced P2 state [OFF])
Storage WD SN850X 2TB + Samsung EVO 2TB (SATA) + Seagate Exos X20 20TB (4Kn mode)
Display(s) LG 27GP950-B
Case Fractal Design Meshify 2 XL
Audio Device(s) Motu M4 (audio interface) + ATH-A900Z + Behringer C-1
Power Supply Seasonic X-760 (760W)
Mouse Logitech RX-250
Keyboard HP KB-9970
Software Windows 10 Pro x64
You could say, those ROPs are there for "cheaper" Anti Aliasing ;)
Last time NV did this with Fermi (IIRC).
 
Joined
Jul 18, 2016
Messages
506 (0.18/day)
System Name Gaming PC / I7 XEON
Processor I7 4790K @stock / XEON W3680 @ stock
Motherboard Asus Z97 MAXIMUS VII FORMULA / GIGABYTE X58 UD7
Cooling X61 Kraken / X61 Kraken
Memory 32gb Vengeance 2133 Mhz / 24b Corsair XMS3 1600 Mhz
Video Card(s) Gainward GLH 1080 / MSI Gaming X Radeon RX480 8 GB
Storage Samsung EVO 850 500gb ,3 tb seagate, 2 samsung 1tb in raid 0 / Kingdian 240 gb, megaraid SAS 9341-8
Display(s) 2 BENQ 27" GL2706PQ / Dell UP2716D LCD Monitor 27 "
Case Corsair Graphite Series 780T / Corsair Obsidian 750 D
Audio Device(s) ON BOARD / ON BOARD
Power Supply Sapphire Pure 950w / Corsair RMI 750w
Mouse Steelseries Sesnsei / Steelseries Sensei raw
Keyboard Razer BlackWidow Chroma / Razer BlackWidow Chroma
Software Windows 1064bit PRO / Windows 1064bit PRO
GPU-Z show more Pixel Fillrate for those gpu?
 
Joined
Jun 3, 2010
Messages
2,540 (0.50/day)
I'm just talking about the pixel filling rate.

the display of gpu-z
It displayed wrong with GTX 970, too. TPU is used to such occurrances.

I mean it depends, anyway. I like the idea though, just as <99p testing we should have rasterizer performance information of individual scene setups like how much of an effect small & big triangles have.

How about a free licence of Beyond3D Suite as well, one cannot run short of such features. Though I doubt TPU has it, not having any associated publishings even after all these years.
 

eraser666

New Member
Joined
Feb 7, 2019
Messages
16 (0.01/day)
It displayed wrong with GTX 970, too. TPU is used to such occurrances.

I mean it depends, anyway. I like the idea though, just as <99p testing we should have rasterizer performance information of individual scene setups like how much of an effect small & big triangles have.

How about a free licence of Beyond3D Suite as well, one cannot run short of such features. Though I doubt TPU has it, not having any associated publishings even after all these years.

Thank you for your reply. According to feedback from B3D subscribers, there will be a public release version of Beyond 3D Suite in recent months, probably similar to that used by Anandtech.

GPU-Z show more Pixel Fillrate for those gpu?

I'm afraid so, including 1060, 1070, 2070, Quadro GV100 (Titan V JHH Special Edition), and maybe even 780.

https://www.anandtech.com/bench/GPU18/2347

:lovetpu:
 

eraser666

New Member
Joined
Feb 7, 2019
Messages
16 (0.01/day)
How about this formula?
pixel fillrate=min(ROP, GPC*16, TPC*4)*frequency/1000

For example:
they have the same pixel fillrate in test.
2070:3GPC/18TPC/64ROP 2060:3GPC/15TPC/48ROP 1660Ti:3GPC/12TPC/48ROP
1080:4GPC/20TPC/64ROP 1070Ti:4GPC/19TPC/64ROP

they have different pixel fillrate in test.
1080:4GPC/20TPC/64ROP 1070:3GPC/15TPC/64ROP
1060:2GPC/10TPC/48ROP 1050Ti:2GPC/6TPC/32ROP 1050:2GPC/5TPC/32ROP
 
Last edited:
Top