Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [KitGuru] AMD Radeon RX Vega caught hiding in plain sight
New Posts  All Forums:Forum Nav:

[KitGuru] AMD Radeon RX Vega caught hiding in plain sight - Page 18

post #171 of 664
Quote:
Originally Posted by RyzenChrist View Post

Quote:
Originally Posted by BiG StroOnZ View Post

Then all NVIDIA has to do is drop 1070 price to $330-350 for most models.

Imagine the 1070 for 299.99

Can I imagine that in €uro ? Thank you. redface.gif
 
Metro 2033 review
Metro 2033
CPUMotherboardGraphicsRAM
Core i7-3820 Asus Sabertooth X79 MSI GTX 750 Ti TF Gaming 16 GB Corsair DDR3 1866 Mhz Dominator 
Hard DriveOptical DriveCoolingOS
Samsung SSD 830 128GB + WD Caviar Black 1TB Sony Optiarc DVD-RW Corsair A70 + Noiseblocker M12-P Windows 7 Home Premium 64-bit 
MonitorKeyboardPowerCase
BenQ RL2455HM Cooler Master Octane Corsair AX750 Professional Modular 80 Plus Gold Cooler Master HAF 912 Plus 
Mouse
Cooler Master Octane 
  hide details  
Reply
 
Metro 2033 review
Metro 2033
CPUMotherboardGraphicsRAM
Core i7-3820 Asus Sabertooth X79 MSI GTX 750 Ti TF Gaming 16 GB Corsair DDR3 1866 Mhz Dominator 
Hard DriveOptical DriveCoolingOS
Samsung SSD 830 128GB + WD Caviar Black 1TB Sony Optiarc DVD-RW Corsair A70 + Noiseblocker M12-P Windows 7 Home Premium 64-bit 
MonitorKeyboardPowerCase
BenQ RL2455HM Cooler Master Octane Corsair AX750 Professional Modular 80 Plus Gold Cooler Master HAF 912 Plus 
Mouse
Cooler Master Octane 
  hide details  
Reply
post #172 of 664
I've noticed that Doom demo on ve.ga website was running on a 4770K with ddr3-1600, and Vega was an engineering sample.
Clocks and drivers were probably terrible back then. And they don't even care.
I'd say get ready for a nice surprise.
Edited by renx - 5/1/17 at 3:19pm
post #173 of 664
Quote:
Originally Posted by tpi2007 View Post

Can I imagine that in €uro ? Thank you. redface.gif
299 + VAT redface.gif

aka 20% atleast so that's 360 for you.
post #174 of 664
Not sure if this was mentioned but those questioning the authenticity of the 3DMark Time Spy results should note that the Doom Vega Demo has the same HW ID Number as on the Time Spy info:




Edited by BiG StroOnZ - 5/1/17 at 3:22pm
post #175 of 664
Quote:
Originally Posted by GoLDii3 View Post

Quote:
Originally Posted by tpi2007 View Post

Can I imagine that in €uro ? Thank you. redface.gif
299 + VAT redface.gif

aka 20% atleast so that's 360 for you.

No, no, that's not imagining. 299 with VAT included.
 
Metro 2033 review
Metro 2033
CPUMotherboardGraphicsRAM
Core i7-3820 Asus Sabertooth X79 MSI GTX 750 Ti TF Gaming 16 GB Corsair DDR3 1866 Mhz Dominator 
Hard DriveOptical DriveCoolingOS
Samsung SSD 830 128GB + WD Caviar Black 1TB Sony Optiarc DVD-RW Corsair A70 + Noiseblocker M12-P Windows 7 Home Premium 64-bit 
MonitorKeyboardPowerCase
BenQ RL2455HM Cooler Master Octane Corsair AX750 Professional Modular 80 Plus Gold Cooler Master HAF 912 Plus 
Mouse
Cooler Master Octane 
  hide details  
Reply
 
Metro 2033 review
Metro 2033
CPUMotherboardGraphicsRAM
Core i7-3820 Asus Sabertooth X79 MSI GTX 750 Ti TF Gaming 16 GB Corsair DDR3 1866 Mhz Dominator 
Hard DriveOptical DriveCoolingOS
Samsung SSD 830 128GB + WD Caviar Black 1TB Sony Optiarc DVD-RW Corsair A70 + Noiseblocker M12-P Windows 7 Home Premium 64-bit 
MonitorKeyboardPowerCase
BenQ RL2455HM Cooler Master Octane Corsair AX750 Professional Modular 80 Plus Gold Cooler Master HAF 912 Plus 
Mouse
Cooler Master Octane 
  hide details  
Reply
post #176 of 664
Quote:
Originally Posted by BiG StroOnZ View Post

Not sure if this was mentioned but those questioning the authenticity of the 3DMark Time Spy results should note that the Doom Vega Demo has the same HW ID Number as on the Time Spy info:




That would mean theres almost no IPC and or architectural improvements if an 1180mhz clocked Fury X scores around ~5700 in time spy. Taking into account bare minimum arch improvements from fiji to polaris, more memory bandwidth, and the new tile base rendering (which should work quite well due to the highly parallel nature of GCN), I see little reason to trust the Timespy result.

