Fenghua, that’s what Xindong / Innosilicon has called its fastest GPU to date, which translated into our language would be Fantasy 1, or Fantasy 1 as the English like to call it. This card aims really high in the market for not having TOP features and also sets quite interesting objectives if we take into account its options and characteristics: could it match an RTX 3060?
The company has been quite concise in terms of architecture and chips, so discerning whether or not it is faster than the competition is the least risky on our part given the lack of available data. We do know that it will have two different chips and that these mark two rather curious different performance options.
Fantasy 1 Type A and Fantasy 1 Type B
Imagination Technologies is undoubtedly the linchpin of these two GPUs since they both include the PowerVR architecture. Under this premise we better understand the two types of chips that Xindong / Innosilicon has arranged:
- Fantasy 1 Type A: corresponds to a mono die GPU that has a performance of 5 TFLOPS in FP32 and 25 INT8 TOPS for Artificial Intelligence and ML.
- Fantasy 1 Type B: it is a double die GPU in the purest AMD and NVIDIA style that achieves performance FP32 of 10 TFLOPS and 50 TOPS in INT8.
To give us an idea of where both GPUs are going in the gaming market, the Fantasy 1 Type A would give a similar performance to the current RTX 2060, while the Type B would rise to surpass an RTX 3060. We agree in that the TFLOPS in FP32 have been just one more metric for years that should not be taken into account as a reference, but there is the data and the performance should be seen.
In terms of VRAM configurations, Fantasy 1 Type A will be equipped with three different options: 4GB, 8GB, and 16GB, which could be GDDR6 or GDDR6X. Instead, Fantasy 1 Type B would go up to 32 GB obviously, where its speed and capacity will be necessary to be able to host the data from both GPUs under PCIe 4.0.
A ridiculous consumption for the acquired power
|Name||Fantasy 1 Type A||Fantasy 1 Type B|
|FP32||5 FP32 TFLOPS (1 GPU)||10 FP32 TFLOPS (2 GPU)|
|INT8||25 TOPS||50 TOPS|
|Pixel rate||160 GPixel / s||320 GPixel / s|
|Video||4x4Kp60, 16x1080p60, 32x720p30||8x4Kp60, 32x1080p60, 64x720p30|
The most curious thing about both GPUs is their consumption, since manufactured with a 12 nm lithographic process it is not well understood how they are capable of achieving 50W and 100W respectively for that FP32 performance.
That is to say, they are much more efficient cards than those of their rivals, much more should be added, to the point that they are impressive if these data are reflected after successive reviews. Likewise, neither the Fantasy 1 Type A nor the Fantasy 1 Type B are commercially available right now and it is possible that they will not do so until they are well debugged in terms of drivers, so it is speculating that they would reach data centers given their capabilities. virtualization and computing.
Of course and as expected there is not a single mention or rumor about the prices that would arrive, so we will have to wait quietly until someone can take the glove and confirm at least the performance and consumption to see if they are real options present and future against NVIDIA, AMD and Intel.