Corsair H170i Elite Capellix XT review
Forspoken: PC performance graphics benchmarks
ASRock Z790 Taichi review
The Callisto Protocol: PC graphics benchmarks
G.Skill TridentZ 5 RGB 6800 MHz CL34 DDR5 review
Be Quiet! Dark Power 13 - 1000W PSU Review
Palit GeForce RTX 4080 GamingPRO OC review
Core i9 13900K DDR5 7200 MHz (+memory scaling) review
Seasonic Prime Titanium TX-1300 (1300W PSU) review
F1 2022: PC graphics performance benchmark review
AMD Radeon Pro Duo Launches April 26th
Let me first state that the release date is a rumor that popped up over the weekend, not a fact. According to sources at HardwareBattle, AMD is to launch the Dual FIJI GPUs based Radeon Pro Duo in roughly three weeks time. AMD recently announced the product on its Capsaicin-event at GDC 2016.
The Radeon Duo Pro with its two FIJI GPUs are tied towards a thick 120mm liquid cooling solution. The product will offer 16 teraflops of performance, indicating two fully enabled Fiji XT GPUs. AMD Radeon Pro Duo (dual-GPU Fiji) was developed under code-name Gemini would have 8 GB HBM1 graphics memory with a whopping 8192 stream processors.
The specs below are not final yet based up-on assumptions (they should be pretty conclusive though). The AMD Radeon Pro Duo is expected to sell at a SEP of $1499 USD.
| Radeon Pro Duo | Radeon R9 Fury X | Radeon R9 Nano | Radeon R9 390X | |
| Fabrication Process | 28nm | 28nm | 28nm | 28nm |
| GPU | Fiji XT x2 | Fiji XT | Fiji XT | Hawaii / Grenada |
| Streaming Processors | 2x 4096 | 4096 | 4096 | 2816 |
| Graphics memory | 2x 4 GB HBM | 4 GB HBM | 4 GB HBM | 8 GB GDDR5 |
| Memory Clock | up-to 500 MHz / 1.0 Gbps | up-to 500 MHz / 1.0 Gbps | up-to 500 MHz / 1.0 Gbps | 6.0 Gbps |
| Core Clock | 1050 MHz | 1050 MHz | up-to 1000 MHz | 1050 MHz |
| Memory Bandwidth | up-to 512 GB/s | up-to 512 GB/s | up-to 512 GB/s | 384 GB/s |
| Power Connectors | 3 x 8-pin | 2 x 8-pin | 1 x 8-pin | 1 x 6-pin - 1 x 8-pin |
| Form Factor | Full Height, Dual slot | Full Height, Dual slot | Full Height, Dual slot | Full Height, Dual slot |
| Freesync | Yes | Yes | Yes | Yes |
| DirectX 12 Support | Yes | Yes | Yes | Yes |
One Radeon Fury X pushes close to 8.6 TFLOPS of performance, the dual-GPU version offers over 16 TFLOPs of single precision performance. And yes a dual-GPU Fiji would have 8 GB HBM1 graphics memory with 8192 stream processors. The product is rated to have a 250 Watt TDP and is powered by three 8-pin PEG connectors.
Rumor: AMD Radeon 490 and 490X Polaris graphics cards launch end of June - 04/07/2016 02:33 PM
In line with expectations today some new rumors surfaced on the mighty WWW. AMD possibly will have its first Polaris based series 400 Radeon graphics cards available in June, this would invoke the R9 ...
Download AMD Radeon Software Crimson 16.4.1 drivers - 04/05/2016 01:24 PM
You can now download the AMD Radeon Software Crimson 16.4.1 driver. This release includes support for Quantum Break and adds Oculus Rift and HTC Vive support. Download...
Download AMD Radeon Software Crimson 16.3.2 driver - 03/29/2016 07:43 AM
You can now download AMD Radeon Software Crimson 16.3.2 driver. This release adds Oculus Rift SDK v1.3 as well as AMD Radeon Pro Duo support. Also there are a number of fixes and improvements. Read m...
Download AMD Radeon Software Crimson Edition 16.3.1 driver - 03/18/2016 09:59 AM
AMD released a new driver, with some interesting fixes. Hitman will be supported in Crossfire (we'll check and update our article today). Added is supper for Need for Speed Crossfire as well. And the...
AMD Radeon Pro Duo (Dual Fiji GPUs) Announced + Initial Perf Numbers - 03/15/2016 07:30 PM
Read it well, an announcement, don't expect any reviews. We do have some photo's though. The DUal Fiji GPU based AMD Radeon Pro Duo has been announced. Imagine two Fury X cards and well, there you ...
Username001
Junior Member
Posts: 6
Joined: 2016-04-10
Junior Member
Posts: 6
Joined: 2016-04-10
#5256302 Posted on: 04/11/2016 02:02 AM
It's in either of these cases that I expect real time procedural effects to take over since we're getting to the point of having a crazy amount of single precision floating point power out of shaders overall, and that also saves quite a lot of video card memory since not as many textures are need to be stored there anymore......