That bench results smells...
Steins Gate
(17 items)
 
  
CPUMotherboardGraphicsRAM
R7 1700 Gigabyte AX370-G5 EVGA GTX 1070 FTW2 2x8GB Trident Z @ 3200 C14 
Hard DriveHard DriveHard DriveCooling
850 Pro 128GB OCZ 460A 240GB WD Blue 1TB Thermalright True Spirit 140 Direct 
MonitorKeyboardPowerCase
BenQ XL2730Z Corsair K70 w/Browns EVGA 750w G2 Fractal Define S 
MouseMouse PadAudioAudio
G900 Corsair MM600 Onboard Logitech Z906 
Audio
HD518 
  hide details  
Reply
Steins Gate
(17 items)
 
  
CPUMotherboardGraphicsRAM
R7 1700 Gigabyte AX370-G5 EVGA GTX 1070 FTW2 2x8GB Trident Z @ 3200 C14 
Hard DriveHard DriveHard DriveCooling
850 Pro 128GB OCZ 460A 240GB WD Blue 1TB Thermalright True Spirit 140 Direct 
MonitorKeyboardPowerCase
BenQ XL2730Z Corsair K70 w/Browns EVGA 750w G2 Fractal Define S 
MouseMouse PadAudioAudio
G900 Corsair MM600 Onboard Logitech Z906 
Audio
HD518 
  hide details  
Reply
post #177 of 664
Quote:
Originally Posted by renx View Post

Actually MI25 is 12.5TF. And it is indeed passively cooled.
It's 25TF-FP16, and 12.5TF-FP32.

The cards are mounted with passive heatsinks but servers have high-power fans that typically have several times the airflow and work at higher static pressure compared to regular fans. I guess you can call it passive but an active cooling solution would probably disrupt or get disrupted by the server fans' airflow.

If it's really just 25TF for the RPM that's a little disappointing considering the TDP is "<300W". It suggests it could be close to 300W for 12.5TF of FP32.
Quote:
Originally Posted by BiG StroOnZ View Post

Not sure if this was mentioned but those questioning the authenticity of the 3DMark Time Spy results should note that the Doom Vega Demo has the same HW ID Number as on the Time Spy info:

Even with that in mind we know in Doom Vega at ~1200MHz will be similar, maybe exceed, the 1080's performance. At the 1500+MHz Vega needs to be running at to hit the 12.5TF of the MI25 it'll be similar to the 1080TI's performance assuming performance scales up linearly with the frequency.
Edited by geoxile - 5/1/17 at 4:17pm
New and Shiny
(18 items)
 
  
CPUMotherboardGraphicsRAM
Rynze 7 1700 ASUS ROG Crosshair VI Hero Sapphire HD7950 G.SKILL TridentZ F4-3200C14D 
Hard DriveHard DriveHard DriveHard Drive
Kingston HyperX 3K Crucial MX300 Western Digital Black Western Digital Green 
Hard DriveCoolingOSMonitor
Western Digital Red Noctua NH-U14S Windows 10 Pro Dell U2414H 
MonitorKeyboardPowerCase
Dell P2414H Ducky One Corsair RM650x NZXT H440 White 
MouseAudio
Logitech G502 Proteus Spectrum Xonar DX 
  hide details  
Reply
New and Shiny
(18 items)
 
  
CPUMotherboardGraphicsRAM
Rynze 7 1700 ASUS ROG Crosshair VI Hero Sapphire HD7950 G.SKILL TridentZ F4-3200C14D 
Hard DriveHard DriveHard DriveHard Drive
Kingston HyperX 3K Crucial MX300 Western Digital Black Western Digital Green 
Hard DriveCoolingOSMonitor
Western Digital Red Noctua NH-U14S Windows 10 Pro Dell U2414H 
MonitorKeyboardPowerCase
Dell P2414H Ducky One Corsair RM650x NZXT H440 White 
MouseAudio
Logitech G502 Proteus Spectrum Xonar DX 
  hide details  
Reply
post #178 of 664
I's May 1st, and this card is still white smoke.
Render Bender
(10 items)
 
