Tuesday, June 20th 2017
![AMD](https://tpucdn.com/images/news/amd-v1723408558483.png)
AMD also Announces Radeon Instinct MI8 and MI6 Machine Learning Accelerators
AMD also announced the Radeon Instinct MI8 and MI6 Machine Learning GPUs based on Fiji and Polaris cores, respectively. These parts comprise the more "budget" part of the still most certainly non-consumer oriented high-end machine learning lineup. Still, with all parts using fairly modern cores, they aim to make an impact in their respective segments.
Starting with the Radeon Instinct MI8, we have a Fiji based core with the familiar 4 GBs of HBM1 memory and 512 GB/s total memory bandwidth. It has 8.2 TFLOPS of either Single Precision of Half Precision floating point performance (so performance there does not double when going half precision like its bigger Vega based brother, the MI25). It features 64 Compute Units.
The Radeon Instinct MI6 is a Polaris based card and slightly slower in performance than the MI8, despite having four times the amount of memory at 16 GBs of GDDR5. The likely reason for this is a slower bandwidth speed, at only 224 GB/s. It also has less compute units at 36 total, with a total of 2304 stream processors. This all equates out to a still respectable 5.7 TFLOPs of overall half or single precision floating point performance (which again, does not double at half precision rate like Vega).Interestingly, the Polaris solution is less energy efficient than the Fiji solution; The Fiji-based Radeon Instinct MI8 rates out at 47 GFLOPS / Watt, while the Polaris-based Radeon Instinct MI6 comes out to a lesser 38 GFLOPS per watt.
The two product slides relevant to these launches can be viewed below:
Starting with the Radeon Instinct MI8, we have a Fiji based core with the familiar 4 GBs of HBM1 memory and 512 GB/s total memory bandwidth. It has 8.2 TFLOPS of either Single Precision of Half Precision floating point performance (so performance there does not double when going half precision like its bigger Vega based brother, the MI25). It features 64 Compute Units.
The Radeon Instinct MI6 is a Polaris based card and slightly slower in performance than the MI8, despite having four times the amount of memory at 16 GBs of GDDR5. The likely reason for this is a slower bandwidth speed, at only 224 GB/s. It also has less compute units at 36 total, with a total of 2304 stream processors. This all equates out to a still respectable 5.7 TFLOPs of overall half or single precision floating point performance (which again, does not double at half precision rate like Vega).Interestingly, the Polaris solution is less energy efficient than the Fiji solution; The Fiji-based Radeon Instinct MI8 rates out at 47 GFLOPS / Watt, while the Polaris-based Radeon Instinct MI6 comes out to a lesser 38 GFLOPS per watt.
The two product slides relevant to these launches can be viewed below:
8 Comments on AMD also Announces Radeon Instinct MI8 and MI6 Machine Learning Accelerators
It says right on it - MI6 is GDDR5. How did you invent a Polaris with HBM2 memory?
Does anyone even review the stuff that goes up on this site?
Oh, wait, nevermind... I see what you're getting at...
Oh, wait... hang on, These numbers are scrambling my brain....
Yeah, finally, third paragraph, I thought I saw that....
It is our policy to correct all major errors though and this qualifies. It is being corrected now as we speak. Have a thanks for your report.
EDIT: Actually, a moderator beat me to the correction, peer review in action I suppose. I accept full responsibility for this writing error, and do apologize. That said, we should now move discussion to the actual device merits.
At the end of the day mining isn't necessarily a sustainable market, so allowing it to become the only sector you're selling GPUs into is a risky move. These cards can help protect AMD against a mining market crash by ensuring it doesn't flood the gaming GPU market.