One example that comes to mind is the damage modeling system used in star citizen, where under the previous system and this was just for a regular fighter sized craft, about 100 MB of textures needed to be stored in local video card memory, while with the new procedural system that dropped to just 8MB for the same ship.
The savings are definitely there and the damage effect looked a lot better as an added bonus, but since for the most part we're playing console ports with a few extra graphics features for the PC release as long as they don't take too long to add them ( costs extra money after all ), most games will keep it simple and stick with what works.
This is true but in most scenes I imagine both GPU's will have the exact same thing in memory anyway, even when they are split. It's still better obviously, but lots of people think it will just double the free memory. Memory in use will nearly double as well.
^ PCIe 3.0 x16 bandwidth is not sufficient for one GPU to use other GPU's vRAM as data storage for 'per frame' required textures anyway.
At time we'll have sufficient bandwidth (like PCIe 5.0 x16) we'll in contrast have stupid high amount of vram like 16GB and it will be again used for overkill textures.
(and it will in return kill benefit of higher PCIe bandwidth.)
At time we'll have sufficient bandwidth (like PCIe 5.0 x16) we'll in contrast have stupid high amount of vram like 16GB and it will be again used for overkill textures.
(and it will in return kill benefit of higher PCIe bandwidth.)
It's in either of these cases that I expect real time procedural effects to take over since we're getting to the point of having a crazy amount of single precision floating point power out of shaders overall, and that also saves quite a lot of video card memory since not as many textures are need to be stored there anymore......
One example that comes to mind is the damage modeling system used in star citizen, where under the previous system and this was just for a regular fighter sized craft, about 100 MB of textures needed to be stored in local video card memory, while with the new procedural system that dropped to just 8MB for the same ship.
The savings are definitely there and the damage effect looked a lot better as an added bonus, but since for the most part we're playing console ports with a few extra graphics features for the PC release as long as they don't take too long to add them ( costs extra money after all ), most games will keep it simple and stick with what works.
Username001
Junior Member
Posts: 6
Joined: 2016-04-10
Junior Member
Posts: 6
Joined: 2016-04-10
#5256309 Posted on: 04/11/2016 02:24 AM
I mean 16 teraflop of single precision floating point math power in this latest Radeon Pro Duo........That's 16 trillion operations per second, theoretical maximum.....Think about it.
So if we take a 4K screen which totals almost 8.3 million pixels and we want 60 Fps, that comes to 498 million pixels and then an incredibly complex shader effect that is totally dependent of that 16 teraflop floating point math power is applied to all those pixels, and no other limitations are in play (which there are), then a shader ( or multiple shaders ) that requires 32 000 floating point calculations can be applied to each pixel and still sustain 60 Fps in the process.
In short, we don't know how good we have it really......
I mean 16 teraflop of single precision floating point math power in this latest Radeon Pro Duo........That's 16 trillion operations per second, theoretical maximum.....Think about it.
So if we take a 4K screen which totals almost 8.3 million pixels and we want 60 Fps, that comes to 498 million pixels and then an incredibly complex shader effect that is totally dependent of that 16 teraflop floating point math power is applied to all those pixels, and no other limitations are in play (which there are), then a shader ( or multiple shaders ) that requires 32 000 floating point calculations can be applied to each pixel and still sustain 60 Fps in the process.
In short, we don't know how good we have it really......
Robbo9999
Senior Member
Posts: 1691
Joined: 2012-10-07
Senior Member
Posts: 1691
Joined: 2012-10-07
#5256345 Posted on: 04/11/2016 07:13 AM
I don't get it, why would AMD bother with this & why would anyone buy it, according to the graph it's only about 25% faster on average than the the 295x2 - lump that in together with the fact that it's not long till Polaris launches...well, I see it as pointless!
I don't get it, why would AMD bother with this & why would anyone buy it, according to the graph it's only about 25% faster on average than the the 295x2 - lump that in together with the fact that it's not long till Polaris launches...well, I see it as pointless!
kobiandy
Junior Member
Posts: 4
Joined: 2013-09-02
Junior Member
Posts: 4
Joined: 2013-09-02
#5256400 Posted on: 04/11/2016 09:50 AM
Nvidia had no answer for 295x2 now this....overkill
Nvidia had no answer for 295x2 now this....overkill
Click here to post a comment for this news story on the message forum.
Senior Member
Posts: 11808
Joined: 2012-07-20
^ PCIe 3.0 x16 bandwidth is not sufficient for one GPU to use other GPU's vRAM as data storage for 'per frame' required textures anyway.
At time we'll have sufficient bandwidth (like PCIe 5.0 x16) we'll in contrast have stupid high amount of vram like 16GB and it will be again used for overkill textures.
(and it will in return kill benefit of higher PCIe bandwidth.)