On The Move
(11 items)
 
 
CPUGraphicsRAMHard Drive
Intel X5690 Xeon (12 Cores @ 3.46) Nvidia Titan Xp 64GB 1333MHz ECC DDR3 Samsung 850 EVO 1TB x2 - RAID 0 
MonitorCaseAudio
Samsung UN40JU6700 40" 4K Curved Apple Mac Pro 2010 Klipsch 2.1 
CPUGraphicsRAMHard Drive
Intel Core 2 Duo Nvidia GeForce 9400M 8GB 1067 MHz DDR3 Toshiba 500GB 
OSMonitorKeyboardCase
Mavericks 10.9 Dell U3011 Rosewill Mech Reds 2010 Unibody 
MouseMouse PadAudio
Logitech G700 Multi-Touch Kilpsch 2.1 
  hide details  
Reply
Render Bender
(10 items)
 
On The Move
(11 items)
 
 
CPUGraphicsRAMHard Drive
Intel X5690 Xeon (12 Cores @ 3.46) Nvidia Titan Xp 64GB 1333MHz ECC DDR3 Samsung 850 EVO 1TB x2 - RAID 0 
MonitorCaseAudio
Samsung UN40JU6700 40" 4K Curved Apple Mac Pro 2010 Klipsch 2.1 
CPUGraphicsRAMHard Drive
Intel Core 2 Duo Nvidia GeForce 9400M 8GB 1067 MHz DDR3 Toshiba 500GB 
OSMonitorKeyboardCase
Mavericks 10.9 Dell U3011 Rosewill Mech Reds 2010 Unibody 
MouseMouse PadAudio
Logitech G700 Multi-Touch Kilpsch 2.1 
  hide details  
Reply
post #179 of 664
Quote:
Originally Posted by geoxile View Post

The cards are mounted with passive heatsinks but servers have high-power fans that typically have several times the airflow and work at higher static pressure compared to regular fans. I guess you can call it passive but an active cooling solution would probably disrupt or get disrupted by the server fans' airflow.

If it's really just 25TF for the RPM that's a little disappointing considering the TDP is "<300W". It suggests it could be close to 300W for 12.5TF of FP32.
Even with that in mind we know in Doom Vega at ~1200MHz will be similar, maybe exceed, the 1080's performance. At the 1500+MHz Vega needs to be running at to hit the 12.5TF of the MI25 it'll be similar to the 1080TI's performance assuming performance scales up linearly with the frequency.


How is that disappointing? What are you comparing that too? It cant be Nvidia P100, because that is a 300w TDP card, with 10.6TFLOPS of FP32 performance, and 21.2 FP16.
http://images.nvidia.com/content/tesla/pdf/nvidia-tesla-p100-datasheet.pdf
post #180 of 664
Quote:
Originally Posted by rv8000 View Post

That would mean theres almost no IPC and or architectural improvements if an 1180mhz clocked Fury X scores around ~5700 in time spy. Taking into account bare minimum arch improvements from fiji to polaris, more memory bandwidth, and the new tile base rendering (which should work quite well due to the highly parallel nature of GCN), I see little reason to trust the Timespy result.

That bench results smells...

It could be a test run to get drivers straight for various programs, Time Spy being one of them.
Quote:
Originally Posted by geoxile View Post

The cards are mounted with passive heatsinks but servers have high-power fans that typically have several times the airflow and work at higher static pressure compared to regular fans. I guess you can call it passive but an active cooling solution would probably disrupt or get disrupted by the server fans' airflow.

If it's really just 25TF for the RPM that's a little disappointing considering the TDP is "<300W". It suggests it could be close to 300W for 12.5TF of FP32.
Even with that in mind we know in Doom Vega at ~1200MHz will be similar, maybe exceed, the 1080's performance. At the 1500+MHz Vega needs to be running at to hit the 12.5TF of the MI25 it'll be similar to the 1080TI's performance assuming performance scales up linearly with the frequency.

From here these most recent results, 1080 Ti seems like a stretch. However, I don't believe performance regressed unless the Doom Demo is the best case scenario.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Rumors and Unconfirmed Articles
Overclock.net › Forums › Industry News › Rumors and Unconfirmed Articles › [KitGuru] AMD Radeon RX Vega caught hiding in plain sight